papers

Zootaxa 3701 (5): 518–550 ISSN 1175-5326 (print edition) www.mapress.com /zootaxa / Copyright © 2013 Magnolia Press Article ZOOTAXA ISSN 1175-5334 (online edition) http://dx.doi.org/10.11646/zootaxa.3701.5.2 http://zoobank.org/urn:lsid:zoobank.org:pub:44F12E00-E270-41B0-ADE9-4FF774C975C3 A molecular phylogeny of African Dainty Frogs, with the description of four new species (Anura: Pyxicephalidae: Cacosternum) ALAN CHANNING1,4 , ANDREAS SCHMITZ2, MARIUS BURGER1 & JOS KIELGAST3 1 Biodiversity and Conservation Biology Department, University of the Western Cape, Private Bag X17, Bellville, 7535, South Africa 2 Department of Herpetology & Ichthyology, Natural History Museum of Geneva, 1 route de Malagnou, CH - 1208 Geneva, Switzerland 3 Zoological Museum, Universitetsparken 15, DK-2100 København, Denmark 4 Corresponding author. E-mail: achanning@uwc.ac.za Abstract We examined specimens from all eleven described species of African Dainty Frogs, Cacosternum. Advertisement calls, 16S and tyr sequences were obtained from voucher specimens of all known species plus undescribed taxa. A phylogenetic analysis indicated that there were 15 species. We describe four new species from South Africa that can be diagnosed by their advertisement calls: Cacosternum aggestum sp. nov. from the interior of the south-western Cape, the large C. nano- gularum sp. nov. from KwaZulu-Natal, C. australis sp. nov. from the Western Cape Province and C. rhythmum sp. nov. from the KwaZulu-Natal midlands. Cacosternum schebeni is confirmed as a junior synonym of C. boettgeri, and we agree that C. poyntoni is a junior synonym of C. nanum. The populations of dainty frogs on the Ethiopian highlands remain to be investigated. Shared tyr haplotypes occur between species that are not necessarily closely related, but always sympatric, at least in the recent past. This is evidence for hybridisation that requires further investigation. A provisional identification key to the species is provided. Key words: Anura, Pyxicephalidae, Cacosternum, new species, advertisement call, phylogeny Introduction Dainty frogs in the genus Cacosternum are found from Ethiopia to South Africa and are presently assigned to 11 species (Frost 2013). Although there are exceptions, the genus consists mostly of cryptic, highly polychromatic species, with green and brown dorsal colouring being a common feature. Most species have small ranges, although "Cacosternum boettgeri" which was until recently considered to have an Africa-wide distribution, is now known to consist of many cryptic forms (Poynton et al. 2004, Channing et al. 2005, Channing & Schmitz 2009). This study derives from fieldwork associated with the Southern African Frog Atlas Project, additional South African projects and during visits to the Democratic Republic of Congo, Kenya and Tanzania. Here we consider the molecular, morphological and vocal differences between populations, and use these data to delimit the known species, and recognise four new species. Detailed distributions of the known southern African species are available in Minter et al. (2004). Material and methods Sampling. Tissues were obtained from museums, or from field-collected specimens. An attempt was made to sample widely across the distribution range of the genus, with two or more samples from each locality where possible. Frogs were located by their vocalisations, and by searching along small streams and around ponds. Voucher material, GenBank numbers and locality details are presented in Appendix 1. 518 Accepted by M. Vences: 18 Jul. 2013; published: 22 Aug. 2013  Voucher specimens referred to in this paper are housed in the Natural History Museum of Geneva, Switzerland (MHNG); National Museum, Bloemfontein, South Africa (NMB); and National Museums of Kenya, Nairobi (NMK). DNA sequences. Specimens were euthanized in MS222, and toe clips or a piece of thigh muscle was preserved in absolute ethanol. DNA was extracted using Qi-Amp tissue extraction kits (Qiagen) and the peqGold Tissue DNA Mini Kit (PEQLAB Biotechnologie GmbH) or High Pure PCR Template Preparation kits (Roche) following the manufacturers' protocols. Alternatively tissues were digested using standard Proteinase-K protocol, and DNA was extracted using phenol-chloroform (Hillis et al. 1996). A 550 bp fragment of the mt 16S gene was amplified using the primers 16sar-L and 16sbr-H of Palumbi et al. (1991). Amplification of 25 µl or 50 µl PCR reactions follow the protocol: initial denaturation at 94°C for 90 s, 30 cycles of denaturation at 94°C for 45 s, annealing at 54°C for 30 s, extension at 72°C for 60 s, followed by a final extension at 72°C for 420 s. Furthermore, we used the primers 16SaR-F and 16SbR-R of Simon et al. (1994), as modified by Bossuyt & Milinkovitch (2000) annealing at 51°C. The nuclear tyrosinase exon 1 Tyr was amplified using the primers TyrC-F and TyrG-R (Bossuyt & Milinkovitch 2000), annealing at 56°C. Primer sequences are shown in Table 1. PCR products were purified using Qiaquick purification kits (Qiagen). TABLE 1. Primer sequences used in this study. Name and source Sequence (5' to 3') 16SaR-F (Bossuyt & Milinkovitch 2000) CGCCTGTTTAYCAAAAACAT 16SbR-R (Bossuyt & Milinkovitch 2000) CCGGTYTGAACTCAGATCAYGT 16sar-L (Palumbi et al. 1991) CGCCTGTTTATCAAAAACAT 16sbr-H (Palumbi et al. 1991) CCGGTCTGAACTCAGATCACGT TyrC-F (Bossuyt & Milinkovitch 2000) GGCAGAGGAWCRTGCCAAGATGT TyrG-R (Bossuyt & Milinkovitch 2000) TGCTGGCRTCTCTCCARTCCCA Sequencing reactions and electrophoresis were carried out by the University of Stellenbosch Central Analytical Facility or in the labs of the MHNG, Geneva. Forward and reverse strands were sequenced for all samples. Both sequences were checked against the chromatograms, trimmed, and combined into a single contig for each fragment using Sequencher 5.1 (GeneCodes Corporation) or BioEdit (Hall 1999). Sequences were checked using BLAST to confirm their placement in the ingroup (http://blast.ncbi.nlm.nih.gov/). All new sequences were deposited in GenBank (Benson et al. 2012). Additional 16S sequences were obtained from GenBank. Sequences were aligned using Clustal W2 (2.0.12) with default settings. JModelTest 0.1.1 (Posada 2008, Guindon & Gascuel 2003) was used to determine the appropriate model of evolution under AIC. The aligned 16S sequences (129 Cacosternum plus five outgroup Microbatrachella, were input into MrBayes 3.2.1 (Ronquist & Huelsenbeck 2003), and run for 5 million generations, with three attempted swaps each iteration, with the temperature set at 0.1, and using the GTR + G + I model. Two independent runs were analysed, each with one hot and three cold chains. The nuclear tyr gene was phased into the most likely two haplotypes for each individual by first submitting the edited sequences including IUPAC polymorphism symbols, to SeqPhase step 1 (Flot 2010) an online service that prepares a simplified output file. The output from SeqPhase is then used as input to PHASE (Stephens & Donnelly 2003), which computes the liklihood of possible haplotypes. The output from PHASE is converted to full sequences through SeqPhase step 2. A haplotype network was constructed for the tyr sequences, using TCS 1.21 (Clement et al. 2000) that implements statistical parsimony to estimate gene genealogies (Templeton et al. 1992). Uncorrected p distances for the 16S gene fragment were determined using PAUP* (Swofford 2002). Maximum likelihood (ML) models were analysed using Garli 2.0 (Zwickl 2006). Bootstrap support was determined using 1000 bootstrap repetitions each with three search repetitions, summarised using DendroPy (Sukumaran & Holder 2010). Advertisement call. Choruses of Cacosternum were investigated in the field, and the following details were recorded at most of the sites: locality name and coordinates, habitat, date, time of recording, air, substrate and water temperatures (°C), species and chorus size. The calls were recorded using a Sony TCM-5000EV stereo cassette recorder or a Marantz PMD 660 digital recorder, both with an Audio Technics directional microphone. Calls were analysed using Raven Pro 1.4 (Bioacoustics Research Program 2011). Since the calls of Cacosternum consist of various arrangements of clicks, a AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 519 Fast Fourier Transform (FFT) size of 512 allowed good temporal analysis. The filter bandwidth was 1398.8 Hz, with a frame length of 128 points. The standard measurements taken were: call duration in milliseconds (ms); number of pulses in a call; pulse rate; and emphasised frequency in kilohertz (kHz). Morphology. All measurements were made with a digital calliper, recorded to an accuracy of 0.1 millimetres. The following measurements were taken: SUL snout-urostyle length from the tip of the snout to the end of the urostyle; TIB the maximum tibia length measured on the bent hind limb; FOT the length of the foot measured from the proximal end of the inner metatarsal tubercle to the tip of the fourth toe on the fully extended foot; EN the eye- nostril distance measured from the anterior corner of the eye to the posterior border of the nostril; SL snout length measured from the tip of the snout to the anterior corner of the eye; EE the distance between the anterior corners of the eyes; NN the inter-nostril distance; ED the horizontal eye diameter; HW head width at the angle of the jaw; TYM horizontal diameter of the tympanum; ET The shortest distance between the eye and the tympanum; RAD forearm length as the maximum distance between the wrist and the elbow; and HAN hand length measured from the tip of the third finger to include the metacarpal tubercle. The following proportions were derived from the measurements for comparative purposes: HW/SUL, EE/SUL, NN/EN, EN/SL, NN/EE, ED, HW, HAN/SUL, HAN/HW, TIB/SUL, FOT/SUL and TIB/FOT. Results The Bayesian analysis showed that the Potential Scale Reduction Factor averaged 1.000, with the maximum PSRF for alpha=1.003. The summary statistics for tree partitions show the average PSRF for parameter values=1.001, with the maximum PSRF for parameter values=1.010. The phylogeny is shown in Fig. 1. Clade credibility and maximum likelihood bootstrap values are reported on the tree. DNA sequences. All sequences generated in this study have been deposited in GenBank (KF144411–144583) (Appendix 1). Uncorrected p distances for 16S are shown in Table 2. The known species differ by 2.3–7.6 p distances for the 16S gene, and (with few exceptions) did not share tyr haplotypes. Four additional clades were discovered that showed p distances for the 16S gene of 1.1–7.3, and also did not share tyr haplotypes (see exceptions below). These four are described as new species. Altogether 60 specimens were sequenced for tyr. The sequences consisted of 84 haplotypes, resulting in an overall H = 1.4. One haplotype was shared between a single specimen of C. boettgeri and one C. striatum, a second haplotype was shared between a single C. nanum and one C. rhythmum sp. nov. and a third haplotype was shared between one C. platys and one C. australis sp. nov. There were 63 unique haplotypes. The most parsimonious haplotype network is shown in Fig. 2. Advertisement call. The calls are described below. Table 3 is a summary of the advertisement call parameters. The calls consist of brief pulses or clicks, which may be run together into trills of various pulse rates and duration. The four new species have remarkably different calls from all other species. Morphology. This genus of frogs consists of very similar species in external morphology, with few diagnostic features. Morphological proportions are given in Appendix 2. Species variation. Diagnostic parameters of the advertisement calls include combinations of duration, pulse rate, number of pulses, and the presence of both rapidly pulsed chirps and slow pulsed creaks. There are a number of call types, but the details of each species call allow the advertisement call to be used for identification. The genetic variation between species is 20–36 times the variation within species for most pairwise comparisons. The minimum difference is 2.2 times, with the smallest differences found in the C. australis/C. platys, C. boettgeri/C. rhythmum and C. platys/C. rhythmum pairs. This is discussed below. Morphologically, many species are very similar. The smallest species (maximum SUL in our sample) are C. platys (15.3), C. parvum (16.5), C. leleupi (16.7) and C. nanum (18.7), while the largest are C. capense (31.3) and C. nanogularum sp. nov. (25). Cacosternum capense possesses large lateral and postero-dorsal skin glands which are diagnostic. Most species are very polymorphic for dorsal colour and pattern, although many species show characteristic (if not unique) pattern features. The colouration of the throat in breeding males may be diagnostic, once larger sample sizes are available. Unpigmented throats are present in C. aggestum sp. nov., C. plimptoni, C. rhythmum sp nov. and C. striatum. Other species have white pigment, or dark pigment anterior (C. capense), posterior (C. leleupi). Some throats have spotted or reticulated patterns (C. karooicum, C. namaquense, C. nanogularum sp. nov. and C. nanum). In combination, these characters provide a clear means of identification of all the species. 520 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. FIGURE 1. Phylogeny of Cacosternum species, based on 16S. Branch support is indicated as Posterior Probability/Maximum Likelihood Bootstrap. Terminal triangles represent collapsed clades, to simplify the tree. Numerals in parentheses represent sample size/number of localities. FIGURE 2. Network of most likely haplotypes of the tyr gene. Black circles represent hypothetical intermediate haplotypes. Circle size is proportional to number of individuals. AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 521 522 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. TABLE 3. Summary of the advertisement call parameters of Cacosternum species. Species and source Individuals/ Duration (ms) Pulses Mean Pulse rate Notes calls/ temp °C (s-1) C. aggestum 3/7/18 50 (34–63) 12 (8–17) 45 (37–65) C. australis 4/12/18 38 (24–87) 9 (7–12) 24 (10–29) Pulse rate accelerates. Some inital pulses paired. C. boettgeri 6/16/24 49 (22–79) 5 (3–7) 9 (7–14) (Channing et al. 2005) C. capense 3/13/18 22 (17–32) 15 (12–23) 67 (59–72) C. karooicum 2/6/22 60 (52–67) 25 (22–29) 40 (38–42) C. kinangopensis 4/19/13 1000 (800–1300 66 (51–77) (Channing & Schmitz 2009) C. leleupi 3/11/24 10 (8–14) 7 (5–8) 61 (40–83 C. namaquense 3/15/22 58 (26–88) 39 (19–57) 66 (48–98) C. nanogularum 10/100/ 89 (46–131) 9 (5–11) 89 (77–102) C. nanum 41/447/ 52 (19–100) 11 (5–18) 200 (95–288) C. parvum 17/172/ 60 (28–109) 6 (3–9) 87 (60–125) C. platys 6/17/10 28 (12–45) 8 (5–10) 26 (16–35) (Channing et al. 2005). Pulse rate accelerates C. plimptoni 11/29/29 55 (37–77) 13 (8–27) 22 (14–34) (Channing et al. 2005, Channing & Schmitz 2009) C. rhythmum 10/20/21 1150 (644–3068) 20 (10–77) 15 (12–25) Without terminal chirps. These have a mean duration of 113 ms C. striatum 7/43/11 41 (28–45) 21 (14–27) 50 (37–70) Taxonomy We recognise 15 species, based on genetics and call data. We review existing species, and describe four new species below. Cacosternum aggestum sp. nov. Klipheuwel Dainty Frog. (Figs. 3A, 4A) Holotype. A male, MHNG 2690.25, collected at Klipheuwel on a flooded field, 13 June 2007, 33° 41' 48.7 " S, 18° 43' 27.6" E, by A. Channing (Fig. 3). Paratypes. One male MHNG 2690.26 and a female MHNG 2690.27, with the same collecting details as the holotype; one male MHNG 2669.40 collected at Vissershok, 24 August 2007, 33 46' 11.3" S, 18 31' 33.6" E, by C. Dorse. Diagnosis. The within-clade uncorrected p distance for this species is 0.0–0.2% for 16S, while the difference between the other 14 species ranges from 3.9–7.1%. The sample has only two likely tyr haplotypes, of which none are shared. Breeding males are less than 20 mm SUL, while C. boettgeri, C. capense, C. karooicum, C. nanogularum sp. nov., range from 21–30 mm. The throat is beige in preservative (Fig. 4), while in C. capense it is black anteriorly, in C. karooicum it is beige anteriorly and white posteriorly, C. leleupi is black anteriorly, C. namaquense it is grey anteriorly, in C. boettgeri it is pale grey, in C. nanum and C. parvum it is dark, and in C. rhythmum sp. nov. the throat is white. This species has no prominent lateral and dorsoventral skin glands as in C. capense. Viewed from above, the nostrils are within one nostril diameter of the anterior snout margin, compared to C. karooicum which has the nostrils at least three nostril diameters from the anterior snout margin. The supratympanic fold continues posteriorly as a distinct thick dark saddle, while the saddle is indistinct or absent in AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 523 C. capense, C. leleupi, C. nanogularum sp. nov., and C. rhythmum sp. nov. Cacosternum aggestum sp. nov. has five small palmar tubercles, as does C. leleupi; all the other species have 0–4. The toes have distinct margins, although this does not form a web between toes 4 and 5, unlike C. leleupi which possesses a small web reaching the proximal subarticular tubercles. The outer metatarsal tubercle is very small, low and pale, but absent in C. capense, C. namaquense, C. australis sp. nov., C. striatum, C. nanum and C. rhythmum sp. nov. The inner metatarsal tubercle is conical, protrudes, and is twice the width of the tip of the first toe. It is narrower, equal to the width of the first toe tip in C. capense, C. karooicum, C. namaquense, C. striatum, C. nanum, and C. platys; it is about equal in size to the proximal subarticular tubercle of the first toe in C. nanogularum sp. nov., C. australis sp. nov. and C. rhythmum sp. nov. The belly pattern consists of large separated pale beige blotches and a few small black marks. The same double colour marking is seen in C. australis sp. nov. All the other species have various arrangements of blotches, spots and lines in one colour. It differs from C. striatum, which has a silvery white abdomen with dark blotches. The advertisement call resembles the sound made by a marble bouncing on a hard floor; a series of clicks that speed up. It differs from the regular single or multiple clicks of C. boettgeri, C. kinangopensis, C. leleupi, C. plimptoni or the creaking calls of C. capense, C. karooicum, C. namaquense, C. nanogularum sp. nov., C. parvum, C. striatum and the brief chirp of C. nanum. The call of C. australis sp. nov. speeds up at the end, but has double pulses. C. platys also has a call resembling a marble bouncing, but the pulse rate is much faster, with more pulses. Finally, it is quite different to the long complex call of C. rhythmum sp. nov. Description of the holotype. A male in breeding condition, SUL 19.2. Body widest at mid-belly, with a narrow head (HW/SUL 0.35). The head is bluntly rounded from above and truncated in profile. Head length measured from the angle of the jaw is about one third of body length (HL/SUL 0.34). Canthus rostralis rounded, straight from eye to nostril, loreal region sloped outwards ventrally; nostrils small, rounded, rimmed, directed laterally and posteriorly. The nostrils are placed slightly closer to the snout than the eye (EN/SL 0.59). Internostril distance is less than distance between eye and nostril (NN/EN 0.94). Eyes directed anterolaterally, the eyes protrude, and are visible from below, relatively small (ED/HW 0.31; ED/SUL 0.11), less than snout (ED/SL 0.78). Distance between anterior corners of eyes is subequal to half inter-nostril distance (NN/EE 0.44). The angle of the jaw is posterior to a line drawn vertically from the back of the eye. The tympanum is not visible. Jaws without dentition; choanae small, round, located at anterior margins of roof of mouth; vomer processes and teeth absent; tongue long, narrow, slightly bifurcated distally. No median lingual papilla present. The dorsal surfaces of the head, trunk and limbs are smooth, with glands and skin folds present but indistinct; the rictal gland is broken, continuing posteriorly as a series of bulges to the arm insertion. The supratympanic fold is thick and glandular to the arm insertion, then forming a weak groove (not visible in life) running over the arm to the leg insertion. Lines of broken skin glands form an hourglass pattern dorsally, narrowest at the scapular region. The underside is smooth, with a slightly yellowish throat. The forelimb is slender, hand small (HAN/SUL 0.27), finger tips bluntly rounded without discs. Relative finger lengths I<II<IV<III; subarticular tubercles distinct, rounded, with one on fingers I and II, two on fingers III and IV. No webbing between fingers, although each digit tapers laterally to a thin margin. Thenar tubercle obscured by nuptial pad that reaches the distal phalanx of the first finger dorsally; palmar tubercles moderate, rounded, outer metacarpal tubercle small, dark and distinct. The inner metatarsal tubercle is obscured by the nuptial pad. There are many supernumerary tubercles on the granular palm. Hind limbs moderately long (TIB/SUL 0.44; FOT/SUL 0.54), foot longer than tibia (TIB/FOT 0.81); thighs are moderately developed, with rough glands on the inner posterior faces; relative toe lengths are I<II<V<III<IV. The toe tips are not expanded; subarticular tubercles: one on toes I and II, two on toes III and V, and three on toe IV. No webbing between the toes. Inner metatarsal tubercle rounded, prominent, outer metatarsal tubercle present as a small bump. Comparative proportions are shown in Appendix 2. Colour in preservative. The holotype has a brown dorsum with darker blotches and some small black flecks. The upper limb surfaces are brown with small black flecks. The sides are lighter with fine pale mottling. The lower lip is pale with small speckles. The throat has very fine speckling around the margins. The belly has about ten large pale brown irregular blotches with a few small black flecks around the margin. The undersurfaces of the limbs are pale with small black flecks. Colour in life. The dorsum is overall pale brown with irregular darker blotches. The iris is golden brown. The palms and soles of the feet are pale brown with the digits dark brown. The rictal gland is white. 524 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL.  Paratype variation. The paratypes are similar in body proportions, with the female lightly larger SUL 21.9. The male from Vissershok has a pale green dorsum overlaying a brown background, with irregular brown makings visible where the green is absent. FIGURE 3. Representatives of species of African Dainty Frogs, Cacosternum. All localities in South Africa unless indicated otherwise. A–C. aggestum sp. nov. Holotype MHNG 2690.25 Klipheuwel; B–C. australis sp. nov. MHNG 2699.34, Robertson; C–C. boettgeri MHNG 2740.23 Matatiele; D–C. capense MHNG 2690.23 Klipheuwel; E–C. karooicum MHNG 2740.67 Vrolijkheid; F–C. kinangopensis holotype NMK A/4372 Murugaru, Kenya; G–C. leleupi Lusinga, Democraic Republic of Congo; H–C. namaquense MHNG 2699.45 Arakoep; I–C. nanogularum sp. nov. holotype MHNG 2750.77 Nkandla; J–C. nanum MHNG 2740.86 Hogsback; K–C. parvum MHNG 2741.2, Mariepskop; L–C. platys MHNG 2709.03 Noordhoek; M–C. plimptoni AC2535 MHNG 2661.23 Serengeti National Park, Tanzania; N–C. rhythmum sp. nov. MHNG 2741.15, Qudeni; O–C. striatum MHNG 2741.20 Mpur Forest. AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 525 FIGURE 4. Ventral views of breeding Cacosternum males. All localities in South Africa unless indicated otherwise. A–C. aggestum sp. nov. MHNG 2690.25, Klipheuwel. B–C. australis sp. nov., MHNG 2699.42 Pearly Beach C–C. boettgeri MHNG 2740.31, The Vale Farm. D–C. capense MHNG 2690.23 Klipheuwel. E–C. karooicum MHNG 2740.67, Vrolijkheid Nature Reserve. F–C. kinangopensis NMK A/4372, Murungaru, Kenya. G–C. leleupi MHNG 2740.69 Lusinga, Democratic Republic of Congo. H–C. namaquense, MHNG 2699.44 Arakoep. I–C. nanogularum sp. nov. MHNG 2740.78, Nkandla. J–C. nanum MHNG 2740.87 Hogsback. K–C. parvum MHNG 2740.66 Maclear. L–C. platys MHNG 2699.37 Kenilworth. M–C. plimptoni MNHG 2661.21, Serengeti National Park, Tanzania. N–C. rhythmum sp. nov., MHNG 2741.12 Harrismith. O–C. striatum MHNG 2741.21 Mpur Forest. Advertisement call. The call consists of a series of pulses that speed up at the end, resembling a bouncing marble. The illustrated call from Klipheuwel (Fig. 5) consists of 18 pulses, with the pulse rate increasing from an initial 10 s-1 to a final 55 s-1. 526 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. FIGURE 5. Representative advertisement calls of C. aggestum sp. nov. Klipheuwel (left), C. boettgeri Harrismith (middle) and C. capense Klipheuwel (right). FIGURE 6. Localities of the molecular samples used in this study. Cacosternum aggestum sp. nov. pale blue circles; C. australis sp. nov. - red squares; C. boettgeri—dark blue circles; C. kinangopensis—yellow square; C. leleupi—green triangle; C. plimptoni—white circle; C. rhythmum—orange triangles. Eggs and tadpoles. Unknown. Distribution. This species is presently only known (Fig. 6) from the molecular samples collected at Vissershok MHNG 2699.40 and Klipheuwel MHNG 2690.25, MHNG 2690.26, MHNG 2690.27, but we expect AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 527 that it occurs more widely. The frogs were found on a flooded wheat field near temporary rain pools, during winter rain. Etymology. The specific epithet is derived from the Latin aggestus meaning "hill". It is a noun in apposition. This refers to the type locality Klipheuwel (rocky hill). Cacosternum australis sp. nov. Southern Dainty Frog. (Figs. 3B, 4B) Holotype. A male, MHNG 2740.19, collected at Vrolijkheid Nature Reserve, Robertson, Western Cape Province, South Africa (33° 54' 55" S, 19° 53' 04" E) by A.L. de Villiers, 19 August 2007. Paratypes. MHNG 2699.34–2699.35, two males, near Vrolijkheid Nature Reserve, collected by A. Channing 11 August 2007; MHNG 2699.39 male, Grootvadersbosch Nature Reserve, collected by C. Dorse; MHNG 2699.41–2699.42 two males, Pearly Beach, collected by A.L. de Villiers. Coordinates are listed in Appendix 1. Diagnosis. The within-clade variation in uncorrected p distances for the 16S fragment varied from 0.0–0.4%, while the species differed by 1.4–6.6% from the other 14 species. The low difference of 1.4% from C. platys is considered in the Discussion. The sample has seven tyr haplotypes with one shared with a C. platys. Breeding males of C. australis sp. nov. possess very lightly speckled throats (Fig. 4). This distinguishes them from species with very dark or black throats (C. capense, C. leleupi, C. nanogularum sp. nov., C. kinangopensis, C. nanum and C. parvum.), species with yellow-beige or white throats (C. karooicum, C. aggestum sp. nov., C. striatum,C. plimptoni and C. rhythmum sp. nov.) and species with grey throats (C. boettgeri and C. platys). It can be distinguished from C. capense which has large dorsolateral glands and a pair of large glands above the vent. Viewed from above, the nostrils are within one nostril diameter from the anterior edge of the snout. This distingushes it from C. karooicum which has nostrils about three diameters behind the snout edge. The rictal gland is smooth and continuous with the upper lip. This distinguishes it from species with an indistinct rictal gland (C. nanogularum sp. nov., C. striatum, C. parvum) and species with interrupted "bumpy" rictal glands (C. capense, C. aggestum, C. striatum, C. nanum and C. platys). The advertisement call resembles the sound of a marble bouncing on a hard surface. It differs from the regular single or multiple clicks of C. boettgeri, C. kinangopensis, C. leleupi, C. plimptoni or the creaking calls of C. capense, C. karooicum, C. namaquense, C. nanogularum sp. nov., C. parvum, C. striatum and the brief chirp of C. nanum. The advertisement call of C. australis has double pulses, which distinguishes it from the 'bouncing marble' call of C. aggestum and C. platys with single pulses. The call of C. platys is also much faster, with about three times more pulses (Table 3). Description of holotype. The holotype is an adult male in breeding condition, SUL 17.7. Body gracile, widest at mid-belly, with a narrow head (HW/SUL 0.38). The head is bluntly rounded from above and in profile. Head length measured from the angle of the jaw is moderate (HL/SUL 0.32). Canthus rostralis rounded, straight from eye to nostril, loreal region sloped outwards; nostrils small, rounded, directed dorsolaterally. The nostrils are placed closer to the snout than the eye (EN/SL 0.64). Internostril distance is less than distance between eye and nostril (NN/EN 0.93). Eyes directed anterolaterally, the eyes protrude, and are visible from below, relatively small (ED/ HW 0.28) (ED/SUL 0.11), less than snout (ED/SL 0.86). Distance between anterior corners of eyes greater than internostril distance (NN/EE 0.42). The angle of the jaw is posterior to a line drawn vertically from the back of the eye. The tympanum is not visible. Jaws without dentition; choanae small, round, located at anterior margins of roof of mouth; vomer processes and teeth absent; tongue long, narrow, slightly bifurcated distally. No median lingual papilla present. The dorsal surfaces of the head, trunk and limbs are smooth, without glands and skin folds present; the rictal gland is smooth, continuing posteriorly unbroken to the arm insertion. The supratympanic fold forms a groove running over the arm to the leg insertion. There are no visible skin glands arranged in the typical hour-glass pattern. The underside is smooth, with a slightly pigmented vocal sac with lateral folds. The forelimb is slender, hand small (HAN/SUL 0.25), finger tips bluntly rounded without discs. Relative finger lengths I<II<IV<III; subarticular tubercles rounded, with one on fingers I and II, two on finger IV with the distal tubercle low and indistinct and three on finger III, with the two proximal subarticular tubercles on finger III 528 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. small but distinct. No webbing between fingers. Thenar tubercle small, elongated, partially obscured by nuptial pad that reaches the distal phalanx of the first finger; palmar tubercles large, rounded, outer metacarpal tubercle absent. There are at least five supernumerary tubercles on the palm. Hind limbs moderately long (TIB/SUL 0.42) (FOT/SUL 0.51), foot longer than tibia (TIB/FOT 0.83); thighs are moderately developed, with rough glands on the inner posterior faces; relative toe lengths are I<II<V<III<IV. The toe tips are not expanded; subarticular tubercles: one on toes I and II, two on toes III and V, and three on toe IV. No webbing between the toes. Inner metatarsal tubercle elongated, prominent, outer metatarsal tubercle present as a small pale raised spot. Morphological proportions are shown in Appendix 2. Colour in preservative. The back has a grey background, with a darker brown V originating behind the eyes, forming an indistinct vertebral band running to the sacral area, from where it is broken into small blotches. Small darker patches are arranged in pairs on either side of the midline. A relatively large single dark patch is present on the dorsal surface of each thigh and tibia. The belly is pale, with many small clouds of speckles arranged in about 25 brown spots. The undersides of the hands and feet are pigmented. Colour in life. The back is pale beige with scattered darker markings. The upper lip is white. The belly has large numbers of very small black spots, with some brown markings joining the spots (Fig. 4B). Paratype variation. The two males from near Vrolijkheid Nature Reserve are similar in overall proportions and dorsal colour pattern to the holotype. The belly spots are also similar, with one possessing additional small black spots. These small dark spots are also present in the males from Pearly Beach (Fig. 4). The male from Grootvadersbosch Nature Reserve has an immaculate belly. Advertisement call. A typical advertisement call (Fig. 8) consists of 12 pulses, with the initial nine arranged into three doublets and three single pulses. The final three pulses are produced at a much faster rate. This call has a duration of 0.55 s, with an initial pulse rate of 10 s-1, and a final rate of 40 s-1. The call sounds like a marble bouncing on a hard surface. Eggs and Tadpoles. Unknown. Distribution. Our molecular records are from Pearly Beach, Grootvadersbosch Nature Reserve and Vrolijkheid Nature Reserve (Fig. 6). We expect that it is widely distributed in the southern Cape and Little Karoo. They are found in temporary rain-filled pools from coastal sands and inland rocky areas. Etymology. This species was initially confused with C. platys, but is now known to occur only in the southern Cape region. The specific epithet is an adjective, derived from the Latin australis (of the south) referring to this distribution. Cacosternum boettgeri (Boulenger, 1882) Boettger's Dainty Frog. (Figs. 3C, 4C) Cacosternum schebeni (Nieden, 1914) Genetics. The within-clade uncorrected p distances for 16S vary from 0–0.5%. The differences between it and the other 14 species range from 1.1–7.1% (Table 2). The 16S differences between C. boettgeri and C. rhythmum sp. nov. are very small, and by themselves might indicate that these are conspecific, but the advertisement calls are different. There are 15 likely tyr haplotypes in the sample, of which one is shared with a specimen of C. striatum. This is considered in the Discussion. Advertisement call. A typical call (Fig. 5) consists of about seven or more pulses produced at a pulse rate of -1 12 s , with a duration of 0.55 s, and dominant energy at 4.5 kHz. Morphology. Females exceed 21 mm SUL. Breeding males have a grey vocal sac. The back is smooth, without large glands. Viewed from above the nostrils are situated at the front of the snout. The supratympanic fold continues posteriorly forming a distinct saddle. The rictal gland is smooth, continuous with the upper lip. There are about two supernumerary tubercles on the palm. The margin notch between the third and fourth toes passes the proximal subarticular tubercle of the third toe. The inner metatarsal tubercle is conical, protruding, twice as wide as the tip of the first toe. The belly possesses a few small spots, leaving at least 95% of the belly pale (Fig. 4). Distribution. Molecular records for this study include Acornhoek, Bamboesberg, Baviaanskloof, Britstown, AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 529 Franklin, Hardap, Jansenville, Keetmanshoop, Kubusi Forest Reserve, Kuruman, Ladismith, Lebombo Mountains, Maselspoort, Matatiele, Maclear and The Vale Farm (Fig. 6). Although the records for this widespread species are confused with other (often sympatric) taxa, it is known to occur from Namibia, South Africa, southern Mozambique, Zimbabwe and southern Zambia (Poynton et al. 2004) with Ethiopian records referring most probably to an undescribed species. It is known from 1027 quarter degree squares in South Africa (Animal Demography Unit 2012). The frogs breed in temporary shallow pools in a range of habitats from arid rocky areas to moist savanna. It may breed in forest clearings (Scott 2004a). Scott (2004a) provides additional life history details. Notes. The calls and 16S sequences of material from Namibia, attributable to Cacosternum schebeni (Nieden, 1914) fall within the variation of C. boettgeri, confirming that C. schebeni is a junior synonym, as proposed by Mertens (1955). The GenBank sequence DQ022352 with the voucher identified as Cacosternum boettgeri (Scott 2005) is C. platys. Cacosternum capense Hewitt, 1925 Cape Dainty Frog. (Figs. 3D, 4D) Genetics. Cacosternum capense varies from 2.1–7.3% for 16S from the other 14 species, while the within-clade p distance is 0 (n=3) (Table 2). The sample has four likely tyr haplotypes, but shares none. Advertisement call. A call from Klipheuwel, (Fig. 5) is an indistinct creak, 0.3 s in duration, consisting of 20 pulses with a mean pulse rate of about 63 s-1. The pulse rate slows at the end of the call. Morphology. This species is unique in possessing a pair of large dorsolateral glands and a pair of glands above the vent. It is the largest species, females exceeding 30 mm SUL. Breeding males have a vocal sac that is black anteriorly, becoming spotted and glandular posteriorly. Viewed from above the nostrils are within one nostril diameter of the anterior snout edge. The supratympanic fold continues posteriorly as an indistinct glandular fold, expanded at the arm insertion. The rictal gland forms an interrupted arc around the angle of the jaw. There are few supernumerary tubercles on the palm. There is no trace of webbing between the toes. The inner metatarsal tubercle is conical, protruding, equal in width to the tip of the first toe. The nuptial pad in breeding males extends to the level of the proximal subarticular tubercle. The belly pattern consists of bold spots, about the same diameter as the lens of the eye, often joined into lines of irregular blotches (Fig. 4). Distribution. This species is endemic to the south-western Cape of South Africa. Molecular records are from Klipheuwel (Fig. 7). It is restricted to low-lying sandy areas, known from 26 quarter degree squares in the south- western Cape (Animal Demography Unit 2012). It breeds in pools on poorly drained soils (De Villiers 2004), in areas that are today mostly agricultural lands. Cacosternum karooicum Boycott, de Villiers & Scott, 2002 Karoo Dainty Frog. (Fig. 3) Genetics. The within-clade difference for C. karooicum for 16S is 0.2% (n=2). This species differs from the other 14 species by 4.0–6.7% (Table 2). Our sample has two likely tyr haplotypes, of which none are shared. Advertisement call. A call recorded at Oukloof (Fig. 7) is a creak about 0.5 s long, consisting of 24 more or less evenly spaced pulses, at a mean pulse rate of 48 s-1. Morphology. Females exceed 22 mm SUL. Male vocal sacs are pale beige anteriorly, becoming white posteriorly. The back is smooth with no protruding glands. The nostrils, viewed from above, are about three nostril diameters back from the snout. The supratympanic fold continues backwards as a distinct saddle. The rictal gland is small, visible at the angle of the jaw, weakly extending from the upper lip. The notch formed by the margins of toes 3 and 4 reaches the proximal subarticular tubercle of the third toe. The inner metatarsal tubercle is small, similar in size to the tip of the first toe. The nuptial pad in breeding males extends from the base of the first finger to the level of the proximal subarticular tubercle. The belly pattern varies, but often consists of small spots that coalesce, resulting in about two-thirds of the belly without pigment. 530 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. FIGURE 7. Representative advertisement calls of Cacosternum karooicum Oukloof (left) and C. kinangopensis Murungaru (right). Distribution. The molecular records are based on vouchers from Vrolijkheid Nature Reserve near Robertson (Fig. 9). This species is known from the drier areas of the south-western Cape, recorded from 11 quarter degree squares (Animal Demography Unit 2012). It occurs in arid areas, and it has been suggested that its flattened body allows it to aestivate in rock cracks (Scott 2004c). Cacosternum kinangopensis Channing & Schmitz, 2009 Kinangop Dainty Frog. (Fig. 3) Genetics. This species differs by 3.2–8.2% for the 16S gene, from the other 14 species (Table 2). The single specimen has two likely tyr haplotypes but shares none. Advertisement call. Channing & Schmitz (2009) illustrate a typical call with nine notes, each consisting of three to four pulses. Parameters from 19 calls include 7–10 notes, a duration of 0.8–1.3 s, and a note rate of 7.7–8.6 s-1 (Fig. 7). The pulses comprising the notes have a pulse rate of 51–77 s-1. Morphology. The male type is 18.8 mm SUL. Male vocal sacs are black. The back is smooth with no protruding glands. The nostrils, viewed from above, are about two nostril diameters back from the snout. The supratympanic fold continues backwards as a distinct saddle. The rictal gland is smooth, distinct, extending from the upper lip. The notch formed by the margins of toes 3 and 4 does not reach the proximal subarticular tubercle of the third toe. The inner metatarsal tubercle is small, wider that the tip of the first toe. The nuptial pad in breeding males extends from the base of the first finger to the level of the proximal subarticular tubercle. The belly pattern of the holotype consists of a few small and large spots on a white background (Fig. 4). The tympanum is visible, at least on one side (Channing & Schmitz 2009). Distribution. This species in only known from the North Kinangop Plateau, Kenya (Channing & Schmitz 2009). It was breeding in a rain-filled quarry in high altitude grassland. Our molecular sample is from the holotype collected at Murungaru (Fig. 6). The species probably occurs more widely. AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 531 Cacosternum leleupi Laurent, 1950 Leleup's Dainty Frog. (Figs. 3G, 4G) Genetics. Poynton & Scott (2004) confirmed that this poorly-known species was valid, citing unpublished work. New material from the type locality shows an uncorrected p distance for 16S of 4.1–7.3% from the other 14 species, with a within-sample p distance of 0.0 (n=2). Our sample has two likely tyr haplotypes, but shares none. Advertisement call. A typical call recorded at Lusinga shows an initial brief buzz followed by 11 notes. The duration is about 0.4 s, with the notes produced at a rate of 33 s-1. The initial buzz consists of about five pulses (Fig. 8). FIGURE 8. Representative advertisement calls of Cacosternum leleupi Lusinga (left), C. namaquense Nuwerus (middle) and C. australis sp. nov. Vrolijkheid (right). Morphology. Females are not known to reach 17 mm SUL, with males around 14 mm. Breeding males have a black vocal sac posteriorly, becoming grey anteriorly. Viewed from above, the nostrils are situated at the anterior margin of the snout. The supratympanic fold is absent or indistinct. The upper lip is continuous with a smooth rictal gland. There are five supernumerary tubercles on the palm. There is distinct webbing which reaches the proximal subarticular tubercles of the third and fourth toes. The inner metatarsal tubercle is low, twice the width of the tip of the first toe. Breeding males have a black nuptial pad that extends to the last phalanx of the first finger. The belly shows small spots and blotches that coalesce, covering about half of the abdomen with pigment. Distribution. The species is recorded from the Upemba and Kundelunga National Parks in Katanga Province, Demographic Republic of Congo. Our molecular sample comes from Lusinga in the Upemba National Park, Haut- Katanga Province, Democratic Republic of the Congo (Fig. 6). This is the locality referred to as "Mukana" by the De Witte expedition (Schmidt & Inger 1959). The species was breeding in a temporary pool in grassland. Cacosternum namaquense Werner, 1910 Namaqua Dainty Frog. (Figs. 3H, 4H) Genetics. The within-clade uncorrected p distance for 16S is 0.2%, with the distances to the other 14 species ranging from 3.0–7.2%. The sample has two likely tyr haplotypes, of which neither are shared. 532 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL.  Advertisement call. The call is a high-pitched creak. The illustrated example (Fig. 8) has a duration of 0.3 s, with 18 pulses at a pulse rate of 56 s-1. Morphology. Females reach 20 mm SUL in our sample. Breeding males have a vocal sac that is diffuse grey anteriorly, with the belly pattern of spots posteriorly. Viewed from above, the nostrils are within one nostril diameter of the snout margin. The supratympanic fold continues posteriorly to form a glandular saddle. The rictal gland is prominent, continuous with the upper lip. There are two supernumerary tubercles on the palm. There is no webbing, although the notch between the third and fourth toes passes the proximal subarticular tubercle of the third toe. The inner metatarsal tubercle is conical, equal to the width of the tip of the first toe. The nuptial pad of breeding males extends to the level of the proximal subarticular tubercle of the first finger. The belly spots are larger than the lens of the eye, many running together to form lines (Fig. 4). Distribution. Molecular samples are from Arakoep (Fig. 7). This species is morphologically distinct, unlikely to be confused with similar species, and is widespread in Namaqualand, known from 41 quarter-degree squares (Animal Demography Unit 2012). The species is known from arid rocky areas, where it breeds in temporary pools (Scott 2004d). FIGURE 9. Localities of the molecular samples used in this study. Cacosternum capense pale pink circle; C. namaquense pale blue square; C. nanum dark green circles; C. platys yellow triangles; C. karooicum yellow circle; C. nanogularum orange squares; C. parvum red triangles; C. striatum white circle. Cacosternum nanogularum sp. nov. Small-throated Dainty Frog. (Figs. 3I, 4I) Holotype. An adult male MHNG 2740.77 collected by M. Burger, 3 December 2005, on the road between Eshowe and Nkandla, KwaZulu-Natal Province, South Africa, 28° 51’ 04” S, 31° 12’ 27” E. The specimen was collected from a roadside puddle. Paratypes. Six males; MHNG 2740.75 from 41 km from Kranskop on road to Qudeni, KwaZulu-Natal Province, South Africa, 28° 43’ 41” S, 30° 56’ 35” E; NMB A7518–20, MHNG 2740.76 from 13 km from AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 533 Kranskop along road to Qudeni, KwaZulu-Natal Province, South Africa, 28° 54’ 14” S, 30° 51’ 15” E; MHNG 2740.78 collected on the road between Eshowe and Nkandla, KwaZulu-Natal Province, South Africa, 28° 51’ 04” S, 31° 12’ 27” E; a female, MHNG 2740.74, from Dukuduku, KwaZulu-Natal Province, South Africa, 28° 21' 09" S, 32° 17' 53" E. Diagnosis. The uncorrected p distances for 16S from the other 14 species ranges from 3.8–6.9%, while the within-clade p distances range from 0–0.4 (n=7). The sample has five likely tyr haplotypes of which none are shared. This species lacks a distinct supratympanic fold, which is present in all the other species. It can be distinguished from C. capense by the absence of large lateral and posterior glands on the dorsum; from C. karooicum which has the nostrils at least three nostril diameters back from the snout margin; from C. leleupi which has a small amount of webbing; from C. namaquense which has a diffuse grey vocal sac anteriorly (dark in C. nanogularum sp. nov.) (Fig. 4). It has three supernumerary tubercles on the palm, less than the five of C. aggestum sp. nov. and C. leleupi. The inner metatarsal tubercle is equal to the proximal subarticular tubercle of the first toe, differing from the wider inner metatarsal tubercle of C. aggestum sp. nov., C. boettgeri and C. parvum. The rictal gland is indistinct or absent, which distinguishes it from the prominent glands of C. aggestum sp. nov., C. boettgeri, C. australis sp. nov., C. striatum, C. nanum, C. parvum, C. platys, C. kinangopensis and C. plimptoni. The advertisement call is a series of chirps, which distinguishes it from the regular single or multiple clicks of C. boettgeri, C. kinangopensis, C. leleupi, C. plimptoni and the brief chirp of C. nanum. It is unlike the 'bouncing marble' calls that consist of a number of pulses that speed up, as C. aggestum sp. nov., C. australis sp. nov. and C. platys, or the long complex call of C. rhythmum sp. nov. The call is not a creak as in C. capense, C. karooicum, C. namaquense, C. parvum, or C. striatum. FIGURE 10. Representative advertisement calls of Cacosternum nanogularum sp. nov. Kranskop (left), C. nanum Beacon Bay (middle) and C. parvum Maclear (right). Description of the holotype. The holotype is an adult male, SUL 19.7. Body widest at mid-belly, with a narrow head (HW/SUL 0.33). The head is bluntly rounded from above and in profile. Head length measured from the angle of the jaw is about one third of body length (HL/SUL 0.31). Canthus rostralis rounded, straight from eye to nostril, loreal region sloped outwards ventrally; nostrils small, rounded, directed laterally, and posteriorly. The nostrils are placed nearly midway between the snout than the eye (EN/SL 0.52). Internostril distance is greater than distance between eye and nostril (NN/EN 1.31). Eyes directed anterolaterally, the eyes protrude, and are visible 534 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. from below, relatively small (ED/HW 0.35; ED/SUL 0.12), less than snout (ED/SL 0.74). Distance between anterior corners of eyes subequal to internostril distance (NN/EE 0.53). The angle of the jaw is posterior to a line drawn vertically from the back of the eye. The tympanum is not visible. Jaws without dentition; choanae small, round, located at anterior margins of roof of mouth; vomer processes and teeth absent; tongue long, narrow, slightly bifurcated distally. No median lingual papilla present. The dorsal surfaces of the head, trunk and limbs are smooth, with glands and skin folds present but indistinct; the rictal gland is smooth, continuing posteriorly as a series of bulges to the arm insertion. The supratympanic fold forms a forms a groove running over the arm to the leg insertion. Lines of broken skin glands form an hourglass pattern dorsally, narrowest at the scapular region. The underside is smooth, with a slightly pigmented vocal sac with lateral folds. The fore limb is slender, hand small (HAN/SUL 0.30), finger tips bluntly rounded without discs. Relative finger lengths I<II<IV<III; subarticular tubercles distinct, rounded, with one on fingers I and II, two on fingers III and IV. No webbing between fingers. Thenar tubercle small, rounded, partially obscured by nuptial pad that reaches the distal phalanx of the first finger; palmar tubercles and inner metatarsal tubercle moderate, rounded, outer metacarpal tubercle present. There is a single supernumerary tubercle on the palm. Hind limbs moderately long (TIB/SUL 0.50; FOT/SUL 0.52), foot subequal to tibia (TIB/FOT 0.95); thighs are moderately developed, with rough glands on the inner posterior faces; relative toe lengths are I<II<V<III<IV. The toe tips are not expanded; subarticular tubercles: one on toes I and II, two on toes III and V, and three on toe IV. Only a trace of webbing between the toes. Inner metatarsal tubercle conical, prominent, outer metatarsal tubercle present as a small pale raised spot. Morphological proportions are shown in Appendix 2. Colour in life. The background colour is beige, overlain with a dark brown interorbital bar with a sharp tip pointing posteriorly, and a black band running from the nostril, through the eye to the arm insertion. The upper lip is white with black speckles, continuing as a white rictal gland. Colour in preservative. The patterns are visible as dark browns on a paler background. Paratype variation. The male paratypes have SUL 20.4–22.5, with the females 23.2–25.0. They are very similar to the holotype in having discontinuous bumpy rictal glands, large subarticular tubercles on the hand, and small conical subarticular tubercles at the base of the fourth toe. The belly spots are similar, with the female throats spotted but less pigmented than the males, while the body proportions show little variation. Advertisement call. The general call structure consists of a series (10 to >20) of regularly repeated pulsed notes (chirps), with a note rate of 0.4–0.6 s-1. The mean note duration is 89 ms, with the mean number of pulses per call of 9. The mean pulse rate is 89 s-1. It has a mean emphasised frequency of 3.02 kHz. The illustrated call (Fig. 10) has a duration of 0.1 s, with 9 pulses at a pulse rate of 80 s-1. Eggs and tadpoles. Unknown. Distribution. Molecular samples come from Nkandla and the Lebombo Mountains, with additional specimens known from Qudeni and Dukuduku (Fig. 9). The full extent of its range is unknown. The frogs were found in flooded temporary pools with emergent vegetation. Etymology. The specific name is derived from the Latin nanus (small), and gula (throat). It is an adjective. The name refers to the very small vocal sac that is produced during calling. Cacosternum nanum Boulenger, 1887 Dwarf Dainty Frog. (Figs. 3J, 4J) Cacosternum poyntoni Lambiris, 1988 Genetics. The uncorrected p distances between this species and the other 14 range from 2.8–6.0%, while the within-clade p distances range from 0.0–0.4 (n=20). The sample has 16 likely tyr haplotypes, of which one is shared between a single C. nanum and one C. rhythmum sp. nov. Advertisement call. Based on an analysis of 447 calls of 41 specimens, a calling bout usually consists of a long (15 to >30) series of pulsed chirps that are rapidly repeated, about 1.2 s-1 (range 0.6–2.0 s-1). The mean call (chirp) duration is 52.6 ms (range 19.5–99.7 ms). The mean number of pulses per call is 11 (range 5–18), with a mean pulse rate of 200 s-1 (range 95–288 s-1). The mean emphasised frequency is 3.6 kHz (range 3.3–4.1 kHz) A AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 535 spectrogram of a typical call is shown in Fig. 10. It has a duration of 0.04 s, consisting of 11 pulses at a pulse rate of 250 s-1. Morphology. Females reach 18 mm SUL. The dorsum lacks protruding glands. Viewed from above, the nostrils are close to the anterior margin of the snout. The supratympanic fold continues as a straight line from eyelid to arm insertion. The rictal gland is interrupted, tapering to the arm insertion. There are four supernumerary tubercles on the palm. The inner metatarsal tubercle is low, with a width equal to the tip of the first toe. Breeding males possess an inconspicuous nuptial pad that extends to the last phalanx. The belly spots are the size of the lens of the eye, dense grey, often running together (Fig. 3). Distribution. This species is known from the south and east of South Africa, occurring in 276 quarter degree squares (Animal Demography Unit 2012). Molecular samples come from Baviaanskloof, Cedarville, Goukamma, Hogsback, Kubusi Forest Reserve, Maclear and Montague Pass in South Africa (Fig. 7). It breeds in a variety of water bodies (Scott 2004b), usually those with emergent vegetation. Note. We regard Cacosternum poyntoni as a junior synonym of C. nanum, following the reasons presented by Minter (2004), Scott (2004b) and Scott & Minter (2004a, b). These include the lack of material apart from the type, after nearly 60 years, from a well-known locality, Pietermaritzburg; the consensus amongst local herpetologists that this is not a valid taxon; and the skeletal similarity to C. nanum from an unpublished report. Cacosternum parvum Poynton, 1963 Mountain Dainty Frog. (Figs. 3K, 4K) Genetics. The within-clade variation in uncorrected p distances for the 16S fragment is 0.2%, with the differences to the other 14 species being 3.9–7.6%. The sample has two tyr haplotypes, of which none are shared. Advertisement call. The call is a brief chirp. The following description is based on an analysis of 172 calls of 17 specimens. A motivated calling bout usually consists of a long (10 to >20) series of pulsed chirps that are repeated at a rate of about 0.6 s-1 (range 0.4–0.7 s-1). The mean call (chirp) duration is 60.2 ms, (range 27.6–109.0 ms). The mean number of pulses per call is 6 (range 3–9) with a mean pulse rate of 86.8 s-1 (range 59.5–125 s-1). The mean emphasised frequency is 3.99 (range 3.55–4.42 kHz). The illustrated call (Fig. 10) has a duration of 0.06 s, with seven almost equally spaced pulses, with a pulse rate of 100 s-1. Morphology. Females do not exceed 16.5 mm SUL in our sample. Breeding males have dark vocal sacs with a paler margin. The dorsum is smooth, and, viewed from above the nostrils are positioned on the anterior edge of the snout. The supratympanic fold continues posteriorly to form a rounded saddle. The rictal gland is indistinct, interrupted. There are no supernumerary tubercles on the palm. There is no webbing, with the notch between the third and fourth toes reaching the proximal subarticular tubercle of the third toe. The outer metatarsal is absent or indistinct, while the inner metatarsal tubercle is low, twice the width of the tip of the first toe. The nuptial pad in breeding males is slightly pigmented but indistinct. The belly markings consist of groupings of brown speckles that form blotches. Distribution. Molecular samples come from Mariepskop (Fig. 9). The species appears to be a highland endemic in eastern South Africa (Scott & Minter 2004b), where it breeds in high altitude grassland pools (Scott 2004e). Note. The GenBank sequence DQ022353 with the voucher identified as Cacosternum nanum parvum (Scott 2005) is a C. nanum sequence. Cacosternum platys Rose, 1950 Flat Dainty Frog. (Figs. 3L, 4L) Genetics. The within-clade variation in uncorrected p distance for the 16S fragment varies from 0.0–0.2%, but between C. platys and the other 14 species 1.8–6.6%. The sample has nine likely tyr haplotypes, of which one is shared with a specimen of C. australis sp. nov. See the Discussion. 536 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL.  Advertisement call. A typical advertisement call (Fig. 11) consists of a long series of pulses that increase in tempo. The duration of the illustrated call is 0.55 s, with 34 pulses. The initial and final pulse rates are 20 s-1 and 80-1. Morphology. This is a small species, with females less than 14 mm in our sample. The vocal sac in breeding males is pale grey (Fig. 4). The dorsum is smooth. Viewed from above, the nostrils are situated on the anterior margin of the snout. The supratympanic fold continues posteriorly in a straight line from the eyelid to the arm insertion. The rictal gland is interrupted. There are two supernumerary tubercles on the palm. There is no webbing. The inner metatarsal tubercle is low, rounded, equal in width to the tip of the first toe. The nuptial pad is a dark swollen mass at the base of the first finger, tapering to the level of the proximal tubercle. The belly pattern consists of very small spots consisting of groups of speckles. FIGURE 11. Representative advertisement calls of Cacosternum platys Kenilworth (left), C. plimptoni Nairobi (center) and C. striatum Singisi (right). Distribution. Molecular evidence shows the species is present on the Cape Peninsula (Noordhoek) and in Cape Town (Kenilworth) (Fig. 7). It is a fynbos species, breeding in low altitude temporary pools. Notes. The GenBank sequence DQ283258 with the voucher identified as Cacosternum platys (Frost et al. 2006) is Microbatrachella capensis. The GenBank sequence DQ022352 with the voucher identified as Cacosternum boettgeri (Scott 2005) is C. platys. Cacosternum plimptoni Channing, Brun, Burger, Febvre & Moyer, 2005 Plimpton's Dainty Frog. (Figs. 3M, 4M) Genetics. This species varies from the other 14 species for the 16S fragment, by uncorrected p values of 2.5–5.9%. A single specimen had only one likely tyr haplotype, that was not shared with any other species. Advertisement call. A series of 29 calls of four individuals from the Serengeti National Park, Tanzania showed a mean duration of 0.55 s (range 0.37–0.77) and the mean number of notes (pulse trains) 6.2 (range 4–8) (Channing et al. 2005). A typical advertisement call from Nairobi (Fig. 11) is a harsh clicking, where each series of notes consists of double pulses, apart from the initial note of three pulses. The call duration is 0.5 s, with six notes at a rate of 10 s-1. Morphology. The female holotype has a SUL 19.7 (Channing et al. 2005). Breeding males have yellow vocal AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 537 sacs without a pale margin (Fig. 4). The dorsum is smooth, and, viewed from above the nostrils are positioned on the anterior edge of the snout. The supratympanic fold is inconspicuous. The rictal gland is indistinct. There are 5 supernumerary tubercles on the palm. There is no webbing, with the notch between the third and fourth toes reaching the proximal subarticular tubercle of the third toe. The outer metatarsal is absent or indistinct, while the inner metatarsal tubercle is conical and protruding, twice the width of the tip of the first toe. The nuptial pad in breeding males is slightly pigmented but massive, reaching the proximal subarticular tubercle. The belly markings consist of groupings of brown speckles that form blotches with spots equal to the size of the eye lens. Distribution. The species appears to be present from northern Tanzania to the highlands of central Kenya (Channing et al. 2005, Channing & Schmitz 2009) (Fig. 6) and may be more widely distributed. It breeds in temporary and permanent pools with little emergent vegetation. Cacosternum rhythmum sp. nov. Rhythmic Dainty Frog. (Figs. 3N, 4N) Holotype. An adult male NMB A7521, collected by M. Burger, 18 December 2001 at Nottingham Road (29° 21’ 21” S, 29° 59’ 36” E), KwaZulu-Natal Province, South Africa. The habitat is flat open terrain with shallow pools and flooded grassy patches. Paratypes. Eight males and one female, NMB A7522–7530, from Nottingham Road, Nottingham Ford, and 28 km S of Harrismith. Diagnosis. This species differs from the other 14 species for the 16S fragment, by uncorrected p values of 1.1– 6.2%. The within-clade variation ranges from 0.0–0.2%. The small difference of 1.1% with C. boettgeri will be considered in the Discussion. The sample had eight likely tyr haplotypes, of which two were shared between one C. nanum and one C. rhythmum sp. nov. The vocal sac of breeding males is white (Fig. 5), which distinguishes it from species where the sac is black, grey or yellowish-beige (C. capense, C. karooicum, C. leleupi, C. namaquense, C. nanogularum sp. nov., C. aggestum sp. nov., C. boettgeri, C. australis sp nov., C. nanum, C. parvum, C. platys, and C. kinangopensis. It is distinguished from C. capense by the absence of protruding dorsolateral glands and a pair just above the vent. Viewed from above, the nostrils are close to the anterior margin of the snout, distinguishing it from C. karooicum, which has the nostrils about three nostril diameters back from the anterior edge of the snout. The supratympanic fold continues posteriorly as a weak saddle, distinguishing it from C. nanogularum sp. nov., which has no saddle. The rictal gland is smooth, tapering to the arm insertion. This distinguishes it from those species where the rictal gland is interrupted (C. capense, C. karooicum, C. striatum, C. nanum, C. parvum and C.platys). There are no supernumerary tubercles on the palm, distinguishing it from species with 2–5 tubercles (C. capense, C. karooicum, C. leleupi, C. namaquense, C. nanogularum sp. nov., C. aggestum sp. nov., C. boettgeri, C. nanum and C. platys). It has no webbing, which distinguishes it from C. leleupi, which has a trace of webbing between the third and fourth toes. The inner metatarsal tubercle is small, conical, equal in width to the proximal subarticular tubercle of the first toe. This distinguishes it from species where the width of the inner metatarsal tubercle is equal to the width of the tip of the first toe (C. capense, C. karooicum, C. namaquense, C. nanogularum sp. nov., C. australis sp. nov., C. striatum, C. nanum, C. plimptoni and C. platys), and from those where the inner metatarsal tubercle is twice the width of the tip of the first toe (C. leleupi, C. aggestum sp. nov., and C. parvum. The belly is pale, with a few very small brown blotches. This distinguishes it from species where the spots are larger than the eye lens, sometimes running together to form blotches (C. capense, C. namaquense, C. nanogularum sp. nov., C. aggestum C. aggestum sp. nov.,) and from C. australis sp. nov. which has irregular yellow-beige blotches overlain with very small black speckles, and from C. striatum which has a silvery-white belly with small black blotches. The advertisement call is complex and long. A typical call has a duration over 1.5 s, and consists of a string of pulses followed by a number of chirps, produced with a distinctive beat. It differs from the simple series of single or multiple clicks of C. boettgeri, C. kinangopensis, C. leleupi, and C. plimptoni or the creaking calls of C. capense, C. karooicum, C. namaquense, C. nanogularum sp. nov., C. parvum, and C. striatum or the brief chirp of C. nanum. It differs from those species producing a 'bouncing marble' call of pulses that speed up such as C. aggestum sp. nov., C. australis sp. nov. and C. platys. Description of the holotype. The holotype is an adult male, SUL 15.0. Body gracile, widest at mid-belly, with 538 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. a narrow head (HW/SUL 0.34). The head is acutely rounded from above and in profile. Head length measured from the angle of the jaw is moderate (HL/SUL 0.32). Canthus rostralis rounded, straight from eye to nostril, loreal region sloped outwards; nostrils small, rounded, directed laterally. The nostrils are placed closer to the snout than the eye (EN/SL 0.61). Internostril distance is less than distance between eye and nostril (NN/EN 0.91). Eyes directed anterolaterally, the eyes protrude, and are visible from below, relatively small (ED/HW 0.35; ED/SUL 0.12), equal to snout (ED/SL 1.0). Distance between anterior corners of eyes greater than internostril distance (NN/ EE 0.36). The angle of the jaw is posterior to a line drawn vertically from the back of the eye. The tympanum is not visible. Jaws without dentition; choanae small, round, located at anterior margins of roof of mouth; vomer processes and teeth absent; tongue long, narrow, slightly bifurcated distally. No median lingual papilla present. The dorsal surfaces of the head, trunk and limbs are smooth, with glands and skin folds present; the rictal gland is smooth, continuing posteriorly unbroken to the arm insertion. The supratympanic fold forms a groove running over the arm to the leg insertion. Lines of broken skin glands form an hourglass pattern dorsally, narrowest at the scapular region, with a second pair of broken skin folds parallel to the midline, running from behind the scapular region nearly to the tip of the urostyle. The underside is smooth, with a slightly pigmented vocal sac with lateral folds (Fig. 4). The forelimb is slender, hand small (HAN/SUL 0.26), finger tips bluntly rounded without discs. Relative finger lengths I<II<IV<III; subarticular tubercles distinct, rounded, with one on fingers I and II, two on finger IV and three on finger III, with the proximal subarticular tubercle on finger III small but distinct. No webbing between fingers. Thenar tubercle small, rounded, partially obscured by nuptial pad that reaches the distal phalanx of the first finger; palmar tubercles and inner metatarsal tubercle small, rounded, outer metacarpal tubercle absent. There are no supernumerary tubercles on the palm. An outgrowth is present on the outside of the fourth finger, at the level of the penultimate phalanx. It resembles a subarticular tubercle. No other growths like this are present in the sample examined. Hind limbs moderately long (TIB/SUL 0.43; FOT/SUL 0.53), foot longer than tibia (TIB/FOT 0.82); thighs are moderately developed, with rough glands on the inner posterior faces; relative toe lengths are I<II<V<III<IV. The toe tips are not expanded; subarticular tubercles: one on toes I and II, two on toes III and V, and three on toe IV. Only a trace of webbing between the toes. Inner metatarsal tubercle conical, prominent, outer metatarsal tubercle present as a small pale raised spot. Colour in preservative. The back has a grey background, with a darker brown vertebral band, within which is a thin pale vertebral line. Small darker patches are arranged in pairs. The belly is immaculate, with a margin of small clouds of speckles arranged in spots. The undersides of the hands and feet are pigmented. Colour in life. The dorsum is tan and brown with small black speckles and white blotches. Paratype variation. The paratypes are similar in body proportions to the holotype. The males range in SUL from 15.3–16.7, with the female 19.3. The rictal gland is identical in all the paratypes, while the subarticular tubercles of the hand vary slightly, from round to a sharper cone-shape. The female has a uniform coloured back, but all the male paratypes have a wide vertebral band, or a narrower vertebral line. Two males have dorsal warts. The belly is immaculate in eight paratypes, with one possessing a few small spots, leaving over 90% of the belly unpigmented. One specimen, although immaculate, possesses light clouds of speckles shaped as spots. The specimens show the darker belly window through which the abdominal muscles can be seen, and all have supernumerary tubercles on the hands. Eggs are visible through the belly skin of the female. Advertisement call. A typical call recorded at Nottingham Road is illustrated in Fig 12. The call consists of an initial phase of single or double clicks lasting about one second, at a rate of 10 notes s-1, followed by three chirps, each with seven pulses. The pulse rate of each chirp is about 60 s-1, with the chirps uttered at a rate of 6 s-1. The duration of such a call is 1.6 s. The call has a distinct repeated beat, characteristic of this species. Eggs and tadpoles. Unknown. Distribution. Our specimens come from the KwaZulu-Natal midlands in South Africa (Fig. 7), but we expect the species to be more widely distributed. It breeds in temporary pools with abundant vegetation. Etymology. The species name is derived from the Latin rhythmus (rhythm) referring to the rhythmic advertisement call. It is a noun in apposition. Note. A call description for Cacosternum cf. striatum from Champagne Castle (Pickersgill 2007) is attributable to C. rhythmum sp. nov. AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 539 FIGURE 12. An example of the advertisement call of Cacosternum rhythmum. Cacosternum striatum FitzSimons, 1947 Striped Dainty Frog. (Figs. 3O, 4O) Genetics. The uncorrected p within-clade variation in 16S ranges from 0.0–0.2% (n= 3). This species differs from the other 14 species by 2.5–7.3%. The sample has four likely tyr haplotypes, of which one is shared with an individual C. boettgeri. Advertisement call. The advertisement call is a brief high-pitched chirp. A typical call (Fig. 11) shows that the pulses are arranged in doublets in the middle, with the initial and final few pulses being produced at a faster rate than the middle. The duration of this call is 0.33 s, with the 21 pulses starting at 100 s-1, slowing to 60 s-1, before speeding up slightly at the end of the call. Morphology. These are small frogs, with females reaching 13.7 mm SUL in our sample (n=7). The vocal pouch of breeding males is pale (Fig. 4). The dorsum is smooth, lacking protruding glands. Viewed from above, the nostrils are within one nostril diameter of the anterior snout margin. The supratympanic fold continues posteriorly to form a saddle. The upper lip continues posteriorly as an indistinct interrupted rictal gland. The palm lacks supernumerary tubercles. The inner metatarsal tubercle is low, inconspicuous, as wide as the tip of the first toe. The nuptial pad of breeding males is low, inconspicuous, extending to the level of the proximal subarticular tubercle. The belly is silvery white, with some small black speckles. Distribution. The species is widespread in KwaZulu-Natal, South Africa, where it has been recorded from 13 quarter-degree cells (Animal Demography Unit, University of Cape Town). Our sequenced specimens come from Mpur Forest (Fig. 9). It breeds in shallow temporary pools that may be overgrown. Discussion Dainty frogs are polymorphic for colour, with tan, brown, green and beige common. They are small, and generally call from concealed sites. This study illustrates the presence of four previously unrecognised species. Wider sampling is sure to discover yet more species new to science. 540 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL.  The generally small size of dainty frogs may be a derived feature for the genus, or may be an ancestral trait. Van der Meijden et al. (2011) suggested that within the pyxicephalids in southern Africa, there have been several instances of reduction in body size, including Cacosternum, Arthroleptella and Microbatrachella. They find Microbatachella immediately basal to Cacosternum for one set of genes, while the relationship is reversed when additional genes are incorporated in the analysis. An earlier study using molecular and morphological data (Scott 2005) found Microbatrachella immediately basal to Cacosternum in a simultaneous analysis. Both these analyses suggest that the ancestor to Microbatrachella and Cacosternum was small. In the phylogeny shown in Fig. 1, one of the smallest species, C. nanum, is basal. However, the very small C. parvum is sister to the largest species in the genus, C. nanogularum. Although polystomatid flatworms, parasitic in the bladder of frogs, are assumed to be host-specific, the recently described Polystoma channingi has been found in both Cacosternum nanum and C. boettgeri where both species are sympatric (Du Preez 2013). As these two species are quite distantly related, it suggests that the polystome parasites are perhaps only restricted to the genus Cacosternum. Cacosternum speciation hypotheses. The mean uncorrected p distances for the 16S gene between the terminal sister-species in Fig. 1 are 5.6% between C. nanogularum (KwaZulu-Natal midlands) and C. parvum (highlands), 3.8% between C. striatum (KwaZulu-Natal) and C. kinangopensis (Kenya highlands), 1.4% between C. boettgeri (widespread) and C. rhythmum (KwaZulu-Natal midlands), 1.7% between C. platys (Cape Peninsula) and C. australis (southern Cape), and 2.4% between C. capense (Western Cape lowlands) and C. namaquense (Namaqualand). The varying genetic differences between the terminal species pairs suggest that they did not arise from common ancestors at about the same time. What speciation drivers might have been responsible? The following speculative scenarios are offered as testable hypotheses once more genetic material is available, and appropriate ingroup calibrations can be discovered. Explicit testing using techniques such as a relaxed-clock analysis are required, which are beyond the scope of this paper. 16S divergence rates of the North African ranids Pelophyylax saharicus and Pelophylax bedriagae are about 1% per my (Avise et al. 1998). Rates for salamanders and newts are slower, between 0.4% (Caccone et al. 1997) and 0.7% per my (Veith et al. 1998). Using the 1% per my rate, suggests that the approximate time of speciation for these pairs is between 5.6–1.4 my. During this period there were large climate changes, many taking place very rapidly (Adams et al. 1999). Divergence of C. nanogularum and C. parvum. Based on the North African ranid calibration, these species last shared a common ancestor about 5.6 mya. Presently C. parvum is found in high altitude grassland (Scott & Minter 2004b), which is adjacent to the area where C. nanogularum is found. This age overlaps with the initiation of the raising of the Karoo Plateau at around 5 mya (DeMenocal 2004), associated with an increase in aridity leading to the spread of open savanna-mosaic habitats (Sepulchre et al. 2006). This drying of the environment may be implicated in the splitting of the common ancestral population, with one part of the population in moist high altitude refugia, and the other in moist forest remnants. Divergence of C. striatum and C. kinangopensis. These species last shared a common ancestor around 3.8 mya. Today these species appear to be separated by over 3000 km, although the intervening areas have not been adequately explored for these small frogs, and the discovery of additional species in the intervening areas might cast new light on the question. C. kinangopensis is known from the high Aberdares above the rift valley of Kenya. The rift reached a maximum uplift at the Plio-Pleistocene, with the major escarpments in Kenya being in place by 3 mya, following major local uplift between 5 and 2 mya in the Lake Tanganyika and Malawi areas. There was a general drying out around 3 mya (DeMenocal 2004), associated with the formation of the high altitude topography (Sepulchre et al. 2006). Further collecting is required to determine the actual ranges of these two species. Divergence of C. boettgeri and C. rhythmum. These species last shared a common ancestor around 1.4 mya. Despite the low p distance, these two species have very different advertisement calls and share no mt or nuc haplotypes. C. boettgeri is widespread, but possibly sympatric with C. rhythmum in the KwaZulu-Natal midlands. There is evidence for an increase in climate variability and aridity, that coincides with the intensification of the northern glacial cycles (DeMenocal 2004) and a changeover in bovid and other fauna (Vrba et al. 1995), at around 1.2–0.8 mya. C. rhythmum may have developed in the remnant forests on the slopes of the Drakensberg Mountains, with C. boettgeri restricted to moist environments elsewhere. Divergence of C. platys and C. australis. These two species last shared a common ancestor about 1.7 mya. C. platys is known from the Cape Peninsula and surrounding lowlands, while C. australis is known from the southern AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 541 Cape. They are presently separated by about 100 km, although more collecting in the intervening areas may show that C. australis occurs much closer to Cape Town. Based on faunal and paleoclimate records, DeMenocal (2004) recognised a variable, drier period around 1.8–1.6 mya. The Cape Peninsula sea levels fluctuated by -120 m to +200 m from the Pliocene and Pleistocene ice-ages. The peninsula would have been a series of islands when the sea covered the Cape Flats. This inundation of the Cape Flats has occurred many times, most recently around 19– 14 ka (Compton 2011). The Cape Peninsula provides sufficient high ground for populations of C. platys to have survived, while the hills and mountains south of the Little Karoo would have provided moist habitat for C. australis. Divergence of C. namaquense and C. capense. These species are now adjacent, separated at about 32° S, roughly by the Olifants River Valley. They last shared a common ancestor 2.4 mya. This coincides with an increasingly variable climate and drier conditions at 2.9–2.4 mya (DeMenocal 2004). This period is also associated with major faunal changes, including bovid turnover. The southern C. capense might have survived in the moist highlands associated with the series of hills and mountains running north, bordering the coastal plain. C. namaquense may have survived in the moist mountains south of Springbok. Divergence of C. platys and C. rhythmum. These species differ by uncorrected p distances of 1.8 to 2.3, suggesting they last shared a common ancestor about 2.3–1.8 mya. Around 2.2 mya Nambia was becoming more arid, associated with increased coastal upwelling (Dupont et al. (2005). These species are found in South Africa in different rainfall regimes: The clade including C. platys is in the winter-rainfall area, while the clade including C. rhythmum is only known from the KwaZulu-Natal midlands, a summer-rainfall area. This is a common distribution pattern among closely related organisms, such as chameleons (Tolley et al. 2008), reed frogs (Poynton 1964), and some species of the ericoid genus Phylica (Richardson et al. 2001) although many plant groups show a much more complicated history in the Cape Floral Region (Linder 2005). The origin of the Cape Flora can be traced back to 8– 7 mya when there was widespread aridification (Richardson et al. 2001). The reasons why dainty frogs should have diverged around 2 mya, with clades in both winter and summer rainfall areas remains unclear. Shared nuclear haplotypes. The very few shared haplotypes we identified may result from incomplete lineage sorting of ancestral polymorphisms, or more recent hybridisation. Regarding the comparatively young age of the recovered lineages this is to expected since it takes much more time to reach complete sorting for nuclear genes than it does for mitochondrial genes (Moore 1995). The species pairs sharing tyr haplotypes (C. nanum/C. rhythmum - 2 haplotypes; C. boettgeri/C. striatum - 1 haplotype; C. australis/C. platys - 1 haplotype) are found in sympatry or near sympatry. Both C. nanum and C. rhythmum are sympatric in the KwaZulu-Natal midlands; C. boettgeri and C. striatum are sympatric in the KwaZulu-Natal midlands; and C. australis and C. platys are separated by a small distance, although further collecting may show that they, too, are sympatric. They may have been sympatric in the recent past before the last inundation of the Cape Flats by high sea levels. Only the latter pair are closely related sister species. This could be evidence of hybridisation in now sympatric species, which cannot be ruled out in environments where populations may have been isolated and reunited for different amount of time in the past. This aspect should be further investigated. There is presently insufficient evidence to confirm or reject these speciation hypotheses, although the climate and faunal data are compelling. Provisional key to males of the species of Cacosternum. Notes: This key needs to be tested against more specimens, as the variation in each species is not well understood. Advertisement calls are a reliable tool for identification. 1 A pair of large glands above the vent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum capense 1' No large skin glands above vent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 Webbing reaches proximal subarticular tubercle of fourth toe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum leleupi 2' No webbing present. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 Tympanum visible, at least on one side. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum kinangopensis 3' Tympanum not visible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 4 From above, nostril situated at least three nostril diameters from anterior margin of snout. . . . . . . . . . Cacosternum karooicum 4' Nostril situated within one nostril diameter of anterior margin of snout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 5 Dorsal skin glands form an hourglass pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 542 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. 5' Skin glands do not form an hourglass pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 6 Inner metatarsal tubercle as wide as tip of first toe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum nanogularum 6' Inner metatarsal tubercle twice width of tip of first toe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum aggestum 7 Dorsal skin glands form longitudinal dorsolateral ridges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 7' No glandular ridges on dorsum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 8 Inner metatarsal tubercle as wide as proximal tubercle of first toe. . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum nanogularum 8' Inner metatarsal tubercle twice width of tip of first toe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Cacosternum parvum 9 Supratympanic fold forms a thickened saddle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 9' Supratympanic fold is straight or indistinct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 10 Inner metatarsal tubercle as wide as tip of first toe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum boettgeri 10' Inner metatarsal tubercle as wide as tip of first toe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 11 Rictal gland prominent and smooth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Cacosternum namaquense 11' Rictal gland indistinct, interrupted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum striatum 12 Inner metatarsal tubercle twice width of tip of first toe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum plimptoni 12' Inner metatarsal tubercle as wide as proximal tubercle of first toe. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 13 Rictal gland smooth, uninterrupted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Cacosternum australis 13' Rictal gland interrupted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 14 Belly with grey spots, throat with dark margin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum nanum 14' Belly with brown spots, no dark margin of throat. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cacosternum platys Acknowledgements We would like to acknowledge the outstanding assistance of Atamato Madrandele, Chief Warden of the Upemba National Park in the DRC. We note with concern that he was subsequently ambushed and killed by a local mai mai rebel group, December 16, 2012. This will set back research in the area. Atherton de Villiers and staff at Vrolijkheid NR from CapeNature are thanked for providing material of C. australis. Ciff Dorse collected material at Montague Pass and Vissershok, James Harvey collected material in the KwaZulu-Natal midlands. AC acknowledges funding from the National Research Foundation and the University of the Western Cape. References Adams, J., Maslin, M. & Thomas, E. (1999) Sudden climate transitions during the Quaternary. Progress in Physical Geography, 23, 136. http://dx.doi.org/10.1191/030913399670425018 Animal Demography Unit (2012) Southern African Frog Atlas Project. Electronic Database accessible at http://adu.org.za (Accessed 21 Aug. 2013) Avise, J.C., Walker, D. & Johns, G.C. (1998) Speciation durations and Pleistocene effects on vertebrate phylogeography. Proceedings of the Royal Society of London B, 265, 1707–1712. http://dx.doi.org/10.1098/rspb.1998.0492 Benson, D.A., Karsch-Mizrachi, I, Clark, K., Lipman, D.J., Ostell, J. & Sayers, E.W. (2012) GenBank. Nucleic Acids Research, 40 (Database issue), D48–53. http://dx.doi.org/10.1093/nar/gkr1202 Bioacoustics Research Program. (2011) Raven Pro: Interactive Sound Analysis Software (Version 1.4). Ithaca, NY. The Cornell Lab of Ornithology. Available from: http://www.birds.cornell.edu/raven (Accessed 21 Aug. 2013) Bossuyt, F. & Milinkovitch, M.C. (2000) Convergent adaptive radiations in Madagascan and Asian ranid frogs reveal covariation between larval and adult traits. Proceedings of the National Academy of Sciences, USA, 97, 6585–6590. http://dx.doi.org/10.1073/pnas.97.12.6585 Caccone, A., Milinkovitch, M.C., Sbordoni, V. & Powell, J.R. (1997) Mitochondrial DNA rates and biogeography in European newts (genus Euproctus). Systematic Biology, 46, 126–144. http://dx.doi.org/10.2307/2413640 Channing, A., Brun, C., Burger, M., Febvre S. & Moyer, D. (2005) A new cryptic dainty frog from East Africa (Anura: Ranidae) Cacosternum). African Journal of Herpetology, 54, 139–148. http://dx.doi.org/10.1080/21564574.2005.9635527 Channing, A. & Schmitz, A. (2009) Hiding in plain sight: another cryptic dainty frog from the highlands of Kenya (Anura: Pyxicephalidae: Cacosternum). African Journal of Herpetology, 57, 75–84. http://dx.doi.org/10.1080/21564574.2008.9635570 Clement, M., Posada, D. & Crandall, K.A. (2000) TCS: a computer program to estimate gene genealogies. Molecular Ecology, 9, 1657–1660. http://dx.doi.org/10.1046/j.1365-294x.2000.01020.x AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 543 Compton, J.S. (2011) Pleistocene sea-level fluctuations and human evolution on the southern coastal plain of South Africa. Quaternary Science Reviews, 30, 506–527. http://dx.doi.org/10.1016/j.quascirev.2010.12.012 DeMenocal, P.B. (2004) African climate change and faunal evolution during the Pliocene–Pleistocene. Earth and Planetary Science Letters, 220, 3–24. http://dx.doi.org/10.1016/s0012-821x(04)00003-2 De Villiers, A.L. (2004) Cacosternum capense Hewitt, 1925. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop, P.J. & Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland. SI/MAB Series #9. Smithsonian Institution, Washington, DC, pp 224–227. Dupont, L.M., Donner, B., Vidal, L., Pérez, E.M. & Wefer, G. (2005) Linking desert evolution and coastal upwelling: Pliocene climate change in Namibia. Geology, 33, 461–464. http://dx.doi.org/10.1130/g21401.1 Du Preez, L.H. (2013) Polystomatidae (Monogenea) of southern African Anura: Polystoma channingi n.sp. parasitic in two closely related Cacosternum species. African Zoology, 48, 64–71. Flot, J.-F. (2010) SeqPHASE: a web tool for interconverting PHASE input/output files and FASTA sequence alignments. Molecular Ecology Resources, 10, 162–166. http://dx.doi.org/10.1111/j.1755-0998.2009.02732.x Frost, D.R. (2013) Amphibian Species of the World: an Online Reference. Version 5.6. American Museum of Natural History, New York, USA. Electronic Database accessible at http://research.amnh.org/vz/herpetology/amphibia/ (Accessed 31 Jan. 2013) Frost, D.R., Grant, T., Faivovich, J., Bain, R.H., Haas, A., Haddad, C.F.B., de S, R.O., Channing, A., Wilkinson, M., Donnellan, S.C., Raxworthy, C.J., Campbell, J.A., Blotto, B.L., Moler, P.E., Drewes, R.C., Nussbaum, R.A., Lynch, J.D., Green, D.M. & Wheeler, W.C. (2006) The amphibian tree of life. Bulletin of the American Museum of Natural History, 297, 1–370. http://dx.doi.org/10.1206/0003-0090(2006)297[0001:tatol]2.0.co;2 Guindon, S. & Gascuel, O. (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology, 52, 696–704. http://dx.doi.org/10.1080/10635150390235520 Hall, T.A. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series, 41, 95–98. Hillis, D.M., Mable, B.K., Larson, A., Davis, S.K. & Zimmer, E.A. (1996) Nucleic acids IV: Sequencing and cloning, In: Hillis, D.M., Moritz, C. &. Mable, B.K (eds) Molecular Systematics 2nd ed. Sinauer Associates, Sunderland, Massachusetts, USA. Linder, H.P. (2005) Evolution of diversity: the Cape flora. Trends in Plant Science, 10, 536–541. http://dx.doi.org/10.1016/j.tplants.2005.09.006 Mertens, R. (1955) Die Amphibien und Reptilien südwestafrika. Aus den Ergebnissen einer im Jahre 1952 ausgeführten Reise. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft, Frankfurt am Main, 490, 1–171. Minter, L.R. (2004) Cacosternum poyntoni Lambiris, 1988. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop P.J. and Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland. SI/MAB Series #9. Smithsonian Institution, Washington, DC, pp 235–236. Moore, W.S. (1995) Inferring Phylogenies from mtDNA variation – mitochondrial-gene trees versus nuclear-gene trees. Evolution, 49, 718–726. http://dx.doi.org/10.2307/2410325 Palumbi, S.R., Martin, A., Romano, S., McMillan, W.O., Stice, L. & Grabowski, G. (1991) The simple fool's guide to PCR. Department of Zoology and Kewalo Marine Laboratory, Hawai, 47 pp. Pickersgill, M. (2007) Frog Search. Edition Chimaira, Frankfurt am Main, 574 pp. Posada, D. (2008) jModelTest: phylogenetic model averaging. Molecular Biology and Evolution, 25, 1253–1256. http://dx.doi.org/10.1093/molbev/msn083 Poynton, J., Largen, M. & Scott, E. (2004) Cacosternum boettgeri. In IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2. Available from: http://www.iucnredlist.org/ (Accessed 20 Nov. 2012) Poynton, J. & Scott, E. (2004) Cacosternum leleupi. In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2 Available from: http://www.iucnredlist.org/ (Accessed 10 Feb. 2013) Richardson, J.E., Weitz, F.M., Fay, M.F., Cronk, Q.C.B., Linder, H.P., Reeves, G. & Chase, M.W. (2001) Rapid and recent origin of species richness in the Cape flora of South Africa. Nature, 412, 181–183. http://dx.doi.org/10.1038/35084067 Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19, 1572–1574. http://dx.doi.org/10.1093/bioinformatics/btg180 Schmidt, K.P. & Inger, R.F. (1959) Exploration du Parc National de l'Upemba. In: Mission G. F. de Witte en collaboration avec W. Adam, A. Janssens, L. van Meel et R. Verheyen (1946-1949), Amphibians. Institut des Parcs Nationaux du Congo Belge, Bruxelles, 56, 1–264. 544 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. Scott, E. (2004a) Cacosternum boettgeri (Boulenger, 1882. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop P.J. and Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland. SI/MAB Series #9. Smithsonian Institution, Washington, DC, pp 222–224. Scott, E. (2004b) Cacosternum nanum nanum Boulenger, 1887. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop P.J. & Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland. SI/MAB Series #9. Smithsonian Institution, Washington, DC, pp 232–235. Scott, E. (2004c) Cacosternum karooicum Boycott, De Villiers & Scott, 2002. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop P.J. & Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland. SI/MAB Series #9. Smithsonian Institution, Washington, DC, pp 227–229. Scott, E. (2004d) Cacosternum namaquense Werner, 1910. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop P.J. & Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland, pp 230–231. Scott, E. (2004e) Cacosternum nanum parvum Poynton, 1963. In: Minter, L.R., Burger, M., Harrison, J.A., Braack, H.H., Bishop P.J. & Kloepfer, D. (eds.) Atlas and Red Data Book of the Frogs of South Africa, Lesotho and Swaziland, pp 233– 235. Scott, E. (2005) A phylogeny of ranid frogs (Anura: Ranoidea: Ranidae), based on a simultaneous analysis of morphological and molecular data. Cladistics, 21, 507–574. http://dx.doi.org/10.1111/j.1096-0031.2005.00079.x Scott, E. & Minter, L. (2004a) Cacosternum poyntoni. In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2. Available from: http://www.iucnredlist.org/ (Accessed 1 Apr. 2013) Scott E. & Minter, L. (2004b) Cacosternum parvum. In IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2. Available from: http://www.iucnredlist.org/ (Accessed 20 Nov. 2012) Sepulchre, P., Ramstein, G., Fluteau, F., Schuster, M., Tiercelin, J.-J. & Brunet, M. (2006) Tectonic uplift and eastern Africa aridification. Science, 313, 1419–1423. http://dx.doi.org/10.1126/science.1129158 Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H. & Flook, P. (1994) Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Annals of the Entomological Society of America, 87, 651–701. Stephens, M. & Donnelly, P. (2003) A comparison of Bayesian methods for haplotype reconstruction from population genotype data. American Journal of Human Genetics, 73, 1162–1169. http://dx.doi.org/10.1086/379378 Sukumaran, J. & Holder, M.T. (2010) DendroPy: a Python library for phylogenetic computing. Bioinformatics, 26, 1569–1571. http://dx.doi.org/10.1093/bioinformatics/btq228 Swofford, D.L. (2002) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. Sinauer Associates, Sunderland, Massachusetts. Templeton, A.R., Crandall, K.A. & Sing, C.F. (1992) A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics, 132, 619–633. Vrba, E. (1995) The fossil record of African antelopes (Mammalia, Bovidae) in relation to human evolution and paleoclimate, In: Vrba, E., Denton, G., Burckle, L. & Partridge, T. (Eds.), Paleoclimate and Evolution with Emphasis on Human Origins, Yale University Press, New Haven, CT, 1995, 385–424. Zwickl, D.J. (2006) Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Ph.D. dissertation, The University of Texas at Austin. Van der Meijden, A., Crottini, A., Tarrant, J., Turner, A. & Vences, M. (2011) Multi-locus phylogeny and evolution of reproductive modes in the Pyxicephalidae, an African endemic clade of frogs. African Journal of Herpetology, 60, 1–12. http://dx.doi.org/10.1080/21564574.2010.523904 Veith, M., Steinfartz, S., Zardoya, R., Seitz, A. & Meyer, A. (1998) A molecular phylogeny of ‘true’ salamanders (family Salamandridae) and the evolution of terrestriality of reproductive modes. Journal of Zoological Systematics and Evolutionary Research, 36, 7–16. http://dx.doi.org/10.1111/j.1439-0469.1998.tb00774.x AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 545 APPENDIX 1. Material examined, listing species, voucher, locality, coordinates and GenBank accession numbers. All specimens are from South Africa unless otherwise indicated. SPECIES SOURCE Latitude; longitude GenBank C. aggestum MHNG 2690.25 Klipheuwel 33° 41' 45" S; 18° 43' 18" E KF144411, KF144571 (tyr) C. aggestum MHNG 2690.26 Klipheuwel 33° 41' 45" S; 18° 43' 18" E KF144412, KF144541 (tyr) C. aggestum MHNG 2690.27 Klipheuwel 33° 41' 45" S; 18° 43' 18" E KF144413, KF144570 (tyr) C. aggestum MHNG 2699.40 Vissershok, Cape Town 33° 46' 11" S; 18° 31' 34" E KF144414, KF144569 (tyr) C. australis MHNG 2699.34 Robertson 33° 57' 15" S; 19° 56' 28" E KF144415 C. australis MHNG 2699.35 Robertson 33° 57' 15" S; 19° 56' 28" E KF144416, KF144580 (tyr) C. australis MHNG 2699.39 Grootvadersbosch 33° 59' 05" S; 20° 48' 39" E KF144417, KF144572 (tyr) C. australis MHNG 2699.41 Pearly Beach 34° 39' 30" S; 19° 29' 19" E KF144418, KF144568 (tyr) C. australis MHNG 2699.42 Pearly Beach 34° 39' 30" S; 19° 29' 19" E KF144419, KF144581 (tyr) C. australis MHNG 2740.19 Vrolijkheid Nature Reserve 33° 57' 15" S; 19° 56' 28" E KF144420, KF144540 (tyr) C. boettgeri MHNG 2709.036 Keetmanshoop, Namibia 26° 34' 59" S; 18° 07' 59" E KF144421 KF144527 (tyr) C. boettgeri MHNG 2740.21 Maclear, 5 km W 31° 02' 05" S; 28° 17' 43" E KF144422 C. boettgeri MHNG 2740.22 Matatiele road stop 1 30° 20' 56" S; 28° 49' 42" E KF144423, KF144534 (tyr) C. boettgeri MHNG 2740.23 Matatiele road stop 1 30° 20' 56" S; 28° 49' 42" E KF144424, KF144528 (tyr) C. boettgeri MHNG 2740.24 Matatiele road stop 1 30° 20' 56" S; 28° 49' 42" E KF144425 C. boettgeri MHNG 2740.25 Matatiele road stop 2 30° 21' 13" S; 28° 51' 35" E KF144426, KF144526 (tyr) C. boettgeri MHNG 2740.26 Matatiele road stop 2 30° 21' 13" S; 28° 51' 35" E KF144427, KF144529 (tyr) C. boettgeri MHNG 2740.27 Maselspoort road 28° 58' 17" S; 26° 21' 17" E KF144428 C. boettgeri MHNG 2740.28 Maselspoort road 28° 58' 17" S; 26° 21' 17" E KF144429 C. boettgeri MHNG 2740.29 Maselspoort road 28° 58' 17" S; 26° 21' 17" E KF144430, KF144539 (tyr) C. boettgeri MHNG 2740.30 Maselspoort road 28° 58' 17" S; 26° 21' 17" E KF144431, KF144530 (tyr) C. boettgeri MHNG 2740.31 Farm The Vale 32° 11' 56" S; 22° 50' 29" E KF144432 C. boettgeri MHNG 2740.32 Farm The Vale 32° 11' 56" S; 22° 50' 29" E KF144433, KF144560 (tyr) C. boettgeri MHNG 2740.33 Farm The Vale 32° 11' 56" S; 22° 50' 29" E KF144434 C. boettgeri MHNG 2709.39 Ladismith 33° 29' 24" S; 21° 16' 00" E KF144435 C. boettgeri MHNG 2709.40 Franklin 6.3 km SE 30° 21' 46" S; 29° 30' 26" E KF144436 C. boettgeri MHNG 2709.41 Franklin 6.3 km SE 30° 21' 46" S; 29° 30' 26" E KF144437 C. boettgeri Hardap Dam, Mariental, 24° 33' 00" S; 17° 55' 59" E AF215414 Namibia C. boettgeri Namibia DQ347299 C. boettgeri MHNG 2740.37 Kubusi Forest, Amatola Mts 32° 34' 00" S; 27° 18' 57" E KF144438 ......continued on the next page 546 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. APPENDIX 1. (Continued) SPECIES SOURCE Latitude; longitude GenBank C. boettgeri MHNG 2740.38 Kubusi Forest, Amatola Mts 32° 34' 00" S; 27° 18' 57" E KF144439 C. boettgeri MHNG 2740.39 Kubusi Forest, Amatola Mts 32° 34' 00" S; 27° 18' 57" E KF144440 C. boettgeri MHNG 2740.40 Baviaanskloof at Enkeldoorn 33° 39' 13" S; 24° 29' 57" E KF144441 C. boettgeri MHNG 2740.41 Lebombo, 11 km SE Big Bend 26° 54' 00" S; 32° 01' 00" E KF144442 C. boettgeri MHNG 2740.42 Lebombo, 11 km SE Big Bend 26° 54' 00" S; 32° 01' 00" E KF144443 C. boettgeri MHNG 2740.43 Lebombo, 11 km SE Big Bend 26° 54' 00" S; 32° 01' 00" E KF144444 C. boettgeri MHNG 2740.36 Acornhoek 24° 35' 55" S; 31° 05' 08" E KF144445 C. boettgeri MHNG 2740.44 Farm Lemoenfontein, 35 km SE 30° 48' 34" S; 23° 39' 25" E KF144446 Britstown C. boettgeri MHNG 2740.45 Farm Lemoenfontein, 35 km SE 30° 48' 34 S; 23° 39' 25 E KF144447 Britstown C. boettgeri MHNG 2740.46 Farm Lemoenfontein, 35 km SE 30° 48' 34 S; 23° 39' 25 E KF144448 Britstown C. boettgeri MHNG 2740.47 Farm Lemoenfontein, 35 km SE 30° 48' 34 S; 23° 39' 25 E Britstown C. boettgeri MHNG 2740.48 Farm Lemoenfontein, 35 km SE 30° 48' 34 S; 23° 39' 25 E KF144449 Britstown C. boettgeri MHNG 2740.49 Farm Lemoenfontein, 35 km SE 30° 48' 34 S; 23° 39' 25 E KF144450 Britstown C. boettgeri MHNG 2740.50 Farm Kareehoek, 30 km N 30° 09' 42" S; 23° 27' 37" E KF144452 Britstown C. boettgeri MHNG 2740.51 Farm Kareehoek, 30 km N 30° 09' 42" S; 23° 27' 37" E KF144453 Britstown C. boettgeri MHNG 2740.52 Farm Kareehoek, 30 km N 30° 09' 42" S; 23° 27' 37" E KF144454 Britstown C. boettgeri MHNG 2740.53 Farm Kareehoek, 30 km N 30° 09' 42" S; 23° 27' 37" E KF144455 Britstown C. boettgeri MHNG 2740.54 Farm Kareehoek, 30 km N 30° 09' 42" S; 23° 27' 37" E KF144456 Britstown C. boettgeri MHNG 2740.55 Kuruman 27° 27' 00" S; 23° 25' 59" E KF144457 C. boettgeri MHNG 2740.56 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144458 C. boettgeri MHNG 2740.57 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144459 C. boettgeri MHNG 2740.58 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144460 C. boettgeri MHNG 2740.59 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144461 C. boettgeri MHNG 2740.65 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144462 C. boettgeri MHNG 2740.60 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144463 C. boettgeri MHNG 2740.61 Bamboesberg 31° 25' 03" S; 26° 05' 00" E KF144464 C. boettgeri MHNG 2740.62 Jansenville 32° 55' 59" S; 24° 40' 00" E KF144465 C. boettgeri MHNG 2740.63 Jansenville 32° 55' 59" S; 24° 40' 00" E KF144466 C. boettgeri MHNG 2740.64 Jansenville 32° 55' 59" S; 24° 40' 00" E KF144467 C. capense MHNG 2690.23 Klipheuwel 33° 41' 45" S; 18° 43' 18" E KF144468, KF144548 (tyr) C. capense MHNG 2690.24 Klipheuwel 33° 41' 45" S; 18° 43' 18" E KF144469, KF144547 (tyr) C. capense TMSA 84242 Klipheuwel 33° 41' 45" S; 18° 43' 18" E DQ022354 C. karooicum MHNG 2740.67 Vrolijkheid Nature Reserve 33° 57' 15" S; 19° 56' 28" E KF144470, KF144574 (tyr) ......continued on the next page AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 547 APPENDIX 1. (Continued) SPECIES SOURCE Latitude; longitude GenBank C. karooicum MHNG 2740.68 Vrolijkheid Nature Reserve 33° 57' 15" S; 19° 56' 28" E KF144471, Tadpole KF144575 (tyr) C. kinangopensis NMK A/4372 Murungaru, North Kinangop, 00° 34' 23" S; 36° 29' 29" E EU978471 Kenya KF144583 (tyr) C. leleupi MHNG 2740.69 3 km NW Lusinga, Democratic 08° 54' 33" S; 27° 11' 35" E KF144472, Republic of Congo KF144531 (tyr) C. leleupi MHNG 2740.70 3 km NW Lusinga, Democratic 08° 54' 33" S; 27° 11' 35" E KF144473 Republic of Congo C. namaquense MHNG 2699.43 Arakoep 30° 07' 00" S; 17° 55' 00" E KF144474, KF144582 (tyr) C. namaquense MHNG 2699.44 Arakoep 30° 07' 00" S; 17° 55' 00" E KF144475, KF144576 (tyr) C. namaquense MHNG 2699.45 Arakoep 30° 07' 00" S; 17° 55' 00" E KF144476, KF144578 (tyr) C. namaquense South Africa HQ014419 C. nanogularum MHNG 2740.77 Nkandla 28 51' 04" S; 31 12' 27" E KF144477 C. nanogularum MHNG 2740.78 Nkandla 28 51' 04" S; 31 12' 27" E KF144478 C. nanogularum MHNG 2740.79 Manyiseni, Lebombo Mts 26° 51' 21" S; 32° 02' 34" E KF144479, KF144559 (tyr) C. nanogularum MHNG 2740.80 Manyiseni, Lebombo Mts 26° 51' 21" S; 32° 02' 34" E KF144480, KF144550 (tyr) C. nanogularum MHNG 2740.81 Manyiseni, Lebombo Mts 26° 51' 21" S; 32° 02' 34" E KF144481, KF144558 (tyr) C. nanogularum MHNG 2740.82 Manyiseni, Lebombo Mts 26° 51' 21" S; 32° 02' 34" E KF144482, KF144546 (tyr) C. nanogularum MHNG 2740.83 Manyiseni, Lebombo Mts 26° 51' 21" S; 32° 02' 34" E KF144483, KF144549 (tyr) C. nanum MHNG 2740.84 Maclear 31° 02' 57" S; 28° 17' 44" E KF144484 C. nanum MHNG 2740.85 Hogsback 32° 35' 36" S; 26° 55' 55" E KF144485 KF144555 (tyr) C. nanum MHNG 2740.86 Hogsback 32° 35' 36" S; 26° 55' 55" E KF144486, KF144551 (tyr) C. nanum MHNG 2740.87 Hogsback 32° 35' 36" S; 26° 55' 55" E KF144487, KF144544 (tyr) C. nanum MHNG 2740.88 Hogsback 32° 35' 36" S; 26° 55' 55" E KF144488 KF144552 (tyr) C. nanum MHNG 2740.89 Maclear 31° 02' 05" S; 28° 17' 14" E KF144489, KF144577 (tyr) C. nanum MHNG 2740.90 Maclear 31° 02' 05" S; 28° 17' 14" E KF144490, KF144542 (tyr) C. nanum MHNG 2740.91 Montague Pass, base 33° 55' 33" S; 22° 25' 03" E KF144491 KF144553 (tyr) C. nanum RCBS2045 Malalotja, Swaziland 26° 05' 53" S; 31° 05' 56" E AY838886 C. nanum TMSA83951 Swaziland DQ022353 C. nanum MHNG 2740.92 Kubusi 32 33' 40" S; 27 18' 56" E KF144492 C. nanum MHNG 2740.93 Kubusi 32 33' 40" S; 27 18' 56" E KF144493 C. nanum MHNG 2740.94 Kubusi 32 33' 40" S; 27 18' 56" E KF144494 C. nanum MHNG 2740.95 Baviaanskloof 33 37' 03" S; 24 25' 48" E KF144495 C. nanum MHNG 2740.96 Goukamma 34 03' 52" S; 22 56' 34" E KF144496 ......continued on the next page 548 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. APPENDIX 1. (Continued) SPECIES SOURCE Latitude; longitude GenBank C. nanum MHNG 2740.97 Baviaanskloof 33 37' 03" S; 24 25' 48" E KF144497 C. nanum MHNG 2740.98 Qudeni 28 30' 22" S; 30 50' 04" E KF144498 C. nanum MHNG 2740.99 Cedarville 25 km SW 30° 35' 08" S; 28° 53' 17" E KF144499, KF144543 (tyr) C. nanum MHNG 2740.100 Cedarville 25 km SW 30° 35' 08" S; 28° 53' 17" E KF144500 C. nanum MHNG 2741.1 High Waters Farm, 19 km SW 30° 31' 41" S; 28° 54' 12" E KF144501, Cedarville KF144556 (tyr) C. parvum MHNG 2741.2 Mariepskop 24° 32' 24" S; 30° 52' 11" E KF144502 C. parvum MHNG 2741.3 Mariepskop near top 24° 32' 24" S; 30° 52' 11" E KF144503, KF144557 (tyr) C. platys MHNG 2709.30 Noordhoek 35° 04' 50" S; 18° 23' 40" E KF144504, KF144573 (tyr) C. platys MHNG 2709.31 Noordhoek 35° 04' 50" S; 18° 23' 40" E KF144505, KF144536 (tyr) C. platys MHNG 2709.32 Noordhoek 35° 04' 50" S; 18° 23' 40" E KF144506, KF144562 (tyr) C. platys MHNG 2709.33 Noordhoek 35° 04' 50" S; 18° 23' 40" E KF144507, KF144564 (tyr) C. platys MHNG 2709.34 Noordhoek 35° 04' 50" S; 18° 23' 40" E KF144508, KF144579 (tyr) C. platys MHNG 2709.35 Noordhoek 35° 04' 50" S; 18° 23' 40" E KF144509, KF144565 (tyr) C. platys MHNG 2699.36 Kenilworth 33° 59' 52" S; 18° 29' 01" E KF144510, KF144566 (tyr) C. platys MHNG 2699.37 Kenilworth 33° 59' 52" S; 18° 29' 01" E KF144511, KF144567 (tyr) C. platys MHNG 2699.38 Kenilworth 33° 59' 52" S; 18° 29' 01" E KF144512, KF144563 (tyr) C. platys ES 745 Kenilworth 33° 59' 52" S; 18° 29' 01" E DQ022352 C. plimptoni Meserani, Tanzania 03° 24' 36" S; 36° 29' 45" E EU978472 C. rhythmum AACRG416 Drakensberg Gardens 29° 27' 04" S; 29° 15' 10" E KF144513 C. rhythmum AACRG417 Drakensberg Gardens 29° 27' 04" S; 29° 15' 10" E KF144514, KF144532 (tyr) C. rhythmum AACRG418 Drakensberg Gardens 29° 27' 04" S; 29° 15' 10" E KF144515, KF144545 (tyr) C. rhythmum MHNG 2741.13 Fort Nottingham 30 50' 04" S; 28 30' 22" E KF144516, KF144535 (tyr C. rhythmum MHNG 2741.14 Fort Nottingham 29 56' 19" S; 29 24' 27" E KF144517, KF144537 (tyr) C. rhythmum MHNG 2741.15 Qudeni 28 30' 22" S; 30 50' 04" E KF144518 C. rhythmum MHNG 2741.16 Qudeni 28 30' 22" S; 30 50' 04" E KF144519 C. rhythmum MHNG 2741.17 Qudeni 30 50' 04" S; 28 30' 22" E KF144520 C. rhythmum MHNG 2741.18 Glencoe, 12 km W 28° 08' 07" S; 30° 02' 45" E KF144521, KF144538 (tyr) C. rhythmum MHNG 2741.19 Glencoe, 12 km W 28° 08' 07" S; 30° 02' 45" E KF144522 C. striatum MHNG 2741.20 Mpur Forest 30 16' 52" S; 29 34' 55" E KF144523, KF144561 (tyr) C. striatum MHNG 2741.21 Mpur Forest 30 16' 52" S; 29 34' 55" E KF144524, KF144533 (tyr) C. striatum MHNG 2741.22 Mpur Forest 30 16' 50" S; 29 34' 56" E KF144525 AFRICAN DAINTY FROGS Zootaxa 3701 (5) © 2013 Magnolia Press · 549 550 · Zootaxa 3701 (5) © 2013 Magnolia Press CHANNING ET AL. 

Animal Conservation. Print ISSN 1367-9430 SUPPLEMENT ARTICLE Widespread occurrence of the amphibian chytrid fungus in Kenya J. Kielgast1, D. Rodder ¨ 2 , M. Veith2 & S. Lotters ¨ 2 1 Department of Biology, University of Copenhagen, Copenhagen, Denmark 2 Department of Biogeography, Trier University, Trier, Germany Abstract Keywords Amphibians at the global scale are dramatically declining and the pathogenic amphibian decline; Batrachochytrium fungus Batrachochytrium dendrobatidis (Bd) has been suggested to be an important dendrobatidis; chytridiomycosis; ‘out of Africa’; endemic; enzootic; Kenya; Africa. driver in this biodiversity crisis. Increasing evidence points towards the global emergence of Bd being a panzootic caused by pathogen pollution. Africa has been Correspondence suggested to be the origin of the pathogen but remains one of the least-studied J. Kielgast, Department of Biology, Section areas. We have conducted the most comprehensive survey on the continent to date for Evolution and Microbiology, University focusing on Kenya for investigating taxonomic and environmental components in of Copenhagen, Universitetsparken 15, the distribution of Bd in tropical Africa. Eleven sites along a 770 km transect from 2100 Copenhagen, Denmark. the coast up to the border of Uganda were surveyed. Using quantitative PCR, we Email: jkielgast@snm.ku.dk screened 861 samples from 23 different species in nine genera. The pathogen was confirmed at all studied sites, with an overall prevalence of 31.5%. No dead or Received 15 February 2009; accepted 15 symptomatic specimens were found and no declines have been reported in the June 2009 region so far. Both prevalence and parasite load ranged from the detection limit to some of the highest ever reported. The parasite load showed a significant doi:10.1111/j.1469-1795.2009.00297.x taxonomic bias and a strong inverse correlation with temperature. Our findings suggest that Bd may be enzootic in the region. We recommend that further research should focus on comparative experimental studies of susceptibility to Bd in African species. Moreover, we stress the need for improved knowledge on the conservation status of the tropical African amphibian fauna to confirm the enzootic nature of widespread Bd infections. spread (James et al., 2009). This is supported by empirical Introduction evidence of a wave-like emergence of disease in the Neo- Amphibians are the most drastically declining vertebrates tropics (Lips et al., 2008). on our planet, with nearly one-third of the c. 6400 known The large-scale dissemination of Bd has been suggested to species being threatened with extinction under the IUCN be human mediated and data on the international trade of Red List (Stuart et al., 2008). This biodiversity crisis is amphibian species underline the potential for pathogen developing globally at a rate only paralleled by mass extinc- pollution (e.g. Daszak et al., 2006; Fisher & Garner, 2007; tion events in geological time (Mendelson III et al., 2006; Walker et al., 2008). The point of origin is therefore a central Wake & Vredenburg, 2008). Epizootics of the disease question towards an understanding of the epidemiology of chytridiomycosis, caused by the parasitic fungus Batracho- this disease. It has been advocated, based on evidence from chytrium dendrobatidis (Bd), have been proposed to play an clawed frogs (genus Xenopus), that an ‘out of Africa’ important role as the proximate cause of the rapid decline of scenario is plausible (Weldon et al., 2004). This was founded more than 200 amphibian species (Skerratt et al., 2007). This on the oldest known record of Bd, a temporally constant Bd is supported by well-documented aetiology in focal studies prevalence in museum collections and disease resistance. of population decline in North America (e.g. Rachowicz Moreover, a dissemination pathway exists via worldwide et al., 2006), Central America (e.g. Lips et al., 2006), Europe trade in these frogs beginning as early as the 1930s – first (e.g. Bosch, Mart ´ınez-Solano & Garc ´ıa-Par ´ıs, 2001) and used for a pregnancy assay and subsequently as a model Australia (e.g. Schloegel et al., 2006). Bd shows a remark- animal in biological and medical teaching and research ably low host specificity and has now been detected in more (Weldon et al., 2004; Weldon, De Villiers & Du Preez, 2007). than 400 different anuran and salamander species on all The ‘out of Africa’ hypothesis has been frequently re- continents on which amphibians occur (Fisher, 2008). How- ferred to in the literature. However, research on Bd has so ever, molecular studies strongly indicate that Bd’s current far suffered from a geographic bias, whereby Africa has geographic distribution is the result of a recent panzootic largely been left unstudied. Owing to the lack of apparent Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 1 Amphibian chytrid fungus in Kenya J. Kielgast et al. barriers to Bd within the continent, we hypothesize that an underlying prevalence of 10% (following Cannon & Roe, ‘out of Africa’ scenario would render widespread occurrence 1982). Only post-metamorphic and adult anurans were of Bd in suitable habitats. Furthermore, if the host–patho- sampled (as caecilians were not found and salamanders are gen system has co-evolved in the African herpetofauna, absent). Recommended disinfection and containment pro- some intrinsic factors characterizing its interactions are cedures were followed to avoid transmission and dissemina- likely to differ from the epizootic range, that is parameters tion of the pathogen (see Speare et al., 2004). Specimens such as virulence and host specificity. Here we examine the were caught by hand or, in a few cases, by dip-nets and host–pathogen system in tropical Africa using landscape- placed individually into plastic bags. Transmission of Bd diverse Kenya as a representative. For the first time, we between individuals was eliminated by collecting the speci- investigate multiple localities and host taxa with quantita- mens wearing a latex glove or a plastic bag on the hand. tive methods and present the most comprehensive survey of Diagnostic sampling was carried out using a diagnostic fine- Bd conducted in Africa so far. tip dry swab (Medical Wire & Equipment, MW-100) by comprehensively swabbing each specimen’s dorsum, ven- trum, both lateral sides, the dorsal and ventral surface of Materials and methods hind limbs, toes and toe webbing. A fresh pair of latex gloves From 16 September to 22 October 2006, fieldwork was was used for every specimen to avoid contamination of carried out at 11 localities along a 770 km transect from the samples. Voucher specimens to validate the taxonomic Kenyan coast up to the border of Uganda covering semi- identification (following Channing & Howell, 2006) of all humid and humid habitats at an altitudinal range of sampled species (Appendix S2) were deposited at National 159–3100 m a.s.l. (Fig. 1, Appendix S1). The time of sam- Museums of Kenya (NMK), Nairobi and Zoologisches pling targeted the beginning of the ‘short rains’ of the Forschungsmuseum Alexander Koenig (ZFMK), Bonn. bimodal precipitation pattern. No information on the sea- Swabs were stored as cool as possible in the field and sonality of Bd infection dynamics in Africa is available but thereafter at 20 1C until processing. sampling took place corresponding well with the recommen- For Bd analysis, 861 samples from 23 different species in dations given for sampling in Australia, for example, focus- nine genera (Appendix S2) were analysed by quantitative ing on the cold season (Kriger & Hero, 2006; Skerratt et al., PCR, following the protocol of Boyle et al. (2004), but 2008). Approximately 1500 specimens were examined in situ running each sample in two replicates. Samples were identi- during the survey. A minimum of 30 individual diagnostic fied as positive for Bd if a clear log-linear amplification was samples for Bd per species and locality was targeted to observed for both replicates and genomic equivalents (GE) enable 95% probability of detecting a positive assuming an quantified according to standards yielded above 0.5. If Figure 1 Study sites in Kenya sampled for Batrachochytrium dendrobatidis: 1, Saiwa Swamp National Park; 2, Mt. Elgon National Park; 3, Kakamega Forest National Reserve; 4, Thompson Falls; 5, Aberdares National Park (moorlands); 6, Aberdares National Park (Salient); 7, Tigoni Dam; 8, Nairobi (Karens); 9, Taita Hills (Mwundanyi); 10, Taita Hills (Mwatate); 11, Shimba Hills National Parc (see Appendix S1). 2 Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London J. Kielgast et al. Amphibian chytrid fungus in Kenya between-replicate standard deviation on the GE was higher w2 = 67.57, Po0.001). Hence, we can conclude that our than either of the quantified replicates, or only one replicate results strongly indicate a taxonomical component to the amplified above the detection threshold, the sample was re- abundance and intensity of infection at the time of sampling. run. If this also yielded equivocal results, the sample was There was no significant difference in the median parasite identified as negative. The mean GE of positive samples was load across all genera containing more than a single infected regarded as an index for parasite load. individual (exact two-sided, P= 0.077). Regression analysis Fisher’s exact test was used for comparing Bd prevalence on climate variables indicates a negative correlation between in the sampled genera. The parasite loads were compared by temperature and prevalence during the sampling period. a rank-based generalized linear model (using SAS 9.1 This pattern strongly co-varies with altitudinal distribution Statistical package). Detected GE loads within a locality and shows an increase in the frequency of Bd infection with were percentile ranked by partitioning them into 100 groups altitude. At the same time, the general precipitation pattern in which the smallest received a value of 0 and the largest was not found to correlate with prevalence (Fig. 2). value received a value of 99. Thus, the influence of sample size and locality-specific parasite abundance was removed, enabling comparison of interspecific parasite load across Discussion localities in a generalized linear model. The GE loads Using landscape-diverse Kenya as a representative, we have present in the investigated genera were further compared examined the Bd host–pathogen system in tropical Africa. by the distribution-free mood’s median test using Quantita- Bd was detected at all investigated sites, indicating that it is tive Parasitology 3.1 (http://www.zoologia.hu/qp/qp.html). ubiquitous in the sampled habitat types and may be wide- In order to assess the relationships between climate spread in tropical Africa. We furthermore found that the conditions during the surveys and Bd prevalence among prevalence and intensity of Bd infection correlated with the sites, we extracted information on current climate (mini- taxonomic and climatic patterns in the sampled region. mum and maximum temperature and precipitation) and Our results suggest that there is a taxonomic component topography from the Worldclim database, version 1.4 to Bd susceptibility in the anuran communities studied. (http://www.worldclim.org). This is a climate model based Susceptibility appeared to be highest in the genus Amietia on the weather conditions recorded between 1950 and 2000 (Ranidae), which consistently showed a high prevalence and with a grid cell resolution of 30 arc sec (Hijmans et al., 2005) parasite load. The genera Ptychadena (Ptychadenidae) and and was created by interpolation using a thin-plate smooth- Hyperolius (Hyperoliidae) exhibited intermediate infection ing spline of observed climate at weather stations with levels and members of the genus Xenopus (Pipidae) were latitude, longitude and elevation as independent variables seldom infected. Interestingly, the median parasite load of (Hutchinson, 1995; 2004). The relationships between local- infected individuals was not taxonomically biased, indicat- ity-level prevalence and environmental factors were assessed ing that the extreme ends of the intensity distribution with simple linear regressions calculated with Xlstat 2009 constitute the difference. (http://www.addinsoft.com). During our survey, despite extensive collection efforts, no dead or moribund anurans were encountered, nor were any clinical symptoms of chytridiomycosis observed. Even the Results most heavily infected individuals carrying a parasite load of All 11 sampled localities were Bd positive, with a prevalence over one million GE appeared to be in a good body from 4 to 71% and an overall mean of 31.5% (Table 1, condition. The high prevalence of Bd and the large number Appendix S2). Parasite load ranged from the detection of sub-clinically infected individuals may suggest that the threshold up to more than one million GE. The highest pathogen is enzootic in and possibly native to the studied parasite load coincided with the highest genus-specific pre- region. However, considering the cryptic disease progres- valence at a stream habitat in the moorlands of Aberdares sion characteristic for chytridiomycosis, it is not possible to National Park (Fig. 1), where only a single species, Amietia exclude that it may be causing mortality in the studied wittei (Ranidae), was sampled. The distribution of Bd populations undetected. Sub-clinical infections can build parasite load was heavily aggregated in a small subset of up to a threshold, with symptoms only occurring in the samples with extremely high GE loads as illustrated by the terminal phase of disease (see e.g. Carey et al., 2006; Voyles variance/mean ratio (Table 1). However, a high parasite et al., 2007), and field observations equivalent to ours have load was generally unusual throughout the survey, as even been made just before mass die-offs (Woodhams et al., indicated by the median mirroring central tendency for GE 2007). Furthermore, an enzootic state of Bd may be a load in the infected fraction of samples (Table 1). Again, the consequence of populations being in a post-decline phase sampled A. wittei from Aberdares National Park displayed and that declines were merely not observed as they hap- unusually high values with a median in the parasitized pened. There are now numerous examples of enzootic Bd fraction of 3966 GE. Evaluating potential taxonomic bias infection in amphibian populations globally (Retallick, in the distribution of Bd, we found a significant difference in McCallum & Speare, 2004; McDonald et al., 2005; Long- the proportion of infected individuals across all sampled core et al., 2007; Brem & Lips, 2008; Woodhams et al., 2008; genera (exact two-sided, Po0.001). The parasite load simi- Padgett-Flohr & Hopkins, 2009), and recent compelling larly showed a significant effect of genus (n = 861, d.f. = 8, evidence suggests that Bd can have a marked impact even Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 3 Amphibian chytrid fungus in Kenya J. Kielgast et al. Table 1 Summary statistics for the distribution of Batrachochytrium dendrobatidis (Bd) across localities (see Fig. 1, Appendix S1) and genera surveyed Specimens Variance/mean sampled Bd positive Prevalence (CI)a Max GE Mean GE Median GE ratio Locality Aberdares National Park (moorlands) 31.00 22.00 0.71 (0.53–0.84) 1 003 737.00 95 146.00 3966.00 739 406.00 Aberdares National Park (Salient) 111.00 43.00 0.39 (0.30–0.48) 120 350.00 7144.00 48.00 73 771.00 Kakamega Forest National Park 72.00 12.00 0.17 (0.10–0.27) 52 232.00 6059.00 96.00 40 414.00 Mt. Elgon National Park 11.00 5.00 0.46 (0.20–0.74) 14 610.00 2938.00 6.00 14 512.00 Nairobi (Karens) 109.00 27.00 0.25 (0.17–0.34) 48 945.00 2543.00 12.00 39 526.00 Saiwa Swamp National Park 126.00 62.00 0.49 (0.40–0.58) 227 779.00 5145.00 24.00 176 607.00 Shimba Hills National Park 150.00 6.00 0.04 (0.02–0.09) 9.00 6.00 6.00 7.00 Taita Hills (Mwatate) 21.00 4.00 0.19 (0.07–0.40) 980.00 296.00 101.00 828.00 Taita Hills (Mwundanyi) 116.00 34.00 0.29 (0.21–0.38) 10 395.00 587.00 7.00 6185.00 Thompson Falls 71.00 47.00 0.66 (0.54–0.76) 20 336.00 1124.00 65.00 9780.00 Tigoni Dam 50.00 9.00 0.18 (0.09–0.31) 565.00 83.00 12.00 443.00 Genus Afrixalus (Hyperoliidae) 18.00 2.00 0.11 (0.02–0.33) 4.00 4.00 4.00 3.00 Amietia (Ranidae) 81.00 49.00 0.61 (0.49–0.71) 1 003 737.00 49 327.00 133.00 680 415.00 Amietophrynus (Bufonidae) 13.00 3.00 0.23 (0.07–0.52) 5.00 5.00 5.00 4.00 Hyperolius (Hyperoliidae) 418.00 180.00 0.43 (0.38–0.48) 120 350.00 2579.00 30.00 56 624.00 Kassina (Hyperoliidae) 30.00 1.00 0.03 (0.00–0.18) 2.00 2.00 2.00 2.00 Leptopelis (Arthroleptidae) 20.00 0.00 0.00 (0.00–0.17) 0.00 0.00 0.00 0.00 Phrynobatrachus (Phrynobatrachidae) 27.00 1.00 0.04 (0.00–0.18) 200.00 200.00 200.00 200.00 Ptychadena (Ptychadenidae) 126.00 31.00 0.25 (0.18–0.33) 48 945.00 2216.00 15.00 37 812.00 Xenopus (Pipidae) 128.00 4.00 0.03 (0.01–0.08) 18.00 7.00 4.00 14.00 a 95% confidence intervals on Bd prevalence in parentheses were constructed using Sterne’s exact method (Reiczigel, 2003). CI, confidence intervals; GE, genomic equivalents. 1 1 (a) (b) 0.8 0.8 0.6 0.6 Prevalence Prevalence 0.4 0.4 0.2 0.2 0 0 0 5 10 15 20 15 20 25 −0.2 −0.2 Minimum temperature (°C) Minimum temperature (°C) 1 Figure 2 Simple linear regressions illustrating 1 (c) (d) the relationships between the mean Bd preva- 0.8 0.8 lence at the studied sites and (a) minimum temperature (r2 = 0.825, Po0.001), (b) maxi- 0.6 0.6 mum temperature (r2 = 0.666, P= 0.002), (c) Prevalence Prevalence 0.4 0.4 precipitation (r2 = 0.002, P= 0.904) and (d) alti- tude (r2 = 0.709, Po0.001); 95% confidence 0.2 0.2 intervals are indicated as vertical bars. The grey lines represent the 95% confidence limits for 0 0 0 50 100 150 200 0 1000 2000 3000 the mean of the prediction of a given value −0.2 −0.2 (dotted line) and the 95% confidence limits on a Precipitation (mm) Altitude (m) single prediction for a given value (solid line). decades after becoming enzootic (Murray et al., 2009). Existing accounts of Bd in African amphibians are Hence, the apparently enzootic state of Bd found in the equivocal regarding pathogenicity and so far little data are present study should be regarded as a conservation concern available. In tropical Africa, Bd has been reported to be and requires further investigation. present in Nigeria (Imasuen et al., 2009), The Democratic 4 Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London J. Kielgast et al. Amphibian chytrid fungus in Kenya Republic of Congo (Greenbaum et al., 2008), Uganda cipitation patterns did not correlate with Bd prevalence (Fig. (Goldberg, Readel & Lee, 2007) and Tanzania (Weldon 2). The species sampled were a priori the most abundant; et al., 2004), while a single study reports not detecting Bd in hence, it is understandable that species compositions varied Cameroon (Doherty-Bone et al., 2008). In Tanzania, Bd was among sampling sites. The correlation detected between suggested to be linked to the decline and disappearance of temperature and prevalence may therefore either reflect the Kihansi spray toad Nectophrynoides asperginis, sup- varying degrees of climatic suitability of the study sites for ported by finding two out of four dead individuals collected Bd (e.g. coincident with the results shown by Piotrowski during the decline to be Bd positive (Lee et al., 2006). The et al., 2004; Kriger et al., 2007; Rodder¨ et al., 2008) or species was categorized under ‘rapid enigmatic declines’ in differences in the susceptibility of the sampled amphibian the IUCN Global Amphibian Assessment 2004 (Stuart communities or both. et al., 2004), which have been suggested to cover over Bd- The question remains open as to whether the Bd panzoo- driven extinction processes (Skerratt et al., 2007). However, tic has its origin in Africa. If this is the case, the observed the Kihansi spray toad was endemic to a single waterfall patterns of high prevalence and apparently low virulence system (c. 2 ha.) that was radically changed by the construc- found in the present study can be explained in a co-evolu- tion of a power plant before declines (Krajick, 2006). The tionary context, that is evolution of host resistance, aviru- aetiology of this decline is therefore highly confounded by lence or attenuation of the local strain of Bd (McCallum, habitat change and the role of Bd is unclear. All other 2005). There is accumulating evidence that Bd exists in studies from tropical Africa have not been coupled with various strains exhibiting markedly different levels of viru- observations of mortality or decline. Goldberg et al. (2007) lence (Berger et al., 2005; Retallick & Miera, 2007; Fisher conducted a single site survey with methodology similar to et al., 2009) although recent evidence points towards no ours and report congruent findings with 22% prevalence in correspondence between the genetic lineage of strain and 109 specimens and no mortality events noted in the region. virulence (James et al., 2009). Comparative clinical studies The remaining reports are accounts of small-scale sampling investigating host–pathogen dynamics by experimental ex- and single Bd-positive individuals. posure of African and non-African amphibian hosts and Bd In the South African region, Bd has been found to be isolates may be a way forward in testing the ‘out-of-Africa’ widespread but no associated population declines have been hypothesis. Here, the conditional nature of pathogenicity observed (Hopkins & Channing, 2003; Weldon, 2002; Wel- should be evaluated, by including a range of climatic don et al., 2004; Smith et al., 2007). However, Bd-associated conditions, to determine the relative importance of the mortality in frogs has been detected twice without presum- intrinsic taxonomic components of susceptibility and cli- ing an aetiological link. The findings of the diagnostic matic factors. Moreover, the ‘out-of-Africa’ hypothesis characteristics of chytridiomycosis (Lane, Weldon & Bing- needs to be addressed by analysing the phylogenetic patterns ham, 2003), and patterns of mortality and infection resem- and global genetic diversity in Bd including a broader bling a die-off (Hopkins & Channing, 2003), call for further coverage of Bd isolates from Africa than are currently investigations, but do not contradict an endemic presence of available (James et al., 2009). Bd in the region. It should be underlined that disease surveys as such have Information on the susceptibility of African amphibians a limited capability of detecting declines and proving patho- to Bd under controlled conditions is limited to pipids and logical causality. This requires clinical experimental studies indicates susceptibility in the western African species Silur- of host–pathogen dynamics (including fulfilment of Koch’s ana tropicalis but resistance in Xenopus laevis (Reed et al., postulates) and long-term disease surveillance coupled with 2000; Parker et al., 2002; Fisher & Garner, 2007). This has studies of host population dynamics. There is an urgent need not yet been verified by experimental infection challenges to improve our knowledge of the conservation status of but is solely based on observations of mortality and aclinical tropical Africa’s amphibian fauna. A crucial step towards infection in captive populations. this is a qualified assessment of the risk imposed by wide- It has repeatedly been shown that the pathogenicity of spread Bd infections found in the present study. As a chytridiomycosis is context dependent and particularly priority, we urge the initiation of population monitoring affected by environmental parameters, for example tem- focusing on susceptible amphibian communities in the high- perature and precipitation (Bosch et al., 2001; Daszak, lands of tropical Africa. Cunningham & Hyatt, 2003; Berger et al., 2004; Kriger & Hero, 2006; Kriger, Peregolou & Hero, 2007; Kriger, 2009; Longo, Burrowes & Joglar, in press). Furthermore, the environmental suitability for a species at a given site may Acknowledgements affect the capacity of the host’s immune system affecting the We are grateful to Trent Garner and Andrew Cunningham prevalence and intensity of Bd infections (Raffel et al., 2006; for making available the facilities and expertise of the Fisher, 2007). Our results indicate that prevalence decreases Institute of Zoology, London. This extraordinary helpful- with higher monthly minimum and maximum temperatures ness was completely essential in realizing the present study. (Fig. 2), thus supporting previous findings (i.e. Berger et al., We thank The National Museums of Kenya and staff for 2004; Piotrowski, Annis & Longcore, 2004; Kriger & Hero, fruitful cooperation. Susanne Schick, Felix Muller, ¨ George 2006; Rodder, ¨ Veith & Lotters, ¨ 2008). In our study, pre- Kennedy and Beryl Bwong were of invaluable assistance Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 5 Amphibian chytrid fungus in Kenya J. Kielgast et al. during fieldwork. The first author would like to express Daszak, P., Cunningham, A.A. & Hyatt, A.D. (2003). Infec- deepest gratitude to Mads Frost Bertelsen and Copenhagen tious disease and amphibian population declines. Divers. Zoo for decisively contributing to the study and Mogens Distrib. 9, 141–150. Andersen and Peter Gravlund for practical support. Finally, Daszak, P., Schloegel, L.M., Maranda, L., Cronin, A., we would like to extend many thanks to two anonymous Pokras, M., Smith, K. & Picco, A. (2006). The global trade reviewers whose helpful comments greatly improved the in amphibians: summary interim report of a ccm study. paper. The project was financed through the Copenhagen Available at http://www.conservationmedicine.org Zoo/Center for Zoo and Wild Animal Health, The Danish (accessed 11 January 2009). WWF and Aase og Ejnar Danielsens Fond, BIOLOG- Doherty-Bone, T.M., Bielby, J., Gonwouo, N.L., LeBreton, BIOTA from the Federal Ministry of Education and Re- M. & Cunningham, A.A. (2008). In a vulnerable position? search (BMB+F, Germany), H.R. Frederiksen og Grete Siim Frederiksen Fond, Fonden Kjebi. D.R. is grateful to Preliminary survey work fails to detect the amphibian the ‘Graduiertenforderung ¨ des Landes Nordrhein-Westfa- chytrid pathogen in the highlands of Cameroon, an am- len’ for financial support. The Kenya Wildlife Service, phibian hotspot. Herp. J. 18, 115–118. KWS, kindly granted permission for this work (No. KWS/ Fisher, M.C. (2007). Potential interactions between amphi- RP/5001). bian immunity, infectious disease and climate change. Conflicts of interest: None. Anim. Conserv. 10, 420–421. Fisher, M.C. (2008). Molecular toolkit unlocks life cycle of the panzootic amphibian pathogen Batrachochytrium den- drobatidis. Proc. Natl. Acad. Sci. USA 105, 17209–17210. References Fisher, M.C., Bosch, J., Yin, Z., Stead, D.A., Walker, J., Selway, L., Brown, A.J.P., Walker, L.A., Gow, N.A.R., Berger, L., Marantelli, G., Skerratt, L.L. & Speare, R. (2005). Stajich, J.E. & Garner, T.W.J. (2009). Proteomic and Virulence of the amphibian chytrid fungus Batrachochy- phenotypic profiling of the amphibian pathogen Batracho- trium dendrobatidis varies with the strain. Dis. Aquat. Org. chytrium dendrobatidis shows that genotype is linked to 68, 47–50. virulence. Mol. Ecol. 18, 415–429. Berger, L., Speare, R., Hines, H., Marantelli, G., Hyatt, A.D., Fisher, M.C. & Garner, T.W.J. (2007). The relationship McDonald, K.R., Skerratt, L.F., Olsen, V., Clarke, J.M., between the emergence of Batrachochytrium dendrobatidis, Gillespie, G., Mahony, M., Sheppard, N., Williams, C. & the international trade in amphibians and introduced Tyler, M. (2004). Effect of season and temperature on amphibian species. Fungal Biol. Rev. 21, 2–9. mortality in amphibians due to chytridiomycosis. Aust. Goldberg, T.L., Readel, A.M. & Lee, M.H. (2007). Chytrid Vet. J. 82, 31–36. fungus in frogs from an Equatorial African montane forest Bosch, J., Mart ´ınez-Solano, I. & Garc ´ıa-Par ´ıs, M. (2001). in western Uganda. J. Wildl. Dis. 43, 521–524. Evidence of a chytrid fungus infection involved in the Greenbaum, E., Chifundera, K., Mwenebatu, M.A. & Reed, decline of the common midwife toad (Alytes obstetricans) K. (2008). Amphibian chytrid fungus infections in Hyper- in protected areas of central Spain. Biol. Conserv. 95, olius (Anura: Hyeproliidae) from Eastern Democratic Re- 331–337. public of Congo. Herp. Rev. 39, 70–73. Boyle, D.G., Boyle, D.B., Olsen, V., Morgan, J.A.T. & Hyatt, Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis, A.D. (2004). Rapid quantitative detection of chytridiomy- A. (2005). Very high resolution interpolated climate surfaces cosis (Batrachochytrium dendrobatidis) in amphibian sam- for global land areas. Int. J. Climatol. 25, 1965–1978. ples using real-time taqman PCR assay. Dis. Aquat. Org. Hopkins, S. & Channing, A. (2003). Chytrid fungus in 60, 141–148. Northern and Western Cape frog populations, South Brem, F.M.R. & Lips, K.R. (2008). Batrachochytrium den- Africa. Herp. Rev. 34, 334–336. drobatidis infection patterns among Panamanian amphi- Hutchinson, M.F. (1995). Interpolating mean rainfall using bian species, habitats and elevations during epizootic and thin plate smoothing splines. IJGIS 9, 385–403. enzootic stages. Dis. Aquat. Org. 81, 189–202. Hutchinson, M.F. (2004). Anusplin version 4.3. Centre for Cannon, R.M. & Roe, R.T. (1982). Livestock disease surveys: resource and environment studies. Canberra: The Australian a field manual for veterinarians. Canberra: Australian National University. Government Publishing Service. Imasuen, A.A., Weldon, C., Aisien, M.S.O. & Dupreez, L.H. Carey, C., Bruzgul, J.E., Livo, L.J., Walling, M.L., Kuehl, (2009). Amphibian chytridiomycosis: first report in Nigeria K.A., Dixon, B.F., Pessier, A.P., Alford, R.A. & Rogers, from the skin slough of Chiromantis rufescens. Froglog 90, K.B. (2006). Experimental exposures of boreal toads (Bufo 6–8. boreas) to a pathogenic chytrid fungus (Batrachochytrium James, T.Y., Litvintseva, A.P., Vilgalys, R.J., Morgan, A., dendrobatidis). EcoHealth 3, 5–21. Taylor, T.J.W., Fisher, M.C., Berger, L., Weldon, C., Channing, A. & Howell, K.M. (2006). Amphibians of East DuPreez, L. & Longcore, J.E. 2009. Rapid global expan- Africa. Ithaca: Cornell University Press. sion of the fungal disease chytridiomycosis into declining 6 Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London J. Kielgast et al. Amphibian chytrid fungus in Kenya and healthy amphibian populations. PLoS Pathog 5, Wake, D.B. & Brodie, E.D.J. (2006). Confronting amphi- e1000458. bian declines and extinctions. Science 313, 48. Krajick, K. (2006). Conservation biology – the lost world of Murray, K.A., Skerratt, L.F., Speare, R. & McCallum, H. the Kihansi toad. Science 311, 1230–1232. 2009. Impact and dynamics of disease in species threatened Kriger, K.M. (2009). Lack of evidence for the drought-linked by the amphibian chytrid fungus, Batrachochytrium den- chytridiomycosis hypothesis. J Wildl. Dis. 45, 537–541. drobatidis. Conserv. Biol. (Online DOI: 10.1111/j.1523- Kriger, K.M. & Hero, J.M. (2006). Large-scale seasonal 1739.2009.01211.x). variation in the prevalence and severity of chytridiomyco- Padgett-Flohr, G.E. & Hopkins, R.L. (2009). Batrachochy- sis. J. Zool. (Lond.) 271, 352–359. trium dendrobatidis, a novel pathogen approaching ende- Kriger, K.M., Peregolou, F. & Hero, J.-M. (2007). Latitudi- mism in central California. Dis. Aquat. Org. 83, 1–9. nal variation in the prevalence and intensity of chytrid Parker, J.M., Mikaelian, I., Hahn, N. & Diggs, H.E. (2002). (Batrachochytrium dendrobatidis) infection in eastern Aus- Clinical diagnosis and treatment of epidermal chytridio- tralia. Conserv. Biol. 21, 1280–1290. mycosis in African clawed frogs (Xenopus tropicalis). Lane, E.P., Weldon, C. & Bingham, J. (2003). Histological Comp. Med. 52, 265–268. evidence of chytridiomycosis in a free-ranging amphibian Piotrowski, J.S., Annis, S.L. & Longcore, J.E. (2004). Phy- (Afrana fuscigula (Anura: Ranidae)) in South Africa. J. S. siology of Batrachochytrium dendrobatidis, a chytrid A. Vet. Assoc. 74, 20–21. pathogen of amphibians. Mycologia 96, 9–15. Lee, S., Zippel, K., Ramos, L. & Searle, J. (2006). Captive- Rachowicz, L.J., Knapp, R.A., Morgan, J.A.T., Stice, M.J., breeding programme for the Kihansi spray toad, Necto- Vredenburg, V.T., Parker, J.M. & Briggs, C.J. (2006). phrynoides asperginis, at the Wildlife Conservation Society, Emerging infectious disease as a proximate cause of am- Bronx, New York. Int. Zoo Yb. 4, 24–25. phibian mass mortality. Ecology 87, 1671–1683. Lips, K.R., Brem, F., Brenes, R., Reeve, J.D., Alford, R.A., Raffel, T.R., Rohr, J.R., Kiesecker, J.M. & Hudson, P.J. Voyles, J., Carey, C., Livo, L., Pessier, A.P. & Collins, J.P. (2006). Negative effects of changing temperature on am- (2006). Emerging infectious disease and the loss of biodi- phibian immunity under field conditions. Funct. Ecol. 20, versity in a neotropical amphibian community. Proc. Natl. 819–828. Acad. Sci. USA 102, 3165–3170. Reed, K.D., Ruth, G.R., Meyer, J.A. & Shukla, S.K. (2000). Lips, K.R., Diffendorfer, J., Mendelson, J.R. III & Sears, Chlamydia pneumoniae infection in a breeding colony of M.W. (2008). Riding the wave: reconciling the roles of African Clawed frogs (Xenopus tropicalis). Emerg. Infect. disease and climate change in amphibian declines. PLoS Dis. 6, 196–199. Biol. 6, 441–454. Reiczigel, J. (2003). Confidence intervals for the binomial Longcore, J.R., Longcore, J.E., Pessier, A.P. & Halteman, parameter: some new considerations. Stat. Med. 22, W.A. (2007). Chytridiomycosis widespread in anurans of 611–621. northeastern United States. J. Wildl. Mgmt. 71, 435–444. Retallick, R.W.R., McCallum, H. & Speare, R. (2004). Longo, A.V., Burrowes, P.A. & Joglar, R.L. (in press) Endemic infection of the amphibian chytrid fungus in a Seasonality of Batrachochytrium dendrobatidis infection in frog community post-decline. Plos Biol. 2, 20351. direct-developing frogs suggests a mechanism for persis- Retallick, R.W.R. & Miera, V. (2007). Strain differences in tence. Dis. Aquat. Org. (Online DOI: 10.3354/dao02054). the amphibian chytrid Batrachochytrium dendrobatidis and McCallum, H. (2005). Inconclusiveness of chytridiomycosis non-permanent, sub-lethal effects of infection. Dis. Aquat. as the agent in widespread frog declines. Conserv. Biol. 19, Org. 75, 201–207. 1421–1430. Rodder, ¨ D., Veith, M. & Lotters, ¨ S. (2008). Environmental ´ McDonald, K.R., Mendez, D., Muller, ¨ R., Freeman, A.B. & gradients explaining the prevalence and intensity of infec- Speare R, . (2005). Decline in the prevalence of chytridio- tion with the amphibian chytrid fungus: the host’s per- mycosis in upland frog populations in North Queensland, spective. Anim. Conserv. 11, 513–517. Australia. Pac. Conserv. Biol. 11, 114–120. Schloegel, L.M., Hero, J.-M., Berger, L., Speare, R., McDo- Mendelson, J.R. III, Lips, K.R., Gagliardo, R.W., Rabb, nald, K. & Daszak, P. (2006). The decline of the sharp- ´ G.B., Collins, J.P., Differdorfer, J.E., Daszack, P., Ibanez, snouted day frog (Taudactylus acutirostris): the first docu- R., Zippel, K.C., Lawson, D.P., Wright, K.M., Stuart, mented case of extinction by infection in a free-ranging S.N., Gascon, C., Silva, H.R., Burrowes, P.A., Joglar, wildlife species? EcoHealth 3, 35–40. R.L., La Marca, E., Lotters, ¨ S., Preez, L.H., Weldon, C., Skerratt, L.F., Berger, L., Hines, H.B., McDonald, K.R., Hyatt, A., Rodriguez-Mahecha, J.V., Hunt, S., Robertson, Mendez, D. & Speare, R. (2008). Surveyprotocol for H., Lock, B., Raxworthy, C.J., Frost, D.R., Lacy, R.C., detecting chytridiomycosis in all Australian frog popula- Alford, R.A., Campbell, J.A., Parra-Olea, G., Bolanos, F., tions. Dis. Aquat. Org. 80, 85–94. Domingo, J.J.C., Halliday, T., Murphy, J.B., Wake, M.H., Skerratt, L.F., Berger, L., Speare, R., Cashins, S., McDonald, Coloma, L.A., Kuzmin, S.L., Price, M.S., Howell, K.M., K.R., Phillott, A.D., Hines, H.B. & Kenyon, N. (2007). Lau, M., Pethiyagoda, R., Boone, M., Lannoo, M.J., Spread of chytridiomycosis has caused the rapid global Blaustein, A.R., Dobson, A., Griffiths, A., Crump, M.L., decline and extinction of frogs. EcoHealth 4, 125–134. Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 7 Amphibian chytrid fungus in Kenya J. Kielgast et al. Smith, K.G., Weldon, C., Conradie, W. & du Preez, L.H. Africa, with implications for biodiversity conservation. (2007). Relationships among size, development, and Ba- Afr. J. Herpetol. 56, 77–83. trachochytrium dendrobatidis infection in African tadpoles. Weldon, C., du Preez, L.H., Hyatt, A.D., Muller, R. & Dis. Aquat. Org. 74, 159–164. Speare, R. (2004). Origin of the amphibian chytrid fungus. Speare, R., Berger, L., Skerratt, L., Alford, R., Mendez, D., Emerg. Infect. Dis. 10, 2100–2105. Cashins, S., Kenyon, N., Hauselberger, K. & Rowley, J. Woodhams, D.C., Vredenburg, V.T., Simon, M.A., Billhei- (2004). Hygiene protocol for handling amphibians in field mer, D., Shakhtour, B., Shyr, Y., Briggs, C.J., Rollins- studies. Available at http://www.jcu.edu.au/school/phtm/ Smith, L.A. & Harris, R.N. (2007). Symbotic bacteria PHTM/frogs/field-hygiene.pdf (accessed 2 January 2009). contribute to innate immune defences of the threatened Stuart, S.N., Chanson, J.S., Cox, N.A., Young, B.E., Rodri- mountain yellow-legged frog, Rana muscosa. Biol. Conserv. gues, A.S.L., Fischman, D.L. & Waller, R.W. (2004). 138, 390–398. Status and trends of amphibian declines and extinctions worldwide. Science 306, 1783–1786. Stuart, S.N., Hoffmann, M., Chanson, J.S., Cox, N.A., Berridge, R.J., Ramani, P. & Young, B.E. (2008). Threa- tened amphibians of the world. Barcelona, Spain: Lynx Supporting Information Edicions, in association with IUCN, Conservation Inter- Additional Supporting Information may be found in the national and NatureServe. online version of this article: Voyles, J., Berger, L., Young, S., Speare, R., Webb, R., Warner, J., Rudd, D., Campbell, R. & Skerratt, L.F. Appendix S1. Localities sampled including altitude in m (2007). Electrolyte depletion and osmotic imbalance in a.s.l. (see Fig. 1). amphibians with chytridiomycosis. Dis. Aquat. Org. 77, Appendix S2. Bd distribution within the sampled am- 113–118. phibian communities (species identification after Channing Wake, D.B. & Vredenburg, V.T. (2008). Are we in the midst & Howell, 2006). Exact Clopper-Pearson 95% Confidence of the sixth mass extinction? A view from the world of intervals are presented for the prevalence. Parasite load amphibians. Proc. Natl. Acad. Sci. USA 105, 11466–11473. summary statistics are based on the infected fraction of the Walker, S.F., Bosch, J., James, T.Y., Litvintseva, A.P., Valls, sample. J.A.O., Pina, S., Garcia, G., Rosa, G.A., Cunningham, A.A., Hole, S., Griffiths, R. & Fisher, M.C. (2008). As a service to our authors and readers, this journal Invasive pathogens threaten species recovery programs. provides supporting information supplied by the authors. Curr. Biol. 18, R853–R854. Such materials are peer-reviewed and may be re-organized Weldon, C. (2002). Chytridiomycosis survey in south africa. for online delivery, but are not copy-edited or typeset. Froglog 51, 1–2. Technical support issues arising from supporting informa- Weldon, C., De Villiers, A.L. & Du Preez, L.H. (2007). tion (other than missing files) should be addressed to the Quantification of the trade in Xenopus laevis from South authors. 8 Animal Conservation 13, Suppl.1 (2010) 1–8 c 2009 The Authors. Journal compilation c 2009 The Zoological Society of London 

Interacting Symbionts and Immunity in the Amphibian Skin Mucosome Predict Disease Risk and Probiotic Effectiveness Douglas C. Woodhams1,2*¤a, Hannelore Brandt1, Simone Baumgartner1¤b, Jos Kielgast3¤c, Eliane Ku¨pfer1,4, Ursina Tobler1,5, Leyla R. Davis1, Benedikt R. Schmidt1,5, Christian Bel1, Sandro Hodel1, Rob Knight6, Valerie McKenzie2 1 Institute of Evolutionary Biology and Environmental Studies, University of Zurich, Zurich, Switzerland, 2 Department of Ecology and Evolutionary Biology, University of Colorado, Boulder, Colorado, United States of America, 3 Section for Freshwater Biology, Department of Biology, University of Copenhagen, Copenhagen, Denmark, 4 Department of Evolutionary Biology, Technical University of Braunschweig, Braunschweig, Germany, 5 KARCH, Neuchaˆtel, Switzerland, 6 Howard Hughes Medical Institute and Department of Chemistry and Biochemistry, BioFrontiers Institute, University of Colorado, Boulder, Colorado, United States of America Abstract Pathogenesis is strongly dependent on microbial context, but development of probiotic therapies has neglected the impact of ecological interactions. Dynamics among microbial communities, host immune responses, and environmental conditions may alter the effect of probiotics in human and veterinary medicine, agriculture and aquaculture, and the proposed treatment of emerging wildlife and zoonotic diseases such as those occurring on amphibians or vectored by mosquitoes. Here we use a holistic measure of amphibian mucosal defenses to test the effects of probiotic treatments and to assess disease risk under different ecological contexts. We developed a non-invasive assay for antifungal function of the skin mucosal ecosystem (mucosome function) integrating host immune factors and the microbial community as an alternative to pathogen exposure experiments. From approximately 8500 amphibians sampled across Europe, we compared field infection prevalence with mucosome function against the emerging fungal pathogen Batrachochytrium dendrobatidis. Four species were tested with laboratory exposure experiments, and a highly susceptible species, Alytes obstetricans, was treated with a variety of temperature and microbial conditions to test the effects of probiotic therapies and environmental conditions on mucosome function. We found that antifungal function of the amphibian skin mucosome predicts the prevalence of infection with the fungal pathogen in natural populations, and is linked to survival in laboratory exposure experiments. When altered by probiotic therapy, the mucosome increased antifungal capacity, while previous exposure to the pathogen was suppressive. In culture, antifungal properties of probiotics depended strongly on immunological and environmental context including temperature, competition, and pathogen presence. Functional changes in microbiota with shifts in temperature provide an alternative mechanistic explanation for patterns of disease susceptibility related to climate beyond direct impact on host or pathogen. This nonlethal management tool can be used to optimize and quickly assess the relative benefits of probiotic therapies under different climatic, microbial, or host conditions. Citation: Woodhams DC, Brandt H, Baumgartner S, Kielgast J, Ku¨pfer E, et al. (2014) Interacting Symbionts and Immunity in the Amphibian Skin Mucosome Predict Disease Risk and Probiotic Effectiveness. PLoS ONE 9(4): e96375. doi:10.1371/journal.pone.0096375 Editor: Carlos A. Navas, University of Sao Paulo, Brazil Received January 9, 2014; Accepted April 4, 2014; Published April 30, 2014 Copyright: ß 2014 Woodhams et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Financial support came from the Zoological Institute and the Forschungskredit of the University of Zurich, Vontobel Stiftung, Janggen-Po¨hn Stiftung, Basler Stiftung fu¨r biologische Forschung, Stiftung Dr. Joachim De Giacomi, Zoo Zu¨rich, Gru¨n Stadt Zu¨rich, European Union of Aquarium Curators, Schweizer Tierschutz, Zu¨rcher Tierschutz, Claraz Foundation, the environment departments of the cantons St. Gallen and Zurich, Swiss National Science Foundation (31- 125099 to DCW), and U.S. National Science Foundation Population and Community Ecology Section (DEB 1146284 to VJM and RK). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors hereby confirm that co-author B. Schmidt is a PLOS ONE Editorial Board member. This does not alter their adherence to PLOS ONE Editorial policies and criteria. * E-mail: dwoodhams@gmail.com ¤a Current address: Department of Biology, University of Massachusetts Boston, Boston, Massachusetts, United States of America ¤b Current address: Department of Aquatic Ecology, Duebendorf, Switzerland, and Institute of Integrative Biology, ETH-Zu¨rich, Zurich, Switzerland ¤c Current address: Section for Freshwater Biology, Department of Biology, University of Copenhagen, Copenhagen, Denmark and Center for Macroecology, Evolution and Climate Natural History Museum of Denmark, Copenhagen, Denmark Introduction range of mechanisms [3,4], and disease ecology studies demon- strate that parasitic and non-parasitic microbes interact with each Probiotic therapies often aim to extend or shape the immune other and with the host immune system such that pathogenicity is function of hosts by altering the symbiotic microbial community. often influenced by environmental conditions [5–8]. Thus, the Probiotics are used in human and veterinary medicine, agriculture environment affects the risk of disease to individuals, populations, and aquaculture, and have been proposed for treatment of and species, and assessing disease risk under changing conditions is emerging wildlife diseases such as those occurring on corals and vital to conservation and infectious disease mitigation and can amphibians [1,2]. Microbiota can mediate pathogenesis through a direct the allocation of resources for most effect [9–12]. PLOS ONE | www.plosone.org 1 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk The microbiota inhabiting skin and mucosal surfaces has a risk and treatment effects in rare amphibians including relict profound impact on host health and immunity [7,13,14], and may populations or captive populations of endangered species intended be predictive of risk for some diseases [15–17]. Amphibian skin is a for reintroduction. model system for diseases affecting vertebrate mucosa. The Typical approaches to compare species susceptibility and to mucosome, or micro-ecosystem of the mucus, as defined here assess disease risk include pathogen exposure experiments [27], or contains interdependent host factors (mucosal antibodies, antimi- field surveys to compare infection prevalence and monitor disease crobial peptides, lysozyme, alkaloids) and microbial-community and population trajectories [28], or modeling environmental and factors (microbiota, antibiotic metabolites). The mucosome has biogeographic risk factors [10,29]. Deficits of conventional various functions potentially including communication, and pathogen exposure experiments include lack of environmental predator and pathogen defense. Here, we develop a non-lethal context when amphibians are exposed under clean laboratory assay and holistic measure referred to as ‘‘mucosome function’’ to conditions. Biodiversity including microbiota and macrobiota can describe the effect of amphibian skin mucus on pathogen viability. influence disease outcome [30], and bacterial community diversity We examine how environmental and immunological contexts may is reduced through time in captivity without natural sources such impact the outcome of host-microbe symbioses, and how as soil for re-inoculating the skin [31]. The exposure history, mucosome function captures the in vivo complexity of the micro- population genetics, and life-history stage of the amphibians used ecosystem and can thus accurately predict susceptibility to in the experiment, as well as the strain and dose of the pathogen infection. We focus on probiotic bacteria and fungi applied to can all affect experimental outcomes, and many threatened species the skin mucosome as biocontrol agents against the emerging are not suitable for such experiments. In addition, growth of Bd is amphibian disease chytridiomycosis. often inhibited by skin microbiota of amphibians [32,33]. Chytridiomycosis is a major cause of global amphibian However, little is known about how protective microbiota differs population declines and species extinctions [18,19]. The disease among host populations or regions, or how mucosome function is is caused by the chytrid fungus Batrachochytrium dendrobatidis, or Bd, altered by enrichment with potential probiotics. and is strongly influenced by climatic conditions [20]. Climate- Our aims in this study were (1) to develop a holistic, simple, linked changes to the entire microbiota, not just Bd, may influence non-invasive, and non-lethal method to measure mucosome disease susceptibility [5]. Current efforts to mitigate chytridiomy- function against Bd. Using this tool, we aimed (2) to test whether cosis in wildlife populations have turned to bioaugmentation, or mucosome function can predict Bd infection prevalence of the use of probiotic therapies [1,21]. The successful prophylactic amphibians in the field and survival in Bd exposure experiments. use of Janthinobacterium lividum was demonstrated against chytridio- While we show that probiotics are influenced by a variety of mycosis in mountain yellow-legged frogs, Rana muscosa [22]. factors including competition, temperature, and innate immunity However, when tested on the endangered Panamanian golden when tested in vitro, we aimed (3) to use mucosome function as an frog, Atelopus zeteki, the probiotic survived briefly on the skin, but ecologically-integrated predictor of probiotic therapy effect so that did not protect the amphibians from disease [23]. Similarly, the future research can test probiotic strategies for conservation and probiotic Pedobacter cryoconitis temporarily reduced infection loads of not lose hope in the potential of probiotic therapy in the face of heavily infected R. muscosa [24]. Each target host may thus require immunological and ecological complexity. We provide a detailed probiotic therapy tailored to that species, population, or life- protocol for measuring mucosome function in File S1. history stage. Screening the various bacteria associated with hosts or their environment to identify effective probiotics is challenging Materials and Methods [25,26]. Thus, probiotic therapies for amphibians must be optimized, and an understanding of which candidate bacteria Ethics statement can establish and persist on the host in its natural environmental Permits to conduct fieldwork were obtained from the Swiss context is urgently needed. cantonal conservation authorities, and from Germany - German To date, all attempts to apply probiotic therapy against federal licence (Rheinland-Pfalz) no. 425-104.143.0904 Struktur- chytridiomycosis have used simple selection criteria for choosing und Genehmigungsdirektion Nord, Koblenz. All animal proce- candidate probiotics. Selection of the most efficient probiotic is dures were approved by the Veterinary Authority of Zurich (110/ challenging because there are hundreds of culturable phylotypes to 2007 and 227/2007) and the Federal Office for the Environment. choose from, either from environmental sources, or more typically, Fieldwork conformed to standard decontamination practices to from tolerant host populations that can persist with nonlethal Bd avoid transport of pathogens between sites. All animals in infections [1]. However, simple co-culture assays to determine experiments were monitored daily for animal welfare and to antifungal capacity have been insufficient to ensure probiotic ameliorate suffering. During experiments, any individual demon- effectiveness [23,24]. Co-factors including interactions of the strating clinical signs of disease including lethargy, abnormal skin probiotic with the microbial community already present on the shedding, and loss of righting reflex were humanely euthanized. At amphibian skin, as well as interactions with host immune defenses, the end of the experiment, all animals were humanely euthanized and effects of environmental conditions, may complicate the by overdose of tricaine methanesulfonate. outcome of biotherapy. Here, we experimentally test the impact of immunological and environmental context on potential probiotic Survey of Bd infection prevalence bacteria both in vitro and in vivo. The tested conditions are To compare Bd infection prevalence among species and life- illustrative rather than comprehensive for potential environmental history stages, we combine previously unpublished results from conditions, community and immunological interactions. Because it field studies in Switzerland with Bd surveys from amphibians is impractical to test all potential interactions before testing across Europe collated by Bd-Maps (www.bd-maps.net, accessed probiotics on amphibians for a disease resistance effect, we suggest September 1, 2013). In addition to data from 5939 sampled a protocol for selecting probiotics with the highest potential amphibians available from Bd-maps, skin swabs were collected benefit, and to test whether the probiotics will likely be effective in from 2591 amphibians from 12 species and from 66 Bd-positive the range of foreseeable conditions on the host. Our non-lethal populations from the northern parts of Switzerland and tested for susceptibility assay of mucosome function can help assess disease Bd between 2007 and 2009 (Table 1). Amphibians were caught by PLOS ONE | www.plosone.org 2 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk dip-netting and swabbed with a sterile cotton swab (Copan Italia common midwife toad, A. obstetricans, is a species of conservation S.p.A., Brescia, Italy). Field material was cleaned and disinfected concern [39] and is particularly sensitive to Bd early in life-history before moving between different sites to avoid contamination and [40]. Host-associated bacteria and fungi were surveyed by spread of Bd and other pathogens. Extraction and analysis for Bd- culturing from populations of midwife toads near Basel, Switzer- DNA were done following the qPCR protocol by Boyle et al. [34] land in May, 2009, including samples from 19 adults, 32 larvae, using Bd-specific primers and standards to quantify the amount of and 9 egg clutches. Although many diverse antifungal bacteria DNA. We ran each sample twice and the PCR was repeated if the have been described in association with skin of some amphibian two wells returned dissimilar results. Reactions below 1 genomic hosts [32,33], we chose eleven bacterial residents isolated from A. equivalent were scored Bd-negative to avoid false positives. Mean obstetricans for the environmental context experiments described infection prevalence with 95% binomial confidence interval was below based on potency against Bd in culture and high prevalence calculated for each species and life stage sampled, and calculated in the populations sampled (L. Davis, unpublished). Two bacterial for both Europe and Switzerland. isolates with high in vitro potency against Bd and the ability to withstand host skin defense peptides, and one fungal isolate, were Bd infection prevalence predicted by skin defenses chosen for applications on recently metamorphosed A. obstetricans. Skin defense peptides and mucosome samples were tested All metamorphs used in the study were raised in captivity from against Bd for comparison of anti-Bd activity with infection wild-caught tadpoles that were naturally exposed to the fungus in prevalence in natural populations by logistic regression in R. their pond of origin, near Zunzgen, Switzerland, but negative for Amphibians sampled for skin peptides and mucosome function Bd by qPCR at the time of the experiment. Toadlets were of (Table 1) were sampled in Switzerland and compared to field similar size (mean6SD: 2.160.3 g; ANOVA F5 = 1.179, infection prevalence from Switzerland and across Europe in P = 0.332) and treated at the same time with one exception. separate analyses. Skin peptides were collected upon induction by Toadlets in the Bd-exposure group were exposed to Bd approx- subcutaneous injection of metamorphosed amphibians with 40 imately 2 months prior to the microbial exposure treatments, and nmole/g body mass norepinephrine (bitartrate salt, Sigma) or the toadlets were smaller (1.560.3 g), and no longer infected at the immersion of larval amphibians in 100 mM norepinephrine, and time of sampling based on qPCR. tested for Bd growth inhibition as previously described [35,36]. We treated recently metamorphosed common midwife toads, Skin peptide samples from post-metamorphic amphibians only Alytes obstetricans (N = 70: 10 per treatment group, 7 treatments), by were used in the logistic regression analyses because different housing them individually at 5uC, 18uC, or 25uC with no microbes methods of peptide induction were used on larval stages. added, or at 18uC with exposure to Bd zoospores (8.56106 Mucosome samples from multiple life-history stages of the same zoospores of global panzootic lineage isolated from a Bufo bufo in species were included and matched to life-history stages sampled the UK [38]), a probiotic fungus Penicillium expansum, or a probiotic for Bd diagnostics (Table 1). Detailed methods for measuring bacterium P. fluorescens or F. johnsoniae. Toadlets were bathed mucosome function against Bd using a fluorescence assay of Bd individually for one hour in water containing the microbes and viability adapted from Stockwell et al. [37] (Fig. S1 in File S1) and after 2 weeks, toadlets from all treatments were sampled on the comparisons of mucosome function and skin peptide defenses same day for mucosome function and subsequently skin peptides, against Bd are presented in Supporting Information (Figs. S4, S5 in sampled as described above. File S1). Temperature, competition of probiotic strains, and Survival predicted by mucosome function co-culture with Bd To examine the relationship between mucosome function To determine the effects of competitive interactions and against Bd and susceptibility to infection and subsequent survival temperature on probiotic potential, 11 common host-associated we performed experimental exposures to Bd on four species. All isolates were chosen. These included two isolates of Serratia animals were exposed to zoospores from Swiss lineage Bd TG 739 plymuthica and one isolate of Janthinobacterium lividum from egg isolated from a moribund A. obstetricans in Gamlikon, Switzerland clutches of midwife toads, three isolates of Flavobacterium johnsoniae in 2007 [38] and cryopreserved until use. Egg clutches were and five species of Pseudomonas isolated from the skin of adults. obtained from P. esculentus (n = 8), B. variegata (n = 8), R. temporaria Based on 16S rRNA gene sequences, all 11 isolates were (n = 45), and A. obstetricans (n = 13) in northern Switzerland or considered unique operational taxonomic units (OTUs) at 99%, southern Germany. Rana temporaria were raised in outdoor but clustered into 7 OTUs at 97% similarity as determined by the mesocosms through metamorphosis before experimental exposure UCLUST algorithm in QIIME. The 16S rRNA gene sequences of of metamorphs to Bd (N = 92 exposed, 94 control). Other species all isolates were deposited in the European Nucleotide Archive were exposed to Bd as tadpoles (N = 80 exposed, 40 control per (Table S1 in File S1). species). All animals were kept in the same laboratory at 19uC In one set of experiments, bacterial isolates were freshly grown during the experiments. We measured the proportion of infected at 18uC on RIIA agar media supplemented with 1% tryptone then metamorphs by qPCR, and determined relative survival (survival transferred to experimental conditions. Bacteria and Bd (Swiss of infected/survival of uninfected controls) at the end of the isolate TG 739) readily grew on the same media. Plate experiments (50–90 d after metamorphosis). Kaplan Meier curves experiments were performed in duplicate. Both isolates of Serratia are presented in the Supporting Information for each species. We plymuthica were grown separately at 18 and 25uC, or at 18uC on examined the relationship between mucosome function against Bd media inoculated with Bd and allowed to dry before streaking the and relative survival and proportion infected using logistic bacteria. Two isolates of F. johnsoniae were grown separately or regression analyses in R. combined on media inoculated with Bd, and grown at 18uC. When combined, each isolate was streaked across the entire plate. Three Host ecological context and skin defenses Pseudomonas isolates were grown either separately, combined, or The in vivo effects of ambient temperature and skin microbiota combined on media inoculated with Bd, and grown at 18uC. on mucosome function against Bd and skin peptide defenses were Control plates of sterile media or Bd-only were also tested. All tested on a focal amphibian species, A. obstetricans. In Europe, the plates were incubated for 3 days, and then rinsed with 2 ml sterile PLOS ONE | www.plosone.org 3 April 2014 | Volume 9 | Issue 4 | e96375  Table 1. Amphibians from Switzerland sampled for skin peptide effectiveness and mucosome function against Bd, and Bd infection prevalence at different life-history stages. Mean mucosome function Switzerland: 95% binomial 95% binomial Peptide against Swiss Percent confidence Europe: Percent confidence Species Life-history stage# effectiveness* (N) SE Bd (N) SE infected (N) interval infected (N) interval Alytes obstetricans Adult/Subadult 15.92 (8) 6.21 0.012 (10) 0.000 4.9 (41) 0.6–16.5 29.7 (209)$ 23.5–36.4 PLOS ONE | www.plosone.org Alytes obstetricans Metamorph 37.75 (5) 12.15 Alytes obstetricans Larvae 48.76 (5) 24.23 2.963 (10) 0.681 45.4 (2111) 43.3–47.6 38.0 (3008) 36.3–39.8 Bombina variegata Adult/Subadult 1.075 (4) 0.081 20.0 (150) 13.9–27.3 21.1 (227) 16.0–27.0 Bufo bufo Adult 16.34 (15) 3.37753 0.117 (9) 0.082 0.0 (22) 0.0–15.4 0.9 (3606) 0.6–1.2 Bufo bufo Larvae 1.284 (5) 0.404 6.7 (45) 1.4–18.3 Hyla arborea Adult 11.42 (7) 2.15210 3.8 (26) 0.1–19.6 12.5 (32) 3.5–29.0 Ichthyosaura alpestris Adult 0.94 (7) .52546 1.361 (20) 0.062 24.8 (629) 21.5–28.4 21.5 (775) 18.7–24.6 Lissotriton vulgaris Adult 1.85 (4) 1.02506 27.3 (22) 10.7–50.2 17.0 (47) 7.7–30.8 Pelophylax lessonae/esculentus Adult 27.27 (10) 3.18 22.4 (170) 16.3–29.4 15.6 (275) 11.6–20.5 Pelophylax lessonae/esculentus Metamorph 5.34 (5) 1.88685 0.545 (10) 0.042 13.0 (69) 6.1–23.3 13.2 (76) 6.5–22.9 Rana temporaria Adult/Subadult 1.97 (13) .62111 0.251 (10) 0.128 0.0 (10) 0.0–30.9 3.1 (129) 0.9–7.8 Rana temporaria Larvae 0.220 (5) 0.120 0.0 (20) 0.0–16.8 0.0 (23) 0.0–14.8 4 Salamandra salamandra Adult 4.92 (9) 1.32654 11.1 (9) 0.3–48.3 Salamandra salamandra Larvae 42.78 (5) 13.35528 23.2 (69) 13.9–34.9 Skin peptide effectiveness is the percent inhibition of Bd zoospore growth caused by 50 mg/ml peptide multiplied by the quantity of peptides (mg) per g amphibian according to Woodhams et al. [11]. The mucosome function against Bd (Swiss isolate TG 739) is a measure of zoospore viability quantified by the ratio of green:red fluorescence as described above. Infection prevalence is the mean from all amphibians in each group from multiple sites and seasons. # Larval and post-metamorphic skin peptide samples extracted by different methods. *Peptide effectiveness = % inhibition of Bd growth at 50 mg/ml * mg peptides/g frog mass. $ Includes samples from chytridiomycosis outbreak sites in Spain (S. Walker, unpubl.), not included in logistic regression. doi:10.1371/journal.pone.0096375.t001 Mucosome Function Predicts Disease Risk April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk Mili-Q water. Rinse water was then filtered through a 0.22 mm within Switzerland (Fig. 1b,d). Prevalence of infection with Bd syringe filter. decreased with peptide efficiency (Fig. 1c,d, logistic regressions: Bacteria were also grown in liquid RIIA media for 4 d at 14, 19, Europe, P = 0.0015; Switzerland, P = 0.0079). While induced and 22uC, and metabolites filtered as above. Metabolites from peptide defenses stored in granular glands were measured here, liquid cultures were added to Bd zoospores (Global panzootic ambient peptides (not induced by norepinephrine) are a natural lineage VMV 813 from a bullfrog, Lithobates catesbeianus tadpole) to component of the mucosome [42,43]. Mucosome function was test for inhibitory effects on pathogen growth. To determine the tightly correlated to Bd prevalence in natural populations of Swiss effect of bacterial filtrate on Bd growth, Bd zoospores were amphibians (Fig. 1b, P,0.0001) and in amphibians across Europe harvested in 1% tryptone and counted under a hemocytometer. (Fig. 1a, P = 0.0020). The odds ratios of Bd colonization in Swiss Wells of a 96-well plate were inoculated with 50 ml zoospores at amphibians was 1.950 (Europe, 2.969) with each unit change in 86106 zoospores per ml. Then, 50 ml of filtrate (or filtrate diluted mucosome function, and 0.839 (Europe, 0.811) with each unit 1:10) from each of the experimental or control plates, or liquid decrease in skin peptide efficiency. Correlations of mucosome cultures, was added to the wells in replicates of four. In addition, 6 function and induced skin peptide efficiency are presented in positive control wells contained Bd and 50 ml sterile water or RIIA Figure S4 in File S1 and suggest that both host and microbial media, and 6 negative control wells contained heat-killed Bd and factors contribute to mucosome function against Bd. 50 ml sterile water or RIIA media. The change in optical density measured at 490 nm absorbance over 7 days growth at 19uC was Survival predicted by mucosome function recorded using a Victor3 multilabel plate reader (PerkinElmer). Pathogen exposure experiments were conducted on four host Standard statistical testing was carried out in IBM SPSS Statistics species with a Swiss isolate of Bd, and relative survival post- 22. Significant Bd growth inhibition (or enhancement) caused by metamorphosis of infected tadpoles differed among species (% bacterial filtrate was determined by t-test, and a repeatable result relative survival, mean6SD days survived): A. obstetricans (0%, (Table S2 in File S1). Percent inhibition depended on filtrate dose 24617.5 d), Bombina variegata (39.0%, 32623.9 d), and Pelophylax (see Results) and was not considered comparable among bacterial esculentus (30.4%, 12612.8 d; Fig. S2 in File S1). Relative survival isolates. of recently metamorphosed Rana temporaria exposed to Bd was 100% (Fig. S3 in File S1), and no colonization by Bd was detected Effects of host skin peptides and Bd metabolites on by qPCR (n = 92). Success of Bd colonization of tadpoles also probiotics in culture differed among species (Pearson x23 = 13.102, P = 0.004): A. To test for the response of bacterial growth upon culture with obstetricans (13.9% infected, n = 36), B. variegata (10.7%, n = 75), either Bd filtrate or host skin peptides, bacteria were grown in and P. esculentus (7.9%, n = 76). Mucosome function predicted RIIA liquid media on 96 well plates. Supernatant from a 2-week survival (logistic regression, P,0.0001; Fig. 2a) and infection with old culture of Bd (type isolate JEL 197) growing in 0.5% tyrptone Bd in these species (P = 0.0106; Fig. 2b). The odds of infection was filtered through a 0.22 mm syringe filter. An equal volume of increased by 1.751 with each unit change in mucosome function, Bd filtrate or sterile media was added to bacterial cultures. To test and the odds of survival decreased by 0.0454. effects of peptides, we added an equal volume of sterile water or natural mixtures of partially-purified skin peptides from A. Host ecological context and skin defenses obstetricans metamorphs at a final concentration of 100 mg/ml in Midwife toads, A. obstetricans, were treated with various water. Growth after 48 hr was measured as change in optical temperature and probiotic therapies and tested for mucosome density measured at 480 nm. Differences between experimental function. Host context significantly affected mucosome permis- and control bacterial growth were tested by t-tests using a siveness or lethality towards Bd (Fig. 3a; ANOVA, F6 = 41.606, P, Bonferroni correction for multiple comparisons. 0.001). Bd viability was similar following incubation with mucosome samples from toads at temperatures ranging from 5– Results 25uC. Mucosome samples from toads previously exposed to Bd were least effective at killing Bd zoospores, while those from toads Survey of Bd infection prevalence treated with probiotics Flavobacterium johnsoniae and Penicillium Surveys of approximately 8500 amphibians (http://www. expansum were most effective at killing zoospores (Fig. 3a). While bd-maps.net/; this study) at different life-history stages for Bd Pseudomonas in general, and the P. fluorescens isolate (76.5c) used in infection based on qPCR indicated high prevalence in larval this study were often effective at inhibiting Bd in co-culture and midwife toads, A. obstetricans (45% infected in Switzerland) and produced antifungal metabolites across a range of temperatures aquatic adult newts Ichthyosaura alpestris (26%), and Lissotriton vulgaris ideal for Bd growth (Fig. 4a, Table S2 in File S1), there was no (27%). Low infection prevalence (,5%) was detected in popula- significant benefit of this probiotic when applied on hosts in terms tions of adult A. obstetricans, Bufo bufo, Rana temporaria, and Hyla of increasing mucosome function and reducing Bd viability arborea (Table 1, Fig. 1). (Fig. 3a). Because one significant antimicrobial component of A. obste- Bd infection prevalence predicted by skin defenses tricans skin mucus is antimicrobial peptides (AMPs) [44], we We examined two non-lethal measures of susceptibility to collected peptide skin secretions, quantified them per surface area infection in pathogen-free Swiss amphibians acclimated to of the toads and measured their ability to inhibit Bd growth at a laboratory conditions. These included testing Bd growth or standardized concentration of 100 mg/ml. On average, toads viability upon exposure to natural mixtures of partially purified produced 0.25 mg peptide per cm2 surface area, and at 100 mg/ml skin defense peptides, and a holistic functional measure of the skin these peptides inhibited Bd growth by 48.7%. These values did not mucosal ecosystem (mucosome function) including ambient skin differ significantly among treatment groups, nor did a combined defenses: peptides, alkaloids, lysozymes, mucosal antibodies, measure of skin peptide effectiveness against Bd (% * mg/cm2, microbiota and microbial metabolites [41]. Both antifungal skin Fig. 3b; Kruskal-Wallis tests, P ’s.0.05). Thus, skin peptides stored peptides and mucosome function were correlated with infection in granular glands were not significantly affected by the 2-week prevalence in natural populations across Europe (Fig. 1a,c) and temperature and microbe treatments including previous exposure PLOS ONE | www.plosone.org 5 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk Figure 1. Infection prevalence (mean, 95% binomial CI) of amphibians sampled across Europe and within Switzerland predicted by mucosome function and skin defense peptide activity against Batrachochytrium dendrobatidis (Bd) zoospores. Mucosome function (mean, SE) indicates Bd viability after a 1 hr exposure to amphibian mucus (a,b) and units represent green:red fluorescence. Peptide efficiency (mean, SE) indicates quantity of natural mixtures of skin peptides induced from granular glands multiplied by activity of a standard concentration of peptides against Bd zoospore growth. Only post-metamorphic amphibians sampled upon subcutaneous injection with norepinephrine are plotted in (c) and (d). Amphibian skin mucosome function is a better predictor of infection prevalence than induced skin peptide efficiency (logistic regression, see text). Summary data for all species and life-history stages are presented in Table 1. doi:10.1371/journal.pone.0096375.g001 to Bd. There was not a significant correlation between peptide Temperature, competition of probiotic strains, and effectiveness and mucosome function against Bd (Fig. S5 in File S1; co-culture with Bd Pearson, x2 = 20.102, P = 0.827). Zoospore viability after expo- Environmental conditions affected the capacity of probiotic sure to mucosome samples was significantly higher in the Bd- bacteria to inhibit the fungal pathogen Bd (Table S2 in File S1). exposure treatment compared to other treatments (Fig. 3a). Two Serratia plymuthica isolates (isolates 27 and 28) were capable of However, skin peptides induced from hosts in the Bd-exposure inhibiting Bd growth when incubated at 18uC. Isolate 27 was treatment were effective at inhibiting Bd growth, and not inhibitory under all tested conditions: 18uC, 25uC, and 18uC co- significantly different than peptides from toads in other treatments cultured with Bd. Isolate 28 significantly enhanced Bd growth at (Fig. 3b). 25uC, and was neither enhancing nor inhibitory at 18uC when co- cultured with Bd (Fig. 4c, Table S2 in File S1). A dose-response of Bd growth inhibition was found such that filtrate diluted 1/10 was PLOS ONE | www.plosone.org 6 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk (.100%, Fig. 4b). Most cultures were more inhibitory of Bd at the lower temperatures, except for J. lividum, (isolate 77.5b) which was most inhibitory at 22uC (Fig. 4a,b). While all bacteria were unique based on 16S rRNA gene sequencing when clustered at 99% similarity, probiotic physiology and function against Bd did not always correspond to OTU clustering at 97% similarity (Table S1 in File S1). In other words, bacterial isolates considered to be the same ‘‘species’’ based on 16S rRNA could have different antifungal function. Here, only one of two Flavobacterium johnsoniae isolates inhibited Bd growth. When grown together, the filtrate remained inhibitory. However, when grown together and co-cultured with Bd, the filtrate was no longer inhibitory. Three Pseudomonas isolates were capable of inhibiting Bd growth, and were inhibitory when combined with or without co- culture with Bd. The above mentioned growth inhibition of Bd caused by bacterial filtrate was significantly different from control bacterial growth with water only added (independent t-tests, P’s, 0.05 and replicated result; all data shown in Table S2 in File S1). These conditions represent infected or uninfected hosts and are illustrative rather than comprehensive of all possible environmen- tal conditions and competitive interactions. Effects of host skin peptides and Bd metabolites on probiotics in culture Amphibian skin defense peptides may regulate the skin microbiota. We found that natural mixtures of skin peptides from A. obstetricans at a concentration of 100 mg/ml significantly inhibited growth of Pseudomonas migulae (73b1) and significantly enhanced growth of P. filiscindens (73c1), Flavobacterium johnsoniae (70d1), and Janthinobacterium lividum (76.5c; t-test, Bonferroni corrected P’s,0.05; Fig. S6 in File S1). We tested for a direct effect of Bd metabolites on bacterial growth, and found that filtrate from two-week old cultures of Bd in 0.5% tryptone significantly inhibited the growth of Serratia plymuthica (5/27b2, 5/28a3), F. johnsoniae (81a1, 70d1), and P. filiscindens (73c1), while significantly enhancing the growth of J. lividum (77.5b1; t-test, Bonferroni corrected P’s,0.05; Fig. S6 in File S1). Figure 2. Relative survival (95% binomial CI; a) and Proportion Discussion of infected frogs (95% binomial CI; b) predicted by Mucosome function. Post-metamorphosis survival was measured from four Swiss We found that a holistic measure of mucosome function against amphibian species after exposure to zoospores of a Swiss Bd isolate, TG Bd is predictive of infection risk in natural populations of 739. Survival curves for each species are presented in Supporting amphibians and survival in laboratory exposure experiments. Information (Figs. S2, S3 in File S1) and relative survival was calculated While induced antimicrobial peptides may explain some variation as the proportion of infected frogs surviving/proportion of unexposed control frogs surviving. Alytes obstetricans showed the highest infection in infection risk (Fig. 1b,d), mucosome function can be altered and mortality, and Rana temporaria the lowest, with Bombina variegata through probiotic therapy (Fig. 3a), and thus microbial commu- and Pelophylax esculentus intermediate. All frogs were raised in captivity nities play a major role in determining susceptibility to infection from egg clutches and had no history of natural exposure to Bd. with Bd. In particular, tadpoles of the endangered midwife toad, A. Mucosome function (mean, SE) indicates Bd viability after exposure to obstetricans may be most at risk of both infection and subsequent amphibian mucus and is a significant predictor of both survival disease-induced mortality upon metamorphosis (Fig. 2), even (binomial logistic regressions, P,0.0001) and infection prevalence (P = 0.0106). though adult toads are well protected by the mucosome and doi:10.1371/journal.pone.0096375.g002 perhaps resistant to colonization with Bd. Similarly, the common frog R. temporaria has strong mucosome activity against Bd, shows Bd colonization resistance, but has relatively poor skin defense significantly less inhibitory than undiluted filtrate (paired t-test, peptides. This suggests that this common species has protective t35 = 9.836, P,0.001), and filtrate from control plates with or microbial communities. Adaptive defenses are not suspected without Bd significantly enhanced Bd growth (Table S2 in File S1). because frogs were raised from eggs and had no history of Testing metabolites of the bacteria growing at 14, 19, and 22uC in exposure to Bd. liquid culture against the global panzootic lineage of Bd showed In this study, we provide several striking examples showing that similar results including a dose-response (Fig. 4a,b, paired t-test, probiotic capacity depends on immunological and environmental t31 = 210.607, P,0.001). In several cases, Bd growth was context. These examples lead to recommendations for choosing enhanced with addition of diluted bacterial metabolites in probiotics based on predictable host conditions. Temperature is comparison to positive control growth with RIIA media only known to influence amphibian host immune function [41] and PLOS ONE | www.plosone.org 7 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk Figure 3. Temperature and probiotic treatments of recently metamorphosed midwife toads, A. obstetricans, influence skin mucosome function (a) but not induced skin peptide defenses (b). (a) Mucosome function indicates B. dendrobatidis (Bd) viability after exposure to amphibian mucus quantified by green: red fluorescence. Significantly different subsets are indicated by letters above bars (Tukey post- hoc test). Bd zoospore viability was reduced after exposure to mucus from frogs treated with the bacterium F. johnsoniae and the fungus P. expansum, and zoospore viability was highest after exposure to mucus from toads previously exposed to Bd. (b) Skin peptide effectiveness against Bd did not differ significantly among treatments (ANOVA, F6 = 0.952, P = 0.466). doi:10.1371/journal.pone.0096375.g003 bacterial growth, metabolism, pigment and antibiotic production of susceptibility related to climate, which have previously been [45]. However, it was surprising that a shift from 18 to 25uC, a limited to empirical observation and pathogen-centered effects typical natural range for midwife toads, caused a common [46–49]. bacterial symbiont of the eggs and skin, Serratia plymuthica, to The microbial interactions we tested also altered antifungal change from inhibiting Bd to enhancing Bd growth (Fig. 4c). effects relative to what would be predicted from individual isolates. Testing metabolites of the bacteria growing at 14, 19, and 22uC in For example, co-culture of Flavobacterium johnsoniae with Bd caused liquid culture against the global panzootic lineage of Bd showed cultures of the bacterium that normally produce antifungal similar results (Fig. 4b). Functional changes in probiotic activity metabolites to switch off antifungal activity: when grown together with shifts in temperature have not previously been reported. Our with Bd, F. johnsoniae filtrate was benign, and indeed Bd filtrate results provide an alternative mechanistic explanation for patterns inhibited the growth of two out of three F. johnsoniae isolates (Fig. PLOS ONE | www.plosone.org 8 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk PLOS ONE | www.plosone.org 9 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk Figure 4. Environmental context determines antifungal capacity of probiotics. Tested temperatures (14, 19, 22uC) significantly affected the production of bacterial metabolites in liquid media that could inhibit B. dendrobatidis (Bd; GPL isolate VMV 813) zoospore growth in a dose- dependent fashion (a = full strength metabolites, b = 1:10 dilution). * indicates that Bd growth differed among metabolite temperature treatments (ANOVA, Bonferroni-corrected P’s,0.05). (c) Representative replicates are shown of two isolates of Serratia plymuthica isolated from egg clutches of common midwife toads, Alytes obstetricans, grown on solid media under different temperature conditions. Filtrate from isolate 27 always inhibited growth of Bd, but filtrate from isolate 28 inhibited Bd growth at 18uC, and enhanced Bd growth at 25uC. Filtrate from sterile media (R2A agar supplemented with 1% tryptone) caused enhanced growth of Bd. Note that colony color can be an indication of antifungal metabolites such as prodiginines from red Serratia spp. [45,67], but are produced only under certain growth conditions. doi:10.1371/journal.pone.0096375.g004 S6 in File S1). Co-evolution of Bd and amphibian hosts is a This organism shows antifungal characteristics including activity postulated driver of pathogenicity factors including compounds against Bd growth [33]. The ability of extracellular products of A. suppressing host immune defenses [43,50,51]. These factors may hydrophila to inhibit amphibian antimicrobial peptides indicates a extend to inhibiting certain antifungal symbionts or altering their co-evolutionary relationship between host and symbionts [59]. In function. addition, Pseudomonas mirabilis and Serratia liquefaciens were found to Myriad microbial and immune interactions occur once be resistant to antimicrobial peptides from several host frog species probiotics are added to living hosts. Thus, testing probiotics in [60]. Here we used probiotics that largely resisted low concentra- vivo is critical for testing the intended antifungal effect of probiotic tions of natural mixtures of host defense peptides (Fig. S6 in File therapy under realistic environmental conditions. We found that S1). Thus, to increase the likelihood of probiotic establishment, use previous exposure to Bd may have a negative effect on host of probiotics with a co-evolutionary relationship with the target immunity or the ability of the mucosome to kill zoospores (Fig. 3A). host may be advantageous. This result is consistent with a study on Australia green-eyed tree While easily cultured, the isolates tested here may not be frogs, Litoria serratia, showing inhibition of ambient skin peptides dominant community members based on culture-independent with Bd infection but no inhibition of inducible stored skin peptides analyses [31,61,62]. Therefore, future studies will benefit by [43]. Because stored skin defense peptides can have potent activity examining the effects of probiotic treatments on the natural against Bd, yet not be active on the skin, induced skin peptides may microbial communities on host amphibians using culture-inde- not accurately predict infection susceptibility. This mystery of how pendent techniques such as next-generation sequencing. While seemingly well-defended species can be affected by chytridiomy- community interactions are difficult to test in vitro and before cosis [52] deserves careful study on the conditions under which probiotics are applied to a host, our results affirm that testing host skin defense peptides are activated. Induced skin defense probiotics under certain foreseeable contexts may increase the peptides were previously used to predict disease susceptibility in pace of biotherapy development. Panama [11] and New Zealand [53]. In Panama, most species had Because potential probiotics that inhibit the growth of Bd only weak peptide defenses and declined after disease emergence while do so under certain conditions, we recommend the following only two species had strong peptide defenses against Bd compared screening criteria (Fig. 5): (1) Candidates for probiotic develop- to reference species of known disease resistance. Of these two ment should be chosen from among the culturable microbiota species, the one with the highest levels of skin peptide defenses locally present on tolerant hosts or populations that are able to persisted at the field site (Espadarana prosoblepon) [54], and the other persist with Bd [32,33]. (2) Candidates should have the capacity to species (Agalychnis lemur) disappeared, but a relict population has inhibit Bd growth when grown in isolation, in co-culture with Bd, been detected nearby (Julie Ray, pers. comm.). In New Zealand, and in an environmental context relevant to the amphibian life- all native species demonstrated high levels of skin peptide defenses cycle, and (3) the ability to resist immune defenses on host skin, and appear to resist chytridiomycosis [53], although populations establish within the microbiota, and contribute to antifungal are in decline [55]. defenses in vivo. Resistance to mucosal immune defenses may be We found that a bacterium F. johnsoniae and a fungal probiotic P. critical for establishment within the microbial community associ- expansum can increase the Bd killing function of the mucosome. The ated with the skin, and critical for long-term persistence. Some bacterium P. fluorescens did not show this effect. Because host AMPs symbionts appear to be assisted in surviving on the host by thriving did not appear to be affected by these treatments (Fig. 3B), the on skin mucosal products. Mucosal oligosaccharides, for example, observed effects are most likely caused by antifungal metabolites differ among hosts and life-history stages, and may be a selective produced by the microbes growing on the amphibian skin [56]. force in structuring the microbiota [63,64]. Amphibian skin Upregulation of host mucosal immunity excluding AMPs is an provides a useful model of host-microbiota interactions to better untested alternative mechanism, and potentially a beneficial host understand mechanisms of microbial community assembly and response to probiotics. A non-responsive immune system when maintenance within vertebrate mucosa. Indeed, these mechanisms given probiotics may be preferred from a conservation manage- underlie strategies to promote human health by manipulating ment standpoint in order for the probiotics to colonize the host, microbial communities - a long-term goal of the Human establish within the microbiota and persist. However, this in not Microbiome Project [7,65]. necessarily common and immune stimulation in response to While screening for candidate probiotics, some beneficial probiotics occurs in other systems [57,58]. organisms may be inadvertently discarded based on tests of An ideal probiotic would produce metabolites that inhibit Bd bacterial filtrate on Bd growth. Microbes producing antifungal growth as shown above, and also be uninhibited by host skin metabolites such as bacteriocins [66] or small molecule antibiotics defense peptides. A literature review demonstrates that skin [56,67] will be detected by this method. However, microbes may peptides can inhibit the growth of some bacteria, but not others, also compete directly for space or resources, and may exclude and suggests that skin defense peptides may be critical in pathogenic fungi by other mechanisms [26,68]. Furthermore, structuring the symbiont community on amphibian skin [52]. microbial secondary metabolites such as prodiginines produced by Rollins-Smith et al. [35] showed that Aeromonas hydrophila, a Serratia spp. can be immunosuppressive [67]. Probiotics may common resident on amphibian skin and also an opportunistic strongly influence host immunity through interactions with host pathogen, could tolerate high levels of host antimicrobial peptides. Toll-like receptors or NOD-like receptors, or through interactions PLOS ONE | www.plosone.org 10 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk infected individuals. The potential for negative biodiversity- function relationships, especially among mixtures of closely related bacteria, cautions against the use of probiotic mixtures that may cause interference competition and reduce host protection [71]. Further refinements to the probiotic screening and discovery process will incorporate next-generation sequencing analyses to target rare or as yet uncultured microbes of interest, and testing microbial consortia that appear linked to disease resistance function. Measuring the effectiveness of applied probiotics is a second step in managing disease risk. No previous studies have attempted to relate skin microbiota or a holistic measure of skin defense function against Bd with disease susceptibility. Given the extreme complexity of the skin micro- ecosystem and interactions described above, the holistic measure of mucosome function presents a significant advance in our capacity to predict relative disease susceptibility, and to measure the success of managed treatments without resorting to infection trials. Here, we examined overall prevalence of infection in Figure 5. Choosing probiotics with the greatest potential Switzerland and Europe and test for correlations at these broad against amphibian chytridiomycosis. Candidate probiotic bacteria scales with innate defenses from selected life-stages and species (or fungi) are isolated from populations of amphibians that are able to persist in the presence of B. dendrobatidis (Bd) [1]. To increase the (Fig. 1). We found a very strong correlation between mucosome chances of successful prophylactic biotherapy, candidate probiotics function against Bd and infection prevalence in the field and upon should be tested for at least three characteristics: (a) capacity to inhibit experimental exposure. Since Bd-naı¨ve amphibians were sampled Bd growth as a pure isolate without specific competitive interactions to for mucosome function, adaptive immunity such as mucosal induce antifungal metabolites, (b) capacity to inhibit Bd at a antibodies is not indicated and antifungal function can be temperature range consistent with host habitat, and (c) resistance to attributed primarily to innate defenses including the microbiota. host skin immune defenses that would complicate probiotic establish- ment. Remedial biotherapy of already infected individuals should Indeed, altering the microbiota through probiotic treatments maintain antifungal capacity when grown in competition with Bd and affected mucosome function against Bd. In addition to assessing withstand the sometimes lethal effects of Bd metabolites (Fig. S6 in File infection risk in natural amphibian assemblages, mucosome S1). Testing probiotic effect in vivo can be accomplished without functional assays can now be used to assess risk in relict resorting to pathogen exposure experiments by using the mucosome populations or in captive colonies slated for reintroduction. While function assay described here. the efficacies of human probiotics are under scrutiny [2], doi:10.1371/journal.pone.0096375.g005 quantifying the effectiveness of amphibian probiotic treatments under scenarios of changing environmental conditions is a tangible with epithelial cells and immune system cells modulating both goal. local and systemic immune responses [69]. The immunomodula- tory effect of probiotics cannot be tested with in vitro Bd growth Supporting Information assays and host trials are necessary to test for these emergent properties of probiotics. File S1 Protocol for determining Bd viability, supplementary Antimicrobial peptides and a range of other defenses protect tables and figures. amphibian skin by synergizing or interacting with microbes (PDF) [41,70]. Thus, a better indication of antifungal effect of probiotics was obtained by testing the mucosome directly on zoospore Acknowledgments viability. In vitro screening cannot incorporate every factor and eventually in vivo trials, both in the lab and under natural We thank the V. McKenzie lab, J. Van Buskirk, M. Becker, S. Bell, J. Daskin, and J. Walke for their thoughtful discussion, T. Garner, V. conditions are necessary to determine if an overall health benefit is Vasquez, L. Reinert, and L. Rollins-Smith for donation of Bd isolates, S. provided. However, beginning with a probiotic that is not likely to Ro¨thlisberger for performing microsatellite analyses on Pelophylax embryos, become an opportunistic pathogen with changing climatic S. Lo¨tters and M. Veith from University of Trierand for help with field conditions may be a consideration. Transmissible probiotics would work, and the S.K. Schmidt lab for use of equipment. This work partially aid disease control at the population level [33], and if able to emerged from the advanced ecology course led by H.-U. Reyer at the persist through metamorphosis when applied to tadpoles, disease University of Zurich. presentation at this critical developmental stage could be avoided for A. obstetricans and other susceptible amphibians [40]. Addition- Author Contributions ally, Bd metabolites are known to be toxic to amphibian Conceived and designed the experiments: DCW SB JK EK UT. lymphocytes [50], and in this study were toxic to certain bacteria Performed the experiments: DCW HB SB JK EK UT LRD CB SH. such as Serratia plymuthica (Fig. S6 in File S1), perhaps prohibiting Analyzed the data: DCW SB JK EK UT BRS. Wrote the paper: DCW the use of certain probiotics intended as remedial biotherapy for BRS RK VM. Performed field work: DCW JK UT LRD. References 1. Bletz MC, Loudon AH, Becker MH, Bell SC, Woodhams DC, et al. (2013) 3. Fukuda S, Toh H, Hase K, Oshima K, Nakanishi Y, et al. (2009) Bifidobacteria Mitigating amphibian chytridiomycosis with bioaugmentation: characteristics of can protect from enteropathogenic infection through production of acetate. effective probiotics and strategies for their selection and use. Ecol Lett 16: 807– Nature 469: 543–547. 820. 4. Britton RA, Young VB (2012) Interaction between the intestinal microbiota and 2. Hoffmann DE, Fraser CM, Palumbo FB, Ravel J, Rothenberg K, et al. (2013) host in Clostridium difficile colonization resistance. Trends Microbiol 20: 313–319. Probiotics: Finding the right regulatory balance. Science 342: 314–315. PLOS ONE | www.plosone.org 11 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk 5. Belden LK, Harris RN (2007) Infectious diseases in wildlife: the community 34. Boyle DG, Boyle DB, Olsen V, Morgan JAT, Hyatt AD (2004) Rapid ecology context. Front Ecol Environ 5: 533–539. quantitative detection of chytridiomycosis (Batrachochytrium dendrobatidis) in 6. Lafferty KD (2010) Interacting parasites. Science 330: 187–188. amphibian samples using real-time Taqman PCR assay. Dis Aquat Org 60: 7. Clemente JC, Ursell LK, Parfrey LW, Knight R (2012) The impact of the gut 141–148. microbiota on human health: An integrative view. Cell 148: 1258–1270. 35. Rollins-Smith LA, Doersam JK, Longcore JE, Taylor SK, Shamblin JC, et al. 8. Daskin JH, Alford RA (2012) Context-dependent symbioses and their potential (2002) Antimicrobial peptide defenses against pathogens associated with global roles in wildlife diseases. Proc Roy Soc B 279: 1457–1465. amphibian declines. Dev Comp Immunol 26: 63–72. 9. Burgman MA, Ferson S, Akc¸akaya HR (1993) Risk assessment in conservation 36. Daum JM, Davis LR, Bigler L, Woodhams DC (2012) Hybrid advantage in skin biology. London: Chapman & Hall. peptide immune defenses of water frogs (Pelophylax esculentus) at risk from 10. Murray KA, Skerratt LF (2012) Predicting wild hosts for amphibian emerging pathogens. Infect Genet Evol 12: 1854–1864. chytridiomycosis: integrating host life-history traits with pathogen environmental 37. Stockwell MP, Clulow J, Mahony MJ (2010). Efficacy of SYBR 14/propidium requirements. Human and Ecological Risk Assessment 18: 200–224. iodide viability stain for the amphibian chytrid fungus Batrachochytrium 11. Woodhams DC, Voyles J, Lips KR, Carey C, Rollins-Smith LA (2006) Predicted dendrobatidis. Dis Aquat Org 88: 177–181. disease susceptibility in a Panamanian amphibian assemblage based on skin 38. Farrer RA, Weinert LA, Bielby J, Garner TWJ, Balloux F, et al. (2011) Multiple peptide defenses. J Wildl Dis 42: 207–218. emergences of genetically diverse amphibian-infecting chytrids include a 12. Gascon C, Collins JP, Moore RD, Church DR, McKay JE, et al, editors. (2007) globalized hypervirulent recombinant lineage. Proc Natl Acad Sci USA 108: Amphibian Conservation Action Plan. Gland: IUCN/SSC Amphibian Special- 18732–18736. ist Group. 39. Barrios V, Olmeda C, Ruiz E, compilers (2012). Action plan for the conservation 13. Grice EA, Segre JA (2011) The skin microbiome. Nat Rev Microbiol 9: 244– of the common midwife toad (Alytes obstetricans) in the European Union. Brussels: 253. European Commission. 14. Rosenthal M, Goldberg D, Aiello A, Larson E, Foxman B (2011) Skin 40. Tobler U, Schmidt BR (2010) Within- and among-population variation in microbiota: Microbial community structure and its potential association with chytridiomycosis-induced mortality in the toad Alytes obstetricans. PLoS ONE 5: health and disease. Infect Gen Evol 11: 839–848. e10927 doi:10.1371/journal.pone.0010927 15. Stecher B, Chaffron S, Ka¨ppeli R, Hapfelmeier S, Freedrich S, et al. (2010) Like 41. Rollins-Smith LA, Woodhams DC (2011) Amphibian immunity: staying in tune will to like: Abundances of closely related species can predict susceptibility to with the environment. In: Demas GE, Nelson RJ, editors.Ecoimmunology.New intestinal colonization by pathogenic and commensal bacteria. PLoS Patho. 6: e1000711. York: Oxford University Press. pp. 92–143. 16. Karlsson FH, Tremaroli V, Nookaew I, Bergstro¨m G, Behre CJ, et al. (2012) 42. Pask J, Woodhams DC, Rollins-Smith LA (2012) The ebb and flow of Gut metagenome in European women with normal, impaired and diabetic antimicrobial skin peptides defends northern leopard frogs, Rana pipiens, against glucose control. Nature 498: 99–103. chytridiomycosis. Global Change Bio 18: 1231–1238. 17. Ross EM, Moate PJ, Marett LC, Cocks BG, Hayes BJ (2013) Metagenomic 43. Woodhams DC, Bell SC, Kenyon N, Alford RA, Rollins-Smith LA (2012) predictions: From microbiome to complex health and environmental phenotypes Immune evasion or avoidance: Fungal skin infection linked to reduced defence in humans and cattle. PLoS ONE 8: e73056. peptides in Australian green-eyed treefrogs, Litoria serrata. Fun Bio 116: 1203– 18. Skerratt LF, Berger L, Speare R, Cashins S, McDonald KR, et al. (2007) Spread 1211. of chytridiomycosis has caused the rapid global decline and extinction of frogs. 44. Conlon JM, Demandt A, Nielsen PF, Leprince J, Vaudry H, et al. (2009) The Ecohealth 4: 125–134. alyteserins: Two families of antimicrobial peptides from the skin secretions of the 19. Fisher MC, Henk DA, Briggs CJ, Brownstein JS, Madoff LC, et al. (2012) midwife toad Alytes obstetricans (Alytidae). Peptides 30: 1069–1073. Emerging fungal threats to animal, plant and ecosystem health. Nature 484: 45. Schloss PD, Allen HK, Klimowicz AK, Mlot C, Gross JA, et al. (2010) 186–194. Psychrotrophic strain of Janthinobacterium lividum from a cold Alaskan soil 20. Murray KA, Skerratt LF, Garland S, Kriticos D, McCallum H (2013) Whether produces prodigiosin. DNA and Cell Biology 29: 533–541. the weather drives patterns of endemic amphibian chytridiomycosis: A pathogen 46. Kiesecker JM, Blaustein AR, Belden LK (2001) Complex causes of amphibian proliferation approach. PLoS ONE 8: e61061. population declines. Nature 410: 681–684. 21. Vredenburg VT, Briggs CJ, Harris RN (2011) Host-pathogen dynamics of 47. Pounds JA, Bustamante MR, Coloma LA, Consuegra JA, Fogden MPL, et al. amphibian chytridiomycosis: the role of the skin microbiome in health and (2006) Widespread amphibian extinctions from epidemic disease driven by disease. In: Olsen L, Choffnes E, Relman DA, Pray L, editors.Fungal diseases: global warming. Nature 439: 161–167. an emerging challenge to human, animal, and plant health. Washington, DC: 48. Woodhams DC, Alford RA, Briggs CJ, Johnson M, Rollins-Smith LA (2008) The National Academies Press. pp. 342–355. Life-history trade-offs influence disease in changing climates: Strategies of an 22. Harris RN, Brucker RM, Walke JB, Becker MH, Schwantes CR, et al. (2009) amphibian pathogen. Ecology 89: 1627–1639. Skin microbes on frogs prevent morbidity and mortality caused by a lethal skin 49. Doddington BJ, Bosch J, Oliver JA, Grassly NC, Garcia G, et al. (2013) Context- fungus. ISME J 3: 818–824. dependent amphibian host population response to an invading pathogen. 23. Becker MH, Harris RN, Minbiole KP, Schwantes CR, Rollins-Smith LA, et al. Ecology 94: 1795–1804. (2011) Towards a better understanding of the use of probiotics for preventing 50. Fites JS, Ramsey JP, Holden WM, Collier SP, Sutherland DM, et al. (2013) The chytridiomycosis in Panamanian golden frogs. Ecohealth 8: 501–506. invasive chytrid fungus of amphibians paralyzes lymphocyte responses. Science 24. Woodhams DC, Geiger CC, Reinert LK, Rollins-Smith LA, Lam B, et al. (2012) 342: 366–369. Treatment of amphibians infected with chytrid fungus: learning from failed trials 51. Voyles J, Rosenblum EB, Berger L (2011) Interactions between Batrachochytrium with itraconazole, antimicrobial peptides, bacteria, and heat therapy. Dis Aquat dendrobatidis and its amphibian hosts: a review of pathogenesis and immunity. Org 98: 11–25. Microbes & Infect 13: 25–32. 25. Dunne C, O’Mahony L, Murphy L, Thornton G, Morrissey D, et al. (2001) In 52. Conlon JM (2011) Structural diversity and species distribution of host-defense vitro selection criteria for probiotic bacteria of human origin: correlation with in peptides in frog skin secretions. Cell Molec Life Sci 68: 2303–2315. vivo findings. Am J Clin Nutr 73: 386S–392S. 53. Meltser S, Bishop PJ (2009) Skin peptide defences of New Zealand frogs against 26. Kesarcodi-Watson A, Kaspar H, Lategan MJ, Gibson L (2008) Probiotics in chytridiomycosis. Anim Cons 13: 44–52. aquaculture: The need, principles and mechanisms of action and screening 54. Crawford AJ, Lips KR, Bermingham E (2010) Epidemic disease decimates processes. Aquaculture 274: 1–14. amphibian abundance, species diversity, and evolutionary history in the 27. Searle CL, Gervasi SS, Hua J, Hammond JI, Relyea RA, et al. (2011) highlands of central Panama. Proc Natl Acad Sci USA 107: 13777–13782. Differential host susceptibility to Batrachochytrium dendrobatidis, an emerging amphibian pathogen. Cons Bio 25: 965–974. 55. Newman DG, Bell BD, Bishop PJ, Burns R, Haigh A, et al. (2010) Conservation 28. Balaz V, Vo¨ro¨s J, Civisˇ P, Vojar J, Hettyey A, et al. (2014) Assessing risk and status of New Zealand frogs, 2009. N Zeal J Zool 37: 121–130. guidance on monitoring of Batrachochytrium dendrobatidis in Europe through 56. Brucker RM, Harris RN, Schwantes CR, Gallaher TN, Flaherty DC, et al. identification of taxonomic selectivity of infection. Cons Bio 28: 213–223. (2008) Amphibian chemical defense: Antifungal metabolites of the microsym- 29. Ro¨dder D, Kielgast J, Bielby J, Schmidtlein S, Bosch J, et al. (2009) Global biont Janthinobacterium lividum on the salamander Plethodon cinereus. J Chem Ecol amphibian extinction risk assessment for the panzootic chytrid fungus. Diversity 34: 1422–1429. 1: 52–66. 57. Salinas I, Abelli L, Bertoni F, Picchietti S, Roque A, et al. (2008) Monospecies 30. Johnson PTJ, Preston DL, Hoverman JT, Richgels KLD (2013). Biodiversity and multispecies probiotic formulations produce different systemic and local reduces disease through predictable changes in host community competence. immunostimulatory effects in the gilthead seabream (Sparus aurato L.). Fish & Nature 494: 230–234. Shellfish Immunol 25: 114–123. 31. Loudon AH, Woodhams DC, Parfrey LW, Archer H, Knight R (2013) 58. Ku¨ng D, Bigler L, Davis LR, Gratwicke B, Griffith E, et al. (2013) Stability of Microbial community dynamics and the effect of environmental reservoirs on microbiota on a Panamanian frog: insights for probiotic therapy against disease. red-backed salamanders (Plethodon cinereus). ISME J doi: 10.1038/ismej.2013.200. PLoS ONE 9(1): e87101. doi:10.1371/journal.pone.0087101. 32. Woodhams DC, Vredenburg VT, Simon MA, Billheimerd D, Shakhtourd B, 59. Schadich E, Cole ALJ (2009) Inhibition of frog antimicrobial peptides by et al. (2007) Symbiotic bacteria contribute to innate immune defenses of the extracellular products of the bacterial pathogen Aeromonas hydrophila. Lett Appl threatened mountain yellow-legged frog, Rana muscosa. Biol Cons 138: 390–398. Microbiol 49: 384–387. 33. Walke JB, Harris RN, Reinert LK, Rollins-Smith LA, Woodhams DC (2011) 60. Schadich E (2009) Skin peptide activities against opportunistic bacterial Social immunity in amphibians: Evidence for vertical transmission of innate pathogens of the African clawed frog (Xenopus laevis) and three Litoria frogs. defenses. Biotropica 43: 396–400. J Herpetol 43: 173–183. PLOS ONE | www.plosone.org 12 April 2014 | Volume 9 | Issue 4 | e96375  Mucosome Function Predicts Disease Risk 61. McKenzie VJ, Bowers RM, Fierer N, Knight R, Lauber CL (2012) Co-habiting 67. Williamson NR, Fineran PC, Gristwood T, Chawrai SR, Leeper FJ, et al. (2007) amphibian species harbor unique skin bacterial communities in wild popula- Anticancer and immunosuppressive properties of bacterial prodiginines. Future tions. ISME J 6: 588–596. Microbiol 2: 605–618. 62. Kueneman JG, Parfrey LW, Woodhams DC, Archer HM, Knight R, McKenzie 68. Wilson M (2005) Microbial inhabitants of humans: their ecology and role in VJ (2013) The amphibian skin microbiome across species, space and life history health and disease. Cambridge: Cambridge University Press. stages. Mol Ecol doi: 10.1111/mec.12510. 69. Hancock REW, Nijnik A, Philpott DJ (2012) Modulating immunity as a therapy 63. Delplace F, Maes E, Lemoine J, Strecker G (2002) Species specificity of O-linked for bacterial infections. Nat Rev Microbiol 10: 243–254. carbohydrate chains of the oviducal mucins in amphibians: structural analysis of 70. Myers JM, Ramsey JP, Blackman AL, Nichols AE, Minbiole KPC, et al. (2012) neutral oligosaccharide alditols released by reductive beta-elimination from the Synergistic inhibition of the lethal fungal pathogen Batrachochytrium dendrobatidis: egg-jelly coats of Rana clamitans. Biochem J 363: 457–471. The combined effect of symbiotic bacterial metabolites and antimicrobial 64. Varki A (1993) Biological roles of oligosaccharides – all of the theories are peptides of the frog Rana muscosa. J Chem Ecol 38: 958–965. correct. Glycobiol 3: 97–130. 71. Becker J, Eisenhauer N, Scheu S, Jousset A (2012) Increasing antagonistic 65. NIH HMP Working Group, Peterson J, Garges S, Giovanni M, McInnes P, interactions cause bacterial communities to collapse at high diversity. Ecol Lett et al(2009) The NIH Human Microbiome Project. Genome Res 19: 2317–2323. 15: 468–474. 66. Cotter PD, Ross RP, Hill C (2013) Bacteriocins - a viable alternative to antibiotics? Nat Rev Microbiol 11: 95–105. PLOS ONE | www.plosone.org 13 April 2014 | Volume 9 | Issue 4 | e96375 

Ecography 36: 001–008, 2013 doi: 10.1111/j.1600-0587.2013.00001.x © 2013 The Authors. Ecography © 2013 Nordic Society Oikos Subject Editor: John R. Spence. Accepted 4 February 2013 How do low dispersal species establish large range sizes? The case of the water beetle Graphoderus bilineatus Lars Lønsmann Iversen, Riinu Rannap, Philip Francis Thomsen, Jos Kielgast and Kaj Sand-Jensen L. L. Iversen (lliversen@bio.ku.dk) and K. Sand-Jensen, Freshwater Biological Laboratory, Biological Inst., Univ. of Copenhagen, Helsingørsgade 51, DK-3400 Hillerød, Denmark. – R. Rannap, Inst. of Ecology and Earth Sciences, Univ. of Tartu, Vanemuise 46, EE-51014 Tartu, Estonia. – P. F. Thomsen and J. Kielgast, Centre for GeoGenetics, Natural History Museum of Denmark, Univ. of Copenhagen, Øster Voldgade 5-7, DK-1350 Copenhagen, Denmark. Species’ dispersal abilities have been considered a major driving force in establishment and maintenance of large range sizes. However, recent studies question the general validity of this relationship because the relationship between dis- persal ability and range size might in some cases be less important than species phylogeny or local spatial attributes. In this study we used the water beetle Graphoderus bilineatus a philopatric species of conservation concern in Europe as a model to explain large range size and to support effective conservation measures for such species that also have limited dispersal. We recorded the presence/absence of G. bilineatus and measured 14 habitat and 20 landscape variables at 228 localities in Estonia, Poland and Sweden within the core range of the species. Using information theory and average multivariate logistic regression models we determined that presence of G. bilineatus depended on landscape connectiv- ity, distance to a possible source habitat, and stability of the site; however, specificity of habitat characteristics was not vital for the species. We reason that the large range of G. bilineatus is best explained by the historical combination of lakes, river systems and wetlands which used to be highly connected throughout the central plains of Europe. Our data suggest that a broad habitat niche can prevent landscape elements from becoming barriers for species like G. bilineatus. Therefore, we question the usefulness of site protection as conservation measures for G. bilineatus and similar philopat- ric species. Instead, conservation actions should be focused at the landscape level to ensure a long-term viability of such species across their range. At the frontiers of ecology, biogeography, and interest in how species with low dispersive power attain large range sizes climate-driven species extinction, there is considerable inter- and what processes drive the persistence of such large ranges. est in the distributions and range sizes of species (Brown Habitat and landscape features, which affect the local et al. 1996, Gaston 2006, Hof et al. 2011). For a broad range distribution of species, are generally regarded as decisive for of taxa, dispersal ability appears to be a major driving the viability of populations at the local scale and, thus, the force in establishment of large range sizes (Brown et  al. distribution of species at the global scale (Joly et al. 2001, 1996, Gaston 1996, Gutierrez and Menendez 1997, Mora Bakker et  al. 2002, Lowe and Bolger 2002, Armstrong and Robertson 2005, Lowry and Lester 2006). However, in 2005). In this study we explored the paradoxical distri­ a recent review, Lester et al. (2007) questioned the general bution pattern of a philopatric but widespread aquatic validity of this relationship. In addition, recent studies insect species, the aquatic water beetle Graphoderus have shown that the positive relationship between dispersal bilineatus, and attempted to understand its range size in rela- ability and range size may, in some cases, be an artifact of tion to the factors that limit local persistence. We examined phylogenetic bias or local spatial attributes (Gove et al. 2009, the habitat and landscape dependencies of this species Garcia-Barros and Benito 2010, Mora et  al. 2011). Thus across three different landscapes in northern and eastern dispersal ability should be regarded as only one of several Europe. Graphoderus bilineatus uses a well-defined habitat factors that may shape range size of a species (Lester et al. (lakes, ponds and other stagnant waters), and has a large 2007). Dennis et  al. (2005), for example, have suggested European range size despite low dispersal capacity and is that species ranges are fundamentally a result of interaction thus an ideal candidate for the purposes of this study. between niche-breath and dispersal ability and niche- Graphoderus bilineatus is considered a threatened species breadth itself is often regarded as an important characteri­ throughout its western distribution range where its stic for establishing large range sizes (Gaston 2006). occurrence is rare and patchy (Nilsson and Holmen 1995, Nonetheless, few studies have empirically addressed the basis Foster 1996, Hendrich and Balke 2000). It is strictly pro- of species range sizes. In particular, little is known about tected within the EU-countries (Council of the European Early View (EV): 1-EV Union 1992) and is thereby among the few invertebrates 1993, Hendrich and Balke 2000, Cuppen et  al. 2006). worldwide receiving substantial attention in terms of active However, reports on specific habitat requirements of the conservation measures. Hence, empirically clarifying species are contradictory and differ across the species range. habitat–landscape relationships for this species will contri­ Habitats assumed to be essential for the species can be bute to successful protection of the species and its habitats divided into two general categories: 1) clear and mesotrophic within the framework of the EU Habitat Directive and the or oligotrophic larger lakes or canals with sparse vegetation NATURA 2000 network. (Galewski 1971, Nilsson and Persson 1989, Cuppen et  al. 2006), or 2) lakes with sun-exposed shallow zones and dense marginal vegetation (Holmen 1993, Nilsson and Material and methods Holmen 1995, Hendrich and Balke 2000). Study species Study areas Graphoderus bilineatus is a medium size, predacious water beetle (body length 14–16 mm) from the family The study was carried out in three areas located in the core Dytiscidae. It has a univoltine life cycle with oviposition in of the European distribution of G. bilinieatus: the flood- spring–early summer, followed by aquatic larvae stages and plain valley of the Narew River in eastern Poland; three a terrestrial pupa stage (Nilsson and Holmen 1995, Foster national parks (Haanja, Karula and Emajõe-Suursoo) in 1996). Although some adults of G. bilineatus fly (Nilsson southeastern Estonia and the forest landscape of north­ and Holmen 1995, Hendrich and Balke 2000) and these eastern Scania in Sweden (Fig. 1). These three areas were beetles have normally developed flight muscles, only a frac- chosen based on differences in landscape shaping processes, tion of specimens in a population seems to do so (Iversen and results gained across these areas form a general pattern unpubl.). Flight studies in areas where G. bilineatus occur of the environmental variables determining the presence of have not documented any flight activity for the species the species. They may be characterized as follows. The river (Lundkvist et  al. 2002). Overall, the descriptions of the valley in Poland is a dynamic landscape with large flood- flight ability of G. bilineatus are rare or anecdotal and the plains dominated by grassland and cultivated rural areas. species is regarded as having low dispersal power (Kehl Beavers are active both in the floodplains and in the and Dettner 2007). Nonetheless, the species has a wide cultivated areas. The aquatic habitats are small to medium- distribution, ranging west-east from France to western sized natural ponds and oxbow lakes and artificial man- Siberia, and north-south from central Finland to northern made ponds used for watering cattle and breeding fish. The Italy (Nilsson and Holmen 1995, Foster 1996). sites in Estonia are situated in a moraine landscape with Several authors have described G. bilineatus as a hilly areas and larger wetlands. The landscape is a mosaic of highly stenoecious species with a distribution restricted to a extensively used agricultural land and mixed or coniferous specific combination of habitat characteristics (Holmen forests. Both large and medium-sized lakes are present in the Figure 1. The geographic location of the 3 study areas. The bar graphs illustrates the percentage occurrence of G. bilineatus at numerous sites (n) within each study area (blue  present, red  absent). 2-EV landscape and flooding because of beaver activity is common Table 1. Habitat variables measured at each of the 228 sites; signifi- in depressions and along streams. Small artificial ponds are cance levels were determined by univariate logistic regression. scattered throughout the landscape, either due to their func- Habitat variables Acronym p-value tion as sauna- or fishponds, or due to extensive conservation 2 Surface area (m ) log(area)  0.001 of small water bodies in this area (Rannap et al. 2009). The Shape: length/width ratio Shape  0.01 Swedish landscape consists of a mixture of conifer-dominated Maximum shallow zone: max.shal  0.05 forests and cattle-grazed farmland. The landscape is formed 0–30 cm (m) by exposed bedrock, creating a myriad of lakes, mires and Minimum shallow zone: min.shal  0.05 0–30 cm (m) bogs. In contrast to the two other areas there are no beavers. Max depth ( 0.5, 0.5, 1 or Depth  0.001 1.5 m) Maximum edge slope: max.slope  0.01 Data collection (5, 10, 25, 45 or 90 degrees) Minimum edge slope min.slope  0.001 A total of 228 localities were investigated in 2010 and (5, 10, 25, 45 or 90 degrees) 2011 within the breeding period of G. bilineatus. A semi- Water color (clear, muddy, factor(water)  0.05 standardized dipnetting method adopted from Nilsson algal-green, brown) and Svensson (1995) was used. The species was actively Grazing (yes/no) factor(grazing) 0.12 Shading from surrounding Shadow  0.001 sought during 45 min at each site by sweeping a hand trees (%) dipnet (40  40 cm frame) through vegetation and detrital Vegetation above 1 m in vega1m 0.48 material and based on the search the species was recorded height (%)* as either present or absent at each site. This survey method Vegetation below 1 m in vegb1m 0.07 has proven to be the most time effective for G. bilineatus height (%) Floating vegetation (%) Fveg  0.01 (Klečka and Boukal 2011), with a coverage as good as or Submerged vegetation (%)* Sveg 0.23 better than other sampling methods (Koese and Cuppen 2006). To rule out skill-related sampling bias, all fieldwork *Variables omitted from multivariate analysis. was carried out by the same person (LLI). We believe that detection of adult specimens serves as a reliable 2002) was used in a multivariate logistic regression model proxy for breeding site status, because G. bilineatus has a environment. Information theory relaxes some of the univoltine lifecycle and therefore it must reproduce every problems induced by stepwise model selection and classic year for a population to persist. Additionally the presence null-hypothesis testing (Johnson and Omland 2004, of the species did not show any signs of spatial autocor­ Whittingham et al. 2006, Stephens et al. 2007). The main relation (Results), which indicate that it is within site effect of the 34 environmental variables was used as reproduction and not yearly migration that is the major explanatory variable. Given that the predictive power of one determinant of the presence of the species. Prior to the single model may not be clearly superior, model averaging fieldwork sites that were directly interconnected by water- was performed across a set of candidate models. Estimates ways, only one of them was randomly selected for the of parameters by model averaging are robust in the sense study, though 228 localities were still examined. Finally, that they reduce model selection bias and account for model to assess the reliability of the survey data, 10 Swedish selection uncertainty (Johnson and Omland 2004). sites were re-sampled in 2011. The surveys generated the Hegyi and Garamszegi (2011) argued that only when a same results as those obtained in 2010 (i.e. the same 5 full model containing all initial variables is analyzed, infor- localities had G. bilineatus and the same 5 did not). mation theory can overcome the errors caused by stepwise In total, 14 habitat characteristics and 20 landscape model reduction. However, creating an initial model from level variables were collected for each site (see Table 1 and 2 all 34 variables produces an overly complex model and for detailed description). Variables assumed to influence 234 potential sub models. Thus, following the recommenda- presence/absence of G. bilineatus were selected based on tion of a prior explorative data search by Grueber et  al. existing literature and results of preliminary fieldwork (2011), two selection steps were applied in order to create conducted on the species. The 20 landscape variables a set of candidate variables. described environmental resistance to dispersal such as con- Firstly, a univariate analysis of each environmental nectivity parameters (number of ponds within a given variable was conducted, testing the effect of the variable on distance and distance to a lake or node in the landscape), the presence of G. bilineatus on a log-odds scale. A logistic land cover types (ranging from open to closed and disturbed regression model with the environmental variable as the to natural land cover classes) and the level of heterogeneity explanatory variable was performed, assessing the effect in landscape structure. These habitat variables were derived of each variable by a c²-test (Table 1 and 2). From the by GIS (ArcGis 9.3, www.ESRI.com) from contempo- univariate tests all variables with a p-value  0.15 were rary land cover maps of each region. omitted from further analysis. Prior to the first selection step, lake area was log-transformed to ensure that the Data analysis probability of G. bilineatus’ presence would be zero when lake area is zero. Additionally, the effect of landscape and To explore the relationship between the presence of latitudinal or longitudinal gradient effects on G. bilineatus’ G. bilineatus and the environmental variables of interest, an presence were tested in the same framework as the environ- information theoretic approach (Burnham and Anderson mental variables. 3-EV Table 2. Landscape variables measured at each of the 228 sites; significance levels were determined by univariate logistic regression. Landscape variables Acronym p-value No. of ponds  100 m from site ponds100m 0.07 No. of ponds between 100 and 200 m from site ponds200m  0.05 No. of ponds between 200 and 800 m from site* ponds800m 0.30 Distance to lake (m) dist.lake  0.001 Distance to stream, ditch or river (m) dist.node  0.001 Habitat 0–500 m. % Forest of total area* forest500 0.50 Habitat 0–500 m. % Open land (grassland and field) of total area* open500 0.31 Habitat 0–500 m. % Urban/farm of total area urban500 0.14 Habitat 0–500 m. % Bogs/swamps of total area bog500 0.13 Habitat 500–1000 m. % Forest of total area* forest1000 0.23 Habitat 500–1000 m. % Open land (grassland and field) of total area open1000 0.15 Habitat 500–1000 m. % Urban/farm of total area* urban1000 0.80 Habitat 500–1000 m. % Bogs/swamps of total area* bog1000 0.56 Habitat 1000–2000 m. % Forest of total area* forest2000 0.34 Habitat 1000–2000 m. % Open land (grassland and field) of total area* open2000 0.20 Habitat 1000–2000 m. % Urban/farm of total area* urban2000 0.33 Habitat 1000–2000 m. % Bogs/swamps of total area* bog2000 0.88 No. of patches 10 000 ha21 (0–500 m) pd500  0.05 No. of patches 10 000 ha21 (500–1000 m) pd1000 0.05 No. of patches 10 000 ha21 (1000–2000 m)* pd2000 0.25 *Variables omitted from multivariate analysis. Secondly, two initial logistic regression models were cre- possible overestimation of the degree that habitat variables ated, containing habitat variables and landscape variables affect species occurrence (Legendre 1993, Keitt et  al. respectively (Table 2). These models were used to find the 2002) – a problem which also exists in parameter estimation habitat and landscape level variables potentially affecting by information theory (Diniz-Filho et  al. 2008). Spatial the presence of G. bilineatus. From these initial models auto-correlation in the distribution of G. bilineatus could second-order Akaike information criterion (AICc) values occur in the data for the following reasons: the descriptor were calculated from all possible submodels. AICc was for species presence may not reflect independent breeding chosen due to the low number of samples in relation to populations, but could be an artifact of temporary emigra- the number of parameters present in the initial model tion from nearby sources. If this was true, auto-correlation (Symonds and Moussalli 2011). From all the possible sub- should occur within close distance of a site with species models the model with the lowest AICc value was identified, presence. Secondly, if long distance movements occur fre- and a group of candidate models selected, containing quently, spatial auto-correlation would be present in the all models within two units of this AICc value (Burnham catchment areas of each site. One way to assess the extent and Anderson 2002). of spatial auto-correlation is correlograms of data, a graphi- These models were averaged by the use of the MuMIn cal representation of the spatial correlation between package in R (Barton 2012), and parameter estimates locations at a range of a given lag distances. In this study were calculated from this average model. Before conducting the presence of spatial auto-correlation was assessed after the analysis, all input variables were standardized to a accounting for the level of spatial auto-correlation explained common scale with a mean of 0, and an SD of 0.5, in order by the explanatory model variables. This was achieved by to ease the parameter interpretation and allow comparisons a spline correlogram produced from the Pearson residuals of of effects. From the average habitat and landscape model, the final average model, using the R package ncf (Bjørnstad explanatory variables with unidirectional parameter estima- and Falck 2001). The x-intercept in the spline correlogram tions (within the 95% confidence intervals) were selected is the distance at which an object is no more similar than as variables potentially affecting the presence of G. bilineatus what is expected by random placement across the region. (Supplementary material Appendix 1, Table A2). A final initial model was then formulated based on these two selection steps (Table 3). The final model underwent Table 3. Variables included in separate habitat and landscape models and a final model combining habitat and landscape the same procedure as the habitat and landscape model, and variables. from the 2ΔAICc models an average model was created. From this averaged final model parameter estimates and Models Initial explanatory variables 95% confidence intervals were determined. The relative Habitat model Grazing, depth, vegb1m, min.shal, max.shal, effect of each explanatory variable was expressed as the sum water, max.slope, min.slope, fveg, shadow, of Akaike weights (wi) in the 2ΔAICc models, where the logarea, shape Landscape urban500, field500, bog500, open1000, variable occurred. The larger the sum of the Akaike weight, model ponds100m, ponds200m, pd500, pd1000, the more important the variable is relative to the other dist.node, dist.lake variables in the average model. Final model logarea, depth, max.slope, shadow, fveg, An increasing amount of literature highlights the ponds200m, dist.lake, dist.node, bog500, pd1000, open1000 presence of spatial auto-correlation in ecological data and 4-EV Possible spatial auto-correlation is assessed using 95% point Table 4. Parameter estimates and relative importance of the land- wise bootstrap confidence intervals calculated from 1000 scape and habitat variables for the final model. The estimates are derived from an average model of five possible models explaining bootstrap samples. the presence–absence of Graphoderus bilineatus. The final model was evaluated for predictive ability against the average habitat model and landscape model using 95% confidence Relative Parameter Estimate intervals importance a ROC curve. The ROC curve represents the relationship between true presences and false presences for a range (Intercept) 20.59 [20.99, 20.19] of threshold values classifying the probability of presence bog500 1.10 [0.03, 2.17] 1 Depth 0.49 [20.38, 1.37] 0.32 based on the predicted probabilities of G. bilineatus occur- dist.lake 1 21.77 [22.82, 20.72] rence in the area (Fielding and Bell 1997). Differences dist.node 21.16 [22, 20.32] 1 between the predictive abilities of the three models were Fveg 20.87 [21.81, 0.07] 0.85 evaluated based on their AIC-values using a method for Logarea 2.50 [1.51, 3.5] 1 paired ROC-curves (DeLong et al. 1988). The ROC-curves max.slope 21.21 [22.1, 20.32] 1 and test statistics were created in the qROC package in R open1000 2.03 [1.05, 3.01] 1 (Robin et al. 2011). pd1000 21.55 [22.42, 20.68] 1 All analysis were conducted in R ver. 2.14.2 ( www. ponds200m 20.74 [21.65, 0.16] 0.63 r-project.org ). shading 21.08 [21.99, 20.18] 1 (with a one unit increase of the two variables odds decreased Results by factors of 0.17 [0.06–0.49] and 0.31 [0.14–0.73] respectively). 2) Type of land-cover. The amount of wetland Graphoderus bilineatus was found in 43% of the 228 locali- within 500 m of each site was positively related to the ties investigated, and site occupation varied between 36% presence of G. bilineatus (the odds changed by a factor of in Estonia and 50% in Poland (Fig. 1 and Supplementary 2.99 [1.03–8.76]). The amount of open landscape between material Appendix 1, Table A1). The chance of finding the 500 and 1000 m from the study site had a positive effect to species did not differ significantly (p  0.19) among the the species (odds changed by a factor of 7.64 [2.86–20.29]). three countries, nor did either longitude or latitude influ- 3) Landscape structure. The extent of landscape hetero­ ence probability of occupation (p  0.79 and p  0.07 geneity between 500 and 1000 m from each site affected the respectively). Thus, the data suggested no evidence of a geo- presence of G. bilineatus negatively (with a one unit increase graphic gradient. in patch density, the odds decreased by a factor of 0.21 Two habitat variables and 10 landscape characteristics [0.09–0.51]). were omitted from further analyses after the univariate There was no evidence of spatial auto-correlation in the analyses (Table 1, 2; see also Methods). From the variables spline corellogram from the Pearson residuals of the final used initially in the habitat and landscape models (Table 3), average model (Fig. 2). Thus, spatial auto-correlation is respectively, six and four models were found to be within not likely to influence parameter estimates for the average 2ΔAICc (Supplementary material Appendix 1, Table A3). model (Table 4). The combination of habitat and landscape Model averaging of these top models identified five habitat variables in one average model (Fig. 3) provided a better variables and six landscape variables with unidirectional predictive model than habitat variables or landscape vari- parameter estimation (Supplementary material Appendix 1, ables alone (Z  2.44, p  0.02 and Z  3.90, p  0.001 Table A2). These eleven variables were thus included in respectively). These results reflect findings of other studies our final model (Table 3). From different combinations of (Richard and Armstrong 2010), highlighting the impor- variables in the final model, five submodels were selected tance of integrating both the landscape characteristics and within the 2ΔAICc (Supplementary material Appendix 1, habitat variables in ecological models. Table A3). The results from the model averaging of these five models and the parameter estimates of the standard- ized variables are shown in Table 4. Three habitat variables Discussion (log (surface area), maximum edge slope and the degree of shadow) had constant parameter direction. With a one Our results explore the factors that potentially determine unit increase in log(area), there was a positive effect on the population viability of G. bilineatus, and illuminate the log-odds of G. bilineatus being present (relative change landscape features that may have lead to its current distribu- in odds by a factor of 12.21 [4.53–33.12] (mean and 95% tion. The weak relationship between species presence and CL)). In contrast, increases in maximum slope and the habitat variables suggests a wide habitat niche and contrasts degree of shadowness at a site negatively affected the with the widespread perception of the species as being log-odds of finding G. bilineatus (odds changed by a factor stenotypic. Interestingly, none of the habitat variables, tradi- of 0.30 [0.12–0.73] and 0.34 [0.14–0.84] by one unit tionally thought to reflect suitability (e.g. vegetation type, increase of the two predictor variables, respectively). water quality, depth of water, presence of shallow areas) Five landscape variables with unidirectional parameter had substantial effect on the presence of G. bilineatus, estimations can be referred to the following three groups: although in the literature they have long been regarded as 1) distance to a possible immigration source. Both increasing being crucial (Hendrich and Balke 2000, Cuppen et  al. distance to nearest lake and nearest node in the landscape 2006). We point out that these presumptions about the had a negative effect on the presence of G. bilineatus nature of typical habitat are at the root of the EU Habitat 5-EV  sauna ponds and fish ponds. This could imply that the nega- tive effect of steeper maximum slope might be solely related to the unsuitable nature of these human created waters. The positive effect of lake area on presence of G. bilineatus reflects two important dependencies of a philopatric species with low mobility: stability and immigration (Gaston 2006). Because filling of lakes to a great extent takes place by in-growth of emergent plants and input of terrestrial material across the periphery, larger lakes have a greater lon- gevity than small lakes of the same mean depth (Wetzel 1983). Greater wind exposure in large lakes will also slow sedimentation rates and enhance gas exchange rates with the atmosphere and, thereby, stabilize oxygen and pH condi- tions of benefit for water beetles (Sand-Jensen and Staehr 2009, Staehr et  al. 2012). Overall, larger lakes are more temporally stable compared to smaller waters, hence the turnover rate is longer and mechanisms of metapopulation dynamics less vital. The positive relationship between nearby dispersal sources Figure 2. Spline correlogram, with 95% point-wise bootstrap con- and the presence of G. bilineatus is doubtlessly driven by fidence intervals of the Pearson residuals from the final averaged the limited dispersal ability of the species (Taylor et  al. model. 1993, Tischendorf and Fahrig 2000). The odds of a locality being occupied are closely related to the distance to a Directive and European NATURA 2000 Nature Conser­ potential source population. This source effect includes vation Network guidelines for conservation of the species direct sources (distance to a lake) but also dispersal corridors across the majority of its range. (rivers and ditches). However, the latter features can poten- Our analysis did identify some features useful for tially act as both dispersal corridors and breeding sites making predictions about site occupancy. For example, the (Vermeulen 1994). The fact that rivers, streams, and smaller negative effect of the degree of shadow from surrounding channels in the landscape are used by G. bilineatus was trees on species presence does seem to indicate that either confirmed in all three countries, where the species was sun-exposed areas or an open flight path are important found in slow flowing canals or in larger meandering for G. bilineatus. The species was also more commonly rivers. Positive association of occupied sites with wetlands found at lakes with natural edges, but not necessarily in within a buffer distance of 500 m likely reflects a tendency lakes with a wide shallow zone. Steep edges were, however, for short-distance movement in G. bilineatus. However, associated with man-made waters such as stone quarries, overall, our results suggest a stronger effect of corridors as dispersal routes, as opposed to stepping stones (in this case as number of ponds within a given distance). Thus, it seems that site occupation depends on a dense network of lakes and temporally stable connectivity corridors. Openness in the landscape up to 1000 m from each site was positively related to the presence of G. bilineatus. When flight does occur, it is assumed that localities in an open landscape are more likely to become occupied than localities in closed forest landscapes. Chin and Taylor (2009) documented that an open landscape enhanced long distance movements in a philopatric dragonfly, and this mechanism could also be important for G. bilineatus. Thus, the dependence of site occupation on an open land- scape at 500–1000 m from each site may reflects an effect on arrival potential of dispersing individuals. This hypothesis is further supported by the negative relationship with land- scape heterogeneity between 500 and 1000 m of the site. These ideas about mechanism behind the patterns revealed by our models should be further tested through empirical studies focused on the potential flying capacity and dispersal range of the species. Clearly, G. bilineatus shows high dependence on land- Figure 3. Smoothed ROC-curves describing the final model scape connectivity, distance to a possible source population (AUC  0.89) and the two sub-models containing habitat and temporal habitat stability in its area of occupancy. variables (AUC  0.85) and landscape variables (AUC  0.80) Importantly, the results suggest that G. bilineatus is some- respectively. thing of a habitat generalist, and that landscape features are 6-EV the major factors associated with the current distribution of Acknowledgements – We thank Michał Maniakowski (FPP the species. This conclusion is equally evident in all three Consulting), Mariusz Sachmaciński (Łomża Landscape Park) and countries and on the different landscape types investigated. Lars Briggs (Amphi Consult) for accommodation and practical Given that G. bilineatus is philopatric with limited assistance during the fieldwork and Christian Bressen Pipper (Faculty of Life Sciences, Univ. of Copenhagen) for comments on dispersal capacity and that the local distribution is highly the applied statistic. We received numerous useful suggestions and dependent on landscape features and structure, our linguistic improvements from John R. Spence that we greatly findings can be evaluated in the context of the abundance- acknowledge. This study was supported by The Natural History distribution range theory. Such theory assumes that Museum of Denmark and LIFE  DRAGONLIFE (LIFE08 abundant species establish large range sizes simply due NAT/EE/000257). to their relation to widely distributed habitats (Gaston 2006). For a species with low dispersal ability such as G. bilineatus, a large range would then depend on presence References of its habitat across the entire range. This pre-condition would in turn be highly influenced by the habitat niche- Araújo, M. B. et  al. 2011. Climate change threatens European breadth of the species because stenotypic species are conservation areas. – Ecol. Lett. 14: 484–492. restricted to a smaller number of suitable habitats than are Armstrong, D. P. 2005. Integrating the metapopulation and habitat paradigms for understanding broad-scale declines habitat generalists. Hence, the chance of high habitat of species. – Conserv. Biol. 19: 1402–1410. connectivity is larger for species with a broad niche. Bakker, K. K. et al. 2002. Incorporating landscape attributes into For G. bilineatus, establishment and maintenance of models for migratory grassland bird conservation. – Conserv. a large range is likely due to generality in habitat demands Biol. 16: 1638–1646. (a broad niche) and landscape connectivity. If the species Barton, K. 2012. MuMIn: multi-model inference. – R package, had been more stenotopic about habitat use, barriers to ver. 1.6.6. establishment of populations would have been more Bjørnstad, O. N. and Falck, W. 2001. Nonparametric spatial probable, thereby restricting the range of the species by covariance functions: estimation and testing. – Environ. Ecol. Stat. 8: 53–70. environmental resistance to effective colonization. High Brown, J. H. et  al. 1996. The geographic range: size, shape, connectivity between lakes, wetlands, and river systems boundaries, and internal structure. – Annu. Rev. Ecol. Syst. throughout its range combined with a broad habitat niche, 27: 597–623. has likely enabled G. bilineatus to occupy a variety of Burnham, K. P. and Anderson, D. R. 2002. Model selection habitats over a vast range in Europe. However, today the and multimodel inference: a practical information-theoretic landscape in the entire western part of the species’ distri­ approach, 2nd ed. – Springer. bution range is influenced by strong anthropogenic Chin, K. S. and Taylor, P. D. 2009. Interactive effects of distance impacts, leaving a less stable and unfavorable habitat struc- and matrix on the movements of a peatland dragonfly. – Ecography 32: 715–722. ture for G. bilineatus, and this is likely to affect its long-term Council of the European Union 1992. Council directive 92⁄43⁄EEC persistence. on the conservation of natural habitats and of wild fauna and Such information can be integrated into an informed flora. – Council of the European Union, Strassbourg, Germany. conservation strategy for the species. The Natura 2000 Cuppen, J. G. M. et  al. 2006. Distribution and biotope of habitat conservation network and the habitat directive Graphoderus bilineatus in the Netherlands (Coleoptera: (Council of the European Union 1992) provides the current Dytiscidae). – Nederl. Faun. Med. 24: 29–40. basis for conservation of biodiversity in Europe. This strat- DeLong, E. R. et  al. 1988. Comparing the areas under two or egy allocates substantial resources to protecting rare habitats more correlated receiver operating characteristic curves: a nonparametric approach. – Biometrics 44: 837–845. which are vital for persistence of rare stenotypic species Dennis, R. L. H. et al. 2005. Does diet breadth control herbivorous in the region. However, based on our study, it seems that insect range size? Predictions and tests using butterflies. taking this approach exclusively is not likely to be successful – J. Insect Conserv. 9: 187–200. in the long term for species such as G. bilineatus. In con- Diniz-Filho, J. et  al. 2008. Model selection and information trast, our results clearly emphasize the importance of inte- theory in geographical ecology. – Global Ecol. Biogeogr. 17: grating the concepts of landscape structure and stability into 479–488. conservation strategies and suggest that such approaches Fielding, A. H. and Bell, J. F. 1997. A review of methods for are important even for conservation on a local scale. The the assessment of prediction errors in conservation presence/ absence models. – Environ. Conserv. 24: 38–49. case of G. bilineatus provides a useful example of how Foster, G. N. 1996. Graphoderus bilineatus (DeGeer, 1774). – In: neglect of geographical pattern and the associated dispersal van Helsdingen, P. J. et  al. (eds), Background information processes may be problematic in conservation planning. on invertebrates of the Habitats Directive and the Bern Such concerns amplify the problem of long-term viability Convention. Part 1 – Crustacea, Coleoptera and Lepidoptera. of species, thought to be protected in the Natura 2000 European Invertebrate Survey Leiden, pp. 40–48. network, in relation to the ongoing concern about climatic Galewski, K. 1971. A study on morphobiotic adaptations of changes (Araújo et  al. 2011). Incorporating predictions European species of the Dytiscidae (Coleoptera). – Pol. Pismo. about long term species persistence in the Natura 2000 net- Entomol. 41: 487–702. Garcia-Barros, E. and Benito, H. R. 2010. The relationship between work will be a daunting task; however, our study suggests geographic range size and life history traits: is biogeographic that for a species like G. bilineatus conservation planning to history uncovered? A test using the Iberian butterflies. achieve long-term persistence must be conducted at a land- – Ecography 33: 392–401. scape level, and not solely through designations of specific Gaston, K. J. 1996. Species-range-size distributions: patterns, protection sites. mechanisms and implications. – Trends Ecol. Evol. 11: 197–201. 7-EV Gaston, K. J. 2006. The structure and dynamics of geographic Lundkvist, E. et al. 2002. Dispersing diving beetles (Dytiscidae) in ranges. – Oxford Univ. Press. agricultural and urban landscapes in southeastern Sweden. Gove, A. D. et  al. 2009. Dispersal traits linked to range size – Ann. Zool. Fenn. 39: 109–123. through range location, not dispersal ability, in Western Mora, C. and Robertson, D. R. 2005. Factors shaping the range- Australian angiosperms. – Global Ecol. Biogeogr. 18: size frequency distribution of the endemic fish fauna of 596–606. the tropical eastern Pacific. – J. Biogeogr. 32: 277–286. Grueber, C. E. et  al. 2011. Multimodel inference in ecology Mora, C. et al. 2011. High connectivity among habitats precludes and evolution: challenges and solutions. – J. Evol. Biol. the relationship between dispersal and range size in tropical 24: 699–711. reef fishes. – Ecography 35: 89–96. Gutierrez, D. and Menendez, R. 1997. Patterns in the distribution, Nilsson, A. N. and Persson, S. 1989. The distribution of abundance and body size of carabid beetles (Coleoptera: predaceous water beetles (Coleoptera: Noteridae, Dytiscidae) Caraboidea) in relation to dispersal ability. – J. Biogeogr. in Sweden. – Entomol. Basiliensia 13: 59–146. 24: 903–914. Nilsson, A. N. and Holmen, M. 1995. The aquatic Adephaga Hegyi, G. and Garamszegi, L. Z. S. 2011. Using information (Coleoptera) of Fennoscandia and Denmark. ii. Dytiscidae. theory as a substitute for stepwise regression in ecology and – Fauna Entomol. Scand. 32. behavior. – Behav. Ecol. Sociobiol. 65: 69–76. Nilsson, A. N. and Svensson, B. W. 1995. Assemblages of Hendrich, L. and Balke, M. 2000. Verbreitung, Habitatbindung, dytiscid predators and culicid prey in relation to environmen- Gefährdung und mögliche Schutzmaßnahmen der ffh- tal factors in natural and clear-cut boreal swamp forest pools. Arten Dytiscus latissimus Linnaeus, 1758 (Der Breitrand) – Hydrobiologia 308: 183–196. un Graphoderus bilineatus (De Geer, 1774) in Deutschland Rannap, R. et al. 2009. Restoring ponds for amphibians: a success (Coleoptera: Dytiscidae). – Insecta Berlin 6: 98–114. story. – Hydrobiologia 634: 87–95. Hof, C. et al. 2011. Rethinking species’ ability to cope with rapid Richard, Y. and Armstrong, D. P. 2010. The importance of climate change. – Global Change Biol. 17: 2987–2990. integrating landscape ecology in habitat models: isolation- Holmen, M. 1993. Fredede insekter 1 Danmark. Del 3: biller driven occurrence of north island robins in a fragmented knyttet til vand. – Entomol. Medd. 61: 117–134. landscape. – Landscape Ecol. 25: 1363–1374. Johnson, J. B. and Omland, K. S. 2004. Model selection in ecology Robin, X. et al. 2011. pROC: an open-source package for R and and evolution. – Trends Ecol. Evol. 19: 101–108. S to analyze and compare ROC curves. – BMC Bioinform. Joly, P. et al. 2001. Habitat matrix effects on pond occupancy in 12: 77. newts. – Conserv. Biol. 15: 239–248. Sand-Jensen, K. and Staehr, P. A. 2009. Net heterotrophy in small Kehl, S. and Dettner, K. 2007. Flugfähigkeit der in Deutschland Danish lakes: a widespread feature over gradients in trophic vorkommenden adephagen Wasserkäfer (Coleoptera, state and land cover. – Ecosystems 12: 336–348. Hydradephaga). – Entomol. Heute 19: 141–161. Staehr, P. A. et  al. 2012. Lake metabolism scales with lake Keitt, T. H. et  al. 2002. Accounting for spatial pattern when morphometry and catchment conditions. – Aquat. Sci. 74: modeling organism–environment interactions. – Ecography 155–169. 25: 616–625. Stephens, P. A. et  al. 2007. Inference in ecology and evolution. Klečka, J. and Boukal, D. S. 2011. Lazy ecologist’s guide to water – Trends Ecol. Evol. 22: 192–197. beetle diversity: which sampling methods are the best? – Ecol. Symonds, M. R. E. and Moussalli, A. 2011. A brief guide to Indic. 11: 500–508. model selection, multimodel inference and model averaging Koese, B. and Cuppen, J. G. M. 2006. Sampling methods for in behavioural ecology using Akaike’s information criterion. Graphoderus bilineatus (Coleoptera: Dytiscidae). – Nederl. – Behav. Ecol. Sociobiol. 65: 13–21. Faun. Med. 24: 41–48. Taylor, P. D. et al. 1993. Connectivity is a vital element of land- Legendre, P. 1993. Spatial autocorrelation: problem or new scape structure. – Oikos 68: 571–573. paradigm? – Ecology 74: 1659–1673. Tischendorf, L. and Fahrig, L. 2000. On the usage and measure- Lester, S. E. et al. 2007. The relationship between dispersal ability ment of landscape connectivity. – Oikos 90: 7–19. and geographic range size. – Ecol. Lett. 10: 745–758. Vermeulen, H. J. W. 1994. Corridor function of a road verge Lowe, W. H. and Bolger, D. T. 2002. Local and landscape- for dispersal of stenotopic heathland ground beetles carabidae. scale predictors of salamander abundance in New Hampshire – Biol. Conserv. 69: 339–349. headwater streams. – Conserv. Biol. 16: 183–193. Wetzel, R. G. 1983. Limnology. – Saunders. Lowry, E. and Lester, S. E. 2006. The biogeography of plant Whittingham, M. J. et  al. 2006. Why do we still use stepwise reproduction: potential determinants of species’ range sizes. modeling in ecology and behavior? – J. Anim. Ecol. 75: – J. Biogeogr. 33: 1975–1982. 1182–1189. Supplementary material (Appendix ECOG-00001 at  www.oikosoffice.lu.se/appendix ). Appendix 1. 8-EV 

The green heart of Africa is a blind spot in herpetology By Jos Kielgast & Stefan Lötters T he world’s second largest For a few years now we have aimed to continuous tropical rain forest contribute towards filling this blind is found in the Congo Basin. spot - improving the understanding of It comprises a vast river drainage the evolution, ecology and systematics area intriguingly similar to that of the of amphibians in the Congo Basin. Amazon and includes more than 15 of We have been involved in different the global WWF terrestrial ecoregions projects in cooperation with a number (Olson et al. 2001; de Wasseige et al. of African and European institutions 2009). Fascinatingly, this notable area including the Belgian Royal Institute for tropical biodiversity harbours one of of Science, the Belgian Royal Museum the least known herpetofaunas on our for Central Africa, Museum Alexander planet (Schiøtz 2006; Andreone et al. Fig. 1, Accumulated number of species described from Brazil (blue) Koenig and Max Planck Institute 2008). Most of the available knowledge and what is today The Democratic Republic of the Congo (red) for Evolutionary Anthropology derives from expeditions in the (Germany), Centre de Recherche beginning of last century and en Sciences Naturelles (DRC), a few prominent collections and Universities of Libreville during the colonial regimes (e.g. (Gabon), Kisangani and Boulenger 1919; Noble 1924; Ahl Kinshasa (DRC). We regard 1931; Laurent 1943, 1950, 1972). faunistic surveys a basic step Comparing the accumulated towards any other question. So number of described species far we have been engaged in through time from the DRC collecting and identifying species with that of Brazil (The majority in Monts Cristal (Gabon), the of the Congo and Amazon Ruwenzori Mountains (Uganda), basins respectively) provide Fig. 2 Amphibian species richness according to the Global Amphibian Assessment Salonga National Park and a (IUCN). Warmer colours indicate higher number of species. The Amazon and an illustration of this (fig. 1). Congo river drainages are outlined in black. stretch of more than 350km of A clearly exponential increase the Congo River (DRC). One of in species numbers since the our focal groups is reed frogs, fifties as observed for Brazil (as well as on global scale) has failed genus Hyperolius. Strictly employing an integrative taxonomic to appear in the DRC. This distinct stagnation in the rate of new approach has revealed several conspicuous new species for the species described may be interpreted as actually having described area as H. veithi (Fig. 3) from central DRC (Schick et al. 2010), everything there. However, a more plausible explanation is among others yet un-described (Fig.4). However, a major part of that the region has simply been out of reach for science due to working in this region consists of understanding names coined poor infrastructure and an unpredictable security situation. with poor descriptions based on single or few specimens only ever Further considering the Congo basin in the context of all current seen in preservation. An example is Hyperolius sankuruensis knowledge of amphibian biodiversity it stands out as a rather (Froglog 95, p.15) which we have now collected on multiple evident research gap in Afrotropical biodiversity (Fig.2). localities. A recent finding of this taxon initially led to the believe Fig. 3 Hyperolius veithi. Photo: Jos Kielgast Fig. 4 Hyperolius sp. Photo: Jos Kielgast 16 | FrogLog Vol. 97 | July 2011 that it was new to science (Schiøtz Literature Cited Ahl, E., 1931, Zur Systematik der 2006) until 2 specimens collected afrikanischen Baumfroschgattung Hyperolius more than 50 years ago (Laurent (Amph. Anur.), Mitteilungen des zoologischen Museums Berlin, 17, 1-132. 1979) were examined in the Royal Andreone, F., et al. (2008) Amphibians Museum for Central Africa and the of the Afrotropical realm. In: Threatened amphibians of the world. Stuart, SN, identity clarified. In other species Hoffmann, M., Chanson, JS, Cox, NA, Berridge, RJ, Ramani, P. & Young, BE (Eds). all type material has presumably Barcelona: Lynx Edicions, in association been lost and the difficult decision with IUCN, Conservation International and NatureServe. remains of coining new names or Boulenger GA (1919) Batraciens et reptiles designating neotypes based on recueillis par le Dr. C. Christy au Congo Belge dans les Districts de Stanleyville, Haut-Uelé et poor descriptions. Hyperolius Ituri en 1912-1914. Revue Zoologique Africaine 7:1-29 cf. brachiofasciatus (fig 5) may De Wasseige, C., D. Devers, P. de Marcken, be considered such an example. R. Eba’a Atyi, R. Nasi & P. Mayaux (2009): The forests of the Congo Basin - state of the forest Strikingly this species appear to be Fig. 5 Hyperolius cf. brachiofasciatus. Photo: Jos Kielgast 2008. Publications Office of the European Union, Luxembourg. both common and widespread in the Laurent, R. F. 1943. Les Hyperolius entire Congo Basin - rather thought- (Batraciens) du Musee Congo. Annales du Musée Royal du Congo Belge. Sciences Zoologiques. Tervuren 4: 61-140. provoking that this is the first ever published photograph of such a Laurent R (1950) Exploration du Parc national Albert, Mission G. F. De Witte distinct species although it is calling loudly to be noticed all along (1933-1935). Genres Afrixalus et Hyperolius (Amphibia Salientia). Institute des Parcs Nationaux du Congo Belge 64:1-120, 127 plates the Congo River. Laurent RF (1972) Amphibiens, Exploration du Parc National des Virunga, Deuxième Série, 22:1-125, L’Institut National pour la Conservation de la Nature de la We call for a collaborative effort in exploring this neglected spot Republic du Zaire, Bruxelles. in global herpetological biodiversity. Even the most basic of Laurent, R. F. 1979. Description de deux Hyperolius nouveaux du Sankuru (Zaïre) (Amphibia, Hyperoliidae). Revue de Zoologie et de Botanique Africaines. Tervuren knowledge is missing and the first steps towards understanding 93: 779-791. the fauna of the region are difficult. However, the exploration of Noble, G. K. (1924) Contributions to the herpetology of the Belgian Congo based on the collection of the American Museum Congo Expedition, 1909-1915. Part III. patterns and processes of amphibian speciation, biogeography and Amphibia. Bulletin of the American Museum of Natural History 49:147-347. Olson, D. M, E. Dinerstein, K. E. Wikramanaya, N.D. Burgess, W. D. N. Powell, ecology in the Congo Basin is sure to provide important insights E.C. Underwood, J.A. D’amico, I. Itoua, H. Strand, J.C. Morrison, C.J. Loucks, T.F. and novel understanding of the African herpetofauna. Allnutt, T. Ricketts, K. Kura, J.F. Lamoreux, W.W. Wettengel, P. Hedao & K.R. Kassem (2001) Terrestrial ecoregions of the world: a new map of life on earth. Bioscience 51(11):933-938. Schick, S., J. Kielgast, D. Rödder, V. Muchai, M. Burger & S. Lötters (2010): New Author details: Jos Kielgast1 and Stefan Lötters2. 1 University species of reed frog from the Congo basin with discussion of paraphyly in Cinnamon- of Copenhagen, Natural History Museum of Denmark, 2100 belly reed frogs. Zootaxa, 2501: 23-36. Copenhagen, Denmark. joskielgast@hotmail.com. 2 Trier Schiøtz, A. (2006) Notes on the genus Hyperolius (Anura, Hyperoliidae) in central University, Biogeography Department, 54286 Trier, Germany République Democratique du Congo. Alytes 24:40-60 loetters@uni-trier.de. Amphibians as indicators for the restoration of degraded tropical forests By Noor de Laat T he Upper Guinean forests in West Africa are known for persist in logged or fragmented forests (Ernst & Rödel, 2005). their high number of endemic animal and plant species Therefore, amphibians could serve as meaningful indicators for (Brooks et al. 2001), but face ongoing biodiversity-loss the restoration of degraded forests and could be a crucial part of a through land conversion and forest degradation (Poorter et al. biodiversity monitoring program, providing important directions 2004; Ernst & Rödel 2005; Ernst et al. 2006; McCullough et al. for conservation. 2007; Hillers et al. 2008). Except for combating continued forest loss, we should not forget that many plant and animal species To gather information on recovery of the species assemblage in will return to regenerating forests at some stage. In a review of secondary forests, I designed a monitoring program for plants 39 published studies, Dunn (2004) concluded that after shifting and amphibians. A first evaluation of the system was conducted cultivation it takes 20-40 years for species richness to recover. in the Ashanti Region of Ghana (West Africa). This project was Unfortunately most studies have been carried out at small spatial conducted in collaboration with FORM Ghana Ltd, a company and temporal scales and they often lack replication (Gardner, that started reforestation projects on a previously highly degraded 2010). This situation restricts our ability to make predictions Forest Reserve seven years ago. Together with local employees about species richness recovery and species conservation in of FORM Ghana I extensively searched along small rivers and in secondary forests. swampy areas for frog and toad specimens. We visited nine such possible microhabitats during one week in November 2010, at the Amphibians are very sensitive to habitat degradation (e.g., Wake end of the raining season. I assumed that differences in results 1991; Blaustein et al. 1994; Ernst et al. 2006). Many endemic caused by variation in sampling technique and effort between and range-restricted forest frogs are for example unable to collectors were smoothed out over time. FrogLog Vol. 97 | July 2011 | 17 

Detection of a Diverse Marine Fish Fauna Using Environmental DNA from Seawater Samples Philip Francis Thomsen1*., Jos Kielgast1., Lars Lønsmann Iversen2, Peter Rask Møller3, Morten Rasmussen1, Eske Willerslev1* 1 Centre for GeoGenetics, Natural History Museum of Denmark, University of Copenhagen, Øster Voldgade, Copenhagen, Denmark, 2 Freshwater Biology Section, Department of Biology, University of Copenhagen, Helsingørgade, Hillerød, Denmark, 3 Vertebrate Department, Natural History Museum of Denmark, University of Copenhagen, Universitetsparken, Copenhagen, Denmark Abstract Marine ecosystems worldwide are under threat with many fish species and populations suffering from human over- exploitation. This is greatly impacting global biodiversity, economy and human health. Intriguingly, marine fish are largely surveyed using selective and invasive methods, which are mostly limited to commercial species, and restricted to particular areas with favourable conditions. Furthermore, misidentification of species represents a major problem. Here, we investigate the potential of using metabarcoding of environmental DNA (eDNA) obtained directly from seawater samples to account for marine fish biodiversity. This eDNA approach has recently been used successfully in freshwater environments, but never in marine settings. We isolate eDNA from K-litre seawater samples collected in a temperate marine ecosystem in Denmark. Using next-generation DNA sequencing of PCR amplicons, we obtain eDNA from 15 different fish species, including both important consumption species, as well as species rarely or never recorded by conventional monitoring. We also detect eDNA from a rare vagrant species in the area; European pilchard (Sardina pilchardus). Additionally, we detect four bird species. Records in national databases confirmed the occurrence of all detected species. To investigate the efficiency of the eDNA approach, we compared its performance with 9 methods conventionally used in marine fish surveys. Promisingly, eDNA covered the fish diversity better than or equal to any of the applied conventional methods. Our study demonstrates that even small samples of seawater contain eDNA from a wide range of local fish species. Finally, in order to examine the potential dispersal of eDNA in oceans, we performed an experiment addressing eDNA degradation in seawater, which shows that even small (100-bp) eDNA fragments degrades beyond detectability within days. Although further studies are needed to validate the eDNA approach in varying environmental conditions, our findings provide a strong proof-of- concept with great perspectives for future monitoring of marine biodiversity and resources. Citation: Thomsen PF, Kielgast J, Iversen LL, Møller PR, Rasmussen M, et al. (2012) Detection of a Diverse Marine Fish Fauna Using Environmental DNA from Seawater Samples. PLoS ONE 7(8): e41732. doi:10.1371/journal.pone.0041732 Editor: Senjie Lin, University of Connecticut, United States of America Received April 5, 2012; Accepted June 25, 2012; Published August 29, 2012 Copyright: ß 2012 Thomsen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This study was funded as basic research by the Danish National Research Foundation and the Aage V. Jensen Charity Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: ewillerslev@snm.ku.dk (EW); pfthomsen@snm.ku.dk (PFT) . These authors contributed equally to this work. Introduction An alternative approach for monitoring marine fish is that of environmental DNA (eDNA), i.e. the extraction and analysis of The marine environment represents considerable value in terms genetic material obtained directly from environmental samples of biodiversity [1] and economics through fisheries and other [13]. For macro-organisms, the approach was first applied to products derived from the sea [2,3]. Fish are the most species-rich terrestrial sediment samples revealing ecosystems of extinct and group of vertebrates and constitute a keystone in present-day extant mammals, birds, and plants [14]. Later the same approach monitoring of environmental health of marine ecosystems. was successfully used on ancient cave sediments [15] and ice cores Nevertheless, fish species and populations worldwide are under [16] as well as ancient and contemporary sediments across a threat and suffer from over-exploitation [4–7] with considerable variety of taxa, habitats and climates [17–28]. Recently, eDNA impact on human health [8]. Contemporary monitoring of marine from Bull frogs was successfully retrieved from contemporary pond fish biodiversity and resources is largely dependent on invasive and water samples [29]. This approach has since been used to detect selective methods, such as bottom trawls and rotenone poisoning other amphibians [30] and invasive fish species [31] in freshwater. [9], which can only be carried out in particular areas where Furthermore, it has been demonstrated that rare and endangered conditions are favourable. Furthermore, correct identification of freshwater insects, crustaceans, amphibians, fish and mammals many species across both non-commercial (e.g. Syngnathidae) and can be monitored and quantified using eDNA, and that such an commercial (e.g. Ammodytidae) groups is problematic using approach can account for entire lake faunas [32]. Despite these traditional methods; leaving databases flawed with errors [10] successful applications, the detection of macro-organisms by and checklists incomplete [11,12]. PLOS ONE | www.plosone.org 1 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples eDNA has to our knowledge never been reported from marine from whales can be obtained from seawater [37]. Targeting DNA water samples. from macro-organisms in environmental water samples is not In this study we present the first recording of marine fish comparable to targeting microbial organisms, as the former is biodiversity using eDNA from seawater samples. present only as true eDNA (cellular debris or free DNA), whereas the latter may be detected by DNA deriving from whole, living Results organisms present in the water samples. The fish eDNA detected in this study most likely derives from intestinal cells, sloughed skin, Three seawater samples were collected in a temperate marine scales or mucus and may consists of both free DNA, cellular debris ecosystem in Denmark (Fig. 1). Samples were filtered, DNA and particle bound DNA. Animal cells exposed to the environ- amplified and sequenced (see Materials and Methods section). A ment will quickly undergo lyses, but the specific source and relative comparison with the GenBank sequence database revealed DNA ratio of cellular bound and free DNA is mostly studied in soil and from 15 different fish species, representing a diversity of 9 orders sediment samples (e.g. [38,39,40]), but remains unclear in aquatic and 11 families (Fig. 1, Table 1). These include both important environmental samples. The filter matrix size of 0.45 mm, used in consumption species, such as Atlantic cod (Gadus morhua), this study and [30] as well as 1.5 mm filters [31] and even 3.0 mm European eel (Anguilla anguilla), European plaice (Pleuronectes platessa) filters [41] have been used previously to isolate eDNA from and Atlantic herring (Clupea harengus), as well as non-commercial freshwater. Considering the size of a DNA molecule, it is thus species like Goldsinny-wrasse (Ctenolabrus rupestris), Shorthorn quite likely that some of the detected eDNA is particle or cellular sculpin (Myoxocephalus scorpius) and Greater pipefish (Syngnathus bound. acus). We also detected DNA from European pilchard (Sardina Despite recent successful applications of eDNA detection in pilchardus) – a vagrant fish species in the region – and 4 species of freshwater systems [29–32,41] we find it surprising how well the birds, including the Red-throated loon (Gavia stellata), which only approach performs on marine water samples considering: i) the passes the area occasionally during migration. There was a small larger water-volume to biomass ratio of marine ecosystems difference in the species composition obtained by eDNA from the compared to that of freshwater, ii) the effects of sea-currents and three samples, with more species on the outer pier, compared to wave action, and iii) the impact of salinity on the preservation and inner pier and open beach (Fig. 1). extraction of eDNA. These factors likely mean that eDNA in As a comparison to the eDNA metabarcoding method, we marine water is much less concentrated, more quickly dispersed, conducted expert surveys in the same area using 9 different and may be less efficiently extracted from the water column. Still, conventional methods, which yielded varying coverage of fish our data reveals that marine water samples of just K litres yield species diversity (Fig. 2, Table S1). Among the conventional eDNA from a variety of fish taxa, ranging from highly abundant surveillance methods, fish pots performed least efficient by species, such as the European plaice, to the rarely recorded uncovering on average only 4.3 fish species per sampling event, vagrant species; European pilchard (Fig. 1). We found a small whereas night-snorkeling and bottom trawl performed the best, by difference in the fish species compositions recovered by eDNA detecting an average of 14.7 and 13.3 species, respectively. from the three different sites sampled at a very localised scale However, all conventional methods were outperformed or (Fig. 1). However, it remains unclear whether these differences equalled by the eDNA approach finding 15 species. were due to stochasticity in PCR amplification, insufficient depth In order to address the potential dispersal of eDNA in oceans, of sequencing or a truly patchy occurrence of fish assemblages and we performed an experiment investigating eDNA degradation. A their eDNA in the environment. 50 L seawater sample was collected and frequently sub-sampled Importantly, when comparing results obtained with eDNA to for 15 days. Species-specific eDNA sequences were amplified by quantitative PCR (qPCR) for two target species (Gasterosteus those obtained from an array of 9 different conventional methods aculeatus and Platichthys flesus) showing initial concentrations of 48 used in fish surveys, the eDNA approach performed remarkably and 214 DNA molecules pr 400 ml seawater, respectively. well (Fig. 2). It should be noted that snorkeling, trawl and seine, Importantly, the results suggest that even very small (100-bp) which represents the methods with efficiencies closest to the eDNA eDNA fragments degrade beyond detectability within few (0.9– approach, are either heavily dependent on competent experts in 6.7) days (Fig. 3), (See also Material and Method section). The fish identification on-site (snorkeling), or only possible where detection threshold below which DNA could no longer be detected seabed conditions allow it (trawl and seine). Stratified randomized was near equivalent for both species (approximately 25 DNA bottom trawl surveys represent a cornerstone for marine molecules pr 400 ml water), indicating that this may be a rough monitoring in the framework of the International Council for general threshold for the applied method. Average concentration the Exploration of the Sea [3]. These surveys cannot be carried of DNA in the three samples used for sequencing, was also out in shallow waters, areas with rocks, reefs, kelp or other quantified for the two target taxa, yielding similar initial obstacles on the seabed, and are also difficult in areas with soft concentrations (446 and 215 molecules pr 400 ml seawater for sediment [42,43]. This leaves a bias in the way marine fish faunas Gasterosteus aculeatus and Platichthys flesus, respectively). are monitored today, excluding important areas for biodiversity All DNA extraction blanks and PCR controls performed during and fisheries. In contrast to many conventional methods, the the experiment turned out negative, leaving no indication of eDNA method can be performed in virtually any marine habitat, contamination. and require little expertise or effort in sampling. Additionally, the molecular identification is more confident and objective than visual identification of species, which is in many cases difficult even Discussion for experts. Conversely, DNA based species identification rely on While it has been widely demonstrated that microbial knowledge of species-specific sequences compiled by taxonomic (prokaryotic and eukaryotic) biodiversity can be studied by experts. However, global initiatives addressing this need have been sequencing DNA from filtered seawater samples (e.g. [33–36]), established, and databases are rapidly growing (http://www. we show here, for the first time, that seawater contain a high boldsystems.org). A specific initiative to provide DNA barcodes of density of detectable eDNA from macro-organisms, such as fish. all the world’s fish species was launched in 2005, and has today At the same time, it has now been demonstrated that also eDNA covered more than a third of all described species [44,45]. It is PLOS ONE | www.plosone.org 2 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples PLOS ONE | www.plosone.org 3 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples Figure 1. Summary of results showing sampling site and panel of fish species recovered by eDNA. Sampling locality (The Sound, Elsinore, Denmark) for this study with the three sampling sites; 1) open beach, 2) outer pier, 3) inner pier. The 15 different fish species obtained by eDNA in this study are shown with colour codes explaining in which of the three sampling sites they were found. All fish drawings by Susanne Weitemeyer ß. doi:10.1371/journal.pone.0041732.g001 clear that this remaining gap in knowledge will for some time sampled in this study), we estimate that eDNA could in this case impair the usefulness of eDNA monitoring in faunas where all travel between ca. 40 km–600 km in the oceans before degraded species have not yet been DNA barcoded. On the other hand, the beyond detectability. most complex and species-rich systems are also the most However, many other factors such as water temperature, wind challenging to monitor with conventional methods, and likely speed, wind direction and local changes in currents will have great where the advantage of eDNA will represent the most significant impact on the potential distance that eDNA can be transported in improvement for future monitoring. oceans. The average initial concentration of DNA molecules pr It is obvious that sea currents may move eDNA beyond the area 400 ml seawater in the three original collected samples used for where species actually occur, leaving the possibility for false sequencing, showed similar (446 vs. 48) or very similar (215 vs. 214) positive records. Also, fish predators such as birds, mammals or values as seen in the eDNA degradation experiment for Gasterosteus other fish species may distribute DNA from prey items across aculeatus and Platichthys flesus, respectively. Given the different time marine localities through defecation. Importantly however, our that the water samples were collected for the two purposes (October results from the fish eDNA degradation experiment convincingly 2011 vs. May 2012), it is obvious that seasonal and yearly variation show, that even small (100-bp) eDNA fragments in seawater as well as species phonology of G. aculeatus could easily account for persists for only a few days above detection threshold of the observed difference in eDNA for this species. approximately 25 molecules pr 400 ml seawater at 15uC (Fig. 3). Most importantly, as a consequence of continuous dilution, the In freshwater, the decay of eDNA beyond the threshold of probability of detecting eDNA in marine waters very likely detectability has been demonstrated to happen at a scale of days or decreases rapidly with distance to its source, making recovery of weeks [32,46]. Notably, however, DNA degradation in seawater eDNA of local origin much more plausible. Therefore, we feel has previously been suggested to be substantially faster with an convinced that eDNA obtained from marine water samples should empirical turnover rate as low as 10 hours [47], which supports represent only local fish fauna. This may also be the reason why our findings and indicate lower probability of long distance we do not detect any truly exotic species (i.e. species living in dispersal of eDNA in marine ecosystems. Using an approximate deeper waters, different salinity or different latitude). Apart from degradation time of eDNA beyond detectability of minimum the European pilchard, we only recovered eDNA from species 12 hours and maximum 1 week, and given a rough average speed resident to the area, suggesting that either there is no eDNA from of ocean currents of 1 m/sec (normal in the Sound of Elsinore, non-resident species present, or that such DNA is too dilute to be Table 1. Summary of species-specific eDNA sequences recovered in this study. Taxon Order Family Species Sequence (59-93) Fish Pleuronectiformes Pleuronectidae Pleuronectes platessa CCGCTCGTCACGCCGCCACACATCAAGCCAGAGTGATACT Pleuronectiformes Pleuronectidae Limanda limanda CCACTTGTTACACCCCCACATATCAAGCCCGAATGATATT Pleuronectiformes Pleuronectidae Platicthys flesus CCACTCGTCACGCCACCACATATTAAGCCAGAGTGATACT Perciformes Zoarcidae Zoarces viviparus CCACTAGTCACCCCACCCCACATCAAGCCCGAGTGGTACT Perciformes Labridae Ctenolabrus rupestris TCGTACTTATGGTGGTCCCCATCCTTCACACATCTA Perciformes Trachinidae Trachinus draco CCCCTAGTAACTCCTCCTCATATTAAGCCTGAATGATACT Anguilliformes Anguillidae Anguilla anguilla CCAATAGTTACTCCGCCACACATTAAGCCAGAGTGGTATT Salmoniformes Salmonidae Salmo trutta AACCCCCTAGTCACCCCACCTCATATCAAGCCCGAATGATACTTCCT Gadiformes Gadidae Gadus morhua CCCATCGTTACCCCACCTCATGTTAAGCCCGAATGATATT Gasterosteiformes Gasterosteidae Gasterosteus aculeatus CCATTAGTCACTCCACCTCACATCAAGCCTGAATGGTACT Gasterosteiformes Gasterosteidae Spinachia spinachia CCATTAATTACTCCTCCTCACATTAAACCTGAATGATATT Syngnathiformes Syngnathidae Syngnathus acus CCTTTAGTTACTCCTCCACATATCAAACCGGAATGATACT Clupeiformes Clupeidae Sardina pilchardus CCCATGGTTACCCCACCACACATTAAGCCGGAGTGATACT Clupeiformes Clupeidae Clupea harengus ATTCCGAACAAGTTGGGAGGAGTGCTTGCTCTCCTATTCTCAATT Scorpaeniformes Cottidae Myoxocephalus scorpius TAGATAACGCTACACTTACCCGCTTTTTTGCC Birds Gaviiformes Gaviidae Gavia stellata CCACTCGTTACACCCCCTCACATTAAGCCAGAGTGATACT Columbiformes Columbidae Columba livia CCTCTAGTTACACCTCCCCATATCAAACCAGAATGATACT Anseriformes Anatidae Cygnus olor AGTATCATTCTGGTTTAATGTGTGGAGGGGTTACTAGAGG Pelecaniformes Phalacrocoracidae Phalacrocorax carbo CTAAAAGACATCCTAGGTTTCACACTCCTACTCCTCCTCCTAACAACAATA All sequences are generated by pyrosequencing using Roche GS FLX 454 platform, except the 5 sequences obtained with species-specific primers (see Table 2), which are generated by cloning and subsequent Sanger sequencing. All sequences are full-length 100% match to the particular species only, identified by BLAST to the Genbank nucleotide database. Sequences are given without primers. doi:10.1371/journal.pone.0041732.t001 PLOS ONE | www.plosone.org 4 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples Figure 2. Number of fish species recorded by 9 different conventional survey methods and eDNA at The Sound of Elsinore, Denmark. Bars show mean number of fish species caught across surveys in 2009, 2010 and 2011 and error bars represent the standard deviation (see also Table S1). The eDNA bar represents the total amount of fish species recorded by this method in 2011. *) Depend heavily on competent experts in fish identification. **) Only possible where seabed conditions allow it. doi:10.1371/journal.pone.0041732.g002 picked up with the applied sampling procedure. The recovery of It also remains untested whether the amount of eDNA eDNA from the European pilchard, which is normally regarded as molecules in marine water reflects population sizes and/or a warm-temperate species, is somewhat surprising given that the biomass of the local fauna as seen is freshwater [32,41]. This species is only rarely sighted in the sampling area. However, this has large applications for monitoring of marine biodiversity and in species is getting increasingly more common in the northern North particular fisheries, where data beyond species presence is Sea and adjacent waters possibly due to warmer climate [48]. essential. Furthermore, it is a species that is easily overlooked by Another potential limitation for the eDNA approach is PCR conventional surveys due to similarity to common resident taxa. primer design. It is inherent to the use of generic primers that Therefore, we find it likely that the eDNA detection of European there is a trade-off between targeting higher taxonomic levels and pilchard is due to authentic occurrence of the species in the area, detecting rare sequences. Primer affinity bias leads to certain rather than eDNA originating far from the sampling site. The sequences (species) amplifying less efficiently than others, poten- recovery of eDNA from Red-throated loon (Gavia stellata) was also tially limiting the monitoring results to species, which are expected unexpected, but this finding could be authenticated by exact to be locally present and are therefore used in primer design, or in records in the national bird watching database (http://www. general simply to species-specific sequences with the best primer dofbasen.dk/) showing the species to be locally present at the time affinity. However, this limitation will continuously become less of sampling. We exclude the possibility of a laboratory crucial due to optimization and publication of primers for eDNA contamination, based on negative PCR controls, extraction blanks studies, as well as significant increase in sequencing depth and and since no work has ever been performed on the particular rapid advances in sequencing technology, some of which are species in the settings where this study was carried out. Hence, independent of initial PCR amplification. these findings illustrate how the eDNA approach may be useful in Regardless of many potential present limitations and a need for detecting unexpected species. more basic knowledge, the eDNA approach in marine environ- Despite our promising findings, it is important to emphasize ments have widespread perspectives in terms of biodiversity that a number of issues need to be thoroughly addressed before monitoring and fisheries. This study provides the first evidence eDNA can be considered a reliable tool for monitoring biodiversity that a very simple eDNA based survey may offer a coverage of in marine ecosystems. In particular the dispersal of eDNA in local marine fish faunas, which is comparatively better than, or at marine water must be better understood. This includes to what least as good as, any single conventional method used here. extent abiotic factors, like temperature and salinity, affect results. Importantly, we also demonstrate experimentally that eDNA Similarly, an understanding of the phenology, changing metabo- degrades rapidly in seawater, indicating that detectable DNA is lism and DNA excretion of target species may well have most likely of local origin. We believe that eDNA based surveys implications for the use of eDNA in monitoring. PLOS ONE | www.plosone.org 5 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples Figure 3. Results from eDNA degradation experiment. eDNA concentration in seawater as a function of time for the two fish species; Platichthys flesus (circles) and Gasterosteus aculeatus (triangles), investigated in a 50 l aquarium. Time points with no detection of eDNA signals are shown in red. The lines show simple exponential decay models, p,0.001 (Platichthys flesus) and p,0.05 (Gasterosteus aculeatus). Dashed line shows the suggested detection threshold of 25 DNA molecules pr 400 ml seawater. Estimated time for eDNA to degrade beyond the detection threshold was estimated to be 0.9 days for Gasterosteus aculeatus and 6.7 days for Platichthys flesus. See also Materials and Methods section. doi:10.1371/journal.pone.0041732.g003 may in the future fill an important gap in broad-scale monitoring pier and on open beach (Fig. 1). Samples were collected from of marine biodiversity and resources. surface water at depths of 1.5–6 m. Each sample was a pool of 30 sub-samples of each 50 ml collected along a 145 m transect, taking Materials and Methods one sub-sample every 5 m. All samples were immediately stored at 220uC until extracted. Sampling locality For the eDNA degradation experiment, a total of 50 l of The study was carried out at The Sound of Elsinore, Denmark seawater was collected as twenty-five 2-l samples May 16th 2012 in (56.04387uN, 12.61309uE) (Fig. 1). The Sound of Elsinore (outer pier and open beach), where the original samples, used for sequencing, were also collected. The Conventional fish surveys samples were pooled into a 54 l aquarium and an initial sub- Occurrence data of fish species in the study area was obtained in sample of 400 ml was taken within one hour after sampling (t = 0). late August in 2009, 2010 and 2011 by experiments led by fish The aquarium was set up to mimic natural conditions, kept at a expert PRM (Table S1). In order to find as many species as constant 15uC, with a 12-hour daylight cycle (standard household possible, a wide range of methods were applied each year: five fish 15 watt neon tube) and equipped with a circulation pump pots, two fyke-nets, one beach-seine (width 6 m) dragged for about powerhead (600 l/hour) ensuring full admixture and oxygenation. 100 m near shore, one multi-mesh gillnet (100*1.5 m, mesh sizes Subsamples of 400 ml water were taken from the aquarium at 6.5–110 mm), two hours of push netting (width 68 cm, mesh size close intervals (hours – days) from May 16th to May 31st 2012, and 8 mm), two hours of angling with lures, two hours of snorkeling all samples were immediately stored at 220uC until DNA during the day, two hours of snorkeling at night, and half an hour extraction. of bottom trawling (width 4 m, height 1,5 m, cod-end mesh size 10 mm) from R/V Ophelia. DNA extraction Permission for scientific fishing was provided by the Danish K litre of each of the three 1.5-litre seawater samples was Ministry for Food, Agriculture and Fishery (journal no. 2009- vacuum-filtered onto 47 mm diameter 0.45-mm pore size nylon 02530-23088). filters (Osmonics, Penang, Malaysia). Immediately after, DNA was extracted from the filters using bead beating and Qiagen DNeasy Water sampling Blood & Tissue Kit (using spin-column protocol). Filters were Three 1.5-litre seawater samples were collected on October 1st rolled up, cut into ca. 1 mm slices and placed in 2 ml tubes. 0.3 g 2011. Samples were collected along the inner pier, along the outer of 0.5 mm Zirconia/Silica Beads (Biospec Products, Bartlesville, PLOS ONE | www.plosone.org 6 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples USA) and 720 ml ATL Buffer were added to each tube, which Table 2. Primers and probe details showing sequences, were then shaken in a Bead Beater 8 (Biospec Products, target taxa and fragment sizes. Bartlesville, USA) with 2800 oscillations/min for 45 sec. After this the tubes were incubated at 56uC for 30 min, followed by another beating and incubation step as above. Then 80 ml of Name Sequence (59-93) Target taxon Fragment Proteinase K were added to each tube followed by a final Fish2bCBR GATGGCGTAGGCAAACAAGA Fish 80 incubation step at 56uC for 2 hours with agitation. Samples were then vortexed for 15 sec and spun for 1 min (6000 g). Each Fish2CBL ACAACTTCACCCCTGCAAAC supernatant (600 ml) was transferred into new 2 ml tubes. Fish2degCBL ACAACTTCACCCCTGCRAAY Fish 80 Hereafter the Qiagen DNeasy Blood & Tissue Kit (manufactures Fish2CBR GATGGCGTAGGCAAATAGGA protocol) was followed for the remaining part of the DNA ClupeaCBL CATACGCCATTCTTCGATCA Clupea harengus 85 extraction, with the following minor adjustments; 600 ml AL ClupeaCBR GGAACAAGCAGAAGGACCAG Buffer, 600 ml Ethanol, and final elution steps of 2650 ml AE MyoxoCBL GATCTGAGGCGGTTTCTCAG Myoxocephalus 72 Buffer for each sample. scorpius Extraction of seawater samples for the eDNA degradation MyoxoCBR AAGGGGAAAAGGAAGTGGAA experiment was performed as above on a total of nineteen 400 ml water samples. SalmoCBL CGGACAATTTTACGCCTGCC Salmo trutta 87 SalmoCBR GAAGGATTGCGTAGGCGAAT PCR amplification LabrusCBL CGCCCTCCTATCCTCTATCC Ctenolabrus 76 For PCR, two generic and four species-specific primer sets were rupestris developed to target small (,100 bp) fragments of the mitochon- LabrusCBR GAAGGTGATGCTCCGTTGTT drial gene cytochrome b (cytb) in fish (Table 2). Part of the cytb gene TrachuCBL CGTTCCACCCATACTTCTCC Phalacrocorax 92 was used, since GenBank had the best coverage of the local fish carbo fauna for this genetic region. This gene has been used successfully TrachuCBR AAGGTTTGGGGAAAATAGTGC for a similar approach in previous studies [32]. The four species- GaacCBL ACGCCACCTTAACACGTTTC Gasterosteus 101 specific primers were applied since PCR using generic primers on aculeatus DNA extracted from fresh tissue, showed less efficient amplifica- GaacCBR AGAGCCTGTCTGGTGAAGGA tion on these particular species, which are known to occur in the Gaac.probe CTGGTGCCACACTTGTTCAC area. 25 ml PCR reactions were performed using 2 ml DNA PlflCBL CCGCAACAGTGATTCACCTA Platichthys flesus 104 extract, 10 ml TaqManH Environmental Master Mix 2.0 (Life Technologies), 1 ml of each primer (10 mM) and 11 ml ddH2O PlflCBR TGTGAAGTAGGGGTGGAAGG under thermal conditions: 95uC for 7 min., followed by 50 cycles PlflCB.probe CCACGAAACGGGCTCAAACA of 94uC for 30 sec., 50–60uC for 30 sec. and 72uC for 20 sec. Fragment sizes are given in base pairs including primers. All primers were completed with a final 72uC for 5 min. PCR products were designed for this study and amplify part of the Cytochrome b (cyt-b) gene. All verified on 2% agarose gels stained with GelRedTM, and purified regular PCRs were performed at 50uC annealing temperature and all qPCRs at using a Qiagen MinElute PCR purification kit or using e-gel 60uC annealing temperature. Probes are Minor Groove Binding (MGB) probes sizeselect 2% (Invitrogen, Life technologies, Denmark). Through- and have the modifications; 59: 6-Fam (D-L-Probe), 39: BHQ-1. doi:10.1371/journal.pone.0041732.t002 out the study we used separate laboratories for pre- and post-PCR procedures, and employed rigorous controls to monitor contam- ination including DNA extraction blanks and PCR blanks. Sequence Identification Extracted sequences (trimmed for primers) were compared with 454 pyrosequencing GenBank Nucleotide database using BLAST [49]. Taxon A total of six samples, each representing a pool of 8 PCR identification was made using MEGAN 4 [50], with following replicates with one of the two generic primer sets performed on LCA settings: Min. Support = 2, Min. Score = 50, Top Per- DNA extracts from each of the three samples, were sequenced cent = 2. using Roche GS FLX 454 pyrosequencing. Library builds on the Only sequences with full-length 100% match to a single species six samples were carried out using custom Y-shaped adaptors with were considered. MID barcode identifiers, and all reactions were performed according to protocol using NEBnext DNA Sample Prep Master Quantitative PCR (qPCR) Mix Set 2 (New England Biolabs, Ipswich, MA). Sequencing was For the eDNA degradation experiment, TaqMan qPCRs were carried out in accordance with manufacturer’s guidelines. A total performed on a Stratagene Mx3000P. of 20,315 sequences were generated on one-half of an XLR70 Two species-specific sets of primers and TaqMan minor groove PTP (Roche, Basel, Switzerland). Sequence files were sorted into binding (MGB) probes were developed to target small (101– separate files, by MID and primer pair, allowing 0 mismatches in 104 bp) fragments of the mitochondrial gene cytochrome b (cytb) in the MID and up to 2 in each primer. Platichthys flesus and Gasterosteus aculeatus, respectively (Table 2). The Sequences from pyrosequencing are uploaded to NCBI SRA: cytb gene was used again for the reasons given above. 25 ml qPCR ERP001563. reactions were performed using 2 ml DNA extract, 10 ml TaqManH Environmental Master Mix 2.0 (Life Technologies), Cloning and Sanger sequencing 1 ml of each primer (10 mM), 1 ml probe (2.5 mM) and 10 ml PCR products from amplifications using species-specific primers ddH2O under thermal cycling: 50uC for 5 min and 95uC for (see Table 2) were purified as above, cloned using Topo TA 10 min, followed by 55 cycles of 95uC for 30 sec and 60uC for cloning kit (Invitrogen), and commercially sequenced (Macrogen, 1 min. Europe). PLOS ONE | www.plosone.org 7 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples The primer/probe systems were both validated, and tested R, resulting in the values N0 = 214 and b = 0.322, for Platichthys negative on a total of 20 common saltwater fish species occurring flesus and N0 = 48 and b = 0.701 for Gasterosteus aculeatus. We find a in the area of sampling, and tested positive on the respective target highly significant (p,0.001) or significant (p,0.05) fit to the decay species. To enable a clear quantitative interpretation of eDNA models for Platichthys flesus and Gasterosteus aculeatus, respectively. degradation, we applied species-specific TaqMan systems, which, Using the parameters to calculate t for N(t) = 25 (i.e. the unlike generic primers, ensures that only eDNA of the two selected empirically observed detection threshold), suggests that eDNA target taxa were amplified. qPCR standards were prepared as a will degrade to sub-detectable levels after approximately 6.7 days dilution series (1026–10210) of purified PCR products on tissue for Platichthys flesus and 0.9 days for Gasterosteus aculeatus, in case of derived DNA with concentration measured on a Qubit fluorom- the observed initial DNA concentrations. eter (Invitrogen). All statistics were performed in R ver. 2.13.1. Each sample was replicated in 8 independent qPCR reactions, and all positive amplifications were used in the estimation of DNA Supporting Information concentrations. Final concentrations in DNA molecules pr 400 ml seawater sample were calculated from the standards, setting the Table S1 Species list and details for conventional fish molecular weight of DNA to 660 g/mol/base-pair. surveys. Efficiency of all qPCR standard curves was 90–100%. (PDF) eDNA decay model for seawater Acknowledgments An exponential decay model was fitted to the qPCR data, as this We thank the Danish National Research Foundation and the Aage V. is the relationship one would expect for molecular decay also used Jensen Charity Foundation for economic support, and Pernille Selmer previously for similar purposes [51]. Olsen, Lillian A. Petersen and the Danish National Sequencing centre for The model is the following: help with laboratory work and sequencing. We thank Johan Wedel Nielsen (AquaMind) for helpful input. We also thank Marcus Krag, Jakob Larsen, dN Thomas Christensen and students from the University of Copenhagen for ~{bN help with conventional fish surveys. dt Solving this gives: Author Contributions Conceived and designed the experiments: PFT JK EW. Performed the N(t)~N0 e{bt experiments: PFT JK PRM MR. Analyzed the data: PFT JK LLI PRM MR. Contributed reagents/materials/analysis tools: PFT JK LLI PRM N(t) is the DNA concentration at time = t days. MR. Wrote the paper: PFT JK PRM EW. The two parameters N0 (initial DNA concentration at time t = 0) and b (decay constant) were estimated by the nls function in References 1. Nielsen C (2012) Animal evolution: interrelationships of the living phyla. 16. Willerslev E, Cappellini E, Boomsma W, Nielsen R, Hebsgaard MB, et al. (2007) Oxford: Oxford University Press. 464 p. Ancient biomolecules from deep ice cores reveal a forested Southern Greenland. 2. Costanza R (1999) The ecological, economic, and social importance of the Science 317: 111–114. oceans. Ecological Economics 31: 199–213. 17. Willerslev E, Hansen AJ, Poinar HN (2004) Isolation of nucleic acids and 3. FAO (2010) The State of World Fisheries and Aquaculture 2010. Rome: FAO. cultures from fossil ice and permafrost. Trends in Ecology & Evolution 19: 141– 197 p. 147. 4. Hutchings JA (2000) Collapse and recovery of marine fishes. Nature 406: 882– 18. Taberlet P, Coissac E, Pompanon F, Gielly L, Miquel C, et al. (2007) Power and 885. limitations of the chloroplast trnL (UAA) intron for plant DNA barcoding. 5. Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, et al. (2001) Nucleic Acids Research 35: e14. Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 19. Haile J, Holdaway R, Oliver K, Bunce M, Gilbert MTP, et al. (2007) Ancient 629–637. DNA chronology within sediment deposits: Are paleobiological reconstructions 6. Pauly D, Christensen V, Guanette S, Pitcher TJ, Sumaila UR, et al. (2002) possible and is DNA leaching a factor? Molecular Biology and Evolution 24: Towards sustainability in world fisheries. Nature 418: 689–695. 982–989. 7. Collette B, Carpenter K, Polidoro B, Juan-Jordai M, Boustany A, et al. (2011) 20. Thomsen PF, Elias S, Gilbert MTP, Haile J, Munch K, et al. (2009) Non- High Value and Long Life – Double Jeopardy for Tunas and Billfishes. Science Destructive Sampling of Ancient Insect DNA. PloS ONE 4(4): e5048. 333: 291–292. 21. Haile J, Froese DG, MacPhee RDE, Roberts RG, Arnold LJ, et al. (2009) 8. Worm B, Barbier EB, Beaumont N, Duffy JE, Folke C, et al. (2006) Impacts of Ancient DNA reveals late survival of mammoth and horse in interior Alaska. biodiversity loss on ocean ecosystem services. Science 314: 787–790. Proceedings of the National Academy of Sciences of the United States of 9. Baldwin CC, Collette BB, Parenti LR, Smith DG, Springer VG (1996) America 106: 22352–22357. Collecting Fishes. Pp. 11–33 in: Lang, M. A. and C. C. . Baldwin (eds.). Methods 22. Sonstebo JH, Gielly L, Brysting AK, Elven R, Edwards M, et al. (2010) Using and Techniques of Underwater Research. Proceedings of the 16th Annual next-generation sequencing for molecular reconstruction of past Arctic Scientific Diving Symposium, American Academy of Underwater Sciences. vegetation and climate. Molecular Ecology Resources 10: 1009–1018. 10. Daan N (2001) The IBTS database: a plea for quality control. ICES CM 3: 1–5. 23. Andersen K, Bird KL, Rasmussen M, Haile J, Breuning-Madsen H, et al. (2012) 11. Rogers S, Ellis J (2000) Changes in the demersal fish assemblages of British Meta-Barcoding of ‘Dirt’ DNA from Soil Reflects Vertebrate Biodiversity. coastal waters during the 20th century. ICES Journal of Marine Science: Journal Molecular Ecology 21: 1966–1979. du Conseil 57: 866–881. 24. Hiiesalu I, Opik M, Metsis M, Lilje L, Davison J, et al. (2012) Plant species 12. Love MS, Lenarz B, Snook L (2010) A survey of the reef fishes, purple richness belowground: higher richness and new patterns revealed by next- hydrocoral (Stylaster californicus), and marine debris of Farnsworth Bank, Santa generation sequencing. Molecular Ecology 21: 2004–2016. Catalina Island. Bulletin of Marine Science 86: 35–52. 25. Anderson-Carpenter LL, McLachlan JS, Jackson ST, Kuch M, Lumibao CY, 13. Taberlet P, Coissac E, Hajibabaei M, Rieseberg LH (2012) Environmental et al. (2011) Ancient DNA from lake sediments: Bridging the gap between DNA. Molecular Ecology 21, 1789–1793. paleoecology and genetics. BMC Evolutionary Biology 11: 30. 14. Willerslev E, Hansen AJ, Binladen J, Brand TB, Gilbert MTP, et al. (2003) 26. Jorgensen T, Kjaer K, Haile J, Rasmussen M, Boessenkool S, et al. (2012) Diverse plant and animal genetic records from Holocene and Pleistocene Islands in the ice: detecting past vegetation on Greenlandic nunataks using sediments. Science 300: 791–795. historical records and sedimentary ancient DNA Meta-barcoding. Molecular 15. Hofreiter M, Mead JI, Martin P, Poinar HN (2003) Molecular caving. Current Ecology 21: 1980–1988. Biology 13: 693–695. PLOS ONE | www.plosone.org 8 August 2012 | Volume 7 | Issue 8 | e41732  Fish eDNA from Seawater Samples 27. Jorgensen T, Haile J, Moller P, Andreev A, Boessenkool S, et al. (2012) A 39. Corinaldesi C, Danovaro R, Dell’Anno A (2005) Simultaneous recovery of comparative study of ancient sedimentary DNA, pollen and macrofossils from extracellular and intracellular DNA suitable for molecular studies from marine permafrost sediments of northern Siberia reveals long-term vegetational stability. sediments. Applied and Environmental Microbiology 71: 46–50. Molecular Ecology 21: 1989–2003. 40. Lydolph MC, Jacobsen J, Arctander P, Gilbert MTP, Gillichinsky DA, et al. 28. Yoccoz NG, Bra˚then NG, Gielly L, Haile J, Edwards ME, et al. (2012) DNA (2005) Beringian paleoecology inferred from permafrost-preserved fungal DNA. from soil mirrors plant functional and structural diversity. Molecular Ecology. Applied Environmental Microbiology 71:1012–1017. DOI: 10.1111/j.1365-294X.2012.05545.x 41. Takahara T, Minamoto T, Yamanaka H, Doi H, Kawabata Z (2012) Estimation 29. Ficetola GF, Miaud C, Pompanon F, Taberlet P (2008) Species detection using of Fish Biomass Using Environmental DNA. PloS ONE 7(4): e35868. environmental DNA from water samples. Biology Letters 4: 423–425. 42. ICES (2010) International Council for the Exploration of the Sea: Manual for 30. Goldberg CS, Pilliod DS, Arkle RS, Waits LP (2011) Molecular detection of the International Bottom Trawl Surveys, Revision VIII. Copenhagen, Denmark, vertebrates in stream water: a demonstration using rocky mountain tailed frogs 73pp. and IDAHO giant salamanders. PloS ONE 6(7): e22746. 43. ICES (2011) International Council for the Exploration of the Sea: Manual for 31. Jerde CL, Mahon AR, Chadderton WL, Lodge DM (2011) ‘‘Sight-unseen’’ the Baltic International Trawl Surveys. Copenhagen, Denmark, 73pp. detection of rare aquatic species using environmental DNA. Conservation 44. Ward R, Hanner R, Hebert P (2009) The campaign to DNA barcode all fishes, Letters 4: 150–157. FISH-BOL. Journal of Fish Biology 74: 329–356. 32. Thomsen PF, Kielgast J, Iversen LL, Wiuf C, Rasmussen M, et al. (2011) 45. Becker S, Hanner R, Steinke D (2011) Five years of FISH-BOL: brief status Monitoring endangered freshwater biodiversity using environmental DNA. Molecular Ecology 21: 2565–2573. report. Mitochondrial DNA 22, S1: 3–9. 33. Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, et al. (2004) 46. Dejean T, Valentini A, Duparc A, Pellier-Cuit Sp, Pompanon Fo, et al. (2011) Environmental genome shotgun sequencing of the Sargasso Sea. Science 304: Persistence of Environmental DNA in Freshwater Ecosystems. PloS ONE 6(8): 66–74. e23398. 34. Sogin ML, Morrison HG, Huber JA, Welch DM, Huse SM, et al. (2006) 47. Dell’Anno A, Corinaldesi C (2004) Degradation and turnover of extracellular Microbial diversity in the deep sea and the underexplored ‘‘rare biosphere’’. DNA in marine sediments: ecological and methodological considerations. Proceedings of the National Academy of Sciences 103: 12115–12121. Applied and Environmental Microbiology 70: 4384–4386. 35. Rusch DB, Halpern AL, Sutton G, Heidelberg KB, Williamson S, et al. (2007) 48. Beare D, Burns F, Jones E, Peach K, Portilla E, et al. (2004) An increase in the The Sorcerer II Global Ocean Sampling expedition: Northwest Atlantic through abundance of anchovies and sardines in the north-western North Sea since 1995. Eastern Tropical Pacific. Plos Biology 5: 398–431. Global Change Biology 10: 1209–1213. 36. Stoeck T, Bass D, Nebel M, Christen R, Jones M, et al. (2010) Multiple marker 49. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local parallel tag environmental DNA sequencing reveals a highly complex eukaryotic alignment search tool. Journal of Molecular Biology 215: 403–410. community in marine anoxic water. Molecular Ecology 19: 21–31. 50. Huson DH, Mitra S, Ruscheweyh HJ, Weber N, Schuster SC (2011) Integrative 37. Foote AD, Thomsen PF, Sveegaard S, Wahlberg M, Kielgast J, et al. (2012) analysis of environmental sequences using MEGAN4. Genome Research 21: Investigating the potential use of environmental DNA (eDNA) for genetic 1552–1560. monitoring of marine mammals. PloS ONE. 51. Schnell IB, Thomsen PF, Wilkinson N, Rasmussen M, Jensen LRD, et al. (2012) 38. Dell’Anno A, Corinaldesi C (2004) Degradation and turnover of extracellular Screening mammal biodiversity using DNA from leeches. Current Biology 22: DNA in marine sediments: ecological and methodological considerations. R262–R263. Applied and Environmental Microbiology 70: 4384–4386. PLOS ONE | www.plosone.org 9 August 2012 | Volume 7 | Issue 8 | e41732 

Short Communications Short Communications SALAMANDRA 45 3 170-171 Rheinbach, 20 August 2009 ISSN 0036-3375 Forest weaverbird nests utilized by foam-nest frogs (Rhacophoridae: Chiromantis) in Central Africa Jos Kielgast & Stefan Lötters Abstract. We report that the Afrotropical anuran Chiromantis rufescens may use empty forest weaver bird nests above water for deposition of foam-nests with eggs. Our observation was made in January 2008 at a temporary pond in primary rainforest of Salonga National Park, Democratic Republic of Con- go. To the best of our knowledge this is the first ever report of utilization of bird nests by amphibians. We expect that bird nests with their tube-like entrance were difficult to access for frogs and that the choice of this oviposition site was non-random. If so, it may be a response to strong egg predation, e.g. through primates, as known in C. rufescens. Key words. Amphibia, Anura, Chiromantis rufescens, Aves, Ploceidae, Malimbus nitens, predation, repro- duction, Democratic Republic of the Congo. Anuran amphibians have evolved a re- nests of a forest weaver bird, the blue-billed markable diversity of reproductive strate- malimbe (Malimbus nitens), contained re- gies. Among the approximately 5,600 spe- mains of foam-nest of C. rufescens (Fig. 2). cies known, currently 39 reproductive modes Malimbus nitens makes elaborately woven have been recognized (Haddad & Prado nests with relatively small entrance tubes, at- 2005, Wells 2007). In Chiromantis, a Pal- tached to thin branches or vines overhanging aeotropical rhacophorid genus, a female and water. We expect that the choice of these sites one to several males aggregate to produce a for foam-nest deposition by C. rufescens was sticky foam-nest attached to leaves, rocks, non-random, as nests seem hardly accessible branches tree trunks or thin branches above for the frogs (Fig. 2). water (Schiøtz 999, Wells 2007). This Chiromantis species are known to oppor- strategy, also known as anuran reproductive tunistically deposit foam-nests (e.g. Schiøtz mode 33 (Haddad & Prado 2005), may have 999, Rödel et al. 2002, Channing & How- developed to limit the risk of predation for aquatic egg and larval stages. We here report that the widespread sub-Saharan C. rufescens (Günther, “868” 869) (Fig. ) makes fac- ultative use of empty weaver bird nests for deposition of foam-nests. To the best of our knowledge this is the first ever report of utili- zation of bird nests by amphibians. Our observation was made in Salonga Na- tional Park, Democratic Republic of Congo. The study site was a temporary pond in pri- mary rainforest (02°45.6’ S; 20°22.7’ E) south of the Lokoro River. In late January 2008, when the breeding period for C. rufescens was about to end, still a number of foam- Fig. 1. Chiromantis rufescens from Salonga Nati- nests were present. In two cases, deserted onal Park. © 2009 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT) 170 http://www.salamandra-journal.com  Short Communications Fig. 2. Empty forest weaver bird nest with remains of a Chiromantis rufescens foam-nest. ell 2006). Although depositing foam-nests Permits to perform fieldwork and collections were above water is conceivably a strategy against issued by the Institut Congolais pour la Conserva- aquatic predators (Haddad & Prado 2005), tion de la Nature (ICCN), Kinshasa. Célio Had- easily accessible foam-nests may attract ter- dad and Mark-Oliver Rödel made helpful comments on the topic. restrial predators. Rödel et al. (2002) report- ed that in West African rainforest monkeys have learned to regularly search for and feed References on C. rufescens eggs. Salonga National Park has a highly diverse and dense primate fau- Channing, A. & K. M. Howell (2006): Amphibi- na which cannot be ruled out to similarly af- ans of East Africa. – Frankfurt am Main (Chi- fect the survival of C. rufescens eggs. If so, our maira). finding may suggest that this species at our Haddad, C. F. B. & C. P. A. Prado (2005): Repro- study site shows a trend to perform a more ductive modes in frogs and their unexpected selective – compared to Chiromantis popula- diversity in the Atlantic Forest of Brazil. – Bio- tions of the same or other species elsewhere – Science, 55: 207-27. choice of deposition site for foam-nests with Rödel, M.-O., F. Range, J.-T. Seppänen & R. Noë the goal to minimize predation risk. (2002): Caviar in the rain forest: monkeys as frog-spawn predators in Tai National Park, Ivory coast. – Journal of Tropical Ecology, 8: Acknowledgements 289-294. Schiøtz, A. (999): Treefrogs of Africa. – Frank- We are grateful to Barbara Fruth and Gott- furt am Main (Chimaira). fried Hohmann of the Max Planck Institut für Wells, K. (2007): The behaviour and ecology of Evolutionäre Anthropologie (Leipzig), who sup- amphibians. – Chicago (University of Chicago plied the logistic framework for our fieldwork. Press). Manuscript received: 5 May 2008 Authors’ addresses: Jos Kielgast, Zoological Museum of Copenhagen, Universitetsparken 16, 2100 Co- penhagen, Denmark; E-Mail: jkielgast@snm.ku.dk; Stefan Lötters, University of Trier, Biogeography Department, 54286 Trier, Germany; E-Mail: loetters@uni-trier.de. 171 

Herpetology Notes, volume 4: 091-092 (2011) (published online on 24 February 2011) Foraging acrobatics of Toxicodryas blandingii in the Democratic Republic of the Congo Zoltán T. Nagy1*, Zacharie Chifundera Kusamba2, Guy-Crispin Gembu Tungaluna3, Albert Lotana Lokasola4, Jonathan Kolby5 & Jos Kielgast6 The two African species of the mainly Asian Boiga Congo along the Congo River and its tributaries. In one complex (Squamata: Serpentes: Colubridae) are of the study sites close to the village of Lieki, along usually considered belonging to the genus Toxicodryas the Lomami river (coordinates: latitude N 0.693765°; Hallowell, 1857, with a name indicating that they are longitude E 24.200084°), a transect study was venomous, rear-fanged colubrids. Blanding’s tree snake, conducted using different trapping strategies to survey Toxicodryas blandingii (Hallowell, 1844) is a large (up the vertebrate fauna, as for example pitfalls with drift to 2.8 m in total length), elongated and agile snake, fences and mist nets. Among others, a bat net of 12 m while the powdered tree snake T. pulverulenta (Fischer, length (mesh size: 16 mm) was installed in secondary 1856) is a middle size snake (up to 1.25 m). rain forest above a small stream. On 30 May 2010 in the Toxicodryas blandingii is a common species inhabiting morning, an adult female of T. blandingii (field number wide regions from the West to the East in gallery and CRT-4182) was encountered alive in the bat net holding rain forests of sub-Saharan Africa. It is primarily a a specimen of Casinycteris argynnis (Golden short- nocturnal snake and may be found active from dusk palated fruit bat, field number: CRT-2059) in its mouth to dawn (Luiselli et al., 1998). The species is largely (Fig. 1). arboreal, and has been recorded more than 20 m above The noteworthy feature of this observation was the forest floor (Pitman, 1974). A wide range of prey the unlikely position of the snake: it was hanging items have been documented for this species including approximately in the middle of the net at a height of birds, arboreal lizards, chameleons, frogs, rodents and around 2 m above the ground (for the position and bats (e.g. Cansdale, 1961; Menzies, 1966; Wickler and dimensions of the net, see Fig. 2). It is unclear whether Uhrig, 1969; Greene, 1989; Luiselli et al., 1998; Akani this relatively large snake reached the bat trapped in the et al., 2008). However, T. blandingii has been suggested net by climbing the vertical pole and extremely thin wire to display an ontogenetic dietary change where juveniles of the mist net or by jumping from a surrounding tree. only feed on lizards of appropriate size but shift to a more Both explanations require rather acrobatic manoeuvring generalist diet from sub-adult stages mainly eating birds demonstrating the extreme motility of this snake species and mammals (Greene, 1989; Luiselli et al., 1998). while foraging. A similar observation was reported by In the framework of the interdisciplinary expedition “Boyekoli Ebale Congo 2010”, zoological inventories were carried out in the Democratic Republic of the 1 Royal Belgian Institute of Natural Sciences, Rue Vautier 29, 1000 Brussels, Belgium; e-mail: lustimaci@yahoo.com 2 Laboratoire d’Herpétologie, Département de Biologie, Centre de Recherche en Sciences Naturelles, CRSN, Lwiro, DR Congo; 3 Faculty of Science, University of Kisangani, DR Congo; 4 Vie Sauvage, Avenue Nguma 80, Kinshasa, DR Congo; 5 1671 Edmund Terrace, Union, NJ 07083, USA; 6 Natural History Museum of Denmark, University of Copen- Figure 1. An adult Toxicodryas blandingii found in a mist net hagen, Universitetsparken 15, 2100 København Ø, Denmark. holding a Casinycteris argynnis in its mouth. The inset picture *Corresponding author. shows the captured specimen. 92 Zoltán T. Nagy et al. Figure 2. The position of the bat net. Figure 3. Another Toxicodryas blandingii with its prey (probably a nestling of a bushshrike). Bräunlich & Böhme (1991) from Tanzania, where References another arboreal colubrid snake specimen (an Eastern Akani, G., Ebere, N., Luiselli, L., Eniang, E. (2008): Commu- vine snake, Thelotornis mossambicanus) was found to nity structure and ecology of snakes in fields of oil palm trees enter a mist net and to kill an Olive sunbird (Nectarinia (Elaeis guineensis) in the Niger Delta, southern Nigeria. Afr. olivacea) caught in the net. J. Ecol. 46: 500-506. Another specimen of T. blandingii was collected Bräunlich, A., Böhme, W. (1991): Ungewöhnlicher Beutefang close-by (coordinates: l����������������������������� atitude N 0.7117°; longitude eines Thelotornis capensis mossambicanus (Bocage, 1895). E 24.2286°���������������������������������������� ) with a bird nestling belonging to the Salamandra 27: 119- 121. bushshrikes (Aves: Malaconotidae: aff. Laniarius) Cansdale, G.S. (1961): West African Snakes, London, Longman. Greene, H.W. (1989): Ecological, evolutionary, and conservation in its stomach (Fig. 3). Our observations confirm the implications of feeding biology in Old World cat snakes, genus knowledge on the feeding ecology of this fascinating Boiga (Colubridae). Proc. Calif. Acad. Sci. 46: 193-207. colubrid. Luiselli, L., Akani, G.C., Barieenee, I.F. (1998): Observations on habitat, reproduction and feeding of Boiga blandingi (Colubri- Acknowledgements. We are grateful to Erik Verheyen dae) in south-eastern Nigeria. Amphibia-Reptilia 19: 430-436. (Brussels) for the organisation of the expedition, and we thank Menzies, J. (1966): The snakes of Sierra Leone. Copeia (1966): Kris Pannecoucke for photos. Jon Fjeldså (Copenhagen) kindly 169-179. assisted in the difficult task of identifying the partly digested bird Pitman, C.R.S. (1974): A guide to the snakes of Uganda. Codi- nestling. We also acknowledge the valuable comments of Ulrich cote, Wheldon and Wesley. Joger (Braunschweig) on the manuscript. Wickler, W., Uhrig, D. (1969): Verhalten und ökologische Nische der Gelbflügelfledermaus, Lavia frons (Geoffroy)(Chiroptera, Megadermatidae). Z. Tierpsychol. 26: 726-736. Accepted by Angelica Crottini 

SALAMANDRA 49(2) 109–113 30 June 2013 Correspondence ISSN 0036–3375 Correspondence 3D reconstruction of fang replacement in the venomous snakes Dendroaspis jamesoni (Elapidae) and Bitis arietans (Viperidae) Zoltán T. Nagy 1, Dominique Adriaens 2, Elin Pauwels 3, Luc Van Hoorebeke 3, Jos Kielgast 4, Chifundera Kusamba 5 & Kate Jackson 6 1) JEMU, Royal Belgian Institute of Natural Sciences, Rue Vautier 29, 1000 Brussels, Belgium 2) Evolutionary Morphology of Vertebrates & Zoology Museum, Ghent University, K. L. Ledeganckstraat 35, 9000 Gent, Belgium 3) UGCT, Department of Physics and Astronomy, Ghent University, Proeftuinstraat 86, 9000 Gent, Belgium 4) Natural History Museum of Denmark, University of Copenhagen, Universitetsparken 15, 2100 København, Denmark 5) Laboratoire d’Herpétologie, Département de Biologie, Centre de Recherche en Sciences Naturelles, CRSN, Lwiro, DR Congo 6) Department of Biology, Whitman College, 345 Boyer Ave, Walla Walla, WA 99362, U.S.A. Corresponding author: Zoltán T. Nagy, e-mail: lustimaci@yahoo.com Manuscript received: 22 March 2013 Venomous snakes use highly specialized teeth, so-called since two functional fangs on one side (i.e., both are firm- fangs, to kill living prey items. The evolution of these fangs ly set in neighbouring sockets) are observed very infre- and that of the associated venom-producing and -delivery quently. system has been the subject of continuous research (e.g., Here we report for the first time on the fang configura- Kochva 1978, Kardong 1982, Knight & Mindell 1994, tion at the stage around fang replacement, using CT-data Jackson 2003, Fry et al. 2008, Vonk et al. 2008). In addi- with 3D visualisation. During our field expeditions in the tion to the evolutionary origin of tubular fangs, their on- Democratic Republic of the Congo, two specimens of ven- togenetic formation has been studied as well (e.g., Tomes omous snakes were found where ‘double fangs’ were visible 1874, Bogert 1943, Klauber 1972, Lake & Trevor-Jones on one side of the maxilla. We use this term to describe 1987, Lake & Trevor-Jones 1995, Jackson 2002). While the stage during fang replacement when both fangs in both our knowledge has significantly increased thanks to the sockets are ‘out’ and clearly visible. However, we could not integrated evaluation of palaeontological, morphological, test whether the new fangs were fully functional, since the physiological, molecular and other data, some details still soft tissues and presence of venom in the venom canals remain unresolved. One such remaining problem con- were not examined. cerns the timing and regulation of fang replacement, i.e., An adult specimen of Jameson’s mamba, Dendroaspis when and how a functional fang is replaced by a new fang. jamesoni (Traill, 1843) (Serpentes: Elapidae), was found This topic has been briefly touched by Jackson (2007). at Bomane on the Aruwimi River on 24 May 2010. It was Snakes replace their teeth, including the fangs, regularly collected and preserved, and is currently deposited in the and continuously, therefore there is always a number of Royal Belgian Institute of Natural Sciences in Brussels replacement fangs posterior to the functional one. The (voucher specimen RBINS:ZTN:CRT4055). Furthermore, maxilla typically has sockets for two fangs in lateral ver- an adult specimen of puff adder, Bitis arietans (Merrem, sus medial positions. At any given time, one would expect 1820) (Serpentes: Viperidae), was found in the Kundelun- one fang to be solidly fused (ankylosed) to the socket and gu National Park around 25 km southwards to Katwe on the other more or less loose and in the process of either 22 November 2011. The snake is now voucher specimen attaching or being shed. However, as a functional safety RBINS:ZTN:UP391. factor, the phase when both fangs are ankylosed should The micro-CT scans of the heads of both snake speci- overlap to some degree, avoiding a time window where mens were performed at the Centre for X-ray Tomogra- both the replacement fang and the one in the process of phy of the Ghent University (http://www.ugct.ugent.be; being shed would not be fixed to the upper jaw. According Masschaele et al. 2007) using the transmission head of to Klauber (1972), this period of overlap must be short, a dual head X-ray tube (Feinfocus FXE160.51) and an a-Si © 2013 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Mannheim, Germany All articles available online at http://www.salamandra-journal.com 109  Correspondence Figure 1. Head of an adult Dendroaspis jamesoni (Serpentes: Elapidae). Left: Head of the preserved specimen with fangs visible on the right-hand side. Right: Frontal ventral view of the skull. flat panel detector (PerkinElmer XRD 1620 CN3 CS). The tube voltage was selected to be between 120 and 130 kV. For the Jameson’s mamba, 1801 projections were recorded, cov- ering 360°, with an exposure time of 2 s per projection, re- sulting in a voxel size of 36 µm. For the puff adder, 1201 pro- jections with an exposure time of 2 s per projection were recorded, resulting in a voxel size of 70 µm. Reconstruction of the tomographic projection data was done using the in- house-developed Octopus-package (http://www.octopus- reconstruction.com; Vlassenbroeck et al. 2007). Volume and surface rendering was performed using Amira 5.4.3 (VSG). These two species allow an interesting comparison. Mambas, like other elapid snakes, have proteroglyph den- tition with relatively short fangs ankylosed to a less mo- bile and longer maxilla, whereas viperids have solenoglyph dentition with large fangs attached to a mobile and signifi- cantly reduced maxilla. In both cases, however, to different grades, the maxillae rotate relative to the ectopterygoid and other cranial bones during a strike. In viperids, the greatly elongated fangs are folded backwards in the resting posi- tion, “with base and point at about the same level, and with the bulge of the fang-curve fitting into a hollow in the low- er jaw” (Klauber, 1972), whereas in elapids, they are not. Figure 2. Top: Lateral view of the complete head of Dendroaspis Mambas (genus Dendroaspis) are large snakes with a jamesoni. Functional and first replacement fangs are marked in maximum total length of > 2 m. They possess compara- dark blue, further replacement fangs in light blue on the right- tively large fangs, with their maxillae being rather mobile hand side. Bottom: Detailed frontolateral view of the fangs. 110  Correspondence Figure 3. Head of an adult Bitis arietans (Serpentes: Viperidae). Left: Head of the specimen with all fangs visible. Right: Frontal ventral view of the skull. compared to other elapids. Fangs of adult Dendroaspis jamesoni typically range between 6.4 and 8.0 mm in length (Bogert 1943). A reconstruction of the Jameson’s mamba skull (Figures 1–2) shows that on the right side, the func- tional fang in lateral position abuts against a medial re- placement fang, with both being similar in length (‘double fangs’ are marked in dark blue in the reconstructions). The first replacement fang appears to be fully developed and almost completely ankylosed to the maxilla (Fig. 2); fur- ther replacement fangs on the right-hand side are marked in light blue in the lateral view. This is different for the me- dial replacement fang on the left-hand side, which is clearly not yet ankylosed. The distal end of that fang points pos- teriorly, and the fang is positioned more horizontally than vertically. Furthermore, both fangs on the right-hand side are visible from the outside, suggesting that in the case of a strike, both fangs would be functional in penetrating the prey. However, the strike angles of these fangs differ slight- ly, with the distal end of the replacement fang still point- ing more caudally. Having two fangs in these unequal posi- tions may not be very efficient during a strike, as there will always be one fang that will not penetrate the prey axially, and thus experience bending forces at the tip. To evaluate this aspect, however, kinematic simulations of the strike Figure 4. Top: Lateral view of the complete head of Bitis arietans. would be necessary. All fangs are in resting position, ‘double fangs’ are marked in Puff adders (Bitis arietans) are heavy ambush predators dark blue, replacement fangs in light blue on the right-hand side. in Africa, and they have very large fangs that may exceed Bottom: Detailed frontolateral view of the fangs on the right side 30  mm in length, fused to short but wide maxillae. The of the skull. 111  Correspondence Figure 5. Frontal (left) and lateral views (right) of the fangs of Bitis arietans. Numbers indicate the sequence of fangs in the replace- ment series. anatomy of the skull and the mass of attached muscles fa- Acknowledgements cilitate a powerful strike, and the amount of injected ven- om can be very high compared to that in other venomous We are grateful to Erik Verheyen, Michel Hasson (Brussels, snakes (see also Fig. 3, left). Our specimen was fixed and Belgium) and Klaas-Douwe B. Dijkstra (Leyden, The Neth- erlands) for their assistance during field expeditions. We thank preserved with the mouth closed, and therefore its fangs Kris Pannecoucke for the photo of the Jameson’s mamba. Field- remained in a resting position during the reconstruction. work was supported by the Belgian National Focal Point to the On the right-hand side, two fangs are visible in a parallel Global Taxonomy Initiative (grants in 2011 and 2012 to ZTN). The position, resulting in equal striking angles (Figures 3–4). Special Research Fund of Ghent University (BOF) is acknowl- From a mechanical point of view, both fangs would be fully edged for their financial support (GOA 01G01008). efficient in penetrating a prey animal. At first sight, both of these fangs seem to be fully functional. They are roughly equal in size and positioned very close to each other. To References have a better view, pieces of bones were digitally cut away Bogert, C. M. (1943): Dentitional phenomena in cobras and oth- around the fangs to make close-up views possible (Fig. 5). er elapids with notes on adaptive modifications of fangs. – Bul- These show that fang No. 1, fixed in a lateral socket of the letin of the American Museum of Natural History, 81: 285–360. maxilla is the functional fang, while the one less firmly an- Fry, B. G., H. Scheib, L. van der Weerd, B. Young, J. Mc- kylosed in the medial socket is the next fang in the replace- Naughtan, S. F. Ryan Ramjan, N. Vidal, R. E. Poelmann ment series. The alternate possible interpretation would & J. A. Norman (2008): Evolution of an arsenal: Structural have been that the medial fang is older than the lateral one and functional diversification of the venom system in the ad- and in the process of loosening prior to being shed. How- vanced snakes (Caenophidia). – Molecular and Cellular Pro- ever, based on a comparison of the angles of the shafts of teomics, 7: 215–246. fangs 1 and 2 (# 1 is in a closer-to-vertical position, and Jackson, K. (2002): How tubular venom-conducting fangs are # 2 lies at an angle intermediate between 1 and replacement formed. – Journal of Morphology, 252: 291–297. fang 3), we interpret fang # 2 as being in the process of at- Jackson, K. (2003): The evolution of venom-delivery systems in taching to the maxilla rather than in the process of detach- snakes. – Zoological Journal of the Linnean Society, 137: 337– ing. According to Klauber’s (1972) explanation of the se- 354. quence of fang replacement, fang No. 3 will replace which- Jackson, K. (2007): The evolution of venom-conducting fangs: ever in-place fang (lateral or medial) is shed first. insights from developmental biology. – Toxicon, 49: 975–981. 112  Correspondence Kardong, K. V. (1982): The evolution of the venom apparatus in snakes from colubrids to viperids and elapids. – Memórias do Instituto Butantan, 46: 105–118. Klauber, L. M. (1972): Rattlesnakes. Their habits, life histories, and influence on mankind. – Second edition, Berkeley & Los Angeles, University of California Press, 1533 pp. Knight, A. & D. P. Mindell (1994): On the phylogenetic relation- ship of Colubrinae, Elapinae, and Viperinae, and the evolution of frontfanged venom systems in snakes. – Copeia, 1994: 1–9. Kochva, E. (1978): Oral glands of the Reptilia. – pp. 43–161 in: Gans, C. & K. Gans (eds): Biology of the Reptilia, Vol. 8. – Academic Press, New York. Lake, A. R. & T. R. Trevor-Jones (1987): Formation of the poi- son fang canal of the puff adder Bitis arietans. – South African Journal of Science, 83: 668–669. Lake, A. R. & T. R. Trevor-Jones (1995): The formation of the poison fang of the boomslang Dispholidus typus. – South Afri- can Journal of Science, 91: 329–330. Masschaele, B. C., V. Cnudde, M. Dierick, P. Jacobs, L. Van Hoorebeke & J. Vlassenbroeck (2007): UGCT: New X-ray Radiography and Tomography Facility. – Nuclear Instruments and Methods in Physics Research Section A, 580: 266–269. Tomes, C. S. (1874): On the structure and development of the teeth of Ophidia. – Philosophical Transactions of the Royal Society of London, 165: 297–302. Vlassenbroeck, J., M. Dierick, B. Masschaele, V. Cnudde, L. Van Hoorebeke & P. Jacobs (2007): Software tools for quan- tification of X-ray microtomography at the UGCT. – Nuclear Instruments and Methods in Physics Research Section A, 580: 442–445. Vonk, F. J., J. F. Admiraal, K. Jackson, R. Reshef, M. A. G. de Bakker, K. Vanderschoot, I. van den Berge, M. van At- ten, E. Burgerhout, A. Beck, P. J. Mirtschin, E. Kochva, F. Witte, B. G. Fry, A. E. Woods & M. K. Richardson (2008): Evolutionary origin and development of snake fangs. – Na- ture, 454: 630–633. 113 

Zootaxa 2501: 23–36 (2010) ISSN 1175-5326 (print edition) www.mapress.com / zootaxa/ Article ZOOTAXA Copyright © 2010 · Magnolia Press ISSN 1175-5334 (online edition) New species of reed frog from the Congo basin with discussion of paraphyly in Cinnamon-belly reed frogs SUSANNE SCHICK1, JOS KIELGAST2, DENNIS RÖDDER1, VINCENT MUCHAI3, MARIUS BURGER4 & STEFAN LÖTTERS1,5 1 Trier University, Department of Biogeography, 54286 Trier, Germany. E-Mail: loetters@uni-trier.de 2 Copenhagen University, Department of Biology, Universitetsparken 15, 2100 Copenhagen, Denmark 3 National Museums of Kenya, Herpetology Section, PO Box 40658-00100, Nairobi, Kenya 4 University of the Western Cape, Zoology Department, Private Bag X17, Bellville 7535, South Africa 5 Corresponding author. E-mail: loetters@uni-trier.de Abstract We describe a new species of Afrotropical reed frog, genus Hyperolius (Hyperoliidae), from Salonga National Park in the central Congo basin, Democratic Republic of Congo. Males and females have similar colour and pattern and are easily distinguished from other taxa by a relatively short and broad, bright yellow (in life), dorsolateral line ending in the sacral region and the presence of a light spot on the heel. In a 16S mitochondrial rRNA phylogeny, it clusters with samples allocable to the Cinnamon-belly reed frog, H. cinnamomeoventris. The new species, along with other morphologically well distinguished taxa, splits H. cinnamomeoventris into different non-sister clades. We discuss paraphyly of this reed frog in a taxonomic framework. Key words: Anura, DNA barcoding, Hyperoliidae, Hyperolius cinnamomeoventris, H. veithi sp. nov., paraphyly, taxonomy Introduction Reed frogs, genus Hyperolius (Hyperoliidae), comprise a diverse group of arboreal anurans from sub-Saharan Africa. More than 120 species are recognized (Frost 2009). Hyperolius taxonomy remains difficult for the following reasons. Species are generally poor in external diagnostic characters, and often typified by high intraspecific colour and pattern variation (e.g. Schiøtz 1999). Second, numerous ‘old’ names are available based on one to a few specimens which in many cases are poorly preserved or are even lost, leaving us with poor original descriptions only, as for example those of Ahl (1924). As pointed out by Schiøtz (1999), an almost exclusive feature of reed frogs is that in many species all females and part of the males undergo colour and pattern change with maturation (phase female, PhF), while the remaining males retain their juvenile colour and pattern (phase juvenile, PhJ). Sexual dichromatism and monochromatism are robust species- specific traits in the genus (Veith et al. 2009). During field work at Salonga National Park, Democratic Republic of Congo (DRC), in the central Congo drainage (Fig. 1), we discovered a species of reed frog (Fig. 2A) which, by colour and pattern, is well distinguished from any of the described taxa, leading us to consider it an undescribed species. We used sequences of the 16S mitochondrial rRNA to ‘barcode’ the new reed frog (Vences et al. 2005) and to run a molecular phylogeny to identify its congeneric relatives. We found that the new taxon is related to the Cinnamon-belly reed frog, H. cinnamomeoventris Bocage, 1866; the latter species displays sexual dichromatism (Fig. 2Bʹ D) with a PhJ somewhat similar in pattern to the new species, which is sexually monochromatic (Fig. 2A). At Salonga National Park, the two species can be found syntopically. Hyperolius cinnamomeoventris is suggested to exhibit a vast geographic range (Fig. 1) over a large Accepted by M. Vences: 6 May 2010; published: 10 Jun. 2010 23 portion of Central and adjacent East Africa (e.g. Perret 1966, Schiøtz 1999, Frost 2009). When comparing genetic samples from different localities of Cinnamon-belly reed frog, the new Congo basin species splits the whole group into different, well supported geographic clades. Paraphyly of H. cinnamomeoventris is further supported by the taxonomic status of H. molleri (Bedriaga, 1892) and H. thomensis Bocage, 1866 from Principe and São Tomé which are sexually monochromatic (Figs. 2EʹF, 3). The objectives of this paper are (i) to describe the new species from Salonga National Park and (ii) to discuss paraphyly of Cinnamon-belly reed frogs in a taxonomic framework. Material and methods Specimens morphologically examined and measured are deposited at the British Museum, London (BM), Musée Royal de l'Afrique Centrale, Tervuren (MRAC), National Museums of Kenya, Nairobi (NMK), Naturkundemuseum Berlin (ZMB), Zoologisches Forschungsmuseum Alexander Koenig, Bonn (ZFMK), and Zoological Museum, University of Copenhagen (ZMUC), as listed in the Appendix; abbreviations "A", "AC" and "SL" refer to field numbers given by J. Kielgast, A. Channing and S. Lötters, respectively. Of the new species described herein, 27 pairs in amplexus were found, making sex allocation easy. All specimens a priori identified as males each had a gular flap and a vocal sac, while those a priori identified as females each lacked gular flap and vocal sac and were larger in body size. Additional specimens found were identified as males based on size and presence of gular flap and vocal sac. Definitions of morphological characters and the diagnostic scheme in the new species (adults only) follow Lötters et al. (2004): (1) snout-vent length (SVL) of males and females; (2) tibia length (TIBL)/SVL, head width at angles of jaws (HW)/SVL; (3) dorsal skin texture: smooth or warty; (4) dorsal and lateral snout shape after Heyer et al. (1990: 409) and if nares visible from above or not; (5) distance from anterior corner of eye to nostril (Eʹ N)/horizontal eye diameter (EYE); (6) tympanum: free and distinct or covered by skin and indistinct, horizontal tympanum diameter (TYMP)/EYE; (7) foot length from proximal edge of outer metatarsal tubercle to tip of toe 4 (FOOT)/TIBL; (8) foot and (if present) hand webbing formula using the system described by Glaw & Vences (2007: 70); (9) dorsal and ventral colour and pattern and iris colour in life; (10) DNA barcode, i.e. the sequence of a fragment of the 16S mitochondrial rRNA; we here provide the GenBank accession number only (Benson et al. 2004). We distinguish between colour and pattern (Wollenberg et al. 2008), the latter is less prone to change with preservative. As an example, the colour of the frogs illustrated in Figure 2A is largely light brown. Their pattern includes a dorsolateral line; its colour is bright yellow. In preservative, brown and yellow fade, but the (faded) dorsolateral line remains. Tissue samples (muscle, toe clips) of Cinnamon-belly reed frogs and the new species were obtained at several localities and stored in 98 % ethanol with voucher specimens deposited in scientific collections (Fig. 1, Table 1). DNA was extracted from tissue fixed in 99 % ethanol using Highpure PCR Template Preparation Kit (Roche Diagnostics). Polymerase Chain Reaction (PCR) was used to amplify approximately 530 bp of the 16 S mitocho ndrial rR NA u sing the fo llow ing prim ers of Palum bi et a l. (19 91): 1 6SA (5’- CGCCTGTTTATCAAAAACAT-3’) and 16SB (5’-CCGGTCTGAACTCAGATCACGT-3’). Amplification followed the standard PCR conditions (Palumbi 1996) with the following thermal cycle profile: 90 sec at 94 °C, followed by 33 cycles of 94 °C for 45 sec, 55 °C for 45 sec, and extension 72 °C for 90 sec. All amplified PCR products were verified using electrophoresis on a 1.4 % agarose gel stained with ethidium bromide. PCR products were purified using Highpure PCR Product Purification Kit (Roche Diagnostics). Single-stranded DNA samples were sequenced using the BigDye Deoxy Terminator cycle-sequencing kit (Applied Biosystems) as well as DYEnamic ET terminator cycle sequencing premixkit (GE Healthcare) on an automated DNA sequencer (ABI PRISM 377 or MEGABACE1000 respectively). Editing was completed in MEGA4 (Tamura et al. 2007). All novel nucleotide sequence data were deposited in GenBank (for accession numbers see Table 1). The obtained sequences were verified as Hyperolius DNA by standard nucleotide- nucleotide BLAST search in GenBank. 24 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL. TABLE 1. Species, their family allocation and vouchers processed in the molecular analysis (Fig. 3), their origin and GenBank accession numbers for the 16S mitochondrial rRNA gene. Abbreviations used in addition to those mentioned in the text are: A + number = field number of material collected by Jos Kielgast; AC + number = field number of material collected by Alan Channing; AMNH = American Museum of Natural History, New York; CAR = Central African Republic; CAS = California Academy of Sciences, San Francisco; AM, ANK, BI, BOB, DS, GRE, LOM + number = field number of material collected by Mark-Oliver Rödel; MVZ = Museum of Comparative Zoology, Harvard University, Cambridge; RSA = Republic of South Africa. Samples of Hyperolius cinnamomeoventris: # from the type locality of H. olivaceus and H. fimbriolatus; ## from the type locality of H. cinnamomeoventris and H. tristis; E Eastern Clade (see text). Hyperolius sp. ‘Salonga’ is H. veithi sp. nov. Species Family Voucher Locality GenBank Acanthixalus spinosus Hyperoliidae ZFMK 72000 Mt. Kupe, Cameroon AF215427 Afrixalus delicatus Hyperoliidae ZFMK 68792 Kwa-Mbonambi, RSA AF215428 Amietophrynus maculatus Bufonidae AMNH A163573 Mali DQ283388 Astylosternus schioetzi Astylosternidae ZFMK 67733 Edib, Cameroon AF124108 Boophis tephraeomystax Mantellidae ZFMK 66690 Kirindy, Madagascar AF215334 Heterixalus alboguttatus Hyperoliidae not preserved Ranomafana, Madagascar AF215433 Hyperolius baumanni Hyperoliidae BI 41 Biakpa, Ghana GU443980 H. bobirensis Hyperoliidae ANK 101 Ankasa, Ghana GU443982 H. castaneus Hyperoliidae CAS 202087 Bwindi, Uganda GU444000 H. chlorosteus Hyperoliidae GRE 46 Grebo, Liberia GU443986 H. cinnamomeoventris # Hyperoliidae ZFMK 73207 Lambaréné, Gabon FJ594077 H. cinnamomeoventris ## Hyperoliidae AC 3008 (Fig. 2B) Kalandula, Angola GU443990 H. cinnamomeoventris ## Hyperoliidae AC 3017 Kalandula, Angola HM064461 H. cinnamomeoventris E Hyperoliidae DS 69 Dzanga-Sangha Reserve, CAR GU443997 H. cinnamomeoventris E Hyperoliidae A 185 Salonga National Park, DRC HM0644613 H. cinnamomeoventris E Hyperoliidae A 186 Salonga National Park, DRC HM0644614 H. cinnamomeoventris E Hyperoliidae A 195 Salonga National Park, DRC HM0644615 H. cinnamomeoventris E Hyperoliidae A 200 Salonga National Park, DRC GU443998 H. cinnamomeoventris E Hyperoliidae A 205 Salonga National Park, DRC HM064466 H. cinnamomeoventrisE Hyperoliidae A 230 Salonga National Park, DRC HM064467 H. cinnamomeoventrisE Hyperoliidae NMK A/3858/2 Kakamega Forest, Kenya AY323925 H. cinnamomeoventrisE Hyperoliidae NMK A/3918/2 Kakamega Forest, Kenya HM064462 H. cinnamomeoventrisE Hyperoliidae SL 326 Kibale, Uganda GU443995 H. cinnamomeoventrisE Hyperoliidae SL 508 Bundibugyo , Uganda HM0644678 H. cinnamomeoventrisE Hyperoliidae SL 509 Bundibugyo , Uganda GQ183576 H. cinnamomeoventrisE Hyperoliidae SL 510 Bundibugyo , Uganda GU443994 H. cinnamomeoventrisE Hyperoliidae SL 565 Kampala, Uganda HM064457 H. cinnamomeoventris E Hyperoliidae SL 566 Kampala, Uganda HM064458 H. cinnamomeoventris E Hyperoliidae SL 570 Kampala, Uganda HM0644589 H. cinnamomeoventris E Hyperoliidae SL 575 Kampala, Uganda HM064460 H. cinnamomeoventris E Hyperoliidae SL 576 Kampala, Uganda GU443996 H. cinnamomeoventris E Hyperoliidae SL 532 Semliki, Uganda GQ183577 H. cinnamomeoventris E Hyperoliidae SL 553 Semliki, Uganda GQ1835769 H. cinnamomeoventris E Hyperoliidae SL 555 (Fig. 1D) Semliki, Uganda GU443993 H. cinnamomeoventrisE Hyperoliidae CAS 202493 Bwindi, Uganda AY603985 continued next page NEW REED FROG FROM CONGO AND PARAPHYLY Zootaxa 2501 © 2010 Magnolia Press · 25 TABLE 1. (continued) Species Family Voucher Locality GenBank H. concolor Hyperoliidae AM 40 Amedzofe, Ghana GU443984 H. cystocandicans Hyperoliidae ZFMK 77611 Mt. Kenya, Kenya FJ594079 H. glandicolor Hyperoliidae SL 590 Runda-Gigiri, Kenya GU443977 H. kivuensis Hyperoliidae SL 471 Bundibugyo, Uganda GU443979 H. laurenti Hyperoliidae ANK 72 Ankasa, Ghana GU443987 H. molleri Hyperoliidae CAS 219125 Principe, São Tomé and Principe AY603990 H. mosaicus Hyperoliidae ZFMK 73140 Mt. Cristal, Gabon AY323923 H. picturatus Hyperoliidae LOM 26 Loma, Sierra Leone GU443983 H. riggenbachi Hyperoliidae MVZ 234752 Ndop, Cameroon GU443976 Hyperolius sp. ‘Salonga’ Hyperoliidae ZFMK 89607 Salonga National Park, DRC GU443988 Hyperolius sp. ‘Salonga’ Hyperoliidae ZFMK 89629 Salonga National Park, DRC GU443975 H. sylvaticus Hyperoliidae BOB 37 Ivory Coast GU44398 H. thomensis Hyperoliidae CAS 218925 São Tomé, São Tomé and Principe AY603991 H. torrentis Hyperoliidae AM 22 Amedzofe, Ghana GU443985 H. viridiflavus Hyperoliidae SL 02/50 Kakamega Forest, Kenya GU443978 Leptopelis natalensis Arthroleptidae ZFMK 68785 Mtunzini, RSA AF215448 Mantidactylus cf. grandisonae Mantellidae ZFMK 66669 Ambato, Madagascar AF215315 Tachycnemis seychellensis Hyperoliidae ZFMK 62879 Praslin, Seychelles AF215452 FIGURE 1. Map of Central Africa and adjacent areas showing Salonga National Park (white contour line), distribution of H. cinnamomeoventris according to the IUCN Red List (http://www.iucnredlist.org) following the 2002ʹ2004 IUCN Global Amphibian Assessment (bold black contour line) and localities of our genetic sampling from type localities (reverse filled triangle—H. cinnamomeoventris and H. tristis, white circle—H. veithi sp. nov., filled square—H. ituriensis, filled triangle— H. olivaceus and H. fimbriolatus) and additional localities (small filled dots). In addition, the type locality of H. wittei (cross) is shown. Note that H. veithi syntopically occurs with H. cinnamomeoventris. 26 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL.  In order to not only obtain DNA barcodes but also information on monophyletic clusters of samples, a phylogenetic analysis was performed. Sequences were aligned using the MUSCLE software (Edgar 2004) and executed for uncorrected p-distances with PAUP*4b10 (Swofford 2001). The GTR model (GTR + I + G) was selected from the data using MrModeltest v2 (Nylander 2004). Bayesian inference was performed using MRBAYES 3.1. (Huelsenbeck & Ronquist 2001, Ronquist & Huelsenbeck 2003). Four simultaneous and completely independent analyses were initiated with random starting trees. A total number of 1,000,000 generations with four independent Markov Chains were started while sampling every 100th tree. The first 1,000 trees (burnin) were excluded from the 50 % majority rule consensus tree. Topologies with posterior probabilities ≥ 95 % were considered well supported (Wilcox et al. 2002). As hierarchical outgroups, sequences of other Hyperolius species as well as representatives of additional hyperoliids and other Afrotropical frog families available at GenBank were included in the study (Table 1). An analysis excluding hypervariable sites (altogether 47 nucleotides) did not affect the topology of the phylogeny. Hence, topology, support values and the uncorrected p-distances given here refer to the complete fragment. Hyperolius veithi sp. nov. (Figs. 2A, 4A) Holotype. ZFMK 89607 (field number A519), adult male, collected by J. Kielgast at a flooded area in the middle of primary forest away from rivers and streams in Salonga National Park (02.88 S, 20.41 E, ca. 415 m above sea level), Province of Bandundu, Equateur Kasaï Oriental and Occidental, Democratic Republic of Congo, 24 January 2008. Paratypes. Fifty-eight adult specimens of both sexes (ZFMK 89608ʹ89645, ZMUC R771393ʹR771412) from the same locality; data as the holotype but collected 24ʹ26 January 2008. Diagnosis. A sexually monochromatic species of Hyperolius based on the following characters: pupil horizontal; tips of toes and fingers broadened; last phalanx of fingers out of alignment; fingers not arranged in opposing pairs; gular flap in males (Schiøtz 1999; Channing & Howell 2006). It is defined by: (1) SVL males 25.7 ± 1.7 mm (22.7ʹ28.4 mm, N = 32), females 33.9 ± 1.3 mm (31.7ʹ36.5 mm, N = 27); (2) TIBL/SVL in males 0.54 ± 0.02 (0.52ʹ0.61, N = 32) and in females 0.51 ± 0.02 (0.48ʹ0.55, N = 27), HW/SVL in males 0.33 ± 0.02 (0.3ʹ0.37, N = 32) and in females 0.33 ± 0.01 (0.29ʹ0.35, N = 27); (3) dorsal surface smooth to coarse with few scattered pointed warts, most prominent and tubercular below eye; (4) snout shape dorsally and laterally rounded, nares visible from above; (5) EʹN/EYE in males 0.78 ± 0.09 (0.53ʹ0.95, N = 32) and in females 0.85 ± 0.09 (0.71ʹ0.99, N = 27), canthus rostralis convex from tip of snout to nostril and straight or slightly concave from nostril to eye; (6) tympanum covered by thick skin but with annulus tympanicus still visible; (7) FOOT/TIBL in males 0.67 ± 0.04 (0.57ʹ0.75, N = 32) and in females 0.75 ± 0.05 (0.67ʹ0.86, N = 27); (8) foot webbing formula 1(1), 2i(1) 2e(½), 3i(1ʹ1½) 3e(½ʹ1), 4i(1ʹ2) 4e(1ʹ1½), 5(1ʹ½); (9) no sexual dichromatism; both sexes dorsally tan, olive brown to dark brown, usually lighter towards the sides and dorsally with more or less intense, diffuse dark markings; there is always a bright yellow dorsolateral line from tip of snout (lines from both sides of the body meet here) to sacral region which is commonly bordered by dark brown; it is broader posterior to eye and often contains tiny dark brown spots which may continue posterior into groin and onto hind limbs; always a cream to yellow spot on the heel; ventral sides are white or cream; a red marking on the inner femur is lacking; the gular flap of males is yellow; the iris is bronze to dark brown; (10) for DNA barcodes access GenBank (http://www.ncbi.nlm.nih.gov): GU443988 (male holotype, ca. 530 bp), GU443975 (female paratype, ZFMK 89629, ca. 480 bp). By being brownish and having a light dorsolateral lines, H. veithi males and females (Fig. 2A) are similar in colour and pattern to numerous Hyperolius species (see Schiøtz 1999), including PhJ H. cinnamomeoventris (Fig. 2CʹD) and its suggested synonyms H. fimbriolatus Buchholz & Peters, 1876 and H. ituriensis Laurent, 1943 (Lötters et al. 2001, Frost 2009). Hyperolius schoutedeni from the Congo basin deserves special attention, as this reed frog, like H. veithi, is also sexually monochromatic. The new species can be distinguished from any other Hyperolius species which are brownish and have a light dorsolateral lines NEW REED FROG FROM CONGO AND PARAPHYLY Zootaxa 2501 © 2010 Magnolia Press · 27 (including H. cinnamomeoventris and H. schoutedeni) as follows: (i) dorsolateral lines relatively short and broad ending in the sacral region (versus in the groin in other species); (ii) dorsolateral lines in life are bright yellow (versus whitish to tan in other species); (iii) dorsolateral lines often contain dark spots (absent in most other species); (iv) presence of a light spot on the heel, cream to yellow in life (absent in other species); (v) absence of a in life red marking on the inner femur (present at least in H. cinnamomeoventris and its suggested synonyms and occasionally in H. schoutedeni; Schiøtz 2006). In addition to (i) to (v), in H. schoutedeni, a mid-dorsal line is common (absent in H. veithi) and the new species is larger than H. schoutedeni and develops less foot webbing. Five individuals (three males, two females from N'Sele and Garamba, DRC; Schiøtz 2006) and the female holotype of H. schoutedeni (Fig. 4B) have maximum SVL in males 22.9 (minimum SVL in H. veithi 22.7 mm) and in females 25.2 mm (minimum SVL in H. veithi 31.7 mm); holotype foot webbing formula is 1(1), 2i(1) 2e(½), 3i(1½) 3e(1), 4i(2) 4e(1½), 5(1) (for H. veithi see above) (note that measurements and foot webbing data for H. schoutedeni differ slightly from those given by Schiøtz 2006). Several additional Hyperolius species have been described from DRC (Schiøtz 1999; Frost 2009) but remain little known as they are based on one to a few specimens only and poor original descriptions. Of these, H. atrigularis Laurent, 1941, H. kibarae Laurent, 1957 and H. polli are similar to H. veithi in colour and pattern in preservative, but none has continuous bright dorsolateral lines (based on holotype examinations and Laurent 1941, 1957). In terms of uncorrected p-distances of the 16S mitochondrial rRNA, H. veithi is little differentiated genetically from some, but not all populations assigned to H. cinnamomeoventris (Table 2). We reject conspecifity, however, as H. veithi syntopically occurs with H. cinnamomeoventris (uncorrected p-distance 2.8ʹ4.4 %, but see comment to Table 2) and amplexus was observed 27 times between H. veithi individuals but never between H. veithi and H. cinnamomeoventris. Additional arguments include presence of monochromatism in the new species versus dichromatism in H. cinnamomeoventris (Fig. 2A versus Fig. 2Bʹ D) and H. veithi having white eggs (Fig. 2A) versus the half-pigmented eggs in H. cinnamomeoventris (Lötters et al. 2004). The genetically related H. molleri and H. thomensis (Table 2) are dorsally brownish to greenish, lack dorsolateral lines (present in the new species) and possess a more coarse skin texture (Fig. 2A versus Fig. 2EʹF). Description. The type series contains 32 males and 27 females. In external morphology, mature sexes differ in size (Table 3) and the presence of gular flap and vocal sac in males only. Body slender with sacrum width about one fifth of SVL; head short and broad with distance from tip of snout to posterior corner of eye about one sixth of SVL; HW about one third of SVL; snout tip in dorsal and lateral views rounded; nostril dorsolateral, protruding and visible from above; choanae rounded; maxillary teeth present, vomerine teeth lacking; tongue about as long as wide, free for half of its length and anteriorly bifurcated; canthus rostralis convex from nostril to tip of snout, straight and longer from nostril to anterior corner of eye; EYE > E ʹ N; loreal area barely concave; tympanic membrane covered by thick skin but tympanic annulus visible, TYMP < EYE. Male gular flap relatively small compared to other Hyperolius species (see Schiøtz 1999). Tibia long, about half of SVL, tibiotarsal articulation extending to loreal region when hind limb adpressed forward along body; FOOT < TIBL; relative length of toes: 1 < 2 < 3 < 5 < 4; metatarsal tubercles ill-defined, rounded, inner somewhat larger in size and more prominent than outer, rest of sole smooth with distinct subarticular tubercles present at joints of phalanges of all toes; tips of toes broadened; foot webbing formula 1(1), 2i(1) 2e(½), 3i(1ʹ1½) 3e(½ʹ1), 4i(1ʹ2) 4e(1ʹ1½), 5(1ʹ½). Relative length of fingers: 1 < 2 < 4 < 3; palmar tubercle ill-defined, rounded, thenar tubercle less ill-defined than palmar tubercle, ovoid; rest of palm wrinkled with well visible subarticular tubercles at joints of phalanges of all fingers; tips of fingers broadened, tips of fingers 3 and 4 more than twice the width of finger base; hand webbing absent. Skin of dorsal surfaces smooth to coarse with few scattered pointed warts, most prominent and tubercular below eye, skin of ventral surfaces wrinkled. For measurements and proportions see Table 3. In life (Fig. 2A), both sexes are dorsally tan, olive brown to dark brown, usually lighter towards the sides and dorsally with more or less intense, diffuse dark markings. A yellow dorsolateral line always runs from the 28 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL. FIGURE 2. Life aspects of: (A) Hyperolius veithi sp. nov. in amplexus (unidentified paratypes; photo J. Kielgast), note that there is a yellow spot on the heel visible in the male frog; (B) H. cinnamomeoventris male in PhJ from the type locality (AC 3008; photo A. Channing); (C) H. cinnamomeoventris female from the Kakamega Forest, Kenya (not collected; photo S. Lötters); (D) H. cinnamomeoventris male in PhJ from Semliki, Uganda (SL 555; photo A. Channing); (E) H. molleri from São Tomé (photo D. Lin, CAS) and (F) H. thomensis from São Tomé in amplexus (photo D. Lin, CAS). NEW REED FROG FROM CONGO AND PARAPHYLY Zootaxa 2501 © 2010 Magnolia Press · 29 TABLE 2. Uncorrected p-distances of the 16S mitochondrial rRNA gene (ca. 530 bp sequences) of populations of Hyperolius cinnamomeoventris and related reed frogs studied in this paper (Fig. 3; Table 1). For meaning of #, ## and E see Table 1. Distinct difference of the two H. veithi sp. nov. samples against the other samples is because the sequence of GU443975 is ca. 480 bp only (versus ca. 530 bp in GU443988). Sample, locality (GenBank accession number) 1 2 3 4 5 6 7 8 9 10 11 12 13 1 E H. cinnamomeoventris , Kenya: Kakamega - Forest (AY323925) 2 H. cinnamomeoventrisE, Uganda: Bundibugyo 1.6 - (GU443994) 3 H. cinnamomeoventrisE, Uganda: Semliki 2.2 1.2 - (GU443993) 4 H. cinnamomeoventrisE, Uganda: Kampala 0.2 1.5 2.2 - (GU443996) 5 H. cinnamomeoventrisE, Uganda: Bwindi 1.6 1.1 1.6 1.5 - (AY603985) 6 H. cinnamomeoventrisE, Uganda: Kibale 0.6 1.8 2.6 0.4 1.9 - (GU443995) 7 H. cinnamomeoventrisE, DRC: Salonga 3.4 3.0 3.4 3.4 4.0 3.8 - National Park (GU443998) 8 H., cinnamomeoventrisE, CAR: Dzanga- 4.4 4.0 4.2 4.3 3.6 4.7 2.1 - Sangha Reserve (GU443997) 9 H. cinnamomeoventris#, Gabon: Lamabréné 8.4 8.5 8.1 8.3 8.6 8.7 7.5 7.6 - (FJ594077) 10 H. cinnamomeoventris##, Angola: Kalandula 3.8 3.8 3.4 3.8 4.2 4.2 4.2 3.8 5.6 - (GU443990) 11 H. molleri, São Tomé (AY603990) 6.0 5.7 5.6 6.1 5.1 6.5 6.2 6.2 6.6 4.7 - 12 H. thomensis, São Tomé (AY603991) 7.0 6.4 6.2 6.8 5.9 5.8 6.7 6.8 7.2 4.9 1.4 - 13 H. veithi sp. nov., DRC: Salonga National 9.4 5.2 8.5 5.4 4.6 5.8 4.4 8.7 8.6 3.8 6.2 6.4 - Park (GU443988) 14 H. veithi sp. nov., DRC: Salonga National 4.2 3.5 2.9 3.9 3.3 4.4 2.8 3.5 7.3 2.2 4.9 4.8 0.0 Park (GU443975) TABLE 3. Measurements (mm) and proportions of Hyperolius veithi sp. nov. The mean is followed by the standard deviation and the range in parentheses. For abbreviations see text. Females (N = 27) Males (N = 32) SVL 33.9 ± 1.3 (31.71–36.5) 25.72 ± 1.72 (22.69–28.35) TIBL 17.4 ± 0.73 (15.55–18.49) 13.97 ± 0.98 (12.15–16.06) HW 11.16 ± 0.42 (10.23–12.12) 8.56 ± 0.60 (7.62–10.06) E–N 3.43 ± 0.27 (2.98–4.05) 2.64 ± 0.22 (1.99–2.96) EYE 3.91 ± 0.39 (3.33–4.45) 3.4 ± 0.31 (2.64–4) TYMP Invisible Invisible FOOT 13.0 ± 0.75 (11.42–15.02) 9.31 ± 0.9 (7.08–10.92) TIBL/SVL 0.51 ± 0.02 (0.48–0.55) 0.54 ± 0.02 (0.52–0.61) HW/SVL 0.33 ± 0.01 (0.29–0.35) 0.33 ± 0.02 (0.31–0.37) E–N/EYE 0.84 ± 0.09 (0.71–0.99) 0.78 ±0.09 (0.53–0.95) FOOT/TIBL 0.75 ± 0.05 (0.67–0.86) 0.67 ± 0.04 (0.57–0.75) 30 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL. tip of snout (lines from both sides of the body meet here) to the sacral region; this line is commonly bordered by dark brown colour and is broader posterior to eye where it often contains tiny dark brown spots. Spots may continue posterior to the dorsolateral lines into groin and onto hind limbs. The heel always has a cream to yellow spot. Ventral sides are white or cream; a red marking on inner femur is lacking; the gular flap of males is yellow. The iris is bronze to dark brown. In preservative, colours fade as follows: tan and brown to greyish, yellow to cream; dark brown, cream and white remain. In characters with intraspecific variation, male holotype conditions are as follows: gular flap and vocal sac present; foot webbing formula 1(1), 2i(1) 2e(½), 3i(1) 3e(½), 4i/e(1), 5(½); SVL 26.3 mm, TIBL 13.9 mm, HW 9.2 mm, EʹN 2.4 mm, EYE 3.7 mm, TYMP 1.4 mm, FOOT 10.6 mm. For individual aspects of colour in preservative and pattern see Figure 4A; individual information on life colour was not recorded. For molecular genetic studies, some muscle tissue had been taken from the ventral side of the left femur. Distribution and natural history. The new species is only known from Salonga National Park in the central Congo basin, DRC (Fig. 1). The habitat is lowland rainforest (ca. 400 m above sea level) with seasonal rainfall (short dry period in February, long dry period in JuneʹAugust); annual precipitation is ca. 1,865 mm and monthly mean temperature ca. 25.0 °C (data obtained from WorldClim; Hijmans et al. 2005). Due to limited sampling efforts in the Congo basin and an expected lack of distribution barriers to Hyperolius veithi, we expect this species to show a geographically wider range than currently known. Twenty-seven amplectant pairs and five single males were found at the end of the rainy season (January) at night on vegetation, ca. 1 m above water in a flooded but apparently permanent pond deep within primary forest. Frogs were not heard calling and some pairs were observed laying small clutches of whitish eggs on vegetation above water (Fig. 2A). It is expected that tadpoles (unknown) develop in water. Syntopic Hyperolius species included H. cinnamomeoventris, H. cf. platyceps and another undescribed species, not closely related to H. cinnamomeoventris and H. veithi. Conservation status. Due to the limited knowledge on both the geographic range encompassed by this species and population trends, we consider it Data Deficient when applying IUCN Red List categories and criteria (IUCN 2008). Due to the apparent lack of physical borders to dispersal and the observation that numerous other amphibian species in the Congo basin show relatively large distributions, we expected that Hyperolius veithi occupies a larger geographic range than currently known in the central Congo basin (see above). If so, it may perhaps later well rank under Least Concern. Also, we do not consider chytridiomycosis, an emerging infectious disease causing amphibian decline in other regions of the planet, a potential threat to H. veithi. The central Congo basin shows limited suitability for the emergence of this fungal pathogen (Rödder et al. 2009) and elsewhere in tropical Africa, Hyperolius species well survive despite high prevalence and individual parasite load (Kielgast et al. 2010). Etymology. The specific name is a patronym for Michael Veith, acknowledging his support of amphibian research in tropical Africa. Paraphyly of Cinnamon-belly reed frogs. As shown in Figure 3, Hyperolius cinnamomeoventris is paraphyletic when H. molleri, H. thomensis and H. veithi are considered distinct species. In addition to H. veithi (see above), we concur with Drewes & Wilkinson (2004) that specific distinctness of the sexually monochromatic H. molleri and H. thomensis is warranted. Although both these reed frogs may show similar colour and pattern (Fig. 2Eʹ F) to PhF H. cinnamomeoventris (Fig. 2C) (which is more applicable to H. molleri), they possess a more coarse skin texture, lack (in life) a red marking on the inner femur (present in H. cinnamomeoventris) and have parts of the toe and fingers tips bright orange (not different to dorsal or ventral colours in H. cinnamomeoventris). Hyperolius thomensis is distinguished further by being larger in body size and having a black, white and orange marbled venter in life (cream in H. cinnamomeoventris), while that of H. molleri is white or red in life and thus can be similar to that of H. cinnamomeoventris. When comparing uncorrected p-distances of H. molleri and H. thomensis with other reed frogs discussed in this paper, divergence is ≥ 4.7 % and ≥ 4.9 %, respectively (Table 2). An operable threshold for the consideration of putative species (to be confirmed by an integrative approach) in the 16S mitochondrial rRNA gene for anuran amphibians is at about 3 % (Fouquet et al. 2007, Vieites et al. 2009) but can even be lower between well accepted species (e.g. Zimkus & Schick 2010). Applying this threshold to reed frogs studied here, specific NEW REED FROG FROM CONGO AND PARAPHYLY Zootaxa 2501 © 2010 Magnolia Press · 31 distinctness of both H. molleri and H. thomensis against H. cinnamomeoventris is supported. Likewise, divergence of H. veithi and H. cinnamomeoventris is ≥ 4.4 % (or ≥ 2.2 % in the shorter H. veithi sequence, but see comment to Table 2). In addition, specific distinctness is still supported by morphology (colour and pattern in life and monochromatism versus dichromatism) and by having white versus half-pigmented eggs (see above). FIGURE 3. Bayesian phylogram of Hyperolius species including Cinnamon-belly reed frogs inferred from nucleotide sequence data from 16S mitochondrial rRNA. Bayesian posterior probabilities > 0.95 each are marked by an asterisk on branch. We here apply species names as given in Table 1; in the Cinnamon-belly reed frog clade, numbers in parentheses give sample size of the same haplotype, which corresponds with localities. Note that Hyperolius cinnamomeoventris is paraphyletic. Topotypic material of H. olivaceus (#) and H. cinnamomeoventris (##) are indicated. Hyperolius sp. ‘Salonga’ is H. veithi sp. nov.  Both the Hierarchical Likelihood Ratio Tests and Akaike Information Criterion implemented in MrModeltest selected a GTR+I+G model with a gamma distribution of 0.5912 and a proportion of invariable sites of 0.2291 (estimated base frequencies: A: 0.3294, C: 0.2260, G: 0.1813, T: 0.2633; rate matrix: A-C: 3.0708, A-G:7.4382, A-T: 5.3616, C-G: 1.5864, C-T: 21.7283, G-T: 1.0000). 32 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL. FIGURE 4. Dorsal and ventral views of preserved (A) male holotype of Hyperolius veithi sp. nov. (photos by F. Feß), SVL 26.3 mm, (B) female holotype of Hyperolius schoutedeni (photos by MRAC, through the courtesy of D. Meirte), SVL 26.4 mm. As clades currently assigned to H. cinnamomeoventris are generally well supported, in a purely cladistic view, different names should be applied. Considering the available names in the synonymy of H. cinnamomeoventris (Frost 2009), the Gabon clade in Figure 3 could be referred to as H. olivaceus Buchholz & Peters, 1876 with H. fimbriolatus treated as a synonym (Lötters et al. 2001), since this genetic sample was obtained at Lambaréné which is the type locality of both these names (Table 1). Likewise, the Angola clade NEW REED FROG FROM CONGO AND PARAPHYLY Zootaxa 2501 © 2010 Magnolia Press · 33 (Fig. 3) is based on samples from Kalandula which is the type locality of H. cinnamomeoventris and its synonym H. tristis Bocage, 1866 (Table 1). Therefore, this clade could be recognized as H. cinnamomeoventris sensu stricto. All samples from the eastern range of Cinnamon-belly reed frogs constitute a monophyletic group, here called Eastern Clade (Figs. 1, 3). Of the available names in the synonymy of H. cinnamomeoventris, two originate from the eastern portion of the Cinnamon-belly reed frog distribution, H. ituriensis Laurent, 1943 and H. wittei Laurent, 1947 (Frost 2009). Consequently, one of them could potentially be applicable to the Eastern Clade, apart from the fact that the type locality of H. wittei lays out of the suggested geographic range of Cinnamon-belly reed frogs, at all (Fig. 1). In spite of this a priori conclusive scenario, we momentarily refrain from taxonomic action for two reasons. First, delimitating and phenotypic diagnosing putative species remains a problem. Not only is the material available limited for comparisons (see Appendix) and call and other natural history data are sparse or lacking (Lötters et al. 2004) so that an integrative approach combining data from different fields to is hampered. It is also true that molecular data demonstrate considerable variation within putative species and likewise overlap between clades (Table 2). Second, with regard to tree-based taxonomy, it has to be considered that only a fragment of one mitochondrial gene and relatively few samples have been involved in our analysis. Although, the 16S mitochondrial rRNA gene is useful for DNA barcoding and tree-based species recognition in anuran amphibians (Vences et al. 2005, Fouquet et al. 2007, Vieites et al. 2009), it has been shown that when analysing mitochondrial DNA sequences, paraphyly does not always uncover distinct taxa. There are various explanations for paraphyly or polyphyly within species when constructing mitochondrial gene trees, in particular introgression, hybrid speciation and (although less of a concern for mitochondrial than for nuclear loci) incomplete lineage sorting (Funk & Omland 2003). In these cases some alleles in a species may appear more close to those of another (related) species than to conspecifics. As a result, intraspecific variation is larger than variation at the species level and may lead to a mistaken consideration of variants for species when following the monophyly criterion (Funk & Omland 2003). Intensive sampling and analyzing phylogeographic data including related species are helpful for an appropriate interpretation of paraphyly. There is on-going debate regarding the reliability of mitochondrial gene trees for species delimitation. For example, McKay & Zink (2010) recently found that paraphyly in birds most commonly results from incorrect taxonomy and concluded that mitochondrial gene trees are rarely misleading. With regard to H. cinnamomeoventris, we conclude that additional sampling and studies are necessary for allowing warranted taxonomic action. Acknowledgements We are grateful to the following individuals and institutions (in alphabetical order) for support including field and lab assistance, support with tissue samples, finance, logistics and permits: Alexander Koenig Stiftung, Bonn, Germany; M. Andersen (ZMUC); M. Behangana (Makerere University, Kampala, Uganda); BIOTA (BMB+F, Germany); W. Böhme (ZFMK); H. Bourubou (Université de Libreville, Gabon); B.A. Bwong (NMK); Centre national de Recherche et Technologie (CENAREST) de la Université de Libreville, Gabon; A. Channing (University of the Western Cape, Bellville, South Africa); Deutscher Akademischer Austauschdienst (DAAD), Bonn, Germany; Deutsche Gesellschaft für Herpetologie und Terrarienkunde (DGHT), Rheinbach, Germany; R.C. Drewes (California Academy of Sciences, San Francisco, USA); F. Feß (Trier University); B. Fruth and G. Hohmann (Max Planck Intitut für Evolutionäre Anthropologie, Leipzig, Germany); R. Günther (ZMB); Institut Congolais pour la Conservacion de la Nature (ICCN), Kinshasa; Kenya Wildlife Service (KWS), Nairobi; S. Kigoolo (Makerere University, Kampala, Uganda); P. Malonza (NMK); D. Meirte (MRAC); M.-O. Rödel (ZMB); National Museums of Kenya (NMK), Nairobi; K. Scheelke (University of Mainz, Germany); Societé d‘Énergie et d‘Eaux du Gabon (SEEG); Systebol (Lompolle Village, DRC); P. Teege (University of Mainz, Germany); A. Kamdem Toham (WWF Central African Regional Program Office, CARPO, Libreville, Gabon); Uganda Wildlife Authority (UWA), Kampala; Université de 34 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL. Libreville, Gabon; A. van der Meijden (Trier University, Germany); M. Veith (Trier University, Germany); D.V. Wasonga (NMK); WWF Central African Regional Program Office, CARPO, Libreville (Gabon). References Ahl, E. (1924) Ueber einige afrikanische Frösche. Zoologischer Anzeiger, 59/60, 269–273. Benson, D.A., Karsch-Mitzrachi, I., Lippman, D.J., Ostell, J. & Wheeler, D.L. (2004) GenBank: update, Nucleic Acids Research, 32, 23–26. Channing, A. & Howell, K.M. (2006) Amphibians of East Africa. Chimaira: Frankfurt/Main; Cornell University Press: Ithaca, New York, 418 pp. Drewes, R. & Wilkinson, J. (2004) The California Academy of Sciences Gulf of Guinea expedition I: the taxonomic status of the genus Nesionixalus with comments on the genus Hyperolius. Proceedings of the California Academy of Sciences, 55, 395–407. Edgar, R.C. (2004) MUSCLE: Multiple sequence alignment with high accuracy and high throughput, Nucleic Acids Research, 32, 1792–1797. Fouquet, A., Gilles, A., Vences, M., Marty, C, Blanc, M. & Gemmell, N.J. (2007) Underestimation of species richness in Neotropical frogs revealed by mtDNA analyses. PLoS ONE 2(10): e1109. Funk, D. J. & Omland, K.E. (2003) Species-level paraphyly and polyphyly: frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annual Reviews in Evolution and Systematics, 34, 397–423. Frost, D.R. (2009) Amphibian Species of the World: an Online Reference. Version 5.3 (12 February, 2009). American Museum of Natural History. New York, USA. Available at: http://research.amnh.org/herpetology/amphibia/ index.html (accessed 6 December 2009). Glaw, F. & Vences, M. (2007) A Field Guide to the Amphibians and Reptiles of Madagascar. Third Edition. Vences & Glaw, Köln, 496 pp. Heyer, W.R., Rand, A.S., da Cruz, C.A.G., Peixoto, O.L. & Nelson, C.E. (1990) Frogs of Boracéia. Arquivos de Zoología, 1, 237–410. Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G. & Jarvis, A. (2005) Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25, 1965–1978. Huelsenbeck, J.P. & Ronquist, F. (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics, 17, 754– 755. IUCN Standards and Petitions Working Group. (2008) Guidelines for using the IUCN Red List Categories and Criteria. Version 7.0. (August 2008). Standards and Petitions Working Groups of the IUCN SSC Biodiversity Assesment Sub-Committe. Available from http://intranet.iucn.org/webfiles/doc/SSC/RedList/RedListGuidelines.pd (accessed 15 June 2009). Kielgast, J., Rödder, D., Veith, M. & Lötters, S. (2010) Widespread occurrence of the amphibian chytrid fungus in Kenya. Animal Conservation, 13, 1–8. Laurent, R.F. (1941) Contributionà la systematique du genre Hyperolius Rapp. (batraciens). Revue de Zoologie et de Botanique africaines, 34, 149–167. Laurent, R.F. (1957) Genres Afrixalus et Hyperolius (Amphibia Salientia). Exploration du Parc national de l'Upemba, Mission G.F. de Witte Fasc., 42, 1–47. Lötters, S., Gossman, V., Obame, F. & Böhme, W. (2001) Zur Herpetofauna Gabuns. Teil I: Einleitung, Untersuchungsgebiet und Methodik, kommentierte Artenliste der gefundenen Froschlurche. Herpetofauna, 133, 19– 34. Lötters, S., Schick, S., Scheelke, K., Teege, P., Kosuch, J., Rotich, D. & Veith, M. (2004) Bio-sketches and partitioning of sympatric reed frogs, genus Hyperolius (Amphibia; Hyperoliidae), in two humid tropical African forest regions. Journal of Natural History, 38, 1969–1997. McKay, B.D. & Zink, R.N. (2010) The causes of mitochondrial DNA gene tree paraphyly in birds. Molecular Phylogenetics and Evolution, 54, 647–650. Nylander, J.A.A. (2004) MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala. Palumbi, S.R. (1996) Nucleic acids II: The polymerase chain reaction. In: Hillis DM, Mable BK, Moritz C, eds. Molecular Systematics. 2nd ed., Sunderland, MA: Sinauer, 205–247. Palumbi, S.R., Martin, A., Romano, S., McMillan, W.O., Stice, L. & Grabowski, G. (1991) The simple fool's guide to PCR. University of Hawaii, Honolulu, 46 pp. Perret, J.-L. (1966) Les amphibiens du Cameroun. Zoologisches Jahrbuch für Systematik, 8, 289–464. Rödder, D., Kielgast, J., Bielby, J., Schmidtlein, S., Bosch, J., Garner, T.W.J., Veith, M., Walker, S., Fisher, M.C. Lötters, S. (2009) Global amphibian extinction risk assessment for the panzootic chytrid fungus. Diversity, 1, 52–66. Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. NEW REED FROG FROM CONGO AND PARAPHYLY Zootaxa 2501 © 2010 Magnolia Press · 35  Bioinformatics, 19, 1572–1574. Schiøtz, A. (1999) Treefrogs of Africa. Chimaira: Frankfurt/Main: 350 pp. Schiøtz, A. (2006) Notes on the genus Hyperolius (Anura, Hyperoliidae) in central Republique Democratique du Congo. Alytes, 24, 40–60. Swofford, D.F.( 2001) PAUP*. Phylogenetic analysis using parsimony (*and other methods). Version 4.0.b.10. Sinauer Associates, Sunderland, Massachusetts. Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007) MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution, 24, 1596–1599. Veith, M., Kosuch J., Rödel, M.O., Hillers, A., Schmitz, A., Burger, M. & Lötters, S. (2009) Multiple evolution of sexual dichromatism in African reed frogs. Molecular Phylogenetics and Evolution, 51, 388–393. Vences, M., Thomas, M., van der Meijden, A., Chiari, A. & Vieites, D.R. (2005) Comparative performance of the 16S rRNA gene in DNA barcoding of amphibians. Frontiers in Zoology, 2, 5. Vieites, D.R., Wollenberg, K.C., Andreone, F., Köhler, J., Glaw, F. & Vences, M. (2009) Vast underestimation of Madagascar's biodiversity evidenced by an integrative amphibian inventory. Proceedings of the National Academy of Sciences of the U.S.A., 106, 8267–8272. Wilcox, T.P., Zwickl, D.J., Heath, T.A. & Hills, D.M. (2002) Phylogenetic relationships of the dwarf boas and a comparison of Bayesian and bootstrap measures of phylogenetic support. Molecular Phylogenetics and Evolution, 25, 361–371. Wollenberg, K.C., Lötters, S., Mora-Ferrer, C.A. & Veith, M. (2008) Disentangling composite colour and pattern ornaments. Biological Journal of the Linnean Society, 93, 433–444. Zimkus, B.M.& Schick, S (2010). Light at the end of the tunnel: insights into the molecular systematics of East African puddle frogs (Anura: Phrynobatrachidae). Systematics and Biodiversity, in press. APPENDIX. Material examined Hyperolius atrigularis: DRC: Kasiki: MRAC 18408 (holotype). Hyperolius cinnamomeoventris: 'cinnamomeoventris sensu stricto': ANGOLA: Andarada: MRAC 60559–60; Anhoca: BM 1904.5.2.116-118; Congulu: BM 1936.8.1.223-224; Dundo: MRCA 60557–58; Kalandula: AC 3008, AC 3017; Matala: MRAC 60291; River Kakueje: MRCA 60556 (all tentatively assigned due to geographic proximity to type locality); 'olivaceus': GABON: Lambarén: ZMB 8830, 65178 (two syntypes of H. fimbriolatus), ZMB 8829, 53264–265 (three syntypes of H. olivaceus), ZFMK 73111–112; Eastern Clade: DRC: Djalasiga: MRAC 2175 (holotype of H. ituriensis); Salonga National Park: A 185, A 186, A 195, A 200, A 201, A 205, A 230 (to be deposited at ZFMK); KENYA: Kakamega Forest: NMK A 3858/1–2, ZFMK 77431-433, ZFMK 77614; NMK A/3918/1–2; North Nandi Forest: NMK A/ 1353/1–3; Chemlit: NMK A/1422; UGANDA: Mabira Forest: NMK A/83/1–2, NMK A/3091, NMK A/2096/1–2; Budongo Forest: NMK A/952/1–3, NMK A/2094/1–11, NMK A/3349/1–6, NMK A/3951; Kampala: SL 565–566, SL 570, SL 575–576 (to be deposited at Makerere University, Kampala); Bundibugyo: SL 508–510 (to be deposited at Makerere University, Kampala); Bwamba: NMK A/963/1–2; Semliki: SL 532, SL 553–555 (to be deposited at Makerere University, Kampala); Sango Bay: NMK A/2092/1–5; Kibale Forest: NMK A/2093/1–5. Hyperolius polli: DRC: Tshimbulu s/Luebi (Kasai): MRAC 656 (holotype). Hyperolius schoutedeni: DRC: Kunungu: MRAC B.39837 (holotype, Fig. 4B); DRC: N’Sele: ZMUC R079832–833, ZMUC R079835. Hyperolius veithi: DRC: Salonga National Park: ZFMK 89607 (holotype, Fig. 4A), ZFMK 89608–645 (paratypes), ZMUC R771393–412 (paratypes). 36 · Zootaxa 2501 © 2010 Magnolia Press SCHICK ET AL. 

Molecular Ecology (2012) 21, 2565–2573 doi: 10.1111/j.1365-294X.2011.05418.x FROM THE COVER Monitoring endangered freshwater biodiversity using environmental DNA PHILIP FRANCIS THOMSEN,1* JOS KIELGAST,1* LARS L. IVERSEN,† CARSTEN WIUF,‡ M O R T E N R A S M U S S E N , * M . T H O M A S P . G I L B E R T , * L U D O V I C O R L A N D O * and E S K E WILLERSLEV* *Centre for GeoGenetics, Natural History Museum of Denmark, University of Copenhagen, Øster Voldgade 5-7, DK-1350 Copenhagen, Denmark, †Freshwater Biology Section, Department of Biology, University of Copenhagen, Helsingørgade 51, DK- 3400 Hillerød, Denmark, ‡Bioinformatics Research Center (BiRC), Aarhus University, C. F. Møllers Alle 8, DK-8000 A˚ rhus, Denmark Abstract Freshwater ecosystems are among the most endangered habitats on Earth, with thousands of animal species known to be threatened or already extinct. Reliable monitoring of threatened organisms is crucial for data-driven conservation actions but remains a challenge owing to nonstandardized methods that depend on practical and taxonomic expertise, which is rapidly declining. Here, we show that a diversity of rare and threatened freshwater animals—representing amphibians, fish, mammals, insects and crustaceans—can be detected and quantified based on DNA obtained directly from small water samples of lakes, ponds and streams. We successfully validate our findings in a controlled mesocosm experiment and show that DNA becomes undetectable within 2 weeks after removal of animals, indicating that DNA traces are near contemporary with presence of the species. We further demonstrate that entire faunas of amphibians and fish can be detected by high-throughput sequencing of DNA extracted from pond water. Our findings underpin the ubiquitous nature of DNA traces in the environment and establish environmental DNA as a tool for monitoring rare and threatened species across a wide range of taxonomic groups. Keywords: biological diversity, molecular detection, pyrosequencing, threatened species, wildlife conservation Received 19 August 2011; revision received 8 November 2011; accepted 17 November 2011 water animal species are threatened or recently Introduction extinct—representing more than a quarter of all freshwa- Monitoring of plant and animal biodiversity is conven- ter animals assessed so far (IUCNredlist 2011). tionally based on visual detection and counting. Such DNA obtained directly from environmental samples data collection is nonstandardized and dependent on (environmental DNA) as a method to assess the diver- practical and taxonomic expertise, which is rapidly sity of macro-organism communities was first applied declining (Hopkins & Freckleton 2002; Wheeler et al. to ancient sediments, revealing the past of extinct and 2004). Freshwater ecosystems are among the Earth’s most extant mammals, birds and plants (Willerslev et al. threatened habitats in terms of anthropogenic impact as 2003). Subsequently, the approach has been successfully well as global and local species loss (Revenga & Mock used on several different modern and ancient environ- 2000; Sala et al. 2000; Dudgeon et al. 2006; Vie´ et al. 2009; mental samples including terrestrial sediments, lake Hambler et al. 2011). Worldwide, more than 4600 fresh- and ice cores, and freshwater lakes and rivers (Hofreiter et al. 2003; Haile et al. 2007, 2009; Willerslev et al. 2007; Correspondence: Eske Willerslev Ficetola et al. 2008; Matisoo-Smith et al. 2008; Jerde E-mail: ewillerslev@snm.ku.dk et al. 2011). Faeces, urine and epidermal cells are 1 These authors contributed equally to this study. believed to be the predominant sources of environmen-  2011 Blackwell Publishing Ltd 2566 P . F . T H O M S E N E T A L . tal DNA (Lydolph et al. 2005; Haile et al. 2009), which from each site were taken to improve coverage of the may survive from hours to thousands of years depend- extent of the freshwater systems and species detection ing on the environmental setting (Willerslev et al. 2004). probability (Fig. S1, Supporting information). All sam- Although of great potential for contemporary biodiver- ples were stored at )20C until processed. A proxy sity monitoring, environmental DNA detection in wild for population density was calculated for the amphibi- populations has so far only been applied to a few com- ans Pelobates fuscus and Triturus cristatus using conven- mon or invasive species of amphibians and fish (Ficet- tional monitoring (based on active dip-netting and ola et al. 2008; Goldberg et al. 2011; Jerde et al. 2011). counting larvae one person-hour pr. pond) and The potential for monitoring rare and threatened spe- assuming a reverse cone shape for the estimation of cies of direct interest in a conservation context remains pond water volume using direct measures. Qualitative unreported. It also remains untested i) whether environ- occurrence data were supplied by taxon specialists mental DNA concentrations reflect species abundance based on fresh tracks or scat for the Eurasian otter in natural freshwater systems and ii) whether the (Lutra lutra), electrofishing with active dip-netting for approach is broadly applicable across taxonomic the European weather loach (Misgurnus fossilis) and groups. Answers to these questions are crucial for the active dip-netting for the large white-faced darter relevance and reliability of environmental DNA detec- (Leucorrhinia pectoralis) larvae and the tadpole shrimp tion to applied conservation biology. (Lepidurus apus). To address the potential of environmental DNA as a tool for monitoring rare and threatened freshwater spe- Mesocosm experiment cies, we conducted comparative surveys in natural lakes, ponds, streams and temporary pools in Europe. Outdoor experiments in aquatic mesocosms were set up We used conventional monitoring methods in parallel (at the Natural History Museum of Denmark) in a full with environmental mitochondrial DNA-based species factorial design for the two amphibian species at larval detection and quantification, by applying quantitative densities 0, 1, 2 or 4 specimens pr. 80 L. Two weeks PCR (qPCR) to DNA extracted from water samples. We prior to the introduction of experimental animals, all specifically surveyed six animal species representing containers were filled with tap water and inoculated different taxonomic groups: the amphibians common with identical quantities of water plants, nonpredatory spadefoot toad (Pelobates fuscus) and great crested newt invertebrates, filamentous algae, phytoplankton and (Triturus cristatus), the fish European weather loach zooplankton to simulate natural biotic complexity. (Misgurnus fossilis), the mammal Eurasian otter (Lutra Newt larvae and tadpoles were fed ad libitum with zoo- lutra), the dragonfly large white-faced darter (Leucorrhi- plankton and algal pellets (Tetra GmBH Plecomin) dur- nia pectoralis) and the crustacean tadpole shrimp (Le- ing the experiment. Water samples of 15 mL were taken pidurus apus). All species are locally rare and occur in before introduction of animals and after 2, 9, 23, 44 and low abundance in their natural environments (Helsdin- 64 days. Hereafter, animals were removed and samples gen et al. 1996; Conroy & Chanin 2000; Edgar & Bird were taken after 2, 9, 15 and 48 days (Table S2, Sup- 2006; Eggert et al. 2006; Brendonck et al. 2008; Hartvich porting information). et al. 2010). All except the tadpole shrimp are listed in the EU Habitat Directive (Council of the European DNA extraction, PCR and sequencing Union 1992) as requiring strict protection in their natu- ral habitats and substantial monitoring efforts in the DNA extraction and post-PCR work were performed in EU. We further examined our findings from wild popu- two separate laboratories assigned for these purposes. lations in a controlled mesocosm experiment and All water samples were centrifuged (35 min, 6C, explored the potential of DNA detection by 454 pyrose- 5000 g), and DNA from the pellet was extracted using quencing of PCR amplicons from environmental water Qiagen DNeasy Blood & Tissue kit (spin-column proto- samples targeting entire faunas of fish and amphibians. col). Extraction blanks were included for all DNA extrac- tions and were tested negative in subsequent PCRs. TaqMan qPCRs were performed on a Stratagene Materials and methods Mx3000P using 3 lL of template DNA, 15 lL of Taq- Man Environmental Master Mix 2.0 (Life Technolo- Field sampling and surveys gies), 4 lL of ddH2O, 1 lL of each primer (10 lM) and 3 · 15 mL water samples were collected in 98 natural 1 lL of probe (2.5 lM) under thermal cycling 50C for ponds, lakes and streams in Europe between 2009 and 5 min and 95C for 10 min, followed by 55 cycles of 2011, following Ficetola et al. 2008 (Fig. 1 and 95C for 30 s. and 50–60C for 1 min. Species-specific Table S1, Supporting information). The three samples primers and minor groove binding probes targeting  2011 Blackwell Publishing Ltd  S P E C I E S M O N I T O R I N G B Y E N V I R O N M E N T A L D N A 2567 Fig. 1 Sampling locations of the 90 European natural freshwater systems targeted in this study. Samples were taken in Denmark (DK), Sweden (S), Germany (D), Poland (PL) and Estonia (EST) and covers Tc (Triturus cristatus, 11 ponds), Pf (Pelobates fuscus, 17 ponds), Lp (Leucorrhinia pectoralis, 11 ponds), Ll (Lutra lutra, 15 streams and lakes), Mf (Misgurnus fossilis, 11 ponds and 15 streams) and La (Lepidurus apus, 10 temporary pools). An additional six ponds were sampled as controls and two additional ponds were sam- pled for 454 pyrosequencing (all in Denmark), giving a total of 98 freshwater systems sampled. For exact positions of all 98 localities see Table S1 (Supporting information). mitochondrial genes (cytochrome oxidase I and measured on a Nanodrop ND-1000. Three independent cytochrome b) were validated with relevant species qPCR replications were performed for each sample. occurring in the area. The amphibian primers ⁄ probe For all species, 25–60% of the positive field samples systems were tested negative for all amphibian species and 20–25% of the positive mesocosm samples were occurring in the sampled area Pelophylax kl. esculentus, validated as authentic by cloning using Topo TA clon- Rana arvalis, R. temporaria, R. dalmatina, Bufo bufo and ing kit (Invitrogen), followed by purification and Lissotriton vulgaris. The system for the fish M. fossilis sequencing of the inserted PCR fragment (Macrogen, was tested negative for Cobitis taenia, Anguilla anguilla, Europe) (Table S4, Supporting information). Final con- Tinca tinca, Carassius carassius, Rutilus rutilus and Cypri- centrations in DNA molecules ⁄ 15 mL of water sample nus carpio; the system for the dragonfly L. pectoralis was were calculated from the standards setting the molecu- tested negative for L. dubia, L. ribicunda, Anax imperator lar weight of DNA as 660 g ⁄ mol ⁄ base pair. Efficiency of and Cordulia aenea; the system for the crustacean L. apus all qPCRs with standards was 80–100%. was tested negative for Daphnia pulex, C. aenea and Dor- cus parallelipipedus; and the system for the mammal 454 Pyrosequencing L. lutra was tested negative for Mustela vison, Neomys fodiens and Homo sapiens. All primers and probes used Roche GS FLX 454 sequencing was performed on PCR and developed in this study are listed in Table S3 (Sup- products pooled from six PCR replicates performed on porting information). Negative controls were included DNA extracts from each pond. DNA extraction was for all PCRs and showed no amplification. identical to the rest of the study. However, 3 · 15 mL qPCR standards for the amphibian species were pre- of water samples were used for the fish community and pared as a dilution series (10)5–10)11) of purified PCR a pooled DNA extraction of 20 · 15 mL subsampled products on tissue-derived DNA with concentration from a 1.5-L water sample for the amphibian communi-  2011 Blackwell Publishing Ltd 2568 P . F . T H O M S E N E T A L . ties. Conventional PCRs were performed using 5 lL of When the animals have been removed from the con- DNA, 25 lL of TaqMan Environmental Master Mix 2.0 tainers, only DNA degradation occurs: (Life Technologies), 16 lL of ddH2O and 2 lL of each primer (10 lM) under thermal cycling conditions: 95C xðtÞ ¼ xðtR Þ expðcðt  tR ÞÞ ðeqn 3Þ for 10 min followed by 45 cycles of 94C for 30 s, 45– 48C for 30 s, 72C for 30 s with a final 72C for 5 min. For primer details see Table S3 (Supporting informa- where tR is the time of removal. Hence, there are three tion). PCR products were tested on 2% agarose gels parameters, a, b and c, to be estimated from the data stained with ethidium bromide and purified using a (Fig. S3, Supporting information). TC4.2 at t = 73 days Qiagen QIAquick PCR purification kit or QIAquick Gel was omitted in parameter estimation as it was not pos- extraction kit. Library builds were carried out using sible to replicate in qPCR. custom Y-shaped adaptors with MID barcode identifi- The observations yij are assumed to be independent ers, and all reactions were performed according to pro- of each other. Here, j = 1,…, n(ti) denotes the jth sample tocol using NEBnext DNA Sample Prep Master Mix Set obtained at time ti. The parameters are estimated using 2 (New England Biolabs, Ipswich, MA). Sequencing ordinary least square, weighted according to the num- was carried out in accordance with manufacturer’s ber of observations available from each container and guidelines. A total of 524 027 sequences were generated time point. Confidence intervals on parameter estimates on three-quarters of an XLR70 PTP (Roche, Basel, Swit- are obtained from the likelihood curve assuming data zerland). GS FLX light intensity files were sorted per are Gaussian distributed. The explained variance is cal- combination of primer and MID in separate files and culated as follows: trimmed accordingly before being used as input for AmpliconNoise and Perseus to remove sequencing P 1 ðyij  xðti ÞÞ2 errors and PCR chimeras (Quince et al. 2011). Given the n3 j;i length of the amplicons, the original procedure that r2 ¼ 1  P ðeqn 4Þ 1 n1 yÞ2 ðyij   keeps only reads where the first noisy flow occurred on i;j or after 360 was relaxed to flow number 100. Parame- ters rp and cp were set at the values 1 ⁄ 60 and 0.01, where n is the total number of observations and y  the respectively. Data were analysed using a custom-made mean of all observations. Perl script (available on request) and compared to the We used a linear mixed model to describe the rela- nt database using BLAST with 7 as word size and 0.001 tionship between DNA concentration, and time and as a maximal expect value and only considering density, respectively. Time and density were set as sequences with 100% identity in full sequence length. fixed effects, while individual containers were set as random effect. Two separate models (one with interac- tion between the factors time and density and one with- Statistical Modelling out interaction) were compared by a likelihood ratio To describe DNA concentration in water through time, test. Data were log10-transformed for Pearson’s prod- a differential equation model was constructed assuming uct–moment correlation to meet the assumption of nor- (i) DNA is generated at constant rate (i.e. secreted from mality (Fig. S4, Supporting information). All statistics the animal) but depends on the size of the animal(s), were performed using R version 2.13.1. here taken to be linear over time aÆt + b in the interval of observation, and (ii) DNA degradation occurs at a Results and discussion constant rate (Fig. S2, Supporting information). Here, cÆx(t), where x(t) is the amount of DNA present at time The success rate of the DNA-based species detection by t. The unit of the parameter c is per molecule per day. qPCR in ponds with known occurrence of the targeted This leads to an equation for the concentration of species was 100% for the fish, 91–100% for the amphib- DNA present at time t: ian species, 82% for the dragonfly and 100% for the tadpole shrimp (Fig. 2). Using the same strategy, nega- dx tive results were recovered for each of the six species ¼ a  t þ b  c  xðtÞand xð0Þ ¼ 0 ðeqn 1Þ dt from three control ponds where the respective species are known to be absent. Interestingly, for an additional It has the following solution x(t): eight sampled ponds with recent historical records of     P. fuscus, the species was not found during conven- a 1 a 1 a tional surveys. However, using the DNA detection xðtÞ ¼ t þ  b expðc  tÞ  b ðeqn 2Þ c c c c c approach, the presence of the species was confirmed in  2011 Blackwell Publishing Ltd  S P E C I E S M O N I T O R I N G B Y E N V I R O N M E N T A L D N A 2569 Fig. 2 Environmental DNA detection rates by qPCR in natural freshwater ponds with 100% occurrence of the species confirmed in the field (dark grey) or larger freshwater systems with known occurrence in the area (light grey). Detection rates are given in per- centage positive localities out of the total number of localities surveyed for each species. Data covers amphibians: Pf (Pelobates fuscus, n = 9) and Tc (Triturus cristatus, n = 11); fish: Mf (Misgurnus fossilis, n = 11 ponds and n = 15 streams—light grey); insects: Lp (Leucor- rhinia pectoralis, n = 11); crustaceans: La (Lepidurus apus, n = 10) and mammals: Ll (Lutra lutra, n = 15 streams and lakes). five of these sites, suggesting that the DNA approach throughout a continuous 225 km2 ditch system of run- may in some cases be more sensitive. Supporting this ning water that is known to be inhabited by the species. view, the respective five sites had lower average DNA The 54% success rate obtained (Fig. 2) was comparable concentration than the sites where the presence of to the results of a conventional expert survey in the P. fuscus was confirmed by expert surveys (P < 0.05, area. Considering water volume per individual and Mann–Whitney U test). For the amphibians, where water retention time, the difference between detection environmental DNA was quantified, we find positive probability in running and stagnant water systems is correlation between DNA concentration and estimated expected. Similarly, we tested the performance of envi- population density based on conventional monitoring ronmental DNA detection in large water volumes using (P. fuscus: P < 0.01, R2 = 0.68; T. cristatus: P < 0.05; streams and lakes inhabited by the Eurasian otter and R2 = 0.40, Pearson’s product–moment correlation) confirmed presence of species-specific DNA in 27% of (Fig. 3). the sampled sites (Fig. 2). The semiaquatic lifestyle and To examine the performance of environmental DNA large territorial range of this mammal can account for detection in running water, the fish M. fossilis was fur- the low detectability compared to the other investigated thermore targeted in independent water samples taken organisms. Nevertheless, for Eurasian otter, the (a) (b) Fig. 3 Environmental DNA quantification in natural ponds with Pelobates fuscus (n = 9) (a) and Triturus cristatus (n = 10) (b). Pear- son’s product moment correlation between average number of DNA molecules and estimated population size in each pond. The line shows linear regression, a: R2 = 0.68, P < 0.01; b: R2 = 0.40, P < 0.05.  2011 Blackwell Publishing Ltd 2570 P . F . T H O M S E N E T A L . environmental DNA approach may still be a valuable likely due to the fact that the herbivorous tadpole is complement to conventional monitoring (based on the substantially larger and more active than the carnivo- identification of tracks and faecal remains), which is rous newt larvae. Immediately after the animals were both resource demanding and error prone (Hansen & removed, we observed a rapid and continuous decrease Jacobsen 1999; Davison et al. 2002). in DNA concentration, until it could no longer be While our population density estimates based on con- detected only 1–2 weeks after removal (Fig. 4). These ventional monitoring methods are robust and compara- results suggest that DNA traces are near contemporary ble relative to each other, they serve only as proxies for with the presence of the species, in agreement with pre- true population densities. We therefore investigated the vious studies observing rapid degradation of DNA in consistency of the observed quantitative trend in the freshwater (Kim et al. 1996; Matsui et al. 2001; England relationship between DNA concentration and popula- et al. 2005; Douville et al. 2007; Dejean et al. 2011). tion density of the two amphibian species under semi- We speculate that the ability to detect and quantify natural conditions, allowing control of absolute animal DNA from a given freshwater animal species is deter- density through time. We quantified DNA concentra- mined as a simple relationship between DNA excretion tions by repeated water sampling from freshwater mes- depending on animal density and size, and degradation ocosms with densities of 0, 1, 2 or 4 larvae in 80 L of of this DNA owing to both microbial ⁄ enzymatic attack water, respectively. We sampled at 2, 9, 23, 44 and and spontaneous chemical reactions such as hydrolysis 64 days after introduction of animals to freshwater con- and oxidation (Lindahl 1993). Based on this general tainers. All animals were removed from the containers assumption, we integrated the observed DNA degrada- after 64 days when metamorphosis initiated, and DNA tion in the examination of the quantitative relation concentration was quantified after additional 2, 9, 15 between animal density and DNA concentration in a and 48 days to investigate DNA persistence (Table S2, simple differential equation. This model was con- Supporting information). structed assuming that DNA is generated at a rate For both species, we observe a highly significant dependent on the animal density and growth and effect of animal density and time on DNA concentration degraded by a constant rate. We find that the model quantified from the freshwater mesocosms as well as an parameters estimated from the data are in concordance interaction of the two factors (P. fuscus, P < 0.001; with each other across both species showing constant T. cristatus, P < 0.001; linear mixed model). This con- degradation and increasing excretion of DNA with firms our field observations in an experimental setting. increased density of animals and animal growth Interestingly, DNA concentrations were consistently (Fig. S3, Supporting information). higher for P. fuscus than for T. cristatus in both the con- The observed trends in both the field and controlled trolled experiment and the field survey (Figs 3 and 4), experiments support the conclusion that, despite rapid (a) (b) Fig. 4 Environmental DNA quantification in controlled mesocosm experiment with Pelobates fuscus (a) and Triturus cristatus (b). Means + 2 · SE of DNA molecules in water samples from freshwater containers with 1 (red), 2 (blue) or 4 (green) individuals in 80 L. After a control sample was taken, animals were introduced at time t = 0 and samples were taken at 2, 9, 23, 44, 64, 66, 73, 79 and 112 days. Animals were removed at t = 64 (after sampling). The lines show a differential equation model fitted to the data (see Materials and methods section), a: R2 = 0.29 (red), 0.50(blue), 0.61(green); b: R2 = 0.49 (red), 0.67 (blue), 0.62 (green).  2011 Blackwell Publishing Ltd  S P E C I E S M O N I T O R I N G B Y E N V I R O N M E N T A L D N A 2571 DNA degradation processes, there is a consistent quan- Table 1 Species of amphibians and fish detected by species titative relation between the density of animals and specific DNA in pond water samples. In each of the four ponds DNA molecules, which can be measured and accounted DNA fragments with 100% sequence match were recovered from all species known to occur, respectively. Sequences were for through time (Fig. 4). Overall, these findings consti- obtained through Roche 454 GS FLX sequencing using generic tute promising evidence that DNA may be not only primers except P. fuscus, T. tinca, P. fluviatilis and L. delinea- applied as an efficient tool to detect species in the envi- tus, which were recovered through PCR using species specific ronment but also used to estimate population densities. primers with subsequent cloning and Sanger sequencing. For However, this will necessitate rigorous species-specific the former three because the applied generic primers do not comparative studies to fine-tune model parameters and amplify tissue derived DNA of these species. JD11 (N55.79799, further validate the approach in natural freshwater E12.58399), HEL56 (N55.98929, E12.20933), ELL1 (N55.842498, E12.534903), BOT1 (N55.68651, E12.57432) (Datum: WGS84) environments. Moreover, the effect of factors such as temperature, pH, conductivity and microbial commu- Species Pond nity composition should be further investigated as these are likely to influence DNA decay and detectability. Amphibians Lissotriton vulgaris JD11, HEL56 Also, the exact cellular origin of environmental DNA in Triturus cristatus JD11 freshwater and the relative contribution of different Pelophylax kl. esculentus JD11, HEL56 states (e.g. free, cellular or particle-bound DNA) remain Rana temporaria HEL56 Rana arvalis HEL56 unclear, and clarification of this may focus future sam- Pelobates fuscus HEL56 pling strategies. Precipitation, as used in this study, Fish Carassius carassius ELL1, BOT1 recovers DNA independent of state but is limited to Carassius auratus ELL1, BOT1 small sample volumes compared to filtering methods Cyprinus carpio ELL1, BOT1 (e.g. Jerde et al. 2011), which accommodate larger sam- Scardinius erythrophthalmus ELL1, BOT1 ples but may fail to recover free DNA. Tinca tinca ELL1, BOT1 Finally, to explore the broad-scale potential of environ- Leucaspius delineatus ELL1 Perca fluviatilis BOT1 mental DNA-based species detection, we investigated the extent to which complete species diversity can be documented by environmental DNA screening. We used water samples from four ponds with well-known ing the taxonomic groups in question. It is inherent to amphibian or fish faunas (updated occurrence data from the use of generic primers that there is a trade-off the Danish freshwater fish atlas project and Amphi-Con- between targeting higher taxonomic levels and detecting sult Aps national amphibian monitoring data) and tar- rare sequences. geted DNA from these groups with a combination of specific and generic primers (Table S3, Supporting infor- Conclusion mation). PCR products were sequenced using the Roche GS FLX 454 platform and Sanger sequencing, generating Faced with a global decline in biodiversity that is 100– a total of 524 027 sequences. 1000 times faster than prehuman rates (Pimm et al. We recovered species-specific DNA fragments with 1995; Barnosky et al. 2011), there is an urgent need for 100% sequence match for all species of amphibians or data-driven prioritization of conservation actions, which fish previously recorded from each of the ponds relies heavily on fast and effective monitoring of threa- (Table 1 and Table S4, Supporting information). Interest- tened species. Environmental DNA monitoring cannot ingly, we furthermore recovered DNA sequences from replace field observations by experienced ecologists and species living in close proximity to the water, including taxon specialists, who retrieve information beyond birds: Eurasian coot (Fulica atra), wood pigeon (Columba quantitative and qualitative records. However, monitor- palumbus) and marsh warbler (Acrocephalus palustris) and ing of threatened species through environmental DNA red deer (Cervus elaphus). These results suggest that the may be a quick, cost-effective and standardized way to success of DNA detection is largely independent of ani- obtain basic data on distribution and abundance, mal species and abundance, as long as DNA is excreted enabling efficient deployment of limited conservation into the water. Furthermore, this illustrates that DNA is resources and taxonomic expertise. Further research on homogeneously distributed in pond water, in stark con- environmental DNA in relation to conservation of rare trast to recent observations of animal DNA in soil, char- and threatened species should focus on large-scale com- acterized by a patchy distribution (Andersen et al. 2011). parative validation and optimization including addi- The ability to exhaustively recover all species in the tional organismal groups and applying the approach investigated faunas of fish and amphibians probably beyond freshwater ecosystems. However, our findings relies on the design of generic primers specifically target- highlight a vast potential for integrating DNA detection  2011 Blackwell Publishing Ltd 2572 P . F . T H O M S E N E T A L . in the tool set of biodiversity field research and conser- major influence on current population structure and status. vation. With DNA sequencing technology advancing at Conservation Genetics, 7, 185–195. rapidly dropping costs (Metzker 2009; Anonymous England L, Pollok J, Vincent M et al. (2005) Persistence of extracellular baculoviral DNA in aquatic microcosms: 2010), environmental DNA research is set to change extraction, purification, and amplification by the from being merely a scientific curiosity to become an polymerase chain reaction (PCR). Molecular and Cellular important tool in applied field biology. Probes, 19, 75–80. Ficetola GF, Miaud C, Pompanon F, Taberlet P (2008) Species detection using environmental DNA from water samples. Acknowledgements Biology Letters, 4, 423–425. We thank Dr. P.R. Møller (Natural History Museum of Den- Goldberg CS, Pilliod DS, Arkle RS, Waits LP (2011) Molecular mark), A. Drews (LLUR, Schleswig-Holstein), A. Linnet (Dan- detection of vertebrates in stream water: a demonstration ish Nature Agency, Thy) and Martin Hesselsøe (Amphi- using rocky mountain tailed frogs and IDAHO giant Consult Aps) for generously supplying monitoring data. All salamanders. PLoS ONE, 6, e22746. work involving live animals was conducted under permit Haile J, Holdaway R, Oliver K et al. (2007) Ancient DNA number SNS-441-00116 of the Danish Nature Agency. This chronology within sediment deposits: are paleobiological study was supported by the University of Copenhagen, The reconstructions possible and is DNA leaching a factor? Natural History Museum of Denmark and the Danish National Molecular Biology and Evolution, 24, 982–989. Research Foundation. Haile J, Froese DG, MacPhee RDE et al. (2009) Ancient DNA reveals late survival of mammoth and horse in interior Alaska. Proceedings of the National Academy of Sciences of the References United States of America, 106, 22352–22357. Hambler C, Henderson PA, Speight MR (2011) Extinction rates, Andersen K, Bird KL, Rasmussen M et al. (2011) extinction-prone habitats, and indicator groups in Britain Metabarcoding of ‘Dirt’ DNA from soil reflects vertebrate and at larger scales. Biological Conservation, 144, 713–721. biodiversity. Molecular Ecology, doi: 10.1111/j.1365-294X. Hansen MM, Jacobsen L (1999) Identification of mustelid 2011.05261.x. species: otter (Lutra lutra), American mink (Mustela vison) Anonymous (2010) Human genome at ten: the sequence and polecat (Mustela putorius), by analysis of DNA from explosion. Nature, 464, 670–671. faecal samples. Journal of Zoology, 247, 177–181. Barnosky AD, Matzke N, Tomiya S et al. (2011) Has the Hartvich P, Lusk S, Rutkayova´ J (2010) Threatened fishes of Earth’s sixth mass extinction already arrived?. Nature, 471, the world: Misgurnus fossilis (Linnaeus, 1758)(Cobitidae). 51–57. Environmental Biology of Fishes, 87, 39–40. Brendonck L, Rogers DC, Olesen J, Weeks S, Hoeh WR (2008) Helsdingen P, Willemse L, Speight M (1996) Background Information Global diversity of large branchiopods (Crustacea: on Invertebrates of the Habitats Directive and the Bern Convention, Branchiopoda) in freshwater. Hydrobiologia, 595, 167–176. Part 2: Mantodea, Odonata, Orthoptera and Arachnida. pp. 1–177. Conroy JWH, Chanin RF (2000) Status of the Eurasian otter Council of the European Union, Strassbourg, Germany. (Lutra lutra) in Europe—a review. Journal of the International Hofreiter M, Mead JI, Martin P, Poinar HN (2003) Molecular Otter Survival Fund, 1, 7–28. caving. Current Biology, 13, 693–695. Council of the European Union (1992) Council Directive Hopkins G, Freckleton R (2002) Declines in the numbers of 92 ⁄ 43 ⁄ EEC on the Conservation of Natural Habitats and of Wild amateur and professional taxonomists: implications for Fauna and Flora. pp. 1–66. Council of the European Union, conservation. Animal Conservation, 5, 245–249. Strassbourg, Germany. IUCNredlist (2011) IUCN Red List of Threatened Species. Version Davison A, Birks JDS, Brookes RC, Braithwaite TC, Messenger 2010.4. IUCNredlist, Webpage: http://www.iucnredlist.org, JE (2002) On the origin of faeces: morphological versus accessed 15 march 2011. molecular methods for surveying rare carnivores from their Jerde CL, Mahon AR, Chadderton WL, Lodge DM (2011) scats. Journal of Zoology, 257, 141–143. ‘‘Sight-unseen’’ detection of rare aquatic species using Dejean T, Valentini A, Duparc A et al. (2011) Persistence of environmental DNA. Conservation Letters, 4, 150–157. environmental DNA in freshwater ecosystems. PLoS ONE, 6, Kim CK, Kwak MJ, Lee SG (1996) Structural and functional e23398. stability of the genetic recombinant plasmid pCU103 in Douville M, Gagne´ F, Blaise C, Andre´ C (2007) Occurrence and different water environments. Journal of Microbiology (Korea persistence of Bacillus thuringiensis (Bt) and transgenic Bt Republic), 34, 241–247. corn cry1Ab gene from an aquatic environment. Ecotoxicology Lindahl T (1993) Instability and decay of the primary structure and Environmental Safety, 66, 195–203. of DNA. Nature, 362, 709–715. Dudgeon D, Arthington AH, Gessner MO et al. (2006) Lydolph MC, Jacobsen J, Arctander P et al. (2005) Beringian Freshwater biodiversity: importance, threats, status and paleoecology inferred from permafrost-preserved fungal conservation challenges. Biological Reviews, 81, 163–182. DNA. Applied and Environmental Microbiology, 71, 1012–1017. Edgar P, Bird DR (2006) Action plan for the Conservation of the Matisoo-Smith E, Roberts K, Welikala N et al. (2008) Recovery Crested Newt Triturus cristatus Species Complex in Europe, pp. of DNA and pollen from New Zealand lake sediments. 1–33. Council of the European Union, Strassbourg, Germany. Quaternary International, 184, 139–149. Eggert C, Cogalniceanu D, Veith M, Dzukic G, Taberlet P Matsui K, Honjo M, Kawabata Z (2001) Estimation of the fate (2006) The declining Spadefoot toad, Pelobates fuscus of dissolved DNA in thermally stratified lake water from the (Pelobatidae): paleo and recent environmental changes as a  2011 Blackwell Publishing Ltd  S P E C I E S M O N I T O R I N G B Y E N V I R O N M E N T A L D N A 2573 stability of exogenous plasmid DNA. Aquatic Microbial logical systems. M.R, M.T.P.G, L.O and E.W (Centre for Goe- Ecology, 26, 95–102. Genetics, Copenhagen University) are developing methods for Metzker ML (2009) Sequencing technologies—the next analyzing ancient and fragmented DNA from various environ- generation. Nature Reviews Genetics, 11, 31–46. mental samples. Pimm SL, Russell GJ, Gittleman JL, Brooks TM (1995) The future of biodiversity. Science, 269, 347–350. Quince C, Lanzen A, Davenport RJ, Turnbaugh PJ (2011) Removing noise from pyrosequenced amplicons. BMC Data accessibility Bioinformatics, 12, 239. Sequences from pyrosequencing are uploaded to NCBI SRA: Revenga C, Mock G (2000) Pilot Analysis of Global Ecosystems: ERP000988. Freshwater Systems. pp. 1–78. World Resource Institute, Washington, District of Columbia, USA. Sala OE, Stuart Chapin F III, Armesto JJ et al. (2000) Global biodiversity scenarios for the year 2100. Science, 287, 1770– Supporting information 1774. Additional supporting information may be found in the online Vie´ J, Hilton-Taylor C, Stuart S (2009) Wildlife in a Changing version of this article. World. An Analysis of the 2008 IUCN Red List of Threatened Species. pp. 1–184, World Conservation Union, Gland, Table S1 Overview of the results of field experiments. Switzerland. Wheeler Q, Raven P, Wilson E (2004) Taxonomy: impediment Table S2 Overview of the results of controlled mesocosm or expedient? Science, 303, 285. experiments. Willerslev E, Hansen AJ, Binladen J et al. (2003) Diverse plant Table S3 Primers and probes designed and used in this study. and animal genetic records from holocene and pleistocene sediments. Science, 300, 791–795. Table S4 Summary of DNA sequences recovered in this study. Willerslev E, Hansen AJ, Poinar HN (2004) Isolation of nucleic Fig. S1 Accumulated probabilities of detecting the targeted acids and cultures from fossil ice and permafrost. Trends in species in the field studies when taking 1, 2 or 3 samples. Ecology & Evolution, 19, 141–147. Willerslev E, Cappellini E, Boomsma W et al. (2007) Ancient Fig. S2 Model fit on the DNA concentration in individual con- biomolecules from deep ice cores reveal a forested Southern tainers of the mesocosm experiment through time. Greenland. Science, 317, 111–114. Fig. S3 Modeled parameter estimates. Fig. S4 QQ-plot for mesocosm experiments. P.F.T, J.K (Centre for GeoGenetics, Copenhagen University) and L.L.I (Freshwater Biology Section, Copenhagen University) Please note: Wiley-Blackwell is not responsible for the content are developing methods for monitoring biodiversity using or functionality of any supporting information supplied by the environmental DNA. C.W (Bioinformatics Research Center, authors. Any queries (other than missing material) should be Aarhus University) is working with statistical modeling of bio- directed to the corresponding author for the article.  2011 Blackwell Publishing Ltd 

Molecular Ecology (2011) doi: 10.1111/j.1365-294X.2011.05418.x FROM THE COVER Monitoring endangered freshwater biodiversity using environmental DNA PHILIP FRANCIS THOMSEN,1* JOS KIELGAST,1* LARS L. IVERSEN,† CARSTEN WIUF,‡ M O R T E N R A S M U S S E N , * M . T H O M A S P . G I L B E R T , * L U D O V I C O R L A N D O * and E S K E WILLERSLEV* *Centre for GeoGenetics, Natural History Museum of Denmark, University of Copenhagen, Øster Voldgade 5-7, DK-1350 Copenhagen, Denmark, †Freshwater Biology Section, Department of Biology, University of Copenhagen, Helsingørgade 51, DK- 3400 Hillerød, Denmark, ‡Bioinformatics Research Center (BiRC), Aarhus University, C. F. Møllers Alle 8, DK-8000 A˚ rhus, Denmark Abstract Freshwater ecosystems are among the most endangered habitats on Earth, with thousands of animal species known to be threatened or already extinct. Reliable monitoring of threatened organisms is crucial for data-driven conservation actions but remains a challenge owing to nonstandardized methods that depend on practical and taxonomic expertise, which is rapidly declining. Here, we show that a diversity of rare and threatened freshwater animals—representing amphibians, fish, mammals, insects and crustaceans—can be detected and quantified based on DNA obtained directly from small water samples of lakes, ponds and streams. We successfully validate our findings in a controlled mesocosm experiment and show that DNA becomes undetectable within 2 weeks after removal of animals, indicating that DNA traces are near contemporary with presence of the species. We further demonstrate that entire faunas of amphibians and fish can be detected by high-throughput sequencing of DNA extracted from pond water. Our findings underpin the ubiquitous nature of DNA traces in the environment and establish environmental DNA as a tool for monitoring rare and threatened species across a wide range of taxonomic groups. Keywords: biological diversity, molecular detection, pyrosequencing, threatened species, wildlife conservation Received 19 August 2011; revision received 8 November 2011; accepted 17 November 2011 water animal species are threatened or recently Introduction extinct—representing more than a quarter of all freshwa- Monitoring of plant and animal biodiversity is conven- ter animals assessed so far (IUCNredlist 2011). tionally based on visual detection and counting. Such DNA obtained directly from environmental samples data collection is nonstandardized and dependent on (environmental DNA) as a method to assess the diver- practical and taxonomic expertise, which is rapidly sity of macro-organism communities was first applied declining (Hopkins & Freckleton 2002; Wheeler et al. to ancient sediments, revealing the past of extinct and 2004). Freshwater ecosystems are among the Earth’s most extant mammals, birds and plants (Willerslev et al. threatened habitats in terms of anthropogenic impact as 2003). Subsequently, the approach has been successfully well as global and local species loss (Revenga & Mock used on several different modern and ancient environ- 2000; Sala et al. 2000; Dudgeon et al. 2006; Vie´ et al. 2009; mental samples including terrestrial sediments, lake Hambler et al. 2011). Worldwide, more than 4600 fresh- and ice cores, and freshwater lakes and rivers (Hofreiter et al. 2003; Haile et al. 2007, 2009; Willerslev et al. 2007; Correspondence: Eske Willerslev Ficetola et al. 2008; Matisoo-Smith et al. 2008; Jerde E-mail: ewillerslev@snm.ku.dk et al. 2011). Faeces, urine and epidermal cells are 1 These authors contributed equally to this study. believed to be the predominant sources of environmen-  2011 Blackwell Publishing Ltd 2 P. F. THOMSEN ET AL. tal DNA (Lydolph et al. 2005; Haile et al. 2009), which from each site were taken to improve coverage of the may survive from hours to thousands of years depend- extent of the freshwater systems and species detection ing on the environmental setting (Willerslev et al. 2004). probability (Fig. S1, Supporting information). All sam- Although of great potential for contemporary biodiver- ples were stored at )20C until processed. A proxy sity monitoring, environmental DNA detection in wild for population density was calculated for the amphibi- populations has so far only been applied to a few com- ans Pelobates fuscus and Triturus cristatus using conven- mon or invasive species of amphibians and fish (Ficet- tional monitoring (based on active dip-netting and ola et al. 2008; Goldberg et al. 2011; Jerde et al. 2011). counting larvae one person-hour pr. pond) and The potential for monitoring rare and threatened spe- assuming a reverse cone shape for the estimation of cies of direct interest in a conservation context remains pond water volume using direct measures. Qualitative unreported. It also remains untested i) whether environ- occurrence data were supplied by taxon specialists mental DNA concentrations reflect species abundance based on fresh tracks or scat for the Eurasian otter in natural freshwater systems and ii) whether the (Lutra lutra), electrofishing with active dip-netting for approach is broadly applicable across taxonomic the European weather loach (Misgurnus fossilis) and groups. Answers to these questions are crucial for the active dip-netting for the large white-faced darter relevance and reliability of environmental DNA detec- (Leucorrhinia pectoralis) larvae and the tadpole shrimp tion to applied conservation biology. (Lepidurus apus). To address the potential of environmental DNA as a tool for monitoring rare and threatened freshwater spe- Mesocosm experiment cies, we conducted comparative surveys in natural lakes, ponds, streams and temporary pools in Europe. Outdoor experiments in aquatic mesocosms were set up We used conventional monitoring methods in parallel (at the Natural History Museum of Denmark) in a full with environmental mitochondrial DNA-based species factorial design for the two amphibian species at larval detection and quantification, by applying quantitative densities 0, 1, 2 or 4 specimens pr. 80 L. Two weeks PCR (qPCR) to DNA extracted from water samples. We prior to the introduction of experimental animals, all specifically surveyed six animal species representing containers were filled with tap water and inoculated different taxonomic groups: the amphibians common with identical quantities of water plants, nonpredatory spadefoot toad (Pelobates fuscus) and great crested newt invertebrates, filamentous algae, phytoplankton and (Triturus cristatus), the fish European weather loach zooplankton to simulate natural biotic complexity. (Misgurnus fossilis), the mammal Eurasian otter (Lutra Newt larvae and tadpoles were fed ad libitum with zoo- lutra), the dragonfly large white-faced darter (Leucorrhi- plankton and algal pellets (Tetra GmBH Plecomin) dur- nia pectoralis) and the crustacean tadpole shrimp (Le- ing the experiment. Water samples of 15 mL were taken pidurus apus). All species are locally rare and occur in before introduction of animals and after 2, 9, 23, 44 and low abundance in their natural environments (Helsdin- 64 days. Hereafter, animals were removed and samples gen et al. 1996; Conroy & Chanin 2000; Edgar & Bird were taken after 2, 9, 15 and 48 days (Table S2, Sup- 2006; Eggert et al. 2006; Brendonck et al. 2008; Hartvich porting information). et al. 2010). All except the tadpole shrimp are listed in the EU Habitat Directive (Council of the European DNA extraction, PCR and sequencing Union 1992) as requiring strict protection in their natu- ral habitats and substantial monitoring efforts in the DNA extraction and post-PCR work were performed in EU. We further examined our findings from wild popu- two separate laboratories assigned for these purposes. lations in a controlled mesocosm experiment and All water samples were centrifuged (35 min, 6C, explored the potential of DNA detection by 454 pyrose- 5000 g), and DNA from the pellet was extracted using quencing of PCR amplicons from environmental water Qiagen DNeasy Blood & Tissue kit (spin-column proto- samples targeting entire faunas of fish and amphibians. col). Extraction blanks were included for all DNA extrac- tions and were tested negative in subsequent PCRs. TaqMan qPCRs were performed on a Stratagene Materials and methods Mx3000P using 3 lL of template DNA, 15 lL of Taq- Man Environmental Master Mix 2.0 (Life Technolo- Field sampling and surveys gies), 4 lL of ddH2O, 1 lL of each primer (10 lM) and 3 · 15 mL water samples were collected in 98 natural 1 lL of probe (2.5 lM) under thermal cycling 50C for ponds, lakes and streams in Europe between 2009 and 5 min and 95C for 10 min, followed by 55 cycles of 2011, following Ficetola et al. 2008 (Fig. 1 and 95C for 30 s. and 50–60C for 1 min. Species-specific Table S1, Supporting information). The three samples primers and minor groove binding probes targeting  2011 Blackwell Publishing Ltd  SPECIES MONITORING BY ENVIRONMENTAL DNA 3 Fig. 1 Sampling locations of the 90 European natural freshwater systems targeted in this study. Samples were taken in Denmark (DK), Sweden (S), Germany (D), Poland (PL) and Estonia (EST) and covers Tc (Triturus cristatus, 11 ponds), Pf (Pelobates fuscus, 17 ponds), Lp (Leucorrhinia pectoralis, 11 ponds), Ll (Lutra lutra, 15 streams and lakes), Mf (Misgurnus fossilis, 11 ponds and 15 streams) and La (Lepidurus apus, 10 temporary pools). An additional six ponds were sampled as controls and two additional ponds were sam- pled for 454 pyrosequencing (all in Denmark), giving a total of 98 freshwater systems sampled. For exact positions of all 98 localities see Table S1 (Supporting information). mitochondrial genes (cytochrome oxidase I and measured on a Nanodrop ND-1000. Three independent cytochrome b) were validated with relevant species qPCR replications were performed for each sample. occurring in the area. The amphibian primers ⁄ probe For all species, 25–60% of the positive field samples systems were tested negative for all amphibian species and 20–25% of the positive mesocosm samples were occurring in the sampled area Pelophylax kl. esculentus, validated as authentic by cloning using Topo TA clon- Rana arvalis, R. temporaria, R. dalmatina, Bufo bufo and ing kit (Invitrogen), followed by purification and Lissotriton vulgaris. The system for the fish M. fossilis sequencing of the inserted PCR fragment (Macrogen, was tested negative for Cobitis taenia, Anguilla anguilla, Europe) (Table S4, Supporting information). Final con- Tinca tinca, Carassius carassius, Rutilus rutilus and Cypri- centrations in DNA molecules ⁄ 15 mL of water sample nus carpio; the system for the dragonfly L. pectoralis was were calculated from the standards setting the molecu- tested negative for L. dubia, L. ribicunda, Anax imperator lar weight of DNA as 660 g ⁄ mol ⁄ base pair. Efficiency of and Cordulia aenea; the system for the crustacean L. apus all qPCRs with standards was 80–100%. was tested negative for Daphnia pulex, C. aenea and Dor- cus parallelipipedus; and the system for the mammal 454 Pyrosequencing L. lutra was tested negative for Mustela vison, Neomys fodiens and Homo sapiens. All primers and probes used Roche GS FLX 454 sequencing was performed on PCR and developed in this study are listed in Table S3 (Sup- products pooled from six PCR replicates performed on porting information). Negative controls were included DNA extracts from each pond. DNA extraction was for all PCRs and showed no amplification. identical to the rest of the study. However, 3 · 15 mL qPCR standards for the amphibian species were pre- of water samples were used for the fish community and pared as a dilution series (10)5–10)11) of purified PCR a pooled DNA extraction of 20 · 15 mL subsampled products on tissue-derived DNA with concentration from a 1.5-L water sample for the amphibian communi-  2011 Blackwell Publishing Ltd 4 P. F. THOMSEN ET AL. ties. Conventional PCRs were performed using 5 lL of When the animals have been removed from the con- DNA, 25 lL of TaqMan Environmental Master Mix 2.0 tainers, only DNA degradation occurs: (Life Technologies), 16 lL of ddH2O and 2 lL of each primer (10 lM) under thermal cycling conditions: 95C xðtÞ ¼ xðtR Þ expðcðt  tR ÞÞ ðeqn 3Þ for 10 min followed by 45 cycles of 94C for 30 s, 45– 48C for 30 s, 72C for 30 s with a final 72C for 5 min. For primer details see Table S3 (Supporting informa- where tR is the time of removal. Hence, there are three tion). PCR products were tested on 2% agarose gels parameters, a, b and c, to be estimated from the data stained with ethidium bromide and purified using a (Fig. S3, Supporting information). TC4.2 at t = 73 days Qiagen QIAquick PCR purification kit or QIAquick Gel was omitted in parameter estimation as it was not pos- extraction kit. Library builds were carried out using sible to replicate in qPCR. custom Y-shaped adaptors with MID barcode identifi- The observations yij are assumed to be independent ers, and all reactions were performed according to pro- of each other. Here, j = 1,…, n(ti) denotes the jth sample tocol using NEBnext DNA Sample Prep Master Mix Set obtained at time ti. The parameters are estimated using 2 (New England Biolabs, Ipswich, MA). Sequencing ordinary least square, weighted according to the num- was carried out in accordance with manufacturer’s ber of observations available from each container and guidelines. A total of 524 027 sequences were generated time point. Confidence intervals on parameter estimates on three-quarters of an XLR70 PTP (Roche, Basel, Swit- are obtained from the likelihood curve assuming data zerland). GS FLX light intensity files were sorted per are Gaussian distributed. The explained variance is cal- combination of primer and MID in separate files and culated as follows: trimmed accordingly before being used as input for AmpliconNoise and Perseus to remove sequencing P 1 ðyij  xðti ÞÞ2 errors and PCR chimeras (Quince et al. 2011). Given the n3 j;i length of the amplicons, the original procedure that r2 ¼ 1  P ðeqn 4Þ 1 n1 yÞ2 ðyij   keeps only reads where the first noisy flow occurred on i;j or after 360 was relaxed to flow number 100. Parame- ters rp and cp were set at the values 1 ⁄ 60 and 0.01, where n is the total number of observations and y  the respectively. Data were analysed using a custom-made mean of all observations. Perl script (available on request) and compared to the We used a linear mixed model to describe the rela- nt database using BLAST with 7 as word size and 0.001 tionship between DNA concentration, and time and as a maximal expect value and only considering density, respectively. Time and density were set as sequences with 100% identity in full sequence length. fixed effects, while individual containers were set as random effect. Two separate models (one with interac- tion between the factors time and density and one with- Statistical Modelling out interaction) were compared by a likelihood ratio To describe DNA concentration in water through time, test. Data were log10-transformed for Pearson’s prod- a differential equation model was constructed assuming uct–moment correlation to meet the assumption of nor- (i) DNA is generated at constant rate (i.e. secreted from mality (Fig. S4, Supporting information). All statistics the animal) but depends on the size of the animal(s), were performed using R version 2.13.1. here taken to be linear over time aÆt + b in the interval of observation, and (ii) DNA degradation occurs at a Results and discussion constant rate (Fig. S2, Supporting information). Here, cÆx(t), where x(t) is the amount of DNA present at time The success rate of the DNA-based species detection by t. The unit of the parameter c is per molecule per day. qPCR in ponds with known occurrence of the targeted This leads to an equation for the concentration of species was 100% for the fish, 91–100% for the amphib- DNA present at time t: ian species, 82% for the dragonfly and 100% for the tadpole shrimp (Fig. 2). Using the same strategy, nega- dx tive results were recovered for each of the six species ¼ a  t þ b  c  xðtÞand xð0Þ ¼ 0 ðeqn 1Þ dt from three control ponds where the respective species are known to be absent. Interestingly, for an additional It has the following solution x(t): eight sampled ponds with recent historical records of     P. fuscus, the species was not found during conven- a 1 a 1 a tional surveys. However, using the DNA detection xðtÞ ¼ t þ  b expðc  tÞ  b ðeqn 2Þ c c c c c approach, the presence of the species was confirmed in  2011 Blackwell Publishing Ltd  SPECIES MONITORING BY ENVIRONMENTAL DNA 5 Fig. 2 Environmental DNA detection rates by qPCR in natural freshwater ponds with 100% occurrence of the species confirmed in the field (dark grey) or larger freshwater systems with known occurrence in the area (light grey). Detection rates are given in per- centage positive localities out of the total number of localities surveyed for each species. Data covers amphibians: Pf (Pelobates fuscus, n = 9) and Tc (Triturus cristatus, n = 11); fish: Mf (Misgurnus fossilis, n = 11 ponds and n = 15 streams—light grey); insects: Lp (Leucor- rhinia pectoralis, n = 11); crustaceans: La (Lepidurus apus, n = 10) and mammals: Ll (Lutra lutra, n = 15 streams and lakes). five of these sites, suggesting that the DNA approach throughout a continuous 225 km2 ditch system of run- may in some cases be more sensitive. Supporting this ning water that is known to be inhabited by the species. view, the respective five sites had lower average DNA The 54% success rate obtained (Fig. 2) was comparable concentration than the sites where the presence of to the results of a conventional expert survey in the P. fuscus was confirmed by expert surveys (P < 0.05, area. Considering water volume per individual and Mann–Whitney U test). For the amphibians, where water retention time, the difference between detection environmental DNA was quantified, we find positive probability in running and stagnant water systems is correlation between DNA concentration and estimated expected. Similarly, we tested the performance of envi- population density based on conventional monitoring ronmental DNA detection in large water volumes using (P. fuscus: P < 0.01, R2 = 0.68; T. cristatus: P < 0.05; streams and lakes inhabited by the Eurasian otter and R2 = 0.40, Pearson’s product–moment correlation) confirmed presence of species-specific DNA in 27% of (Fig. 3). the sampled sites (Fig. 2). The semiaquatic lifestyle and To examine the performance of environmental DNA large territorial range of this mammal can account for detection in running water, the fish M. fossilis was fur- the low detectability compared to the other investigated thermore targeted in independent water samples taken organisms. Nevertheless, for Eurasian otter, the (a) (b) Fig. 3 Environmental DNA quantification in natural ponds with Pelobates fuscus (n = 9) (a) and Triturus cristatus (n = 10) (b). Pear- son’s product moment correlation between average number of DNA molecules and estimated population size in each pond. The line shows linear regression, a: R2 = 0.68, P < 0.01; b: R2 = 0.40, P < 0.05.  2011 Blackwell Publishing Ltd 6 P. F. THOMSEN ET AL. environmental DNA approach may still be a valuable likely due to the fact that the herbivorous tadpole is complement to conventional monitoring (based on the substantially larger and more active than the carnivo- identification of tracks and faecal remains), which is rous newt larvae. Immediately after the animals were both resource demanding and error prone (Hansen & removed, we observed a rapid and continuous decrease Jacobsen 1999; Davison et al. 2002). in DNA concentration, until it could no longer be While our population density estimates based on con- detected only 1–2 weeks after removal (Fig. 4). These ventional monitoring methods are robust and compara- results suggest that DNA traces are near contemporary ble relative to each other, they serve only as proxies for with the presence of the species, in agreement with pre- true population densities. We therefore investigated the vious studies observing rapid degradation of DNA in consistency of the observed quantitative trend in the freshwater (Kim et al. 1996; Matsui et al. 2001; England relationship between DNA concentration and popula- et al. 2005; Douville et al. 2007; Dejean et al. 2011). tion density of the two amphibian species under semi- We speculate that the ability to detect and quantify natural conditions, allowing control of absolute animal DNA from a given freshwater animal species is deter- density through time. We quantified DNA concentra- mined as a simple relationship between DNA excretion tions by repeated water sampling from freshwater mes- depending on animal density and size, and degradation ocosms with densities of 0, 1, 2 or 4 larvae in 80 L of of this DNA owing to both microbial ⁄ enzymatic attack water, respectively. We sampled at 2, 9, 23, 44 and and spontaneous chemical reactions such as hydrolysis 64 days after introduction of animals to freshwater con- and oxidation (Lindahl 1993). Based on this general tainers. All animals were removed from the containers assumption, we integrated the observed DNA degrada- after 64 days when metamorphosis initiated, and DNA tion in the examination of the quantitative relation concentration was quantified after additional 2, 9, 15 between animal density and DNA concentration in a and 48 days to investigate DNA persistence (Table S2, simple differential equation. This model was con- Supporting information). structed assuming that DNA is generated at a rate For both species, we observe a highly significant dependent on the animal density and growth and effect of animal density and time on DNA concentration degraded by a constant rate. We find that the model quantified from the freshwater mesocosms as well as an parameters estimated from the data are in concordance interaction of the two factors (P. fuscus, P < 0.001; with each other across both species showing constant T. cristatus, P < 0.001; linear mixed model). This con- degradation and increasing excretion of DNA with firms our field observations in an experimental setting. increased density of animals and animal growth Interestingly, DNA concentrations were consistently (Fig. S3, Supporting information). higher for P. fuscus than for T. cristatus in both the con- The observed trends in both the field and controlled trolled experiment and the field survey (Figs 3 and 4), experiments support the conclusion that, despite rapid (a) (b) Fig. 4 Environmental DNA quantification in controlled mesocosm experiment with Pelobates fuscus (a) and Triturus cristatus (b). Means + 2 · SE of DNA molecules in water samples from freshwater containers with 1 (red), 2 (blue) or 4 (green) individuals in 80 L. After a control sample was taken, animals were introduced at time t = 0 and samples were taken at 2, 9, 23, 44, 64, 66, 73, 79 and 112 days. Animals were removed at t = 64 (after sampling). The lines show a differential equation model fitted to the data (see Materials and methods section), a: R2 = 0.29 (red), 0.50(blue), 0.61(green); b: R2 = 0.49 (red), 0.67 (blue), 0.62 (green).  2011 Blackwell Publishing Ltd  SPECIES MONITORING BY ENVIRONMENTAL DNA 7 DNA degradation processes, there is a consistent quan- Table 1 Species of amphibians and fish detected by species titative relation between the density of animals and specific DNA in pond water samples. In each of the four ponds DNA molecules, which can be measured and accounted DNA fragments with 100% sequence match were recovered from all species known to occur, respectively. Sequences were for through time (Fig. 4). Overall, these findings consti- obtained through Roche 454 GS FLX sequencing using generic tute promising evidence that DNA may be not only primers except P. fuscus, T. tinca, P. fluviatilis and L. delinea- applied as an efficient tool to detect species in the envi- tus, which were recovered through PCR using species specific ronment but also used to estimate population densities. primers with subsequent cloning and Sanger sequencing. For However, this will necessitate rigorous species-specific the former three because the applied generic primers do not comparative studies to fine-tune model parameters and amplify tissue derived DNA of these species. JD11 (N55.79799, further validate the approach in natural freshwater E12.58399), HEL56 (N55.98929, E12.20933), ELL1 (N55.842498, E12.534903), BOT1 (N55.68651, E12.57432) (Datum: WGS84) environments. Moreover, the effect of factors such as temperature, pH, conductivity and microbial commu- Species Pond nity composition should be further investigated as these are likely to influence DNA decay and detectability. Amphibians Lissotriton vulgaris JD11, HEL56 Also, the exact cellular origin of environmental DNA in Triturus cristatus JD11 freshwater and the relative contribution of different Pelophylax kl. esculentus JD11, HEL56 states (e.g. free, cellular or particle-bound DNA) remain Rana temporaria HEL56 Rana arvalis HEL56 unclear, and clarification of this may focus future sam- Pelobates fuscus HEL56 pling strategies. Precipitation, as used in this study, Fish Carassius carassius ELL1, BOT1 recovers DNA independent of state but is limited to Carassius auratus ELL1, BOT1 small sample volumes compared to filtering methods Cyprinus carpio ELL1, BOT1 (e.g. Jerde et al. 2011), which accommodate larger sam- Scardinius erythrophthalmus ELL1, BOT1 ples but may fail to recover free DNA. Tinca tinca ELL1, BOT1 Finally, to explore the broad-scale potential of environ- Leucaspius delineatus ELL1 Perca fluviatilis BOT1 mental DNA-based species detection, we investigated the extent to which complete species diversity can be documented by environmental DNA screening. We used water samples from four ponds with well-known ing the taxonomic groups in question. It is inherent to amphibian or fish faunas (updated occurrence data from the use of generic primers that there is a trade-off the Danish freshwater fish atlas project and Amphi-Con- between targeting higher taxonomic levels and detecting sult Aps national amphibian monitoring data) and tar- rare sequences. geted DNA from these groups with a combination of specific and generic primers (Table S3, Supporting infor- Conclusion mation). PCR products were sequenced using the Roche GS FLX 454 platform and Sanger sequencing, generating Faced with a global decline in biodiversity that is 100– a total of 524 027 sequences. 1000 times faster than prehuman rates (Pimm et al. We recovered species-specific DNA fragments with 1995; Barnosky et al. 2011), there is an urgent need for 100% sequence match for all species of amphibians or data-driven prioritization of conservation actions, which fish previously recorded from each of the ponds relies heavily on fast and effective monitoring of threa- (Table 1 and Table S4, Supporting information). Interest- tened species. Environmental DNA monitoring cannot ingly, we furthermore recovered DNA sequences from replace field observations by experienced ecologists and species living in close proximity to the water, including taxon specialists, who retrieve information beyond birds: Eurasian coot (Fulica atra), wood pigeon (Columba quantitative and qualitative records. However, monitor- palumbus) and marsh warbler (Acrocephalus palustris) and ing of threatened species through environmental DNA red deer (Cervus elaphus). These results suggest that the may be a quick, cost-effective and standardized way to success of DNA detection is largely independent of ani- obtain basic data on distribution and abundance, mal species and abundance, as long as DNA is excreted enabling efficient deployment of limited conservation into the water. Furthermore, this illustrates that DNA is resources and taxonomic expertise. Further research on homogeneously distributed in pond water, in stark con- environmental DNA in relation to conservation of rare trast to recent observations of animal DNA in soil, char- and threatened species should focus on large-scale com- acterized by a patchy distribution (Andersen et al. 2011). parative validation and optimization including addi- The ability to exhaustively recover all species in the tional organismal groups and applying the approach investigated faunas of fish and amphibians probably beyond freshwater ecosystems. However, our findings relies on the design of generic primers specifically target- highlight a vast potential for integrating DNA detection  2011 Blackwell Publishing Ltd 8 P. F. THOMSEN ET AL. in the tool set of biodiversity field research and conser- major influence on current population structure and status. vation. With DNA sequencing technology advancing at Conservation Genetics, 7, 185–195. rapidly dropping costs (Metzker 2009; Anonymous England L, Pollok J, Vincent M et al. (2005) Persistence of extracellular baculoviral DNA in aquatic microcosms: 2010), environmental DNA research is set to change extraction, purification, and amplification by the from being merely a scientific curiosity to become an polymerase chain reaction (PCR). Molecular and Cellular important tool in applied field biology. Probes, 19, 75–80. Ficetola GF, Miaud C, Pompanon F, Taberlet P (2008) Species detection using environmental DNA from water samples. Acknowledgements Biology Letters, 4, 423–425. We thank Dr. P.R. Møller (Natural History Museum of Den- Goldberg CS, Pilliod DS, Arkle RS, Waits LP (2011) Molecular mark), A. Drews (LLUR, Schleswig-Holstein), A. Linnet (Dan- detection of vertebrates in stream water: a demonstration ish Nature Agency, Thy) and Martin Hesselsøe (Amphi- using rocky mountain tailed frogs and IDAHO giant Consult Aps) for generously supplying monitoring data. All salamanders. PLoS ONE, 6, e22746. work involving live animals was conducted under permit Haile J, Holdaway R, Oliver K et al. (2007) Ancient DNA number SNS-441-00116 of the Danish Nature Agency. This chronology within sediment deposits: are paleobiological study was supported by the University of Copenhagen, The reconstructions possible and is DNA leaching a factor? Natural History Museum of Denmark and the Danish National Molecular Biology and Evolution, 24, 982–989. Research Foundation. Haile J, Froese DG, MacPhee RDE et al. (2009) Ancient DNA reveals late survival of mammoth and horse in interior Alaska. Proceedings of the National Academy of Sciences of the References United States of America, 106, 22352–22357. Hambler C, Henderson PA, Speight MR (2011) Extinction rates, Andersen K, Bird KL, Rasmussen M et al. (2011) extinction-prone habitats, and indicator groups in Britain Metabarcoding of ‘Dirt’ DNA from soil reflects vertebrate and at larger scales. Biological Conservation, 144, 713–721. biodiversity. Molecular Ecology, doi: 10.1111/j.1365-294X. Hansen MM, Jacobsen L (1999) Identification of mustelid 2011.05261.x. species: otter (Lutra lutra), American mink (Mustela vison) Anonymous (2010) Human genome at ten: the sequence and polecat (Mustela putorius), by analysis of DNA from explosion. Nature, 464, 670–671. faecal samples. Journal of Zoology, 247, 177–181. Barnosky AD, Matzke N, Tomiya S et al. (2011) Has the Hartvich P, Lusk S, Rutkayova´ J (2010) Threatened fishes of Earth’s sixth mass extinction already arrived?. Nature, 471, the world: Misgurnus fossilis (Linnaeus, 1758)(Cobitidae). 51–57. Environmental Biology of Fishes, 87, 39–40. Brendonck L, Rogers DC, Olesen J, Weeks S, Hoeh WR (2008) Helsdingen P, Willemse L, Speight M (1996) Background Information Global diversity of large branchiopods (Crustacea: on Invertebrates of the Habitats Directive and the Bern Convention, Branchiopoda) in freshwater. Hydrobiologia, 595, 167–176. Part 2: Mantodea, Odonata, Orthoptera and Arachnida. pp. 1–177. Conroy JWH, Chanin RF (2000) Status of the Eurasian otter Council of the European Union, Strassbourg, Germany. (Lutra lutra) in Europe—a review. Journal of the International Hofreiter M, Mead JI, Martin P, Poinar HN (2003) Molecular Otter Survival Fund, 1, 7–28. caving. Current Biology, 13, 693–695. Council of the European Union (1992) Council Directive Hopkins G, Freckleton R (2002) Declines in the numbers of 92 ⁄ 43 ⁄ EEC on the Conservation of Natural Habitats and of Wild amateur and professional taxonomists: implications for Fauna and Flora. pp. 1–66. Council of the European Union, conservation. Animal Conservation, 5, 245–249. Strassbourg, Germany. IUCNredlist (2011) IUCN Red List of Threatened Species. Version Davison A, Birks JDS, Brookes RC, Braithwaite TC, Messenger 2010.4. IUCNredlist, Webpage: http://www.iucnredlist.org, JE (2002) On the origin of faeces: morphological versus accessed 15 march 2011. molecular methods for surveying rare carnivores from their Jerde CL, Mahon AR, Chadderton WL, Lodge DM (2011) scats. Journal of Zoology, 257, 141–143. ‘‘Sight-unseen’’ detection of rare aquatic species using Dejean T, Valentini A, Duparc A et al. (2011) Persistence of environmental DNA. Conservation Letters, 4, 150–157. environmental DNA in freshwater ecosystems. PLoS ONE, 6, Kim CK, Kwak MJ, Lee SG (1996) Structural and functional e23398. stability of the genetic recombinant plasmid pCU103 in Douville M, Gagne´ F, Blaise C, Andre´ C (2007) Occurrence and different water environments. Journal of Microbiology (Korea persistence of Bacillus thuringiensis (Bt) and transgenic Bt Republic), 34, 241–247. corn cry1Ab gene from an aquatic environment. Ecotoxicology Lindahl T (1993) Instability and decay of the primary structure and Environmental Safety, 66, 195–203. of DNA. Nature, 362, 709–715. Dudgeon D, Arthington AH, Gessner MO et al. (2006) Lydolph MC, Jacobsen J, Arctander P et al. (2005) Beringian Freshwater biodiversity: importance, threats, status and paleoecology inferred from permafrost-preserved fungal conservation challenges. Biological Reviews, 81, 163–182. DNA. Applied and Environmental Microbiology, 71, 1012–1017. Edgar P, Bird DR (2006) Action plan for the Conservation of the Matisoo-Smith E, Roberts K, Welikala N et al. (2008) Recovery Crested Newt Triturus cristatus Species Complex in Europe, pp. of DNA and pollen from New Zealand lake sediments. 1–33. Council of the European Union, Strassbourg, Germany. Quaternary International, 184, 139–149. Eggert C, Cogalniceanu D, Veith M, Dzukic G, Taberlet P Matsui K, Honjo M, Kawabata Z (2001) Estimation of the fate (2006) The declining Spadefoot toad, Pelobates fuscus of dissolved DNA in thermally stratified lake water from the (Pelobatidae): paleo and recent environmental changes as a  2011 Blackwell Publishing Ltd  SPECIES MONITORING BY ENVIRONMENTAL DNA 9 stability of exogenous plasmid DNA. Aquatic Microbial logical systems. M.R, M.T.P.G, L.O and E.W (Centre for Goe- Ecology, 26, 95–102. Genetics, Copenhagen University) are developing methods for Metzker ML (2009) Sequencing technologies—the next analyzing ancient and fragmented DNA from various environ- generation. Nature Reviews Genetics, 11, 31–46. mental samples. Pimm SL, Russell GJ, Gittleman JL, Brooks TM (1995) The future of biodiversity. Science, 269, 347–350. Quince C, Lanzen A, Davenport RJ, Turnbaugh PJ (2011) Removing noise from pyrosequenced amplicons. BMC Data accessibility Bioinformatics, 12, 239. Sequences from pyrosequencing are uploaded to NCBI SRA: Revenga C, Mock G (2000) Pilot Analysis of Global Ecosystems: ERP000988. Freshwater Systems. pp. 1–78. World Resource Institute, Washington, District of Columbia, USA. Sala OE, Stuart Chapin F III, Armesto JJ et al. (2000) Global biodiversity scenarios for the year 2100. Science, 287, 1770– Supporting information 1774. Additional supporting information may be found in the online Vie´ J, Hilton-Taylor C, Stuart S (2009) Wildlife in a Changing version of this article. World. An Analysis of the 2008 IUCN Red List of Threatened Species. pp. 1–184, World Conservation Union, Gland, Table S1 Overview of the results of field experiments. Switzerland. Wheeler Q, Raven P, Wilson E (2004) Taxonomy: impediment Table S2 Overview of the results of controlled mesocosm or expedient? Science, 303, 285. experiments. Willerslev E, Hansen AJ, Binladen J et al. (2003) Diverse plant Table S3 Primers and probes designed and used in this study. and animal genetic records from holocene and pleistocene sediments. Science, 300, 791–795. Table S4 Summary of DNA sequences recovered in this study. Willerslev E, Hansen AJ, Poinar HN (2004) Isolation of nucleic Fig. S1 Accumulated probabilities of detecting the targeted acids and cultures from fossil ice and permafrost. Trends in species in the field studies when taking 1, 2 or 3 samples. Ecology & Evolution, 19, 141–147. Willerslev E, Cappellini E, Boomsma W et al. (2007) Ancient Fig. S2 Model fit on the DNA concentration in individual con- biomolecules from deep ice cores reveal a forested Southern tainers of the mesocosm experiment through time. Greenland. Science, 317, 111–114. Fig. S3 Modeled parameter estimates. Fig. S4 QQ-plot for mesocosm experiments. P.F.T, J.K (Centre for GeoGenetics, Copenhagen University) and L.L.I (Freshwater Biology Section, Copenhagen University) Please note: Wiley-Blackwell is not responsible for the content are developing methods for monitoring biodiversity using or functionality of any supporting information supplied by the environmental DNA. C.W (Bioinformatics Research Center, authors. Any queries (other than missing material) should be Aarhus University) is working with statistical modeling of bio- directed to the corresponding author for the article.  2011 Blackwell Publishing Ltd 

SALAMANDRA 48(1) 58–62 30 April 2012 Correspondence ISSN 0036–3375 Correspondence Absence of infection with the amphibian chytrid fungus in the terrestrial Alpine salamander, Salamandra atra Stefan Lötters 1, Jos Kielgast 1,2, Marc Sztatecsny 3, Norman Wagner 1, Ulrich Schulte 1, Philine Werner 1,3,4, Dennis Rödder 5, Johannes Dambach 6, Timo Reissner 4, Axel Hochkirch 1 & Benedikt R. Schmidt 4,7 1) Trier University, Biogeography Department, 54286 Trier, Germany 2) Department of Biology, Copenhagen University, Universitetsparken 15, 2100 Copenhagen, Denmark 3) Department of Evolutionary Biology, University of Vienna, Vienna, Austria 4) KARCH, Passage Maximilien-de-Meuron 6, 2000 Neuchâtel, Switzerland 5) Herpetology Department, Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany 6) Department for Molecular Biodiversity, Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany 7) Institut für Evolutionsbiologie und Umweltwissenschaften, Universität Zürich, Winterthurerstrasse 190, 8057 Zürich, Switzerland Corresponding author: Stefan Lötters, e-mail: loetters@uni-trier.de Manuscript received: 15 December 2011 Amphibian declines and species extinctions are worrying ops symptoms of chytridiomycosis, it may eventually die conservationists around the globe, and the emerging infec- from a breakdown of neurological functions (Voyles et al. tious disease chytridiomycosis is suggested to play a key 2009). role in these processes (Fisher et al. 2009). The disease’s To better understand possible consequences of Bd etiological agent, the chytridiomycete fungus Batracho­ spread, it is important to know which species are suscep- chytrium dendrobatidis (Bd), has been reported to be tible to Bd infection and chytridiomycosis. Bielby et al. present on all continents inhabited by amphibians (Fisher (2008) provided evidence that, at least in anuran amphibi­ et al. 2009). Amphibian mass mortalities, however, seem to ans, Bd susceptibility is related to life history. They found be geographically restricted, and it has recently been sug- that species from high altitudes within a geographically ‘re- gested that one particular currently emerging, globalised stricted’ range and having an aquatic life stage accompa- and highly virulent strain of Bd is responsible for the most nied by low fecundity suffer from higher risk of Bd-related dramatic consequences of the disease (Farrer et al. 2011). decline. Woodhams et al. (2007a, b) provided evidence Besides the strain, the specific susceptibility of host spe- that susceptibility can also depend on species-specific skin cies or populations as well as host-environment interac- peptides or skin bacteria. Bd susceptibility may further- tions might play a role in the outcome of an infection (e.g., more be related to the environment in which amphibians Woodhams et al. 2007a, Tobler & Schmidt 2010, Savage live. Based on the pathogen’s temperature sensitivity, Röd- & Zamudio 2011, Searle et al. 2011). der et al. (2009) identified worldwide regions in which Bd is known to infect more than 400 amphibian species climatic conditions are most suitable to Bd and concluded of both anurans and salamanders, and the most dramatic that at lower latitudes higher elevations and at higher lati- mass mortalities have occurred in mountainous areas of tudes lower elevations would provide the best environment the Americas, Australia, and southern Europe (Berger et for the survival of Bd. al. 1998, Bosch & Martinez-Solano 2006). The patho- Bd infection is widespread among European amphibi- gen spreads through motile infectious zoospores released ans including those occurring in the Alps (Garner et al. from zoosporangia growing on keratinised parts of the am- 2005, Sztatecsny & Glaser 2011). While it has caused phibian skin. Despite this aquatic transmission stage, Bd is mortality and population extinctions in some mountain known also to infect purely terrestrial amphibian species ranges (Bosch & Martinez-Solano 2006, Bielby et al. (Weinstein 2009). If an individual is infected and devel- 2009, Walker et al. 2010), Bd apparently leaves many Eu- © 2012 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Mannheim, Germany 58 articles available online at http://www.salamandra-journal.com All  Correspondence mycosis may occur (see Walker et al. 2010), (iii) it inhab- its mountain ranges climatically suitable to Bd (Rödder et al. 2009) and where this fungus occurs (Sztatecsny & Glaser 2011), and (iv) Bd-associated mass mortality has been observed in the congeneric Salamandra salamandra (Bosch & Martinez-Solano 2006). We tested for Bd infection 310 Alpine salamanders liv- ing at different altitudes in nine separated populations well spaced over the species’ geographic range (Table 1, Fig. 2). For sampling, we used sterile cotton swabs (Copan Italia S.p.A., Brescia, Italy; Medical Wire & Equipment, Wilt- shire, England) to swab ventral surfaces of body, hands and feet of salamanders. To avoid that the same individu- als would be tested twice, one site within a population was only sampled once and specimens were released only after Figure 1. Alpine salamander from the Hinterstein Valley, Bavari­ all specimens had been swabbed. Afterwards, swabs were an Alps, Germany (not collected). Photo: U. Schulte frozen as quickly as possible upon return from the field trip (Hyatt et al. 2007). For Bd screening, we used quantita- tive real-time PCR (polymerase chain reaction) of the ITS- 1/5.8S ribosomal DNA region of Bd (Boyle et al. 2004) with internal positive control (Hyatt et al. 2007). Bd data ropean species and populations unaffected. Here, we re- has been made available to the global Bd mapping project port the results of a study on Bd infection in the viviparous at http://www.bd-maps.net/maps/. and entirely terrestrial Alpine salamander, Salamandra Bd was detected in none of our samples (Table 1), indi- atra Laurenti, 1768 (Fig. 1). This caudate is endemic to the cating that none of the S. atra specimens sampled were in- Alps and the Dinaric Alps (Griffiths 1996). fected. To obtain a Bayesian 95% credible interval for prev- We suggest that there is reason for concern that this spe- alence, we used WinBUGS to estimate the posterior distri- cies may be at risk of Bd infection because (i) it has a low bution of prevalence (Kéry 2010, see Appendix). Poste­rior fecundity (Bielby et al. 2008, by implication), (ii) it occurs distributions were left-skewed towards zero and all 95% under climatic conditions where outbreaks of chytridio- credible intervals included a prevalence of zero. Table 1. Details of Alpine salamander and accompanying amphibian species sampling (see Fig. 2). Altitude in metres above sea level. Country State, locality, Approximate Number Observed prevalence Date Additional species altitudinal range coordinates of indi- (Bayesian 95% sampled (n) viduals Credible Interval) Austria Salzburg, Hagengebirge 13.1 E, 47.5 N 35 0% (0.00, 0.10) 7 July 2009 (Schlumsee), 490–1,200 m Austria Salzburg, Krimmler Achental 12.19 E, 47.14 N 20 0% (0.00, 0.16) 14 June 2009 (NP Hohe Tauern), 1,622 m Austria Steiermark, Wörschach 14.13 E, 47.60 N 8 0% (0.00, 0.35) 11-12 June 2009 (Totes Gebirge), 1,715 m Austria Tirol, Imst (Lechtaler 10.6 E, 47.26 N 10 0% (0.00, 0.28) 30 July 2009 Alpen), 1,700–1,800 m Austria Vorarlberg: Schoppenau 10.03 E, 47.31 N 8 0% (0.00, 0.35) 31 July 2009 (Bregenzer Wald), 915–1,000 m Germany Bayern, Hintersteiner Tal 10.4 E, 47.4 N 120 0% (0.00, 0.03) 9-12 July 2009 Bufo bufo (13), Ichthyo­ (Allgäu), 850–1,825 m saura alpestris (59) Switzerland Nidwalden, near 8.38 E, 46.9 N 53 0% (0.00, 0.07) 24-25 July, Bufo bufo (2), Ichthyo­ Wolfen­schiessen, 10 August 2009 saura alpestris (3), Sala­ 550–1,705 m mandra salamandra (2) Switzerland St. Gallen, Murgtal, 9.11 E, 47.03 N 41 0% (0.00, 0.09) 2 August 2009 Bufo bufo (1), Ichthyo­ 1,160–1,604 m saura alpestris (4) Switzerland Glarus, near Braunwald, 8.98 E, 46.93 N 15 0% (0.00, 0.20) 8 August 2008 1,500 m 59  Correspondence Figure 2. Distribution of the Alpine salamander (solid red line, taken from www.iucn.org) and populations sampled (red squares, Table 1). How can we explain the apparent absence of Bd infec- syntopic amphibian species (Table 1) for Bd and they all tion in the Alpine salamander? We here discuss four pos- tested negative either. sible explanations. (3) Another explanation could be that the risk of Bd in- (1) The simplest explanation would be that we failed fection is minimized in this species as a result of its strict- to detect Bd when it was in fact present. Given our sam- ly terrestrial life cycle. Under this assumption, the Alpine ple sizes, we may have missed Bd in some localities (Di- sala­mander might be susceptible to Bd, but in practice does Giacomo & Koepsell 1986, Marti & Koella 1993). The not, or rarely becomes, infected and/or has a low intraspe- range of possible prevalences is given by the 95% credible cific transmission rate. Several studies of life history traits intervals (Table 1). However, because all 95% credible in- and Bd susceptibility suggest that it is more likely to af- tervals included zero and because total sample size was 310 fect species linked to permanent water bodies (Bielby et (Table 1), our results suggest an absence or at least a very al. 2008, Bancroft et al. 2011). However, experimental in- low prevalence of Bd in the populations studied (DiGia­ fection trials conducted on strictly terrestrial salamanders como & Koepsell 1986, Marti & Koella 1993). Peyer clearly demonstrated susceptibility to both Bd infection (2010) tested 52 museum specimens of S. atra for Bd and and clinical chytridiomycosis (Chinnadurai et al. 2009, none tested positive (one specimen collected in 1972 gave Vasquez et al. 2009, Weinstein 2009). Moreover, a wealth an equivocal result, but this also occurred in other species of studies have provided field records of Bd-infected terres- that were tested by Peyer [2010]). Although it is clear that trial salamanders and anurans both in temperate and trop- Bd may occur in very low prevalence in nature, our data ical zones (Bell et al. 2004, Cummer et al. 2005, Kolby et support that Bd was most likely truly absent rather than al. 2009, Weinstein 2009, Becker & Harris 2010, Longo not detected. & Burrowes 2010). Thus, a strictly terrestrial life history (2) One might also argue that Bd was simply not present does not per se exclude or reduce the likelihood of infec- in the general area of our tested salamander populations. tion by Bd. This, however, seems unlikely, as Bd is known to occur at (4) We favour an alternative explanation. We suggest high elevations and in cold climates (Seimon et al. 2007, that it is plausible that S. atra is resistant to Bd because of Knapp et al. 2001, Muths et al. 2008) including the Swiss innate immunity caused by skin peptides or skin micro­ and Austrian Alps (Peyer 2010, Sztatecsny & Glaser biota in the manner observed in a number of other am- 2011). We note, however, that at some localities we tested phibians (Woodhams et al. 2007a, b). This hypothesis 60  Correspondence should be tested through experimental infection trials on tion in susceptibility of frog tadpoles to the pathogenic fungus S. atra involving infection of salamanders under natural Batrachochytrium dendrobatidis. – Conservation Biology, 19: environmental conditions with and without suppressed 1460–1468. immune function. In principle, a species that is immune Bosch, J. & I. Martinez-Solano (2006): Chytrid fungus infec- because of skin peptides or microbiota should become sus- tion related to unusual mortalities of Salamandra salamandra ceptible by a combination of disinfection using antimicro- and Bufo bufo in the Penalara Natural Park, Spain. – Oryx, 40: bials and mechanical or chemical depletion of skin peptide 84–89. reservoirs. Additionally, it could be studied in vitro wheth- Boyle, D. G., D. B. Boyle, V. Olsen, J. A. T. Morgan & A. D. er the salamander’s skin peptides and bacteria inhibit the Hyatt (2004): Rapid quantitative detection of chytridiomy- growth of Bd. If such anti-Bd properties were found, they cosis (Batrachochytrium dendrobatidis) in amphibian samples using real-time Taqman PCR assay. – Diseases of Aquatic Or- might be used as part of a strategy to mitigate the effects of ganisms, 60: 141–148. Bd on wild amphibians (Woodhams et al. 2011). Chinnadurai, S. K., D, Cooper, D. S. Dombrowski, M. F. Poore & M. G. Levy (2009): Experimental infection of native North Carolina salamanders with Batrachochytrium dendro­ Acknowledgements batidis. – Journal of Wildlife Diseases, 45: 631–636. Cummer, M. R., D. E. Green & E. M. O’Neill (2005): Aqua­tic Funding was obtained from the Hans-Schiemenz-Fonds of the chytrid pathogen detected in terrestrial plethodontid sala- DGHT (to BRS, DR, MS) and Stiftung Artenschutz (to BRS). Per- mander. – Herpetological Review, 36: 248–249. mits to conduct fieldwork were kindly issued by the provincial governments of Salzburg, Styria and Vorarlberg in Austria (21301- DiGiacomo, R. F. & T. D. Koepsell (1986): Sampling for detec- RI/548/57/5-2009, FA13C-53S7/59-2008, IVe-123/61) and the pro- tion of infection or disease in animal populations. – Journal vincial governments of Swabia/Bavaria (55.1-8622.002/94) in Ger- of the American Veterinary Medical Association, 189: 22–23. many. For support in the lab, we are grateful to Karin Fischer Farrer, R. A., L. A. Weinert, J. Bielby, T. W. J. Garner, F. Bal- and Ursina Tobler; for help during sampling we thank Martin loux, F. Clare, J. Bosch, A. A. Cunningham, C. Weldon, Berg and Denise J. Ellwein. We thank G.F. Ficetola and an L. H. Du Preez, L. Anderson, S. L. K. Pond, R. Shahar-Go- anonymous referee for their comments. lan, D. A. Henk & M. C. Fisher (2011): Multiple emergences of genetically diverse amphibian-infecting chytrids include a globalized hypervirulent recombinant lineage. – Proceedings References of the National Academy of Sciences of the USA, 108: 18732– 18736. Bancroft, B. A., B. A. Han, C. L. Searle, L. Biga, D. H. Olson, Fisher M. C, T. W. J. Garner & S. F. Walker (2009): Global L. B. Kats, J. J. Lawler & A. R. Blaustein (2011): Species- emergence of Batrachochytrium dendrobatidis and amphibian level correlates of susceptibility to the pathogenic amphibian chytridiomycosis in space, time and host. – Annual Reviews in fungus Batrachochytrium dendrobatidis in the United States. – Microbiology, 63: 291–310. Biodiversity and Conservation, 20: 1911–1920. Garner, T. W. J., S. Walker, J. Bosch, A. D. Hyatt, A. A. Cun- Becker, M. H. & R. N. Harris (2010): Cutaneous bacteria of ningham & M. C. Fisher (2005): Chytrid fungus in Europe. the redback salamander prevent morbidity associated with – Emerging Infectious Diseases, 11: 1639–1641. a lethal disease. – PLoS ONE, 5: e10957. doi: 10.1371/journal. Griffiths, R. A. (1996): The newts and salamanders of Europe. – pone.0010957 London: Academic Press. Bell, B. D., S. Carver, N. J. Mitchell & S. Pledger (2004): Hyatt, A. D., D. G. Boyle, V. Olsen, D. B. Boyle, L. Berger, D. The recent decline of a New Zealand endemic: how and why Obendorf, A. Dalton, K. Kriger, M. Hero, H. Hines, R. did populations of Archey’s frog Leiopelma archeyi crash over Phillott, R. Campbell, G. Marantelli, F. Gleason, & A. 1996–2001? – Biological Conservation, 120: 189–199. Colling (2007): Diagnostic assays and sampling protocols for Berger, L., R. Speare, P. Daszak, D. E. Green, A. A. Cun- the detection of Batrachochytrium dendrobatidis. – Diseases of ningham, C. L. Goggin, R. Slocombe, M. A. Ragan, A. D. Aquatic Organisms, 73: 175–192. Hyatt, K. R. McDonald, H. B. Hines, K. R. Lips, G. Mar- Kéry, M. (2010): Introduction to WinBUGS for ecologists: a antelli & H. Parkes (1998): Emerging infectious disease and Bayesian approach to regression, ANOVA, mixed models and the loss of biodiversity in a Neotropical amphibian commu- related analyses. – Burlington: Academic Press. nity. – Proceedings of the National Academy of Sciences of the Knapp, R. A., C. J. Briggs, T. C. Smith & J. R. Maurer (2011): USA, 95: 9031–9036. Nowhere to hide: impact of a temperature-sensitive amphibi­ Bielby, J., N. Cooper, A. A. Cunningham, T. W. J. Garner & A. an pathogen along an elevation gradient in the temperate Purvis (2008): Predicting susceptibility to future declines in zone. – Ecosphere, 2: art93. doi:10.1890/ES11-00028.1 the world’s frogs. – Conservation Letters, 1: 82–90. Kolby, J. E., G. E. Padgett-Flohr & R. Field (2009): Amphib- Bielby, J., S. Bovero, G. Sotgiu, G. Tessa, M. Favelli, C. An- ian chytrid fungus Batrachochytrium dendrobatidis in Cusuco gelini, S. Doglio, F. C. Clare, E. Gazzaniga, F. Lapietra National Park, Honduras. – Diseases of Aquatic Organisms, & T. W. J. Garner (2009): Fatal chytridiomycosis in the 92: 245-251. Tyrrhenian painted frog. – EcoHealth, 6: 27-32. Longo, A. V. & P. A. Burrowes (2010): Persistence with chytri­ Blaustein, A. R., J. M. Romansic, E. A. Scheessele, B. A. Han, dio­mycosis does not assure survival of direct developing frogs. A. P. Pessier & J. E. Longcore (2005): Interspecific varia- – EcoHealth, 7: 185–195. 61  Correspondence Marti, H. & J. C. Koella (1993): Multiple stool examinations for Woodhams, D. C., V. T. Vredenburg, M. A. Simon, D. Bill­hei­ ova and parasites and rate of false-negative results. – Journal mer, B. Shakhtour, Y. Shyr, C. J. Briggs, L. A. Rol­lins- of Clinical Microbiology, 31: 3044–3045. Smith & R. N. Harris (2007b): Symbiotic bacteria contrib- Muths, E., D. S. Pilliod & L. J. Livio (2008): Distribution and ute to innate immune defenses of the threatened mountain environmental limitations of an amphibian pathogen in the yellow-legged frog, Rana muscosa. – Biological Conservation, Rocky Mountains, USA. – Biological Conservation, 141: 1484– 138: 390–398. 1492. Woodhams, D. C., J. Bosch, C. J. Briggs, S. Cashins, L. R. Da­ Peyer, N. F. (2010): Historical evidence for the presence of the vis, A. Lauer, E. Muths, R. Puschendorf, B. R. Schmidt, emerging amphibian pathogen Batrachochytrium dendrobati­ B. Sheafor & J. Voyles (2011): Mitigating amphibian disease: dis (Longcore et al. 1999) in Switzerland. – Unpubl. MSc the- strategies to maintain wild populations and control chytridio­ sis: University of Zurich. mycosis. – Frontiers in Zoology, 8: 8. doi:10.1186/1742-9994- 8-8 Rödder, D., J. Kielgast, J. Bielby, S. Schmidtlein, J. Bosch, T. J. W. Garner, M. Veith, S. Walker, M. Fisher & S. Lötters (2009): Global amphibian extinction risk assessment for the panzootic chytrid fungus. – Diversity, 1: 52-66. doi:10.3390/ Appendix d1010052 WinBUGS code to compute Bayesian 95% credible intervals Savage, A. E. & K. R. Zamudio (2011): MHC genotypes associate for prevalence with resistance to a frog-killing fungus. – Proceedings of the National Academy of Sciences of the USA, 108: 16705–16710. We assume that the reader will access WinBUGS from the statis- tical software R (see Kéry 2010 for an introduction to both R and Searle, C. L., S. S. Gervasi, J. Hua, J. I. Hammond, R. A. Relyea, WinBUGS). In R, if there are, say, 20 individuals that tested nega- D. H. Olson & A. R. Blaustein (2011): Differential host sus- tive for Bd, data could be entered using the command ceptibility to Batrachochytrium dendrobatidis, an emerging amphibian pathogen. – Conservation Biology, 25: 965–974. data <- c(rep(0,20)) Seimon, T. A., A. Seimon, P. Daszak, S. P. R. Halloy, L. M. Schloegel, C. A. Aguilar, P. Sowell, A. D. Hyatt, B. The code for the WinBUGS model is Konecky & J. E Simmons (2007): Upward range extension of Andean anurans and chytridiomycosis to extreme elevations prevalence ~ dunif(0,1) # uniform, non-informative prior in response to tropical deglaciation. – Global Change Biology, for (i in 1:n.ind) {# n.ind is the number of individuals in the data set 13: 288–299. data[i] ~ dbern(prevalence)} Sztatecsny, M. & F. Glaser (2011): From the eastern lowlands to the western mountains: first records of the chytrid fungus We ran three parallel Markov chains with 2,000 iterations each Batra­chochytrium dendrobatidis in wild amphibian popula- and discarded the first 1,000 iterations as burn-in; we did not thin tions from Austria. – Herpetological Journal, 21: 87–90. the chains. Vazquez, V. M., B. B. Rothermel & A. P. Pessier (2009): Exper- imental infection of North American plethodontid salaman- ders with the fungus Batrachochytrium dendrobatidis. – Dis- eases of Aquatic Organisms, 84: 1–7. Tobler, U. & B. R. Schmidt (2010): Within- and among-popu- lation variation in chytridiomycosis-induced mortality in the toad Alytes obstetricans. – PLoS ONE, 5: e10927. doi:10.1371/ journal.pone.0010927 Voyles, J., S. Young, L. Berger, C. Campbell, W. F. Voyles, A. Dinudom, D. Cook, R. Webb, R. A. Alford, L. F. Skerratt & R. Speare (2009): Pathogenesis of chytridiomycosis, a cause of catastrophic amphibian declines. – Science, 326: 582–585. Walker, S. F., J. Bosch, V. Gomez, T. W. J. Garner, A. A. Cun- ningham, D. S. Schmeller, M. Ninyerola, D. A. Henk, C. Ginestet, C. P. Arthur & M. C. Fisher (2010): Factors driving pathogenicity vs. prevalence of amphibian panzootic chytridiomycosis in Iberia. – Ecology Letters, 13: 372–382. Weinstein, S. B. (2009): An aquatic disease on a terrestrial sala- mander: individual and population level effects of the amphi­ bi­an chytrid fungus, Batrachochytrium dendrobatidis, on Batra­choseps attenuatus (Plethodontidae). – Copeia, 2009: 653–660. Woodhams, D. C., K. Ardipradja, R. A. Alford, G. Maran- telli, L. K. Reinert & L. A. Rollins-Smith (2007a): Resist- ance to chytridiomycosis varies among amphibian species and is correlated with skin peptide defenses. – Animal Conserva- tion, 10: 409–417. 62 

Lötters, S., D. Rödder, J. Bielby, J. Bosch, T.W.J. Garner, J. Kielgast. S. Schmidtlein, M. Veith, S. Walker, C. Weldon, D.M. Aansen & M.C. Fisher (2008): Meeting the challenge of conserving Madagascar's megadiverse amphibians: addition of a risk-assessment for the chytrid fungus. PLoS Biology, 6: http://biology.plosjournals.org/perlserv/?request=read-response&doi=10.1371/journal.pbio.0060118 Responses To This Article The Challenge of Conserving Amphibian Megadiversity in Madagascar Andreone F, Carpenter AI, Cox N, du Preez L, Freeman K, et al. PLoS Biology Vol. 6, No. 5, e118 doi:10.1371/journal.pbio.0060118 • Meeting the challenge of conserving Madagascar's megadiverse amphibians: addition of a risk- assessment for the chytrid fungus Stefan Lotters, Dennis Rodder1, Jon Bielby2, Jaime Bosch3, Trenton J.W. Garner2, Jos Kielgast1,2, Sebastian Schmidtlein4, Michael Veith1, Susan Walker5, Che Weldon6, David M. Aanensen5, Matthew C. Fisher5 (04 July 2008) Meeting the challenge of conserving Madagascar's megadiverse amphibians: addition of a risk- assessment for the chytrid fungus Stefan Lotters Researcher and lecturer Biogeography Department, Trier University E-mail Additional Authors: Dennis Rodder1, Jon Bielby2, Jaime Bosch3, Trenton J.W. Garner2, Jos Kielgast1,2, Sebastian Schmidtlein4, Michael Veith1, Susan Walker5, Che Weldon6, David M. Aanensen5, Matthew C. Fisher5 Competing Interests: None Submitted Date: June 18, 2008 Published: July 04, 2008 Andreone et al. [1] highlighted the need for pro-active conservation action to prevent Madagascar's megadiverse amphibian fauna from loss. The threat of extinction to it is twofold; habitat loss and emerging infectious disease. Of these, the latter presents the least-quantified threat. The amphibian chytrid fungus, Batrachochytrium dendrobatidis (Bd) is a globally emerging pathogen that is known to cause extinction of naïve species even in protected areas (2). In this case 'classical' conservation tools are insufficient (3). Bd has not yet been detected in Madagascar (4). However, its introduction there is possible via the international amphibian trade (5) and Bd's arrival in Madagascar is predicted to have a catastrophic effect; endemics are susceptible to Bd infection (6). Our response to this threat should address in situ and ex situ measures, such as prophylactic conservation breeding, according to the IUCN Amphibian Conservation Action Plan [1,5]. Only a few Madagascan species are currently in captive breeding programs [6], compared to the more than 400 amphibian species known [7]. This demonstrates the necessity of identifying which amphibians are most threatened by Bd when prioritizing species for conservation breeding (http://zims.isis.org/aark/). In order to assess the level of threat that the introduction of Bd poses to Madagascar's amphibians, we ran a risk-assessment with Maxent 3.1.2 ('excellent' results, AUC25% 0.969) [8], based on 'bioclimate' (annual mean, maximum of warmest and minimum of coldest month temperature; annual, wettest and driest month precipitation; http://worldclim.org) and 365 globally distributed Bd presence points (http://www.spatialepidemiology.net/bd/). Results (http://www.spatialepidemiology.net/bd/supplemental/) revealed that there is a high risk of Bd spreading post-introduction over a major portion of Madagascar and areas most suitable for Bd largely overlap with both areas of highest amphibian species richness [9] and those identified as in situ conservation priorities for amphibians [10]. As no meaningful in situ conservation action is available for the emergence of Bd, captive breeding remains the best short-term option for ensuring the survival of Madagascar’s amphibians. Author Affiliations 1 Dpt. Biogeography, Trier University, Trier, Germany. 2 Institute of Zoology, London, UK. 3 Museo Nacional de Ciencias Naturales, Madrid, Spain. 4 Dpt. Geography, Bonn University, Bonn, Germany. 5 Dpt. Infectious Disease Epidemiology, Imperial College London, UK. 6 Potchefstroom University, South Africa References 1. Andreone F, Carpenter AI, Cox N, et al. (2008) The challenge of conserving amphibian megadiversity in Madagascar. PLoS Biology Vol. 6, No. 5, e118 doi:10.1371/journal.pbio.0060118. 2. Skerrat LF, Berger L, Speare R, et al. (2007) Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs. EcoHealth 4: 125-134. 3. Fisher MC, Garner TWJ (2007) The relationship between the emergence of Batrachochytrium dendrobatidis, the international trade in amphibians and introduced amphibian species. Fungal Biology Reviews 21: 2-9. 4. Oevermann A, Schildger B, Feldman S, Robert N (2005) Chytridiomykose bei Tomatenfroschen (Dyscophus antongilii) in der Schweiz. Tierarztliche Umschau 60, 211-217. 5. Mendelson J, Lips KR, Gagliardo RW, et al. (2006) Confronting amphibian declines. Science 313: 48. 6. L0tters S (2008) Afrotropical amphibians in zoos and aquariums: will they be on the ark? International Zoo Yearbook 42, 136-142. 7. Glaw F, Vences M (2007) A field guide to the amphibians and reptiles of Madagascar. 3rd edition. Cologne: Vences & Glaw Publishers. 8. Phillips SJ, Anderson RP, Shapire RE (2006) Maximum entropy modeling of species geographic distributions. Ecological Modelling 190: 231-259. 9. Andreone F, Cadle JE, Cox N, et al. (2005) Species review of amphibian extinction risk in Madagascar: conclusions from the Global Amphibian Assessment. Conservation Biology 19, 1790-1802. 10. Kremen C, Cameroon A, Moilanen A, et al. (2008) Aligning conservation priorities across taxa in Madagascar with high-resolution planning tools. Science 320, 222-226. Supplementary material from http://www.spatialepidemiology.net/bd/supplemental/ "Meeting the challenge of conserving Madagascar's megadiverse amphibians: addition of a risk- assessment for the chytrid fungus" by Stefan Lotters1, Dennis Rodder1, Jon Bielby2, Jaime Bosch3, Trenton J.W. Garner2, Jos Kielgast1,2, Sebastian Schmidtlein4, Michael Veith1, Susan Walker5, Che Weldon6, David A. Aanensen5, Matthew C. Fisher5 1 Dpt. Biogeography, Trier University, Trier, Germany. 2 Institute of Zoology, London, UK. 3 Museo Nacional de Ciencias Naturales, Madrid, Spain. 4 Dpt.. Geography, Bonn University, Bonn, Germany. 5 Dpt. Infectious Disease Epidemiology, Imperial College London, UK. 6 Potchefstroom University, South Africa submitted to PLoS Biology as an addition to Andreone F, Carpenter AI, Cox N, et al. (2008) The challenge of conserving amphibian megadiversity in Madagascar. PLoS Biology Vol. 6, No. 5, e118 doi:10.1371/journal.pbio.0060118. Figure and figure legend: Risk-assessment for Bd in Madagascar with higher Maxent values suggesting higher Bd risk [reference 8]. Both (A) amphibian species richness (from lighter to darker stippling are shown > 18, > 36 > 63 species per grid cell of 0.1°, respectively [reference 9]) and (B) amphibian top 10 % conservation priority areas (hatched [reference 10]) largely overlap with high Bd risk. 

Preprint, 2010 DISEASES OF AQUATIC ORGANISMS Published online April 7, 2010 doi: 10.3354/dao02197 Dis Aquat Org Contribution to DAO Special 4 ‘Chytridiomycosis: an emerging disease’ Future potential distribution of the emerging amphibian chytrid fungus under anthropogenic climate change Dennis Rödder1, 2, Jos Kielgast3,*, Stefan Lötters2 1 Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany 2 Department of Biogeography, Trier University, Am Wissenschaftspark 25-27, 54296 Trier, Germany 3 Natural History Museum of Denmark, Zoological Museum, Universitetsparken 15, 2100 Copenhagen, Denmark ABSTRACT: Anthropogenic climate change poses a major threat to global biodiversity with a poten- tial to alter biological interactions at all spatial scales. Amphibians are the most threatened verte- brates and have been subject to increasing conservation attention over the past decade. A particular concern is the pandemic emergence of the parasitic chytrid fungus Batrachochytrium dendrobatidis, which has been identified as the cause of extremely rapid large-scale declines and species extinc- tions. Experimental and observational studies have demonstrated that the host –pathogen system is strongly influenced by climatic parameters and thereby potentially affected by climate change. Herein we project a species distribution model of the pathogen onto future climatic scenarios gener- ated by the IPCC to examine their potential implications on the pandemic. Results suggest that pre- dicted anthropogenic climate change may reduce the geographic range of B. dendrobatidis and its potential influence on amphibian biodiversity. KEY WORDS: Amphibia · Anthropogenic future climate change · Batrachochytrium dendrobatidis · Bioclimate · Chytridiomycosis · Global warming · Maxent · Species distribution model Resale or republication not permitted without written consent of the publisher INTRODUCTION ian species across the globe (Skerratt et al. 2007, Stu- art et al. 2008). It has been shown that susceptibility Increasing anthropogenic activity is causing envi- to Bd is widely distributed among the Amphibia, both ronmental change and severely degrading global bio- geographically and taxonomically (Bielby et al. 2008). diversity. It has been suggested that we are currently Although some evidence is controversial (Rohr et al. witnessing the onset of a sixth mass extinction (Wake 2008), there is general consensus that the host – & Vredenburg 2008). Amphibians are among the pathogen system is strongly influenced by climatic most dramatically affected organisms, with about parameters (Woodhams & Alford 2005, Pounds et al. one-third of the approximately 6500 described spe- 2006, Rachowicz et al. 2006, Alford et al. 2007, Bosch cies threatened with extinction under the IUCN Red et al. 2007, Kriger & Hero 2007, Kriger et al. 2007, List of Threatened Species (Gascon et al. 2007, Stuart Andre et al. 2008, Laurance 2008, Lips et al. 2008, et al. 2008). The emerging infectious disease chytrid- Pounds & Coloma 2008, Rödder et al. 2008, Kielgast iomycosis, caused by the chytrid fungus Batra- et al. 2009). Furthermore, Bd’s thermal tolerance and chochytrium dendrobatidis (Bd), has been given par- growth rate, determined in laboratory studies ticular attention (Mendelson et al. 2006, Lips et al. (Piotrowski et al. 2004), are well reflected in preva- 2008). The pandemic spread of this pathogen (from a lence and intensity of Bd infection in the wild (Kriger yet unknown source) is suggested to be the proxi- & Hero 2007, Kriger et al. 2007, Rödder et al. 2008, mate cause of rapid decline and extinction of amphib- Kielgast et al. 2009). *Corresponding author. Email: jkielgast@snm.ku.dk © Inter-Research 2010 · www.int-res.com 2 Dis Aquat Org: Preprint, 2010 Species distribution models (SDMs) are a widely used Maxent has been developed within the machine tool in contemporary data-driven conservation biology learning community and implements an algorithm for (Araújo et al. 2004, Kremen et al. 2008) and can be suc- making predictions and inferences from incomplete cessfully applied to predict potential distributions of information. This algorithm estimates geographic dis- diseases (e.g. Levine et al. 2007, Lafferty 2009). In tributions of species from environmental conditions as SDMs, predictions of a species’ potential distribution observed from species records and random back- can be made using GIS technology and environmental ground data by finding the maximum entropy distribu- parameters such as climate, characterizing the species’ tion (Phillips et al. 2006, Phillips & Dudík 2008). In the niche. The SDM result is a map indicating the environ- present study, the logistic output format was chosen, mental (e.g. climatic) suitability at given sites for the which produces continuous, linear scaled maps rang- target species. In order to uncover Bd’s invasive geo- ing from 0 (unsuitable) to 1 (optimal) showing the graphic potential including uninfected regions, some potential distribution of the target species. This authors have already successfully applied SDMs on Bd method allows for more fine-scale distinction of the using climatic predictors (Ron 2005, Lötters et al. 2008, potential distributions than what can be achieved with Puschendorf et al. 2009, Rödder et al. 2009a). However, similar methods such as GARP (genetic algorithm for predictions on future scenarios for the panzootic includ- rule-set production) models (Phillips et al. 2006). ing anthropogenic climate change are lacking. We sug- Also, in numerous comparative studies, Maxent has gest that such an approach is intriguing and relevant, achieved better results than other presence-only meth- as alteration of species distributions related to climate ods (Elith et al. 2006, Hernandez et al. 2006). It has change is expected to happen on a large scale over the been suggested that ensemble model predictions may next decades (Parmesan 2006). Additionally, the partic- enhance the reliability and robustness of SDM results ular impact of future anthropogenic climate change on (Araújo & New 2007). Therefore, 100 models were Bd spread and chytridiomycosis epizootics has already computed and subsequently integrated into a map been advocated by multiple authors (e.g. Pounds et al. indicating the average Maxent value per grid cell. 2006, 2007, Alford et al. 2007, Bosch et al. 2007, Lau- As environmental predictors, 6 bioclimate variables rance 2008, Pounds & Coloma 2008). In the present were used: annual mean temperature, maximum tem- study, we develop an SDM to characterize Bd’s climate perature of the warmest month, minimum temperature niche and project it onto currently expected anthro- of the coldest month, annual precipitation, precipita- pogenic future climate change scenarios. tion of the wettest month and precipitation of the driest month (Hijmans et al. 2005a, www.worldclim.org). Laboratory studies (Piotrowski et al. 2004, Woodhams MATERIALS AND METHODS et al. 2008), field studies (Kriger & Hero 2007, Kriger et al. 2007, Rödder et al. 2008, Kielgast et al. 2009) and We ran a Bd SDM (Fig. 1A) using Maxent 3.3.1 previous attempts to map Bd’s potential distribution at (Phillips et al. 2006, Phillips & Dudík 2008; www.cs. a local scale (Lötters et al. 2008, Puschendorf et al. princeton.edu/~schapire/maxent), in the manner of 2009) have shown that these parameters are physiolog- Rödder et al. (2009a). It was based on 365 globally ically relevant for this pathogen and correlated with Bd distributed Bd records taken from www.spatialepide prevalence and infection rates. miology.net/bd-maps; species records posted on this Ideally, the area from which random background webpage were compiled from numerous scientific data is obtained for model building reflects those publications (cited on the webpage). Bd surveys were regions accessible to the target species, since areas not conducted randomly over the world (Fig. 1A), leav- with climate conditions not analogous to those repre- ing the possibility of a sample selection bias violating sented by background data may lead to uncertainties SDM assumptions (Phillips 2008). Therefore, 19 bio- in model predictions (Phillips 2008). To account for clim values (Nix 1986, Busby 1991) provided by World- this, background samples were restricted to areas from clim (Hijmans et al. 2005a) records were extracted which Bd was detected in the wild. Maxent automati- from Bd records and a cluster analysis based on Euclid- cally allows an identification of the degree of uncer- ean distances was performed. Resulting classes were tainty when projecting models (clamping), which was blunted at a threshold of 200 classes, and only one ran- removed from our Bd SDM. domly chosen record per class was used for further Model validation was performed by using the area processing. This method reduces the amount of dupli- under the curve (AUC), referring to the receiver oper- cate information in the feature space and thereby the ating characteristic (ROC) curve, a threshold-indepen- impact of samples clumped in geographic space dent index widely used in ecological modelling (e.g. (Rödder et al. 2009a). All calculations were performed Manel et al. 2001, Elith et al. 2006). Two tests were per- with XLSTAT 2008 (Addinsoft, www.xlstat.com). formed: (1) the model’s ability to distinguish between  Rödder et al.: Amphibian chytrid fungus under climate change 3 Fig. 1. (A) Potential distribution of Batrachochytrium dendrobatidis (Bd) under current climatic conditions based on a Maxent species distribution model (SDM) and Bd presence localities used for SDM building (red dots), with warmer Maxent colour indi- cating higher suitability to the fungus. (B) Change in the potential distribution of Bd under future climatic scenarios (i.e. the year 2080) relative to current conditions based on mean values per grid cell of the A2a family when projected onto the CCCMA, CSIRO and HADCM3 models. (C) Change computed with the mean predictions of the climate change models assuming B2a con- ditions (for particular 2080 SDMs each for CCCMA, CSIRO and HADCM3 see Figs. S2 & S3 in the supplement, available at www.int-res.com/articles/suppl/dao02197_app.pdf) Bd records and randomly chosen background points culates the mean minimum training presence and the (referred to as AUCtraining) and (2) its ability to predict a mean lowest 10th percentile training presences at each subset of records, setting aside 25% randomly chosen run; values greater than these may be interpreted as Bd records as test points and using the remaining ones reasonable approximation of a species’ potential distri- as training points (referred to as AUCtest). Maxent cal- bution. Nevertheless, the higher a Maxent value the 4 Dis Aquat Org: Preprint, 2010 better the climatic suitability for the modelled species Rödder et al. 2009a), the areas of highest suitability to (Phillips et al. 2006, Phillips & Dudík 2008). For quanti- the pathogen have a patchy distribution (Fig. 1A). In tative comparisons between current-day and future the tropics, Bd is largely associated with higher alti- scenarios, we counted the number of SDM grid cells of tudes including the Mexican Meseta, the Andean mas- different Maxent values with DIVA-GIS 5.4 (Hijmans sif, Pantepui and the Mata Atlantica in the Neotropics, et al. 2005b). the Cameroon Mountains, Ethiopian highlands and For future projections, we used anthropogenic future higher portions of the Rift Valley and eastern Mada- climate change scenarios for the year 2080 based the gascar in the Aethiopis, and the Sumatran and New CCCMA, CSIRO and HADCM3 models (Flato et al. Guinean highlands in the Orientalis. Towards the 2000, Gordon et al. 2000) and the emissions scenarios poles, in the Northern Hemisphere the highest suit- reported in the IPCC Special Report on Emissions Sce- ability for Bd occurs over almost all of western Europe, narios (SRES) (IPCC 2000). A set of different families of the southern Himalayan slopes and adjacent areas in emission scenarios was formulated based on future China to the east, while in the Southern Hemisphere, production of greenhouse gases and aerosol precursor large portions of southern South America and south- emissions by the IPCC, which distinctly address demo- eastern Africa as well as New Zealand and southern graphic, politico-economic, social and technological and southeastern Australia are most suitable for Bd. developmental aspects of climate change. The SRES Many of these regions coincide with hotspots of families A2a and B2a were used in the present study. amphibian species richness and endemism expected to Each family described one possible demographic, be susceptible to Bd infection (Bielby et al. 2008, Röd- politico-economic, social and technological future. B2a der et al. 2009a). emphasizes more environmentally conscious, region- The projected Maxent SDMs in Fig. 1B,C show the alized solutions to economic, social and environmental potential distribution of Bd as expected for 2080. Since sustainability. Compared to B2a, scenarios of the A2a our SDM projections onto the CCCMA, CSIRO and family also emphasize regionalized solutions to eco- HADCM3 scenarios merely revealed minor differ- nomic and social development, but it is less environ- ences, only mean values are given in Fig. 1B,C (for par- mentally conscious. Data sets were also obtained via ticular A2a and B2a SDM maps of each scenario see WorldClim (www.worldclim.org) and the current-day Figs. S2 & S3 in the supplement). Both maps are Bd SDM was projected onto these data sets in order to largely similar, identifying the following as the most assess the future potential distribution of Bd. suitable regions for Bd (in terms of highest Maxent val- ues): the Mexican Meseta, the Andean region, western Europe, part of the East African Rift Valley, China’s RESULTS AND DISCUSSION Yunnan Province, part of New Guinea, the southern edges of South America, Africa and Australia, and The conducted Bd model received ‘good’ to ‘excel- New Zealand. Large portions of North America, west- lent’ AUC values following previously given defini- ern Europe, southern Asia and the southernmost re- tions (Swets 1988), suggesting a high quality of our gions of the Southern Hemisphere are also within the SDM output (mean AUCtraining = 0.937; mean AUCtest = potential distribution of the amphibian chytrid fungus, 0.910). Analysis of the relative contribution of each but at a lower level of suitability. The tropics are least variable to the 100 models revealed that annual mean suitable for Bd, with the exception of highland regions. temperature had the highest explanatory power (28.7 Compared to the SDM based on current climate to 53.5%, mean ± SD = 44.29 ± 4.79%), followed by (Fig. 1A), regions suitable for Bd in the future largely maximum temperature of the warmest month (13.70 to remain suitable, except for most of the tropical low- 36.10%, mean ± SD = 23.15 ± 4.31%). Mean contribu- lands (e.g. Amazon and Congo River basins). It is note- tions of the other variables were below 10%. Model- worthy that climatic suitability for Bd in terms of Max- based uncertainties in future predictions in terms of ent values per grid cell decreases under anthropogenic standard deviations of Maxent values per grid cell future climate change scenarios with the A2a family were comparatively low (Fig. S1 in the supplement, (i.e. less environmentally conscious) (Fig. 2). This is available at www.int-res.com/articles/suppl/dao02197_ particularly applicable to tropical and subtropical app.pdf). The mean minimum training presence was regions, for example, the Mexican Meseta, the north- 0.05 (0.01 to 0.18, SD = 0.03) and the mean lowest 10th ern Andes, the southeastern coastal region of South percentile training presence was 0.28 (0.21 to 0.35, America, Africa (southwest and Rift Valley highlands), SD = 0.03). Madagascar, southern Asia and Australia (Fig. 1B,C). As pointed out in previous studies aiming to model There are 3 possible alternatives to how species Bd’s potential distribution based on current climate ranges may respond to climate change when niche (Ron 2005, Lötters et al. 2008, Puschendorf et al. 2009, conservatism is expected: they may (1) increase, (2)  Rödder et al.: Amphibian chytrid fungus under climate change 5 et al. 2006, Seimon et al. 2007) and may also hold truth for others. SDM future projections also indicate that in temperate regions, Bd suitability, as a general trend compared to today, will decrease. However, especially in the Northern Hemisphere (e.g. northern North America and western Europe), this is contrasted by an increase in area highly suitable for the fungus (Fig. 1B,C). This does not compensate for the general area decrease, however (Fig. 2). Even if climatic suitability for Bd decreases with anthropogenic future climate change, this does not mean per se that the risk for amphibian species to be affected by chytridiomycosis will respond correspond- ingly. This central question remains unstudied, as it is still unclear what exact mechanisms trigger clinical chytridiomycosis (Fisher et al. 2009). However, the development in the distribution of chytridiomycosis may very well have a trend parallel to that of the pre- Fig. 2. Potential distribution area size of Batrachochytrium dicted optimal conditions for Bd. Under the current cli- dendrobatidis (Bd) distribution in terms of number of 2.5 min mate, Rödder et al. (2009a) assessed the extinction risk grid cells and Maxent value classes derived from species for more than 3500 species of anuran amphibians by distribution models under current climatic conditions and combining their expected susceptibility to Bd and Bd future anthropogenic climate change families A2a and B2a (Fig. 1). For A2a and B2a, the mean and range (error bars) probability within species ranges. However, (potential) are shown from projections onto the CCCMA, CSIRO and geographic ranges of most of these species under HADCM3 models anthropogenic future climate change cannot currently be predicted. For successful model building, detailed decrease or (3) geographically shift in (potential) distri- information on species’ natural history is pivotal (Röd- bution (Parmesan & Yohe 2003, Parmesan 2006). All 3 der et al. 2009b,c), and such information is unavailable responses occur in the conducted SDM for the future for the bulk of the global amphibian taxa. Only when potential distribution of Bd, with geographical range knowledge on species-specific responses to particular shift being predicted as the most common mode of climate variables is accessible (as for Bd; Woodhams & change, i.e. towards higher latitudes and altitudes. Alford 2005, Pounds et al. 2006, Rachowicz et al. 2006, Lafferty (2009) emphasized that, as all species, in- Kriger et al. 2007, Lips et al. 2008, Rödder et al. 2008, fectious diseases have upper and lower limits of their Kielgast et al. 2009) can reliable future SDM projec- temperature tolerance. This may be expected to lead tions be made (Rödder & Lötters 2009, Rödder et al. to shifts in their (potential) distributions if climate 2009b). changes. The thermal limit of Bd is reached at 28°C, where growth is inhibited, whereas prolonged periods Acknowledgements. We are grateful to J. Bielby, J. Bosch, M. above 30°C are lethal (Piotrowski et al. 2004). Further- Fisher, T. Garner, S. Schmidtlein, M. Veith and S. Walker for more, clinical studies have shown that transmission collaboration and fruitful discussion. Additionally, we thank A. Cunningham and anonymous reviewers for many helpful efficiency is likely to decrease with a rise in tempera- comments. D.R. thanks the Graduiertenförderung des Landes ture, as this reduces both the fecundity per fungal spo- Nordrhein-Westfalen (Bonn University). rangium and the period that zoospores are infectious (Woodhams et al. 2008). Currently, many tropical low- LITERATURE CITED land areas in which Bd has been detected may at times reach temperatures at or above its thermal limit. The ➤ Alford RA, Bradfield KS, Richards SJ (2007) Ecology: global expected temperature increase will possibly render warming and amphibian losses. Nature 447:E3–E4 such regions unsuitable, e.g. by moving the tempera- ➤ Andre SE, Parker J, Briggs CJ (2008) Effect of temperature on host response to Batrachochytrium dendrobatidis infec- ture of colder microclimates above a critical threshold tion in the mountain yellow-legged frog (Rana muscosa). where Bd can persist. On the other hand, our SDM J Wildl Dis 44:716–720 future projections indicate that the expected tempera- ➤ Araújo MB, New M (2007) Ensemble forecasting of species ture increase may enhance Bd spread and successful distributions. Trends Ecol Evol 22:42–47 establishment at higher altitudes in the tropics, espe- ➤ Araújo MB, Cabeza M, Thullier W, Hannah L, Williams PH (2004) Would climate change drive species out of re- cially in the Andes of South America. These results are serves? An assessment of existing reserve-selection consistent with observations in this region (e.g. Pounds methods. Glob Change Biol 10:1618–1626  6 Dis Aquat Org: Preprint, 2010 ➤ Bielby J, Cooper N, Cunningham AA, Garner TWJ, Purvis A doi:10.1371/journal.pbio.0060072 (2008) Predicting susceptibility to future declines in the Lötters S, Rödder D, Bielby J, Bosch J and others (2008) Meet- world’s frogs. Conserv Lett 1:82–90 ing the challenge of conserving Madagascar’s megadi- ➤ Bosch J, Carrascal LM, Duran L, Walker S, Fisher MC (2007) verse ampibians: addition of a risk assessment for the Climate change and outbreaks of amphibian chytridiomy- chytrid fungus. Available at http://biology.plosjournals. cosis in a montane area of Central Spain; is there a link? org/perlserv/?request=read-response&doi=10.1371/journal. Proc R Soc Lond B Biol Sci 274:253–260 pbio.0060118 Busby JR (1991) BIOCLIM: a bioclimatic analysis and predic- ➤ Manel SP, Williams HC, Ormerod SJ (2001) Evaluating pres- tion system. In: Margules CR, Austin MP (eds) Nature con- ence-absence models in ecology: the need to account for servation: cost effective biological surveys and data analy- prevalence. J Appl Ecol 38:921–931 sis. CSIRO, Melbourne, p 64–68 ➤ Mendelson JR III, Lips KR, Gagliardo RW, Rabb GB and oth- ➤ Elith J, Graham CH, Anderson RP, Dudík M and others (2006) ers (2006) Confronting amphibian declines and extinc- Novel methods improve prediction of species’ distribu- tions. Science 313:48 tions from occurrence data. Ecography 29:129–151 Nix H (1986) A biogeographic analysis of Australian elapid ➤ Fisher MC, Garner TWJ, Walker SF (2009) Global emergence snakes. In: Longmore R (ed) Atlas of elapid snakes of Aus- of Batrachochytrium dendrobatidis and amphibian chytri- tralia. Bureau of Flora and Fauna, Canberra, p 4–15 diomycosis in space, time, and host. Annu Rev Microbiol ➤ Parmesan C (2006) Ecological and evolutionary responses to 63:291–310 recent climate change. Annu Rev Ecol Evol Syst ➤ Flato GM, Boer GJ, Lee WG, McFarlane NA, Ramsden D, 37:637–669 Reader MC, Weaver AJ (2000) The Canadian Centre for ➤ Parmesan C, Yohe G (2003) A globally coherent fingerprint of Climate Modelling and Analysis global coupled model climate change impacts across natural systems. Nature and its climate. Clim Dyn 16:451–467 421:37–42 Gascon C, Collins JP, Moore RD, Church DR, McKay JE, ➤ Phillips SJ (2008) Transferability, sample selection bias and Mendelson JR III (2007) Amphibian conservation action background data in presence-only modelling: a response plan. IUCN, Conservation International, Gland, Cambridge to Peterson et al. (2007). Ecography 31:272–278 ➤ Gordon C, Cooper C, Senior CA, Banks HT and others (2000) ➤ Phillips SJ, Dudík M (2008) Modelling of species distributions The simulation of SST, sea ice extents and ocean heat with Maxent: new extensions and comprehensive evalua- transports in a version of the Hadley Centre coupled tion. Ecography 31:161–175 model without flux adjustments. Clim Dyn 16:147–168 ➤ Phillips SJ, Anderson RP, Schapire RE (2006) Maximum ➤ Hernandez PA, Graham CH, Master LL, Albert DL (2006) The entropy modeling of species geographic distributions. effect of sample size and species characteristics on perfor- Ecol Model 190:231–259 mance of different species distribution modeling methods. ➤ Piotrowski JS, Annis SL, Longcore JE (2004) Physiology of Ecography 29:773–785 Batrachochytrium dendrobatidis, a chytrid pathogen of ➤ Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A amphibians. Mycologia 96:9–15 (2005a) Very high resolution interpolated climate surfaces Pounds JA, Coloma LA (2008) Beware the lone killer. Nature for global land areas. Int J Climatol 25:1965–1978 2:57–59 Hijmans RJ, Guarino L, Jarvis A, O’Brien R and others (2005b) ➤ Pounds JA, Bustamante MR, Coloma LA, Consuegra JA and DIVA-GIS version 5.2 Manual. www.diva-gis.org/docu- others (2006) Widespread amphibian extinctions from epi- mentation demic disease driven by global warming. Nature 439: IPCC (Intergovernmental Panel on Climate Change) (2000) 161–167 Nakicenovic N, Swart R (eds) Special report on emissions ➤ Pounds JA, Bustamante MR, Coloma LA, Consuegra JA and scenarios. Cambridge University Press, Cambridge others (2007) Global warming and amphibian losses; the ➤ Kielgast J, Rödder D, Veith M, Lötters S (2009) Widespread oc- proximate cause of frog declines? (Reply) Nature 447: currence of the amphibian chytrid fungus in Kenya. Anim E5–E6 Conserv (in press) doi:10.1111/j.1469-1795.2009.00297.x ➤ Puschendorf R, Carnaval AC, VanDerWal J, Zumbado-Ulate ➤ Kremen C, Cameron A, Moilanen A, Phillips SJ and others H, Chaves G, Bolaños F, Alford RA (2009) Distribution (2008) Aligning conservation priorities across taxa in models for the amphibian chytrid Batrachochytrium den- Madagascar with high-resolution planning tools. Science drobatidis in Costa Rica: proposing climatic refuges as a 320:222–226 conservation tool. Divers Distrib 15:401–408 Kriger KM, Hero JM (2007) Large-scale seasonal variation in ➤ Rachowicz LJ, Knapp RA, Morgan JAT, Stice MJ, Vreden- the prevalence and severity of chytridiomycosis. J Zool burg VT, Parker JM, Briggs CJ (2006) Emerging infectious 271:352–359 disease as a proximate cause of amphibian mass mortality. ➤ Kriger KM, Peregolou F, Hero JM (2007) Latitudinal variation Ecology 87:1671–1683 in the prevalence and intensity of chytrid (Batra- ➤ Rödder D, Lötters S (2009) Niche shift versus niche conser- chochytrium dendrobatidis) infection in eastern Australia. vatism? Climatic characteristics of the native and invasive Conserv Biol 21:1280–1290 ranges of the Mediterranean house gecko (Hemidactylus ➤ Lafferty KD (2009) The ecology of climate change and infec- turcicus). Glob Ecol Biogeogr 18:674–687 tious diseases. Ecology 90:888–900 ➤ Rödder D, Veith M, Lötters S (2008) Environmental gradients ➤ Laurance WF (2008) Global warming and amphibian extinc- explaining the prevalence and intensity of infection with tions in eastern Australia. Austral Ecol 33:339–408 the amphibian chytrid fungus: the host’s perspective. ➤ Levine RS, Peterson AT, Yorita KL, Carroll D, Damon IK, Anim Conserv 11:513–517 Reynolds MG (2007) Ecological niche and geographic ➤ Rödder D, Kielgast J, Bielby J, Schmidtlein S and others distribution of human monkeypox in Africa. PLoS ONE 2: (2009a) Global amphibian extinction risk assessment for e176, doi:10.1371/journal.pone.0000176 the panzootic chytrid fungus. Diversity 1:52–65 ➤ Lips KR, Diffendorfer J, Mendelson III Jr, Sears MW (2008) ➤ Rödder D, Schmidtlein S, Veith M, Lötters S (2009b) Alien Riding the wave: reconciling the roles of disease and invasive slider turtle in unpredicted habitat: a matter of climate change in amphibian declines. PLoS Biol 6:e72, niche shift or predictors studied? PLoS ONE 4:e7843,  Rödder et al.: Amphibian chytrid fungus under climate change 7 doi:10.1371/journal.pone.0007843 Spread of chytridiomycosis has caused the rapid global Rödder D, Kwet A, Lötters S (2009c) Translating natural his- decline and extinction of frogs. EcoHealth 4:125–134 tory into geographic space: a macroecological perspective Stuart S, Hoffmann M, Chanson J, Cox N, Berridge R, Ramani on the North American slider, Trachemys scripta (Reptilia, P, Young B (2008) Threatened amphibians of the world. Cryptodira, Emydidae). J Nat Hist 43:2525–2536 Lynx Edicions, Barcelona ➤ Rohr JR, Raffel TR, Romansic JM, McCallum H, Hudson PJ ➤ Swets JA (1988) Measuring the accuracy of diagnostic sys- (2008) Evaluating the links between climate, disease tems. Science 240:1285–1293 spread, and amphibian declines. Proc Natl Acad Sci USA ➤ Wake DB, Vredenburg VT (2008) Are we in the midst of the 105:17436–17441 sixth mass extinction? A review from the world of amphib- ➤ Ron SR (2005) Predicting the distribution of the amphibian ians. Proc Natl Acad Sci USA 105:11466–11473 pathogen Batrachytrium dendrobatidis in the New World. ➤ Woodhams DC, Alford RA (2005) Ecology of chytridiomycosis Biotropica 37:209–221 in rainforest stream frog assemblages of tropical Queens- ➤ Seimon TA, Seimon A, Daszak P, Halloy SRP and others land. Conserv Biol 19:1449–1459 (2007) Upward range extension of Andean anurans and ➤ Woodhams DC, Alford RA, Briggs CJ, Johnson M, Rollins- chytridiomycosis to extreme elevations in response to Smith LA (2008) Life-history trade-offs influence disease tropical deglaciation. Glob Change Biol 13:288–299 in changing climates: strategies of an amphibian patho- ➤ Skerratt LF, Berger L, Speare R, Cashins S and others (2007) gen. Ecology 89:1627–1639 Editorial responsibility: Andrew Cunningham, Submitted: April 3, 2009; Accepted: January 11, 2010 London, UK Proofs received from author(s): March 31, 2010 

Diversity 2009, 1, 52-66; doi:10.3390/d1010052 OPEN ACCESS diversity ISSN 1424-2818 www.mdpi.com/journal/diversity Article Global Amphibian Extinction Risk Assessment for the Panzootic Chytrid Fungus Dennis Rödder 1,2,*, Jos Kielgast 1,3, Jon Bielby 4,5, Sebastian Schmidtlein 6, Jaime Bosch 7, Trenton W.J. Garner 4, Michael Veith 1, Susan Walker 8, Matthew C. Fisher 8 and Stefan Lötters 1 1 Department of Biogeography, Trier University, Am Wissenschaftspark 25-27, D-54296 Trier, Germany; E-Mails: joskielgast@hotmail.com (J.K.); veith@uni-trier.de (M.V.); loetters@uni-trier.de (S.L.) 2 Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, D-53113 Bonn, Germany 3 Department of Biology, Copenhagen University, Universitetsparken 15, DK-2100 Copenhagen, Denmark 4 Institute of Zoology, Zoological Society of London, Regent's Park, London, NW1 4RY, UK; E-Mails: bielby04@imperial.ac.uk (J.B.); trent.garner@ioz.ac.uk (T.W.J.G.) 5 Department of Biology, Imperial College London, Silwood Park Campus, Ascot Berkshire SL5 7PY, UK 6 Department of Geography, Bonn University, Meckenheimer Allee 166, D-53115 Bonn, Germany; E-Mail: s.schmidtlein@uni-bonn.de (S.S.) 7 Museo Nacional de Ciencias Naturales, Calle José Gutiérrez Abascal 2, E-28006 Madrid, Spain; E-Mail: bosch@mncn.csic.es (J.B.) 8 Department of Infectious Disease Epidemiology, Imperial College London, Silwood Park Campus, Ascot Berkshire SL5 7PY, UK; E-Mails: susan.walker@imperial.ac.uk (S.W.); matthew.fisher@imperial.ac.uk (M.C.F.) * Author to whom correspondence should be addressed; E-Mail: roedder@uni-trier.de; Tel.: +49-651-201-4617; Fax: +49-651-201-3851. Received: 30 July 2009 / Accepted: 7 September 2009 / Published: 11 September 2009 Abstract: Species are being lost at increasing rates due to anthropogenic effects, leading to the recognition that we are witnessing the onset of a sixth mass extinction. Emerging infectious disease has been shown to increase species loss and any attempts to reduce extinction rates need to squarely confront this challenge. Here, we develop a procedure for identifying amphibian species that are most at risk from the effects of chytridiomycosis by Diversity 2009, 1 53 combining spatial analyses of key host life-history variables with the pathogen's predicted distribution. We apply our rule set to the known global diversity of amphibians in order to prioritize species that are most at risk of loss from disease emergence. This risk assessment shows where limited conservation funds are best deployed in order to prevent further loss of species by enabling ex situ amphibian salvage operations and focusing any potential disease mitigation projects. Keywords: amphibians; Batrachochytrium dendrobatidis; chytridiomycosis; Maxent; Species Distribution Model 1. Introduction The IUCN Red List of Threatened Species reports one third of the about 6,500 extant amphibian species as threatened with extinction [1, for species list see ESI1]. Chytridiomycosis, a disease caused by the amphibian chytrid fungus, Batrachochytrium dendrobatidis (Bd), plays a decisive role in this global biodiversity crisis [1-6] by driving rapid declines and species extinctions in pristine protected areas. The balance of evidence shows that Bd is spreading globally and, in response, this pathogen has been included as a notifiable disease in the Aquatic Animal Health Code of the World Organization for Animal Health (OIE) (Article 2.4.1.2; www.oie.int/fr/normes/fcode/fr_index.htm; latest access 8 July 2009). An IUCN 'Amphibian Conservation Action Plan' has been developed [3,4] with its goal being the prevention of further loss of global amphibian biodiversity. This plan states that short-term ex situ breeding is the primary available conservation strategy for species under immediate threat from Bd to prevent further dramatic amphibian species loss owing to this panzootic. Antifungal bacteria reducing susceptibility of amphibians to Bd in nature may provide an additional solution to solve the amphibian crisis [7]. Beyond this, any attempt to limit further spread of the pathogen requires intimate knowledge of the current distribution of Bd and identification of areas and amphibian species assemblages not currently affected by chytridiomycosis. There is a pressing need for tools to prioritize candidate species for targeted population monitoring, ex situ breeding as well as other treatments, to identify critical areas for the implementation of measures to limit the spread of the pathogen and, when possible, target disease mitigation. Infection by Bd is expected to be particularly sensitive to environmental influences because the pathogen exclusively occurs in ectothermic hosts which are physiologically affected by ambient conditions [8]. Furthermore, the pathogen is known to be particularly temperature and moisture dependent [e.g., 9,10]. The in vitro growth optimum of Bd is at 17–25 °C, whereas temperatures higher than 29 °C, freezing and desiccation are lethal [11], findings that are supported by observations in the field [12-16]. Hence, the geographic extent of Bd's climatic niche can be readily assessed by Species Distribution Models (SDMs). These models give a prediction of potential distributions of species derived from their associated environmental parameters. In the case of pathogens, such models can provide predictions on impending epizootics in uninfected regions [e.g., 17,18]. Diversity 2009, 1 54 Herein we undertake a worldwide risk assessment for the potential impact of Bd with the goal to identify amphibian species that are most at risk of future declines as a result of Bd invasion and infection. We did so by combining known amphibian species ranges with a prediction of the distribution of Bd based on a SDM integrated with the biological characteristics of host susceptibility. 2. Material and Methods 2.1. Prediction of Bd Distribution Climate information for Bd SDM building was obtained from Worldclim, version 1.4 [19], which is based on weather conditions recorded between 1950 and 2000 at spatial resolution of about 1 × 1 km². It was created by interpolation using a thin-plate smoothing spline of observed climate at weather stations using latitude, longitude and elevation as independent variables [20]. The climate data set was obtained from the DIVA-GIS homepage (www.diva-gis.org; downloaded 15 May 2007), i.e., 36 monthly mean variables (each minimum and maximum temperature and precipitation, respectively). Based on these we calculated six 'bioclimate' variables for further processing with DIVA-GIS 5.4 [21]: 'annual mean temperature', 'maximum temperature of the warmest month', 'minimum temperature of the coldest month', 'annual precipitation', 'precipitation of the wettest month' and 'precipitation of the driest month'. Since it can be expected that Bd zoospores are able to survive in unfrozen microhabitat, water or on hosts during winter air freezing [11], subzero values in the 'minimum temperature of the coldest month' grids were pooled and set to 0 °C. More 'bioclimate' variables than the six used here can be obtained from Worldclim. However, to avoid model 'overfitting' [22] and multicollinearity of predictors, we restricted our study to the six 'bioclimate' variables mentioned which can be considered as biologically relevant parameters to Bd [10,11,14,16]. Also, these variables have been suggested to perform well in SDMs [23-25] including those for Bd [26]. We obtained 365 globally scattered Bd records (latitude/longitude) from www.spatialepidemiology.net/bd/ (accessed 15 August 2008). Records posted on this webpage were compiled from numerous scientific publications (a detailed list of references is also posted). Bd records were not randomly distributed over the world (Figure 1A), leaving the problem of possible sample selection bias which may violate SDM assumptions [27]. To account for this, we extracted all 'bioclimatic' values at Bd records and performed a cluster analysis based on Euclidean distances, whereby resulting classes were blunted at a threshold leaving 200 classes. Only one record per class was used for further processing. This method reduces the amount of duplicate information in the feature space and thereby the impact of samples clumped in geographic space. Calculations were performed with XLSTAT 2008 (Addinsoft, www.xlstat.com; downloaded 1 July 2007). GARP was used in an earlier model focusing on Bd's potential distribution in the New World [28]. However, advances in SDM methodology [29] and current availability of Bd presence records enable modelling at a much finer spatial scale. Herein, Maxent 3.3.1 (www.cs.princeton.edu/~shapire/maxent; downloaded 25 May 2009) was used for Bd SDM calculation and mapping [30,31]. Maxent has been developed within the machine learning community and implements a general purpose algorithm for making predictions and inferences from incomplete information. The Maxent algorithm estimates Diversity 2009, 1 55 geographic distributions of species from locality point data and random background data by finding the maximum entropy distribution [30]. Ideally, the area from which background data is obtained reflects those regions accessible to the target species [32]. Therefore, we restricted the background samples to areas from which Bd was detected in the wild. Areas with climate conditions not analogous to those represented by background data may lead to uncertainties in model predictions. Maxent automatically allows an identification of the degree of uncertainty when projecting models ('clamping'). In our SDM, the degree of 'clamping' was removed from the model prediction using the 'fade by clamping' option. In numerous comparative studies, Maxent has achieved better results than other presence only methods [summarized by 29,33]. It has been suggested that ensemble model predictions may enhance the reliability and robustness of SDM results [34]. Therefore, we computed 100 models each trained with randomly chosen 75% of the 200 records for model training and subsequently integrated all results into a map indicating the average Maxent value per grid cell. The remaining 25% of the records were used for model evaluation through calculation of the Area Under the Curve (AUC) in Receiver Operating Characteristic curves [35], a threshold-independent index widely used in ecological modelling [35,36]. In ROCs, the sensitivity values, the true-positive fraction against 1-specificity and the false positive fraction for all available probability thresholds are plotted [36,37]. AUC values may range from 0.5 (random accuracy) to 1.0 (perfect discrimination). We received 'good' to 'excellent', following previously given definitions [37], AUC values, suggesting a high quality of our SDM output: mean AUCtraining data = 0.937; mean AUCtest data = 0.910 (Figure 1A). Maxent allows to trace the relative contribution of each variable to the model. Herein, the 'annual mean temperature' had the highest explanatory power, followed by the 'maximum temperature of the warmest month' (Figure 1B). This pattern was consistent in all 100 models computed. Figure 1. Variation of AUCtraining and AUCtest (A), and variable contribution (B), minimum and 10 percentile training presence (C) in 100 Maxent models. Abbreviations are: Bio1 = 'annual mean temperature'; Bio5 = 'maximum temperature of the warmest month'; Bio6 = 'minimum temperature of the coldest month'; Bio12 = 'mean annual precipitation'; Bio13 = 'maximum precipitation of the wettest month'; Bio14 = 'minimum precipitation of the driest month'. In boxplots, the minimum and maximum values are indicated as blue dots, 95% confidence intervals as short horizontal bars; the 1st and 3rd quartiles and the median are indicated as broad horizontal bars and means as red crosses. Diversity 2009, 1 56 The logistic output of Maxent chosen by us is a continuous, linear scaled map which allows fine distinctions to be made between the modelled probability of habitat suitability for Bd. Generally, the higher a logistic Maxent value the better the prediction and therefore the climatic suitability for a species under study. It has been proposed that this relationship can be directly related to a species' maximum possible abundance [38]. Maxent calculates several threshold values at each run and values exceeding them may be interpreted as reasonable approximation of—in this case—Bd's potential distribution pending on the question at hand. We used the minimum training presence (mean = 0.049) as a strict threshold and the 10 percentile training presence (mean = 0.277; Figure 1C) as a more liberal threshold as suggested by Phillips et al. [30]. In general, the mean Maxent value at the input records is typically 0.5 [30,31], which was therefore selected as third threshold as recommended by Liu et al. [39]. Furthermore, the uppermost 25% of the logistic value was chosen as fourth threshold (logistic Maxent value of 0.75). 2.2. Identification of the Most Threatened Species Geographic ranges of 6,156 amphibian species of all three orders were adopted from the IUCN Red List of Threatened Species, as categorized during the former IUCN/Conservation International/NatureServe 'Global Amphibian Assessment' ([1], www.natureserve.org/getdata/amphibianmaps.jsp; accessed 7 July 2009; see ESI1, whereby taxonomy follows [40,41]). In order to assess Bd suitability within the geographic range of all amphibian species, we performed an overlay analysis of our Bd world map obtained from the SDM and the distributions of the 6,156 amphibian species. Higher logistic Maxent values at a given site are associated with a higher climatic suitability and most likely a higher maximum abundance for Bd [30,38]. Since the mean minimum training presence and the lowest 10 percentile training presences in our model were 0.049 and 0.227 (Figure 1C) and in accordance with previous studies [39], we regard these as minimum thresholds for the environmental suitability to Bd. We calculated the total geographic range encompassed by each amphibian species and quantified the fraction suitable for Bd at four Maxent thresholds (>0.049, >0.227, >0.50, >0.75; see ESI1). Computations were performed with DIVA-GIS 5.4 and ESRI ArcMap 9.2. 2.3. Bd Risk Factor for Anuran Amphibians In an approach to identify anuran species (from a total of 3,976) which exhibit highest probabilities of Bd-related decline or extinction, biological and life history information of Bd infected species showing rapid declines was used as explanative variables in a previously published study [42]. Life history information comprised the degree of aquatic life-stage, the mean snout-vent length and the mean clutch size and geographic range size. Representative environmental values for each species were added using spatial datasets of altitude, annual actual evapotranspiration, net primary productivity, isothermality (a measure of annual temperature consistency), maximum temperature of the warmest month, precipitation seasonality, precipitation in the driest quarter and human population density [43-45]. For model building, so called Generalized Linear Models specifying the link function as either 'logit' or complementary 'log-log' were applied, whereby the link with the lowest residual Diversity 2009, 1 57 deviance was preferred [42]. Expected effects of strong phylogenetic signals were removed using generalised estimating equations and Holm-Bonferroni corrections were made to account for Type I errors. According to the most predictive multi-predictor model (brier score = 0.06), species with aquatic life stages occupying small geographic ranges at high altitudes were most susceptible for Bd related declines. This final model was projected onto a larger data set, whereby a standardized value between 0 and 1 for each species was calculated, hereafter termed 'Biotic Index' (BI) (compare species list in ESI1 and see [42] for elaborate description of methods). In order to generate a Bd 'Risk Factor' (RF) for anurans, we combined BI and the results of our quantification of species' distributions with high suitability to Bd (Maxent value >0.50, which is the mean score at the Bd records used for model building which was suggested as a suitable threshold [39]; ESI1) as: ( BI *100) * areabd RF  (1) 10, 000 where areabd = the percentage of a species' distribution >0.50 Bd suitability. RF equally weights BI and areabd ranges between 0 and 1, whereby higher values reflect a higher threat balancing equally the fraction of the species area suitable to Bd and the BI. 3. Results and Discussion 3.1. Prediction of Bd Distribution Our SDM suggests that highest suitability for Bd occurs in temperate and subtropical regions of both hemispheres, often near coasts; in the tropics, for instance, montane regions are identified to show a high suitability for Bd (Figure 2A). The potential distribution of Bd suggested by our SDM generally resembles that obtained through an earlier GARP approach [28] with fine scale patterns better depicted in our approach. The currently known distribution of Bd records in the world is patchy (Figure 2A). Even when some of these gaps may be the result of limited collection efforts for Bd, as in Western Europe [4], it is remarkable that there are island-like high risk regions in which chytridiomycosis and its effects have not yet been observed [1,4]. These include the Ethiopian Highlands, eastern Madagascar, the southern versant of the Himalaya, China's Yunnan province and considerable portions of tropical Asia (Figure 2A). The importance of these highly suitable locations being apparently free of the effects of the chytrid fungus is magnified considering that these regions harbour high levels of amphibian species richness and endemism (Figure 2B). Evidence for extraordinary dispersal speed of Bd has been recorded in some regions [9] and apparently anthropogenic dissemination, especially through the international trade, has contributed decisively to the speed and extent of Bd's spread [46,47]. Therefore, biosecurity measures and baseline surveys should be initiated at those areas highlighted through our analysis. Diversity 2009, 1 58 Figure 2. (A) Worldwide potential distribution of the amphibian chytrid fungus (Bd) and 200 out of 365 records (black dots) of this pathogen (see methods). The map is derived from a Maxent SDM projected into geographic space based on six 'bioclimate' variables. Warmer colors are associated with higher Maxent values suggesting higher suitability for Bd, whereas grey areas are below the minimum training threshold and therefore considered to be unsuitable. This is equivalent to higher risk of invasion in regions from which Bd is currently absent. (B) Overlay highlighting worldwide regions of both high Bd suitability and high amphibian species richness. Warmer colors suggest higher overlap of Bd suitability (based on our SDM) and number of known amphibian species (see ESI1). (C) 'Risk Factor' (RF) for anuran amphibian species, calculated by combination of species life history traits linked with Bd-caused declines [42] and climatic suitability for Bd within the species' range (Appendices ESI1, ESI2). Warmer colors equal a higher proportion of species with a high RF per 0.5° grid cell. Diversity 2009, 1 59 3.2. Identification of the Most Threatened Species The proportion of species within each amphibian order that exhibit a high overlap with areas that are environmentally suitable to Bd is summarized in Table 1 (for a detailed species list including results of the presented analysis, IUCN conservation status and Bd presence see ESI1). Approximately one sixth of all known amphibian species fall with their total distributions into regions potentially suitable to Bd (Maxent value >minimum observed training presence; Figure 1A) and more than 50% of all species exhibits an overlap of over half of their known geographic range with regions showing high Bd suitability. This number drastically increases to almost 6,000 (i.e., roughly all known amphibians) when over half of a species' range overlaps regions with a Maxent value >0.5 (Table 1). Table 1. Number of species in the known amphibian orders which are most threatened by Bd through range overlap. Suitability for the pathogen is high at different percentages of their known geographic ranges. We here provide species numbers referable to environmentally suitable (minimum observed training presence) or highly suitable to Bd (values > 0.5) (Figure 2A). ESI1 provides detailed species-level information. Portion of species' range environmentally suitable (>0.5/minimum observed training presence) 100% 75% 50% Caecilians (172 species) 10/90 27/132 38/138 Salamanders (551 species) 129/498 195/543 226/545 Frogs and toads (5,532 species) 961/3,905 1,756/4,953 2,259/5,211 Total (6,142 species) 1,100/4,493 1,978/5,628 2,523/5,894 3.3. Bd Risk Factor for Anuran Amphibians Most of the 833 anuran species which by their biology and life history show a high predicted susceptibility to Bd [42] occur in regions which at the same time are characterized by high suitability for Bd (listed in ESI1). Of these, we identified 379 species in which the entire geographic range is, in terms of climate, of high suitability to Bd (Maxent value > 0.5; for detailed list see ESI2). We consider these amphibians to be the most threatened by the emergence of Bd. As shown in Figure 2C, they are distributed all over the world including regions from where Bd is currently unknown (Figure 2A). So far, little is known about the current infection or population status of most of the 'Top 379'. Perhaps due to the circumstance that many of them occur in regions from where Bd has not been recorded only seven of these species are reported to be infected with Bd in nature (list provided in ESI2). We suggest this is the result of limited surveillance for disease rather than the occurrence of healthy populations, as at least 42 species of the 'Top 379' (ESI2) have undergone so called 'rapid enigmatic declines' likely caused by the spread of Bd and the effect of chytridiomycosis [1,5]. Generally, the 'Top 379' priority species outlined in ESI2 should be considered as the 'top of the iceberg'. The threshold of RF = 1 used for their identification was perhaps chosen in an arbitrary way since all species with RF > 0 may be affected in one or the other way. Available conservation funds should be invested for targeted population monitoring, compilation of life history and ecology Diversity 2009, 1 60 information, and—if necessary—further in situ and ex situ efforts in descending order of RF, since the higher the RF value the higher is the potential threat caused by Bd. The importance of emerging infectious diseases and anthropogenic dissemination of pathogens for biodiversity has in recent years been receiving increasing attention especially due to a suite of zoonotic diseases with impact on humans and economically important domestic animals. However, the integration of wildlife diseases in strategies for halting the loss of biodiversity is still remarkably premature [48,49]. Of the 379 species identified under high risk of decline due to chytridiomycosis, 40% are categorized as 'Data Deficient' under the IUCN Red List of Threatened Species, whereas a compelling 94% of the species with sufficient information for proper assessment are categorized as threatened with extinction. However, it needs to be noted that species with restricted distributions and a combination of ongoing habitat threats would at the same time qualify as threatened based on IUCN's criteria B and as susceptible for Bd related declines according to the BI. Herein, we highlight the need to integrate novel approaches in the tool-set of conservation biology to mitigate the threat posed by pathogens. 3.4. Methodological Considerations When interpreting our results, some issues related to the data and methods used herein need to be considered. Information on the distribution of the world's amphibians was adopted via the IUCN Red List of Threatened Species from the former 'Global Amphibian Assessment'. These data provide the most comprehensive and up to information currently available. Range information of each species is based on expert opinions; therefore the polygons describe in most cases the extent of occurrence and sometimes the area of occupancy of each species rather than providing an exact summary of existing populations [see 50]. This may lead to an overestimation of the actually suitable areas in widely occurring species which may not be homogenously distributed throughout their general extent of occurrence (e.g., due to specific habitat requirements). Worldwide, these potential errors may not be homogenously distributed. Before conducting specific management actions, this should be acknowledged. However, since these polygons represent the best information available also used to derive the actual IUCN Red List status of amphibian species, they provide a suitable basis for our assessment. Methodical inherent uncertainties can be assessed comparing multiple SDMs in ensembles of models providing more robust predictions [34]. In our 100 models combined in the final model we observed rather low standard deviations of predictions per grid cells indicating low model variability and a generally good predictive ability (Figure 3). SDMs rely on specific assumptions including that environmental conditions are range limiting and that the range of the target species is in equilibrium with climate [51]. Looking at the requirements of Bd regarding certain temperature and humidity conditions proven by both laboratory and field studies (see above), as well as the nearly world-wide distribution of Bd, these assumptions appear to be fairly met. Furthermore, SDMs commonly ignore biotic interactions which may lead to an exclusion of the modeled species from certain areas [52]. Regarding Bd, the most obvious biotic interaction include the pathogens' relationship to amphibians. Diversity 2009, 1 61 However, since nearly all areas predicted to be suitable for Bd harbor at least one amphibian species, this requirement appears also to be fulfilled. Figure 3. Standard deviation (SD, upper panel) per grid cell of all of the 100 Maxent models used for the ensemble prediction (Figure 2A). In the vast majority of grid cells, the SD is rather low < 0.05 (lower panel). 4. Conclusions This research has shown that prioritization of species based on predictive modelling is possible given our knowledge of the distribution and drivers of a largely cryptic wildlife disease. Given that such diseases are being recognized in other wildlife species [e.g., 49,53,54], such predictive approaches are going to be increasingly necessary to manage the global species rescue that will be necessary if we are to not lose an irreplaceable sector of our biodiversity. Acknowledgements We are grateful to the numerous helpful comments of three anonymous reviewers which helped to improve this paper. The work of the first author was funded by the 'Graduiertenförderung des Landes Nordrhein-Westfalen'. Supplementary Material The following supplementary material is available at http://dx.doi.org/10.3390/d1010052. ESM 1: Table. Overview of amphibian species conservation status as listed by the IUCN Red List status and their geographic range size (downloaded 7 July 2009 from www.redlist.org), followed for each species by the portion of its geographic range suitable to Bd and, if applicable, the 'Biotic Index' Diversity 2009, 1 62 (BI) and the 'Risk Factor' (RF) estimated in this paper. Furthermore species that were encompassed under 'rapid enigmatic declines' in the Global Amphibian Assessment 2004 (updated under the IUCN Red List 2009) and species in which Bd has been reported in the literature are pointed out. ESM 2: List of most threatened species. The 379 species identified as most threatened through Bd, i.e. RF = 1.0 (BI = 1.0 and Maxent values > 0.50 at 100 % of their geographic range, see text). The 216 species listed as 'threatened with extinction' (i.e. IUCN Red List categories 'Vulnerable', 'Endangered' or 'Critically Endangered') are in bold while species that are too poorly known to asses the conservation status (i.e. categorised as 'Data Deficient' in the IUCN Red List) are italicised. A total number of 187 species that are assessed by the IUCN to be declining are marked with asterisks. Species in which Bd have been detected [55-58] are underlined. Entries are by family, following the taxonomy used by IUCN in the 2009 assessment [1,ESM 1]. References 1. Stuart, S.; Chanson, J.S.; Cox, N.A.; Young, B.E.; Rodrigues, A.S.L.; Fischman, D.L.; Waller, R.W. Status and trends of amphibian declines and extinctions worldwide. Science 2004, 306, 1783-1786. 2. Berger, L.; Speare, R.; Daszak, P.; Green, D.E.; Cunningham, A.A.; Goggin, C.L.; Slocombe, R.; Ragan, M.A.; Hyatt, A.D.; McDonald, K.R.; Hines, H.B.; Lips, K.R.; Marantelli, G.; Parkes, H.T. Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America. Proc. Natl. Acad. Sci. USA 1998, 95, 9031-9036. 3. Mendelson, J.R., III; Lips, K.R.; Gagliardo, R.W.; Rabb, G.B.; Collins, J.P.; Diffendorfer, J.E.; Daszack, P.; Ibáñez, R.; Zippel, K.C.; Lawson, D.P.; Wright, K.M.; Stuart, S.N.; Gascon, C.; Silva, H.R.; Burrowes, P.A.; Joglar, R.L.; La Marca, E.; Lötters, S.; Perez, L.H.; Weldon, C.; Hyatt, A.; Rodriguez-Mahecha, J.V.; Hunt, S.; Robertson, H.; Lock, B.; Raxworthy, C.J.; Frost, D.R.; Lacy, R.C.; Alfrod, R.A.; Campbell, J.A.; Parra-Olea, G.; Bolaños, F.; Domingo, J.J.C.; Halliday, T.; Murphy, J.B.; Wake, M.H.; Coloma, L.A.; Kuzmin, S.L.; Price, M.S.; Howell, K.M.; Lau, M.; Pethiyagoda, R.; Boone, M.; Lannoo, M.J.; Blaustein, A.R.; Dobson, A.; Griffiths, A.; Crump, M.L.; Wake, D.B.; Brodie, E.D.J. Confronting amphibian declines and extinctions. Science 2006, 313, 48. 4. Gascon, C.; Collins, J.P.; Moore, R.D.; Church, D.R.; McKay, J.E.; Mendelson, J.R., III. Amphibian Conservation Action Plan; IUCN: Gland, Switzerland, 2007. 5. Skerratt, L.F.; Berger, L.; Speare, R.; Cashins, S.; McDonald, K.R.; Phillott, A.D.; Hines, H.B.; Kenyon, N. Spread of chytridiomycosis has caused the rapid global decline and extinction of frogs. EcoHealth 2007, 4, 125-134. 6. Wake, D.B.; Vredenburg, V.T. Are we in the midst of the sixth mass extinction? A review from the world of amphibians. Proc. Natl. Acad. Sci. USA 2008, 105, 11466-11473. 7. Harris, R.N.; Lauer, A.; Simon, M.A.; Banning, J.L.; Alford, R.A. Addition of antifungal skin bacteria to salamanders ameliorates the effects of chytridiomycosis. Dis. Aquat. Org. 2009, 83, 11-16. Diversity 2009, 1 63 8. Deutsch, C.A.; Tweksbury, J.J.; Huey, R.B.; Sheldon, K.S.; Ghalambor, C.K.; Haak, D.C.; Martin, P.R. Impacts of climate warming on terrestrial ectotherms across latitude. Proc. Natl. Acad. Sci. USA 2008, 105, 6668-6672. 9. Lips, K.R.; Diffendorfer, J.; Mendelson, J.R., III; Sears, M.W. Riding the wave: reconciling the roles of disease and climate change in amphibian declines. PLoS Biol. 2008, 6, 441-454. 10. Woodhams, D.C.; Alford, R.A.; Briggs, C.J.; Johnson, M.; Rollins-Smith, L.A. Life-history trade-offs influence disease in changing climates: strategies of an amphibian pathogen. Ecology 2008, 89, 1627-1639. 11. Piotrowski, J.S.; Annis, S.L.; Longcore, J.E. Physiology of Batrachochytrium dendrobatidis, a chytrid pathogen of amphibians. Mycologia 2004, 96, 9-15. 12. Kriger, K.M.; Hero, J.-M. Large-scale seasonal variation in the prevalence and severity of chytridiomycosis. J. Zool. 2007, 271, 352-359. 13. Kriger, K.M.; Peregolou, F.; Hero, J.-M. Latitudinal variation in the prevalence and intensity of chytrid (Batrachochytrium dendrobatidis) infection in eastern Australia. Conserv. Biol. 2007, 21, 1280-1290. 14. Rödder, D.; Veith, M.; Lötters, S. Environmental gradients explaining prevalence and intensity of infection with the amphibian chytrid fungus: The host's perspective. Anim. Conserv. 2008, 11, 513-517. 15. Woodhams, D.C.; Alford, R.A.; Marantelli, G. Ecology of chytridiomycosis in rainforest stream frog assemblages of tropical Queensland. Conserv. Biol. 2005, 19, 1449-1459. 16. Kielgast, J.; Rödder, D.; Veith, M.; Lötters, S. Widespread occurrence of the amphibian chytrid fungus in Kenya. Animal Cons. 2009, (in press). 17. Blackburn, J.K.; McNyset, K.M.; Curtins, A.; Hugh-Jones, M.E. Modeling the geographic distribution of Bacillus anthracis, the causative agent of Anthrax disease, for the contiguous United States using predictive ecological niche modeling. Am. J. Trop. Med. Hyg. 2007, 77, 1103-1110. 18. Ostfeld, R.S.; Glass, G.E.; Keesing, F. Spatial epidemiology: an emerging (or re-emerging) discipline. TREE 2005, 20, 328-336. 19. Hijmans, R.J.; Cameron, S.E.; Parra, J.L.; Jones, P.G.; Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 2005, 25, 1965-1978. 20. Hutchinson, M.F. Anusplin version 4.3. Centre for Resource and Environment Studies. The Australian National University: Canberra, Australia, 2004. 21. Hijmans, R.J.; Cruz, J.M.; Rojas, E.; Guarino, L. Diva-GIS, version 1.4. A Geographic Information System for the Management and Analysis of Genetic Resources Data. Manual. International Potato Center and International Plant Genetic Resources Institute: Lima, Peru, 2001. 22. Heikkinen, R.K.; Luoto, M.; Araújo, M.B.; Virkkala, R.; Thuiller, W.; Sykes, M.T. Methods and uncertainties in bioclimatic envelope modeling under climate change. Progr. Phys. Geogr. 2006, 30, 751-777. 23. Hijmans, R.J.; Graham, C.H. The ability of climate envelope models to predict the effect of climate change on species distributions. Glob. Change Biol. 2006, 12, 2272-2281. Diversity 2009, 1 64 24. Ficetola, G.F.; Thuiller, W.; Miaud, C. Prediction and validation of the potential global distribution of a problematic alien species—the American bullfrog. Divers. Distrib. 2007, 13, 476-485. 25. Fitzpatrick, M.C.; Weltzin, J.F.; Sanders, N.J.; Dunn, R.R. The biogeography of prediction error: Why does the introduced range of the fire ant over-predict its native range? Glob. Ecol. Biogeogr. 2007, 16, 24-33. 26. Puschendorf, R.; Carnaval, A.C.; Vanderwal, J.; Zumbado-Ulate, H.; Chaves, G.; Bolaños, F.; Alford, R.A. Distribution models for the amphibian chytrid Batrachochytrium dendrobatidis in Costa Rica: proposing climatic refuges as a conservation tool. Divers. Distrib. 2009, 15, 401-408. 27. Dormann, C.F.; McPherson, J.; Araújo, M.B.; Bivand, R.; Bollinger, J.; Carl, G.; Davies, R.G.; Hirzel, A.; Jetz, W.; Kissling, W.D.; Kühn, I.; Ohlemüller, R.; Peres-Neto, P.R.; Reineking, B.; Schröder, B.; Schurr, F.M.; Wilson, R. Methods to account for spatial autocorrelation in the analysis of species distributional data: A review. Ecography 2007, 30, 609-628. 28. Ron, S.R. Predicting the distribution of the amphibian pathogen Batrachochytrium dendrobatidis in the new world. Biotropica 2005, 37, 209-221. 29. Wisz, M.S.; Hijmans, R.J.; Peterson, A.T.; Graham, C.H.; Guisan, A.; NPSDW Group. Effects of sample size on the performance of species distribution models. Divers. Distrib. 2008, 14, 763-773. 30. Phillips, S.J.; Dudík, M. Modelling of species distributions with Maxent: new extensions and comprehensive evaluation. Ecography 2008, 31, 161-175. 31. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231-259. 32. Phillips, S.J. Transferability, sample selection bias and background data in presence-only modelling: a response to Peterson et al. (2007). Ecography 2008, 31, 272-278. 33. Elith, J.; Graham, C.H.; Anderson, R.P.; Dudik, M.; Ferrier, S.; Guisan, A.; Hijmans, R.J.; Huettmann, F.; Leathwick, J.R.; Lehmann, A.; Li, J.; Lohmann, L.G.; Loiselle, B.A.; Manion, G.; Moritz, C.; Nakamura, M.; Nakazawa, Y.; Overton, J.M.M.; Peterson, A.T.; Phillips, S.J.; Richardson, K.; Scachetti-Pereira, R.; Shapire, R.E.; Soberón, J.; Williams, S.; Wisz, M.S.; Zimmermann, N.E. Novel methods improve prediction of species´ distributions from occurrence data. Ecography 2006, 29, 129-151. 34. Araújo, M.B.; New, M. Ensemble forecasting of species distributions. TREE 2007, 22, 42-47. 35. Fielding, A.H.; Bell, J.F. A review of methods for the assessment of prediction errors in conservation presence/absence models. Environ. Conserv. 1997, 24, 38-49. 36. Manel, S.; Williams, H.C.; Ormerod, S.J. Evaluating presence-absence models in ecology: The need to account for prevalence. J. Appl. Ecol. 2001, 38, 921-931. 37. Swets, K. Measuring the accuracy of diagnostic systems. Science 1988, 240, 1285-1293. 38. Vanderwal, J.; Shoo, L.P.; Johnson, C.N.; Williams, S.E. Abundance and the environmental niche: environmental suitability estimated from niche models predicts the upper limit of local abundance. Am. Nat. 2009, 174, 282-291. 39. Liu, C.; Pam, M.; Dawson, T.P.; Pearson, R.G. Selecting thresholds of occurrence in the prediction of species distributions. Ecography 2005, 28, 385-393. Diversity 2009, 1 65 40. Frost, D.R.; Grant, T.; Faivovich, J.; Bain, R.H.; Haas, A.; Haddad, C.F.B.; de Sá, R.O.; Channing, A.; Wilkinson, M.; Donnellan, S.C.; Raxworthy, C.J.; Campbell, J.A.; Blotto, B.L.; Moler, P.; Drewes, R.C.; Nussbaum, R.A.; Lynch, J.D.; Green, M.D.; Wheeler, W.C. The amphibian tree of life. Bull. Am. Mus. Nat. Hist. 2006, 297, 1-370. 41. Grant, T.; Frost, D.R.; Caldwell, J.P.; Gagliardo, R.; Haddad, C.F.B.; Kok, P.J.R.; Means, D.B.; Noonan, B.P.; Schargel, W.E.; Wheeler, W.C. Phylogenetic systematics of dart-poison frogs and their relatives (Amphibia, Athesphatanura, Dendrobatidae). Bull. Am. Mus. Nat. Hist. 2006, 299, 1-262. 42. Bielby, J.; Cooper, N.; Cunningham, A.A.; Garner, T.W.J.; Purvis, A. Predicting susceptibility to rapid declines in the world’s frogs. Conserv. Lett. 2008, 1, 82-90. 43. Hijmans, R.J.; Cameron, S.E.; Parra, J.L.; Jones, P.G.; Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 2005, 25, 1965-1978. 44. Willmott, C.J.; Kenji, M. Terrestrial Air Temperature and Precipitation: Monthly and Annual Time Series (1950-1999). Available online: http://climate.geog.udel.edu/~climate/html_pages/ README.ghcn_ts2.html (access 8 July 2009). 45. CIESIN. Center for International Earth Science Information Network, Columbia University, New York, NY, USA, 2000; Available online http://www.ciesin.columbia.edu/ download_data.html (access on 8 July 2009). 46. Fisher, M.C.; Garner, T.W.J. The relationship between the emergence of Batrachochytrium dendrobatidis, the international trade in amphibians and introduced amphibian species. Fungal Biol. Rev. 2007, 21, 2-9. 47. Weldon, C.; du Preez, L.H.; Hyatt, A.D.; Muller, R.; Spears, R. Origin of the amphibian chytrid fungus. Emerg. Infect. Dis. 2004, 10, 2100-2105. 48. Cunningham, A.A. A walk on the wild side - emerging wildlife diseases. Br. Med. J. 2005, 331, 1214-1215. 49. Cunningham, A.A.; Daszak, P.; Rodriguez, J.P. Pathogen pollution: defining a parasitological threat to biodiversity conservation. J. Parasitol. 2003, 9, S78-S83. 50. Gaston, K.J.; Fuller, R.A. The sizes of species' geographic ranges. J. Appl. Ecol. 2009, 46, 1-9. 51. Araújo, M.B.; Pearson, R.G. Equilibrium of species' distribution with climate. Ecography 2005, 28, 693-695. 52. Guisan, A.; Zimmermann, N. Predictive habitat distribution models in ecology. Ecol. Model. 2000, 135, 147-186. 53. Blehert, D.S.; Hicks, A.C.; Behr, M.; Meteyer, C.U.; Berlowski-Zier, B.M.; Buckles, E.L.; Coleman, J.T.H.; Darling, S.R.; Niver, R.; Okoniewski, J.C.; Rudd, R.J.; Stone, W.B. Bat white-nose syndrome: An emerging fungal pathogen? Science 2009, 323, 227. 54. Rahbek, C. Disease ecology: the silence of the robins. Nature 2007, 447, 652-653. 55. La Marca, E.; Lips, K.R.; Lötters, S.; Puschendorf, R.; Ibáñez, R.; Rueda-Almonacid, J.V.; Schulte, R.; Marty, C.; Castro, F.; Manzanilla-Puppo, J.; García, J.E.; Bolaños, F.; Chaves, G.; Pounds, J.A.; Toral, E.; Young, B.E. Catastrophic population declines and extinctions in Neotropical harlequin frogs (Bufonidae: Atelopus). Biotropica 2005, 37, 190-201. Diversity 2009, 1 66 56. Lampo, M.; Rodríguez, A.; La Marca, E.; Daszak, P. A chytridiomycosis epidemic and a severe dry season precede the disappearance of Atelopus species from the Venezuelan Andes. J. Herpetol. 2007, 16, 395-402. 57. Barrionuevo, S.; Mangione, S. Chytridiomycosis in two species of Telmatobius (Anura: Leptodactylidae) from Argentina. Dis. Aquat. Org. 2006, 73, 171-174. 58. Kriger, K.M.; Hero, J.-M. The chytrid fungus Batrachochytrium dendrobatidis is non-randomly distributed across amphibian breeding habitats. Divers. Distrib. 2007, 13, 781-788. © 2009 by the authors; licensee Molecular Diversity Preservation International, Basel, Switzerland. This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/). 

Investigating the Potential Use of Environmental DNA (eDNA) for Genetic Monitoring of Marine Mammals Andrew D. Foote1*., Philip Francis Thomsen1., Signe Sveegaard2, Magnus Wahlberg3,4, Jos Kielgast1, Line A. Kyhn2, Andreas B. Salling1, Anders Galatius2, Ludovic Orlando1, M. Thomas P. Gilbert1 1 Centre for GeoGenetics, Natural History Museum of Denmark, University of Copenhagen, Copenhagen, Denmark, 2 Department of Bioscience, Aarhus University, Roskilde, Denmark, 3 Fjord&Bælt, Kerteminde, Denmark, 4 Marine Biological Laboratory, University of Southern Denmark, Kerteminde, Denmark Abstract The exploitation of non-invasive samples has been widely used in genetic monitoring of terrestrial species. In aquatic ecosystems, non-invasive samples such as feces, shed hair or skin, are less accessible. However, the use of environmental DNA (eDNA) has recently been shown to be an effective tool for genetic monitoring of species presence in freshwater ecosystems. Detecting species in the marine environment using eDNA potentially offers a greater challenge due to the greater dilution, amount of mixing and salinity compared with most freshwater ecosystems. To determine the potential use of eDNA for genetic monitoring we used specific primers that amplify short mitochondrial DNA sequences to detect the presence of a marine mammal, the harbor porpoise, Phocoena phocoena, in a controlled environment and in natural marine locations. The reliability of the genetic detections was investigated by comparing with detections of harbor porpoise echolocation clicks by static acoustic monitoring devices. While we were able to consistently genetically detect the target species under controlled conditions, the results from natural locations were less consistent and detection by eDNA was less successful than acoustic detections. However, at one site we detected long-finned pilot whale, Globicephala melas, a species rarely sighted in the Baltic. Therefore, with optimization aimed towards processing larger volumes of seawater this method has the potential to compliment current visual and acoustic methods of species detection of marine mammals. Citation: Foote AD, Thomsen PF, Sveegaard S, Wahlberg M, Kielgast J, et al. (2012) Investigating the Potential Use of Environmental DNA (eDNA) for Genetic Monitoring of Marine Mammals. PLoS ONE 7(8): e41781. doi:10.1371/journal.pone.0041781 Editor: Senjie Lin, University of Connecticut, United States of America Received April 5, 2012; Accepted June 29, 2012; Published August 29, 2012 Copyright: ß 2012 Foote et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: A.D.F. was funded by a Marie Curie Intra-European Fellowship no. 272385. L.O. was supported by a EU Marie Curie Career Integration Grant (293845). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: Ludovic Orlando and Tom Gilbert are PLoS ONE editorial board members. Magnus Walhberg is an employee of a commercial company, Fjord&Baelt. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials. * E-mail: footead@gmail.com . These authors contributed equally to this work. Introduction (eDNA), and can provide a record of the species’ presence over the period that the DNA persists in the environment [7–14]. The use of molecular genetic markers for monitoring biodiver- In freshwater aquatic environments the use of eDNA for genetic sity and detecting and identifying species, individuals or measuring monitoring has been tested by a number of recent studies (e.g. population genetic parameters can provide valuable information [9,11–14]), which suggest that eDNA is homogenously distributed for the management and conservation of species and ecosystems within freshwater systems and can be effectively used to detect and [1]. Non-invasive sampling using hair or scat has been successfully even quantify species presence [11]. Here, we investigate whether used in genetic monitoring programs of wide-ranging terrestrial or eDNA from the water column can be used to detect target species semi-aquatic species, which often occur at low density [1] and can occurrence of mammals in the marine environment. The use of be a lower cost approach than direct sampling as the individual or seawater samples for eDNA analysis is likely to be more species does not have to be directly encountered and greater challenging than freshwater due to the larger body of source number of samples can be collected. However, this is offset by water, strong tide and current action, which will rapidly dilute and what can be increased DNA extraction and sequencing costs and disperse the eDNA. Further, the high salinity of the samples may also a decrease in returns to the potentially large numbers of also render amplification of eDNA by polymerase chain reactions duplicate samples [2]. For fully aquatic species such as cetaceans, (PCRs) more prone to inhibition [15]. Despite these potential the collection of non-invasive samples is less straightforward and drawbacks, quantification of DNA in the sea has been used to often (but not always; see ref [3]) requires the sampler to directly provide an indicator of biomass in marine ecosystems [16], and encounter the target species to collect non-invasive samples such as sampling and sequencing of intracellular DNA of microorganisms feces, sloughed skin or exhalation blow [3–6]. Current non- sampled from seawater has enabled the metagenomic investigation invasive sampling of marine species therefore has many of the of their biodiversity and community structure [17,18]. same logistical and financial costs as biopsy sampling. However, In this study, we test the potential for using eDNA to detect the biological excretory processes such as the sloughing of skin, presence of marine mammals, by using as a model a small urination and defecation can be sources of ‘environmental’ DNA cetacean species, the harbor porpoise Phocoena phocoena. The harbor PLOS ONE | www.plosone.org 1 August 2012 | Volume 7 | Issue 8 | e41781  Molecular Detection of Cetaceans from eDNA porpoise is the only regularly occurring cetacean species in the recorded by the C-PODs were analyzed with the software western Baltic, the region where this study took place, although the CPOD.exe (v 2.021) with the filters ‘NBHF’ and ‘Other species is rare in the inner Baltic [19,20]. Static acoustic Cetaceans’ to search for porpoise clicks and clicks from other monitoring devices which log detected echolocation click trains cetaceans, respectively. The percentage of porpoise positive days of harbor porpoises provide a record of occurrence and relative (i.e. days in which porpoise clicks were detected) were calculated density at each site [21–23] and reliable field validation of our for each site during the three month recording period prior to the eDNA based tests. Here, we experimentally demonstrate, using date of water sampling (Table 1). No specific permits were both controlled conditions and natural populations, the feasibility required for the described field studies. of targeted eDNA based animal detection from seawater samples. DNA from seawater samples was extracted in a dedicated clean lab in a building separate from the location of post PCR work and Methods extraction of DNA from the epidermis samples. Rigorous controls for preventing and monitoring contamination adopted from Seawater samples were collected under both controlled condi- ancient DNA protocols were employed. Seawater samples were tions, and from natural field sites. The controlled site was a sea pen centrifuged at 6000 g for 10 min to pellet any precipitated DNA. in a natural harbor basin at Fjord&Bælt (F&B) in Kerteminde, One blank extraction using molecular biology grade water was Denmark (Fig. 1a). The pen holds four harbor porpoise in included for every nine seawater samples to monitor possible approximately 4 million liters of seawater, which is flushed daily by contamination. Following centrifugation the supernatant was the tidal water movements in the harbor basin, that enter at the discarded and DNA was extracted from the pellet using the netted ends of the pen. Five 15 ml water samples were collected at Qiagen DNeasy (Qiagen DNeasy, Valencia, CA, USA) kit a depth of approximately 50 cm from different points around the following the manufacturer’s guidelines and eluted in 100 ml of perimeter of the sea pen in a sterile container, which was sealed buffer. The ethanol wash step of the extraction process is expected until just prior to sampling and handled using unused sterile latex to remove most of the Na+ monovalent ions and therefore reduce gloves, which were discarded after the collection of each sample PCR inhibition due to salinity. DNA was also extracted from the (Fig. 1b). An additional 45 samples of 15 ml were collected at epidermis samples using the Qiagen DNeasy kit. varying distances (,0–1 km) from the pen in the direction of the The PrimerBlast software (NCBI, http://www.ncbi.nlm.nih. ebbing tide. After collection, 1.5 ml of 3 M sodium acetate and gov/tools/primer -blast/) was used to design primers unique to the 33 ml absolute ethanol was added to the water samples to harbor porpoise that would result in amplicons ranging between precipitate any extracellular DNA (final concentrations 0.09 M, 60–80 bp in size, based on records in GenBank. The primers 59- and 66% respectively), which were then stored at 220uC until CGCCCCATCAACACAAAGGTTTGG-39 and 59-ACTGG- DNA extraction. As a control, a small layer of epidermis was also GATGCGGATGCTTGC-39 flanked the region corresponding collected from each of the four porpoises using Scotch tape. to sites 82–119 of the harbor porpoise mitochondrial genome Epidermal samples were stored in 20% dimethyl sulphoxide (GenBank: AJ554063; [24]) and resulted in a 62 bp amplicon of (DMSO) saturated with NaCl at 220uC until DNA extraction. the 12S region of the mitochondrial genome. This 38 bp intra- Seawater samples were also collected during August 2011 at 8 primer sequence was monomorphic in all four of the F&B harbor locations in the western Baltic at sites where static acoustic porpoise, as well as an additional four North Sea and Baltic harbor monitoring devices, C-PODs (Cetacean and POrpoise Detector, porpoises from the Natural History Museum of Denmark’s tissue Chelonia Ltd., U.K.), were deployed as part of the project Static archive. Furthermore, the sequence differs by at least one base pair Acoustic Monitoring of the Baltic Sea Harbor Porpoise (SAM- (either a C-A transversion or C-T transition), from the next most BAH). The water samples were collected approximately 50 cm similar sequence present in GenBank, which was shared by several below the surface as above, employing the same protocols to cetacean species, none of which are likely to occur at our study reduce the risk of contamination. Three 50 ml samples were sites [25], and differed at 6 nucleotide sites from humans, and collected at each acoustic datalogger site and a 15 ml aliquot was harbor and grey seals, which are the only other common marine taken from each of these samples and treated as above. The data mammals in the area, and thus represent possible sources of Figure 1. The sea pen at Fjord&Bælt in Kerteminde, Denmark. Figure 1a shows the sea pen containing approximately 4 million liters of seawater and 4 harbor porpoise. The two ends of the pen are comprised of netting, allowing tidal seawater to move through the pen (photograph ß Solvin Zankl). Figure 1b shows the sampling of 15 ml of seawater from the sea pen (photograph ß Ross Culloch). doi:10.1371/journal.pone.0041781.g001 PLOS ONE | www.plosone.org 2 August 2012 | Volume 7 | Issue 8 | e41781  Molecular Detection of Cetaceans from eDNA Table 1. Detection of harbor porpoise DNA using qPCR at a controlled site (Fjord&Belt pen) and at natural sites. Acoustic detection Genetic detection Location % Porpoise positive days Positive PCRs Cycle threshold Positive control (DNA extracted from skin) 3/3 18, 18, 18 Fjord&Bælt pen 3/3 34, 35, 35 ,10 m from F&B pen 1/3 49 .10 m from F&B pen 0/3 - Site 1 94 1/3 49 Site 2 42 0/3 - Site 3 63 0/3 - Site 4 6 0/3 - Site 5 0 0/3 - Site 6 0 0/3 - Site 7 0 2/3 38*, 50* Site 8 79 0/3 - Genetic detections at the eight natural sites are compared with acoustic detection rates based on data from static acoustic monitoring devices over the three months prior to eDNA sampling. *sequencing of PCR clones indicates these were genetic detections of long-finned pilot whale and not genetic detection of harbor porpoise at this site. doi:10.1371/journal.pone.0041781.t001 mammalian DNA in the water. To increase sensitivity and to investigate species identity in one sample, which did not match specificity of porpoise detection we performed TaqMan qPCRs the porpoise reference sequence. detection assays. A TaqMan probe specific for the target sequence was designed (59-TCCTGGCCTTTCTATTAGTTCTTAGCA- Results 39), modified with 6-Fam dye at the 59 end and a BHQ1 quencher at the 39 end. Porpoise DNA was successfully amplified from a pooled sample Taqman qPCRs were performed on a Stratagene Mx3000P and of the 5 DNA extracts from each of the 15 ml seawater samples each 25 ml reaction contained 10 ml of DNA extract, 16 PCR collected from the Fjord&Bælt sea pen (Table 1). The cloned buffer, 2.5 mM MgCl2, 1 mM of each primer, 0.1 mM mixed sequences were a 100% match with the reference sequence from dNTPs, 2.5 mM of probe and 0.2 ml AmpliTaq Gold enzyme GenBank and the sequences generated from skin samples of the (Applied Biosystems) under thermocycling 50uC for 5 min and four harbor porpoise in the Fjord&Bælt sea pen and the four wild 95uC for 10 min, followed by 55 cycles of 95uC for 30 sec and porpoise (Fig. 2). The number of qPCR cycles required for detection was also consistently between 34–35 cycles for all three 56uC for 1 min. To guard against the incorporation of erroneous eDNA PCR replicates, as oppose to 18 for positive controls data derived from contamination, and to investigate the amplifying tissue-derived DNA (Table 1). Assuming optimal PCR stochasticity of successful amplifications, the PCR amplification efficiency, this suggests a minimal difference of 4–5 orders of was replicated three times for each sample. One PCR blank magnitude in DNA concentration, as expected for eDNA extracts (containing ddH2O instead of sample) was included for every five generated from such dilute environmental samples. Additionally, PCRs to further monitor for contamination during PCR set up. porpoise DNA was successfully amplified in all triplicates on For initial investigation we pooled DNA extracts: DNA extracts individual 15 ml samples. Harbor porpoise eDNA was amplified from the five samples collected inside the F&B pen were pooled, in one out of three qPCRs on a pooled DNA extract from DNA extracts from the five samples collected at less than 5615 ml samples collected at a distance of less than 10 m from the 10 meters from the pen were pooled, and DNA extracts from pen. Beyond 10 m from the pen, we were unable to detect thirty-eight samples collected at distances greater than 10 meters porpoise eDNA (Table 1). from the pen were pooled. Only if DNA was successfully amplified The porpoise genetic detection rate from natural field sites from at least one of the triplicate PCRs of the pooled extracts were within the western Baltic was validated by comparing with the individual extracts included in subsequent PCRs. The detections of harbor porpoise echolocation clicks detected by the amplified PCR products were purified using a Qiagen MinElute C-PODs. Harbor porpoises were only genetically detected by PCR purification kit. The species origin of positive PCRs were eDNA at the site (site 1) with the highest percentage (94%) of days validated as authentic by cloning using the Topo TA cloning kit that porpoises were acoustically detected by the C-PODs (Table 1). (Invitrogen), followed by purification and sequencing of the Cloned sequences of qPCR products were a 100% match for the inserted PCR fragment (Macrogen, Europe). Additional PCRs of GenBank reference sequence (Fig. 2). Long-term acoustic detec- 43 bp of the cytochrome b gene (fwd primer: 59-ACACACC- tion rates at the remaining sampled sites suggest that the lack of CACTAATAAAAAT-39; rev primer: 59-AGCCAAAATTTCAT- genetic detections at some of these sites were false negatives CATGAGGA-39) and the 53 bp of the hypervariable region of the (Table 1). However, it was not possible to confirm this as the C- d-loop (fwd primer: 59-ACACATACCAATATC- PODs ran out of battery charge 3–4 weeks before the seawater TAGTCTTTCCTT-39; rev primer: 59-CGGGCTTTAACT- samples were collected and eDNA typically persists for less than TATCGTATGG-39) using the conditions above were conducted one week (based on studies in freshwater [11,14] and seawater [26]), and extracellular eDNA can have a turnover time of only a PLOS ONE | www.plosone.org 3 August 2012 | Volume 7 | Issue 8 | e41781  Molecular Detection of Cetaceans from eDNA Figure 2. Chromatograms of 12S region mtDNA sequences amplified from two locations in the Baltic aligned to reference sequences. Sites are numbered 82-119 corresponding with the reference harbor porpoise mitochondrial genome (GenBank: AJ554063; [24]). doi:10.1371/journal.pone.0041781.g002 few hours in seawater [27]. Thus it is not known if porpoises were study was less than found in previous studies using a comparable present in the area in the hours prior to sampling. protocol to sample freshwater ecosystems (e.g. [9,11]). There may Environmental DNA was amplified at a site with no acoustic be several reasons for this including those identified above, for detections of porpoises (Site 7, Table 1). Cloned sequences from example, the greater dispersal and dilution of eDNA in marine the two successful PCRs differed by 2 base pairs from harbor ecosystems compared to lakes and ponds. Porpoise eDNA was porpoise (Fig. 2), and additional PCRs using primers to target successfully amplified from five different 15 ml water samples variable regions of the mtDNA cytochrome b gene and d-loop collected at different points around the perimeter of the confirmed the detection as long-finned pilot whale (Globicephala Fjord&Bælt sea pen and therefore appears to be relatively melas) (Table 2), a species occasionally sighted in the Baltic [25]. homogenously distributed throughout the 4 million liter sea pen. This is consistent with findings from studies on pond water [9,11], Discussion and demonstrates that species detection of marine mammals at high density is possible using eDNA from seawater samples. Our study indicates that species detection of marine mammals However, these controlled conditions differ in several respects using eDNA in seawater samples is possible and thus could have from natural conditions. In particular, the animal density in the potential use in future genetic monitoring programs. However, it sea pen is higher than in most natural populations. Furthermore, has to be acknowledged that the rate of successful detections in this the sheltered pen is less subject to wave and wind action and Table 2. Summary of additional PCRs carried out on seawater samples from site 7. Target gene cloned PCR product sequence (excluding primers) GenBank Blast results cytochrome b CATCAATGACACATTCATTGACCTACCCACTCCATCTAACATC 100% G. melas D-loop CTTATAAATATATATATATACATGCTATGTATTACTGTGCATTCATTTATTTT 100% G. melas Both primer sets produced a sequence with 100% query coverage and 100% maximum identity match with long-finned pilot whale Globicephala melas and no other species. doi:10.1371/journal.pone.0041781.t002 PLOS ONE | www.plosone.org 4 August 2012 | Volume 7 | Issue 8 | e41781  Molecular Detection of Cetaceans from eDNA eDNA from porpoises within the pen may not be dispersed to as and travelled with sea currents, or originated from the remains of a great an extent as under natural conditions and is thus more long dead animal (e.g. DNA from mammoths has been successfully concentrated here. Comparing the number of qPCR cycles amplified from riverine sediment samples [28]). required for detection at the natural site (site 1) with the sea pen Two lines of further investigation could be used to clarify these would suggest a minimal difference of 4–5 orders of magnitude in possible scenarios. Firstly, determining if the detected eDNA is DNA concentration. Increasing the sample volume would be cellular, (for example contained within slough skin cells or fecal expected to reduce the rate of false negatives, if the DNA could be material), or extracellular DNA by using extraction methods concentrated (e.g., [26]). Thomsen et al., [26] were able to designed to isolate extracellular DNA from DNA extracted by cell genetically detect several fish species by passing half a liter of a lysis (e.g. [29]). As extracellular DNA degrades much faster (hours) 1.5 liter pool of 30 ml Baltic seawater samples through a nylon than cellular DNA (days) [27], it would be more likely to have filter and then extracting the DNA collected from the filter. Such originated from a living animal recently in the area than have an approach may improve the detection rate and the applicability travelled on currents or be from remains of dead specimens. of this approach to species detection using seawater samples. The Secondly, the controlled release of DNA of a non-native species of detection probability is likely to be dependent upon density of the known concentration could be used to experimentally investigate target species, the amount of DNA released by the organism the longevity of DNA in the water column and its propagation by through excretory processes, and the amount of degradation by wind, wave and ocean currents. local environmental factors such as endogenous nucleases, Our results suggest that, as expected, species detection using hydrolysis, UV radiation and bacterial action [14], it is therefore eDNA is less reliable in the marine ecosystem than in freshwater likely to be vary greatly amongst target species and study locations. ecosystems. However, our results do suggest that marine mammal Despite these limitations, the successful genetic detection of detection by amplifying eDNA from seawater samples and using harbor porpoise at one location where the species was also short species-specific DNA sequences as DNA barcodes [30] is acoustically detected indicates that, with optimization, genetic possible. However, it is likely to be dependent upon the size, detection of marine mammals could provide a useful non-invasive behavior and density of the target species. The volumes of genetic monitoring tool that could compliment visual and acoustic seawater sampled in this study were small to allow comparison surveys. The genetic detection of long-finned pilot whale from with previous freshwater studies [9,11], which sampled similar seawater collected at our most easterly study site, about 75 km volumes of water. With optimization and larger volumes of southeast of Bornholm, highlights the potential for this method to seawater this method could have potential to compliment current detect species that are rare visitors to an area. No other genetic visual and acoustic methods of species detection of marine work on this species has been conducted in the laboratories in mammals and provide a low-cost, logistically simple method of which the work was undertaken, which, in addition to the rigorous obtaining basic genetic data such as species presence or even clean lab procedures employed, makes laboratory contamination intraspecific variation in short diagnostic fragments. an unlikely source of the DNA. Long-finned pilot whales are infrequently sighted in the Baltic, however, there were two Acknowledgments potential but unconfirmed sightings of pilot whales in the western Baltic in July 2011 (www.hvaler.dk). The genetic detection We would like to thank Tina Brand, Pernille Olsen and Jesper Stenderup presented here remains the only confirmed detection of this for laboratory assistance, Jakob H Kristensen, Janni Damsgaard Hansen, and Camilla Eriksson at Fjord&Baelt, Ross Culloch and the crew of the species in the Baltic during this period. C-PODs can also detect RV Skagerak for help with sample collection, the SAMBAH project (www. clicks in the frequency range produced by pilot whales. However, sambah.org) for allowing the use of the unpublished C-pod data. The the C-POD file contained no such click detections. As acoustic animals are maintained by Fjord&Bælt, Kerteminde, Denmark, under monitoring ended one month prior to eDNA water sampling, the permits no. SN 343/FY-0014 and 1996-3446-0021 from the Danish Forest acoustic data cannot act to validate the presence of pilot whales and Nature Agency, Danish Ministry of Environment. contemporaneously with eDNA sampling at this site. Pilot whales are much larger than harbor porpoise and are typically more Author Contributions gregarious, and they would therefore be expected to excrete and Conceived and designed the experiments: ADF PFT LO MTPG. shed more eDNA and be more easily detectable using this method. Performed the experiments: ADF SS MW ABS AG. Analyzed the data: However, the indirect method of sampling DNA used here means ADF PFT JK LK SS ABS. Contributed reagents/materials/analysis tools: that it is not possible to determine whether the animals were LK SS MTPG. Wrote the paper: ADF PFT LK JK SS MW ABS AG LO recently in the area, or whether the eDNA originated elsewhere MTPG. References 1. Schwartz MK, Luikart G, Waples RS (2007) Genetic monitoring as a promising 8. Willerslev E, Hansen AJ, Binladen J, Brand TB, Gilbert MTP, et al. (2003) tool for conservation and management. Trends Ecol Evol 22: 25–33. Diverse plant and animal genetic records from Holocene and Pleistocene 2. Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look sediments. Science 300: 791–795. before you leap. Trends Ecol Evol 14: 323–327. 9. Ficetola GF, Miaud C, Pompanon F, Taberlet P (2008) Species detection using 3. Gillet RM, Frasier TR, Rolland RM, White BN (2010) Molecular identification environmental DNA from water samples. Biol Lett 4: 423–425. of individual North Atlantic right whales (Eubalaena glacialis) using free-floating 10. Andersen K, Bird KL, Rasmussen M, Haile J, Breuning-Madsen H, et al. (2012) feces. Mar Mamm Sci 26: 917–936. Meta-barcoding of ‘dirt’ DNA from soil reflects vertebrate biodiversity. Mol Ecol 4. Parsons KM, Durban JW, Claridge DE (2003) Comparing two alternative doi: 10.1111/j.1365-294X.2011.05261.x methods for sampling small cetaceans for molecular analysis. Mar Mamm Sci 11. Thomsen PF, Kielgast J, Iversen LL, Wiuf C, Rasmussen M, et al. (2012) 19: 224–231. Monitoring endangered freshwater biodiversity using environmental DNA. Mol 5. Rolland RM, Hamilton PK, Kraus SD, Davenport B, Gillett RM, et al. (2006) Ecol doi: 10.1111/j.1365-294X.2011.05418.x Faecal sampling using detection dogs to study reproduction and health in North 12. Goldberg CS, Pilliod DS, Arkle RS, Waits LP (2011) Molecular detection of Atlantic right whales (Eubalaena glacialis). J Cetacean Res Manage 8: 121–125. vertebrates in stream water: a demonstration using Rocky Mountain tailed frogs 6. Fre`re CH, Krzyszczyk E, Patterson EM, Hunter S, Ginsburg A, et al. (2010) and Idaho giant salamanders. PLoS One 6: e22746. Thar She Blows! A Novel Method for DNA Collection from Cetacean Blow. 13. Jerde CL, Mahon AR, Chadderton WL, Lodge DM (2011) ‘‘Sight-unseen’’ PLoS ONE 5: e12299. detection of rare aquatic species using environmental DNA. Conserv Lett 4: 7. Hofreiter M, Mead JI, Martin P, Poinar HN (2003) Molecular caving. Curr Biol 150–157. 13: R693–R695. PLOS ONE | www.plosone.org 5 August 2012 | Volume 7 | Issue 8 | e41781  Molecular Detection of Cetaceans from eDNA 14. Dejean T, Valentini A, Duparc A, Pellier-Cuit S, Pompanon F, et al. (2011) 22. Kyhn LA, Tougaard J, Teilmann J, Wahlberg M, Jørgensen PB, et al. (2008) Persistence of environmental DNA in freshwater ecosystems. PLoS one 6: Harbour porpoise (Phocoena phocoena) static acoustic monitoring: Laboratory e23398. detection thresholds of T-PODs are reflected in field sensitivity. J Mar Biol Assoc 15. Wilson IG (1997) Inhibition and facilitation of nucleic acid amplification. Appl UK 88: 1085–1091. Environ Microbiol 63: 3741–3751. 23. Bailey H, Clay G, Coates EA, Lusseau D, Senior B, et al. (2009) Using T-PODs 16. Holm-Hansen O, Sutcliffe Jr WH, Sharp J (1968) Measurement of deoxyribo- to assess variations in the occurrence of coastal bottlenose dolphins and harbour nucleic acid in the ocean and its ecological significance. Limnol Oceanogr 13: porpoises. Aquat Conserv: Mar Freshw Ecosyst 20: 150–158. 507–514. 24. Arnason U, Gullberg A, Janke A (2004) Mitogenomic analyses provides new 17. Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, et al. 2004 insights into cetacean origin and evolution. Gene 333: 27–34. Environmental genome shotgun sequencing of the Sargasso Sea. Science 304: 25. Kinze CC, Jensen T, Skov R (2003) Focus on whales in Denmark 2000–2002. 66–74. Biological writings, Nr.2, Esbjerg, Denmark: Fisheries and Maritime Museum. 18. Rusch DB, Halpern AL, Sutton G, Heidelberg KB, Williamson S, et al. (2007) 26. Thomsen PF, Kielgast J, Lønsmann Iversen L, Rask Møller P, Ramussen M, et The Sorcerer II global ocean sampling expedition: Northwest Atlantic through al. (2012) Detection of a diverse marine fish fauna using environmental DNA eastern tropical Pacific. PLoS Biol 5: e77. from seawater samples. PLoS one 19. Verfuss UK, Honnef CG, Meding A, Dahne M, Mundry R, et al. (2007) 27. Dell’Anno A, Corinaldesi C (2004) Degradation and turnover of extracellular Geographical and seasonal variation of harbour porpoise (Phocoena phocoena) DNA in marine sediments: ecological and methodological considerations. Appl presence in the German Baltic Sea revealed by passive acoustic monitoring. Environ Microbiol 70: 4384–4386. J Mar Biol Assoc UK 87: 165–176. 28. Haile J, Froese DG, MacPhee RDE, Roberts RG, Arnold LJ, et al. (2009) 20. Scheidat M, Gilles A, Kock K, Siebert U (2008) Harbour porpoise Phocoena Ancient DNA reveals late survival of mammoth and horse in interior Alaska. phocoena abundance in the southwestern Baltic Sea. Endang Spec Res 5: 215– Proc Natl Acad Sci 106: 22352–22357. 223. 29. Corinaldesi C, Danovaro R, Dell’Anno A (2005) Simultaneous recovery of 21. Carstensen J, Henriksen OD, Teilmann J (2006) Impacts of offshore wind farm extracellular and intracellular DNA suitable for molecular studies from marine construction on harbour porpoises: acoustic monitoring of echolocation activity sediments. Appl Environ Microbiol 71: 46–50. using porpoise detectors (T-PODs). Mar Ecol Prog Ser 321: 295–308. 30. Gross M (2012) Barcoding biodiversity. Curr Biol 22: R73–R76. PLOS ONE | www.plosone.org 6 August 2012 | Volume 7 | Issue 8 | e41781 

The following supplement accompanies the article Future potential distribution of the emerging amphibian chytrid fungus under anthropogenic climate change Dennis Rödder1, 2, Jos Kielgast3,*, Stefan Lötters2 1 Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany 2 Department of Biogeography, Trier University, Am Wissenschaftspark 25-27, 54296 Trier, Germany 3 Natural History Museum of Denmark, Zoological Museum, Universitetsparken 15, 2100 Copenhagen, Denmark *Corresponding author. Email: jkielgast@snm.ku.dk Diseases of Aquatic Organisms: dao02197 (2010) Supplement Maxent species distribution models for the A2a and B2a anthropogenic future climate change scenarios for the 2080 CCCMA, CSIRO and HADCM3 models and corresponding standard deviations Fig. S1. Standard deviation (SD) per grid cell of each of 100 Maxent models projected onto (A) A2a and (B) B2a scenarios 2 Supplement (continued) Fig. S2. Species distribution model projection onto the A2a family. Warmer colours indicate higher climatic suitability for Batrachochytrium dendrobatidis (Bd)  Supplement (continued) 3 Fig. S3. Species distribution model projection onto the B2a family. Warmer colours indicate higher climatic suitability for Batrachochytrium dendrobatidis (Bd) 

EcoHealth 6, 358–372, 2009 DOI: 10.1007/s10393-010-0281-6 Ó 2010 International Association for Ecology and Health Short Communication The Link Between Rapid Enigmatic Amphibian Decline and the Globally Emerging Chytrid Fungus Stefan Lo¨tters,1 Jos Kielgast,2 Jon Bielby,3 Sebastian Schmidtlein,4 Jaime Bosch,5 Michael Veith,1 Susan F. Walker,6 Matthew C. Fisher,6 and Dennis Ro¨dder1 1 Department of Biogeography, Trier University, Am Wissenschaftspark 25-27, 54296 Trier, Germany 2 Department of Biology, Section for Microbiology and Evolution, Copenhagen University, Universitetsparken 15, 2100 Copenhagen, Denmark 3 Department of Biology, Imperial College London, Silwood Park Campus Ascot, London SL5 7PY, UK 4 Department of Geography, Bonn University, Meckenheimer Allee 166, 53115 Bonn, Germany 5 Museo Nacional de Ciencias Naturales, CSIC, C/Jose´ Gutie´rrez Abascal 2, 28006 Madrid, Spain 6 Department of Infectious Disease Epidemiology, Imperial College London, St. Mary’s Hospital, Norfolk Place, London W2 1PG, UK Abstract: Amphibians are globally declining and approximately one-third of all species are threatened with extinction. Some of the most severe declines have occurred suddenly and for unknown reasons in apparently pristine habitats. It has been hypothesized that these ‘‘rapid enigmatic declines’’ are the result of a panzootic of the disease chytridiomycosis caused by globally emerging amphibian chytrid fungus. In a Species Distribution Model, we identified the potential distribution of this pathogen. Areas and species from which rapid enigmatic decline are known significantly overlap with those of highest environmental suitability to the chytrid fungus. We confirm the plausibility of a link between rapid enigmatic decline in worldwide amphibian species and epizootic chytridiomycosis. Keywords: Batrachochytrium dendrobatidis, bioclimate, chytridiomycosis, IUCN Red List, MaxEnt, species distribution model The importance of infectious diseases in animal population A pertinent example concerns the status of the world’s dynamics and anthropogenic dissemination of pathogens amphibians (Wake and Vredenburg, 2008). According has long been recognized (Anderson and May, 1979; Scott, to the IUCN Red List of Threatened Species (www. 1988; McCallum and Dobson, 1995). However, within the iucnredlist.org), approximately one-third of the more last decade there has been an increasing focus on wildlife than 6,500 described amphibian species are threatened with diseases in the context of biodiversity conservation extinction making them the most threatened of all verte- (Cleaveland et al., 2002; Lafferty and Gerber, 2002; Cunn- brates (Stuart et al., 2004, 2008). Enzootics of epidermal ingham et al., 2003; Daszak et al., 2004; Rahbek, 2007; chytridiomycosis, caused by the globally emerging McCallum, 2008; Blehert et al., 2009; Smith et al., 2009). amphibian chytrid fungus (Batrachochytrium dendrobatidis, Bd), has been pointed out as a central phenomenon in this ongoing biodiversity crisis. The origin of the pathogen is Published online: March 12, 2010 still unknown, but it is now found in different ecosystems on all continents where amphibians occur (Fisher et al., Correspondence to: Stefan Lo¨tters, e-mail: loetters@uni-trier.de  Rapid Enigmatic Amphibian Decline of Chytrid Fungus 359 2009) and chytridiomycosis has been identified as the key Our current knowledge about this pathogen (Gascon et al., causal mechanism in a few well-documented large-scale 2007) and field observations during periods of sudden declines (Bosch et al., 2001; Rachowicz et al., 2006; Lips decline of amphibians (Lips et al., 2008) support this view. et al., 2006; Schloegel et al., 2006). There is growing con- To test the hypothesis, we ran a global Species Distribution sensus in the scientific community that this disease should Model (SDM) to identify the potential distribution of Bd, be considered a potential driver of decline and extinctions, based on suitability to its specific climate envelope, and although far from all species infected by the pathogen are compared it with the geographic distribution of rapid susceptible to clinical chytridiomycosis (Gascon et al., 2007; enigmatic amphibian decline. Bielby et al., 2008; Stuart et al., 2008). In SDMs, environmental information is extracted from The first exhaustive assessment of the conservation databases at localities where the species under study occurs status of global amphibian diversity (2002–2004 Global to simulate its ecological niche. Commonly, climatic data Amphibian Assessment, GAA) categorized 207 species are used aiming on the species’ climate envelope. Different (Figure 1a) as undergoing rapid enigmatic decline, i.e., algorithms are available for this process. The climate shift to a higher IUCN Red List category between 1980 and envelope is subsequently projected onto maps denoting the 2004 for unknown reason (Stuart et al., 2004, 2008). It has climatic suitability to the target species, hereby inferring its been hypothesized that these declines are all results of the potential distribution (Guisan and Zimmermann, 2000; pandemic amphibian chytrid fungus (Skerratt et al., 2007). Jeschke and Strayer, 2008). Figure 1. a Distribution of worldwide rapid enigmatic amphibian are found in tropical mountains, western Europe, southern China, decline (i.e., sum of geographic ranges of all 207 amphibian species and along east coasts on the southern hemisphere. In some of these, undergoing rapid enigmatic decline). b Worldwide potential Bd is currently unknown and perhaps absent (e.g., the Ethiopian distribution of the amphibian chytrid fungus (based on a MaxEnt Highlands, eastern Madagascar, the southern versant of the SDM and 6 ‘‘bioclimate’’ variables), with higher values indicating Himalaya, the Yunnan Province of China; see Lo¨tters et al., 2008; higher suitability. Of 365 records for this pathogen, the 200 used for Ro¨dder et al., 2009). SDM calculation are indicated by dots. Areas of highest Bd suitability 360 Stefan Lo¨tters et al. For SDM computation, we obtained 365 globally dis- The MaxEnt software (version 3.3.1, available from tributed Bd records (latitude/longitude data from www.cs.princeton.edu/*shapire/maxent) was used for www.spatialepidemiology.net/bd-maps). Records posted SDM calculation and mapping (Phillips et al., 2006). This is on this webpage were compiled from numerous scientific a machine-learning algorithm following the principle of publications (a detailed list of references also is posted). Bd maximum entropy to compute the likelihood of the species records were not randomly distributed globally (Figure 1a), in question occurring at any point in the area based on resulting in a possible sample selection bias and violation of environmental conditions observed at species’ records and SDM assumptions (Dormann et al., 2007; Phillips, 2008). contrasted by random background. Regions accessible to To account for this, we extracted all ‘‘bioclimatic’’ (see the target species ideally define the area from which back- below) values at Bd records and performed a cluster anal- ground data should be obtained (Phillips, 2008). We ac- ysis based on Euclidean distances. The resulting classes counted for this by restricting the background samples to were blunted at a threshold leaving 200 classes, whereby areas from which Bd was detected in the wild. In our SDM, only one record per class was used for further processing. the degree of ‘‘clamping’’ (i.e., negative effects caused by This method reduces the amount of duplicate information nonanalogous climate) was removed from the model pre- in the feature space and thereby the impact of samples diction using the ‘‘fade by clamping’’ option. MaxEnt has clumped in geographic space. Calculations were performed been demonstrated to perform better than comparable with XLSTAT 2008 (Addinsoft; www.xlstat.com). methods (Elith et al., 2006) as, for instance, GARP, which The climate dependence of the Bd host-pathogen sys- was used in an earlier Bd SDM by Ron (2005) for the New tem has been confirmed consistently in all continents where Word. MaxEnt also performed well in recently published it occurs (Woodhams and Alford, 2005; Rachowicz et al., Bd SDMs by Puschendorf et al. (2009) and Ro¨dder et al. 2006; Pounds et al., 2006; Alford et al., 2007; Bosch et al., (2009) with goals different to our paper. We computed a 2007; Kriger et al., 2007; Kriger and Hero, 2007; Andre logistic output, which produces continuous, linear scaled et al., 2008; Laurance, 2008; Lips et al., 2008; Ro¨dder et al., maps and allows for fine distinction of modeled suitability 2008; Kielgast et al., 2010). As environmental predictors, six values for Bd. Model validation was performed by calcu- grid-based bioclimate variables at resolution 2.5 min lating the Area Under the ROC (Receiver Operating (Hijmans et al., 2001) describing physiologically relevant Characteristics) Curve (AUC), a threshold-independent parameters for this pathogen (Piotrowski et al., 2004; index widely used in ecological modeling (Manel et al., Woodhams and Alford, 2005): annual mean temperature, 2001; Elith et al., 2006). For this purpose, we set aside 25% maximum temperature of the warmest month, minimum of all chytrid fungus records (randomly chosen) as test temperature of the coldest month, annual precipitation, points and the remaining ones as training points in our precipitation of wettest month, precipitation of driest SDM. This procedure was repeated 100 times to evaluate month, calculated with DIVA GIS 5.4. (www.diva-gis.org) possible variation. Values received (mean AUC, and based on the Worldclim 1.4. dataset (www. 0.910 ± 0.018; range, 0.861–0.954) suggest that the com- worldclim.org), representing weather conditions recorded puted SDM is robust with performance of high quality between 1950 and 2000 at spatial resolution of about (Elith et al., 2006; Phillips et al., 2006). MaxEnt allows 1 9 1 km2 (Hijmans et al., 2005). These parameters have assessing the relative contribution of each variable to the performed well in previous Bd SDMs (Ron, 2005; Lo¨tters model. Herein, the ‘‘annual mean temperature’’ had the et al., 2008; Puschendorf et al., 2009; Ro¨dder et al., 2009). highest explanatory power (mean, 44.3 ± 4.7%; range, Because we expect that the Bd zoospores are able to survive 28.7–53.5%), followed by the ‘‘maximum temperature of in unfrozen microhabitat, water or on hosts during winter the warmest month’’ (mean, 23.15% ± 4.3%; range, 13.7– air freezing (Piotrowski et al., 2004), the temperature 36.1%). The mean contribution of all other variables was minima of the coldest month grids below freezing tem- <10%. MaxEnt calculates a series of possible threshold perature were pooled and set to 0°C. This approach is values at each run and values greater than it may be reasonable because such buffered microhabitat climate interpreted as reasonable approximation of a species’ po- cannot be addressed using Worldclim data, but by doing so tential distribution. Generally, the mean MaxEnt value at physiological relevant temperature patterns are still the training records is 0.5 (Phillips and Dudı´k, 2008), reflected. which was used as first threshold as recommended by Liu  Rapid Enigmatic Amphibian Decline of Chytrid Fungus 361 et al. (2005). Furthermore, the uppermost 25% of the lo- unconstrained randomizations of the observed proportions gistic value was chosen as second threshold (logistic Max- in the two categories using the statistical language R (Team Ent value of 0.75). The worldwide potential distribution of RDC, 2009). The proportion of enigmatic declines and Bd that we obtained (Figure 1b) coincides with previous species with total range within Bd suitable area were similar approaches (Ron, 2005; Lo¨tters et al., 2008; Ro¨dder shuffled across all known species and the number of et al., 2009). enigmatically declining species falling within Bd suitable Geographic ranges of 207 amphibian species (see areas were counted. The actual number of rapid enigmatic Appendix 1) undergoing rapid enigmatic decline were declining species with a range climatically suitable to Bd downloaded from the GAA (http://www.iucnredlist.org/ was then compared with the null frequency distribution amphibians/download_gis_page; Figure 1a). One of the and the coincidence regarded significantly higher than ex- original 207 species (Stuart et al., 2004, 2008), the Ama- pected by chance if the actual number of species fitting zonian toad Atelopus spumarius, was excluded due to tax- both criteria was higher than the 2.5% right tail of the null onomic confusion (see Appendix 2). Using ArcGIS 9.2, the frequency distribution. geographic ranges (GAA shapefiles) of these amphibians The relationship between rapid enigmatic decline and were overlaid with the computed SDM for the chytrid climatic suitability for Bd was highly significant fungus. Subsequently, the portion of every species’ distri- (P < 0.0001) for both MaxEnt thresholds (Figure 2). bution with high climatic suitability (i.e., MaxEnt values A fraction of the Bd records used for modelling 0.5 and 0.75, respectively; Figure 1b) for Bd was cal- occurred in areas of rapid enigmatic amphibian decline culated (see online supplemental materials). We found that thereby potentially leading to circular conclusion. To test the worldwide distribution of rapid enigmatic amphibian this possibility we created a SDM using 175 Bd presence decline (Figure 1a) entirely falls within the potential dis- points situated outside these areas. The resulting map tribution of Bd. (see Appendix 3) coincides with that shown in Figure 1b The coincidence between rapid enigmatic declines and supporting our conclusion. A statistical comparison areas climatically suitable to Bd was statistically examined among both maps using a modified Hellinger distance by a permutation test. A null frequency distribution for the (I, range 0–1 for no to total coincidence), as proposed number of rapid enigmatic declines falling within Bd by Warren et al. (2008), revealed a high similarity suitable areas (MaxEnt 0.5) was obtained by 100.000 (I = 0.90). Figure 2. Frequency distribution of randomized overlap between from current proportions 100,000 times. The black bar indicates the rapid enigmatic amphibian decline and MaxEnt values  0.5 (left) actual number of rapid enigmatic declining species with total range and 0.75 (right). The bars in the histogram indicate frequencies of falling within the Bd suitable areas (i.e., MaxEnt > 0.5: suitable, respective numbers of species fulfilling both criteria when simulated MaxEnt > 0.75: highly suitable). 362 Stefan Lo¨tters et al. Some issues concerning the data and methods used >0.75 Table 1. List of all amphibian species undergoing rapid enigmatic decline as listed in the GAA with their IUCN Red List status and the geographic range size encompassed by them Portion of species geographic range exhibiting a MaxEnt 0 96 86 100 100 100 100 100 100 herein need to be considered. Our information on the distribution of the world’s amphibians was obtained from >0.5 the GAA. Although these data provide the most com- 100 100 100 100 100 100 100 100 100 prehensive and up-to-date information currently avail- able, the range information of several species is based on Min 10% training expert opinions only. The GAA species distribution record (0.277) value higher than threshold (%) polygons in many cases describe the ‘‘extent of occur- rence’’ and sometimes the ‘‘area of occupancy’’ of each (www.iucnredlist.org), followed for each species by the portion of its geographic range at the same time suitable to the amphibian chytrid fungus species rather than providing an exact summary of 100 100 100 100 100 100 100 100 100 existing populations (Gaston and Fuller, 2008). This may cause an overestimation of the actually suitable areas in record (0.049) Min training widely occurring species, because they may not be hom- ogenously distributed throughout their range. Worldwide, 100 100 100 100 100 100 100 100 100 these potential errors may not be homogenously distrib- uted. However, the GAA data represent the best available Geographic knowledge. We have for the first time provided a macro-ecological (km2) range 147.4 185.0 191.2 80.0 91.5 38.6 68.7 73.3 67.7 test of whether rapid enigmatic amphibian decline is likely to have been caused by chytridiomycosis. Our results A2ace; B1ab(iii,v) + 2ab(iii,v) A2ace; B1ab(iii,v) + 2ab(iii,v) strongly support the hypothesis and impact of the amphibian chytrid fungus on global biodiversity. We con- clude that rapid enigmatic amphibian declines are heavily linked to Bd, thus adding data-driven support to previous A2ace; B2ab(iii,v) Red List criteria A2ace; B2ab(v) A2ace; B2ab(v) A2ace; B2ab(v) convictions (Skerratt et al., 2007; Wake and Vredenburg, (see footnote) A2a; B2ab(v) 2008). However, we also have identified areas showing high climatic suitability for Bd from which rapid enigmatic de- A3ce cline is unknown (Figure 1b). This may be related to knowledge gaps in particular areas (Gascon et al., 2007) (see footnote) and/or to differences between species susceptibility to Bd (Bielby et al., 2008). Knowledge gaps are likely because Red List category long-term amphibian monitoring data are, for the most DD CR CR CR CR CR CR CR part of the world, unavailable. Hence, if rapid enigmatic CR declines have taken place in these under-studied areas, they Nymphargus oreonympha are unlikely to have been noticed. Aromobates leopardalis Atelopus mucubajiensis Atelopus chrysocorallus Atelopus carbonerensis Aromobates nocturnus Atelopus sorianoi Atelopus angelito Atelopus arthuri ACKNOWLEDGMENTS Species The authors thank Ariadne Angulo and Mike Hoffmann of Conservation International for being helpful in making available requested IUCN data. Aromobatidae Aromobatidae Centrolenidae Bufonidae Bufonidae Bufonidae Bufonidae Bufonidae Bufonidae Family APPENDIX 1 Anura Anura Anura Anura Anura Anura Anura Anura Anura Order See Table 1. Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a MaxEnt category range value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Anura Bufonidae Atelopus ebenoides CR A2ace + 3ce 1915.1 100 100 100 57 Anura Bufonidae Atelopus erythropus CR A3ce 281.3 100 100 100 1 Anura Bufonidae Atelopus eusebianus CR A3ce 728.6 100 100 100 77 Anura Bufonidae Atelopus exiguus CR A3ce 761.5 100 100 100 100 Anura Bufonidae Atelopus farci CR A2ace; B2ab(iii,v) 99.9 100 100 100 100 Anura Bufonidae Atelopus guanujo CR A2ace 17.9 100 100 100 100 Anura Bufonidae Atelopus guitarraensis CR A3ce 65.8 100 100 100 100 Anura Bufonidae Atelopus laetissimus CR A3ce 151.3 100 100 100 64 Anura Bufonidae Atelopus lozanoi CR A2ace; B2ab(v) 446.3 100 100 100 100 Anura Bufonidae Atelopus lynchi CR A3ce; B1ab(iii,iv,v) 42.2 100 100 100 100 Anura Bufonidae Atelopus mandingues CR A3ce 288.3 100 100 100 100 Anura Bufonidae Atelopus minutulus CR A3ce 23.5 100 100 100 100 Anura Bufonidae Atelopus monohernandezii CR A2e 102.3 100 100 100 29 Anura Bufonidae Atelopus muisca CR A2ace; B2ab(v) 137.0 100 100 100 100 Anura Bufonidae Atelopus nahumae CR A3ce 292.2 100 100 100 79 Anura Bufonidae Atelopus nanay CR A2ace; B2ab(v) 8.2 100 100 100 82 Anura Bufonidae Atelopus nicefori CR A3ce 91.6 100 100 100 91 Anura Bufonidae Atelopus oxyrhynchus CR A2ace 350.3 100 100 100 100 Anura Bufonidae Atelopus pedimarmoratus CR A3ce 97.2 100 100 100 100 Anura Bufonidae Atelopus petriruizi CR A3ce 311.8 100 100 100 100 Anura Bufonidae Atelopus pictiventris CR A3ce 299.6 100 100 100 54 Anura Bufonidae Atelopus pinangoi CR A2ac; B1ab(iii,v) 48.0 100 100 100 100 Anura Bufonidae Atelopus simulatus CR A2ace 351.1 100 100 100 68 Anura Bufonidae Atelopus sonsonensis CR A3ce 356.6 100 100 100 100 Anura Bufonidae Atelopus subornatus CR A3ce 307.7 100 100 100 100 Anura Bufonidae Atelopus tamaensis CR A3ce 66.9 100 100 100 100 Anura Bufonidae Atelopus walkeri CR A3ce 196.0 100 100 100 100 Anura Centrolenidae Centrolene ballux CR A2ac; B2ab(iii,iv,v) 235.2 100 100 100 86 Anura Craugastoridae Craugastor greggi CR A3e 63.3 100 100 100 45 Rapid Enigmatic Amphibian Decline of Chytrid Fungus Anura Eleutherodactylidae Eleutherodactylus unicolor VU D2 5.0 100 100 100 0 363  364 Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a category range MaxEnt value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Stefan Lo¨tters et al. Anura Amphignathodontidae Gastrotheca splendens EN A3c 41.3 100 100 100 100 Anura Ceratophryidae Telmatobius colanensis EN A3e 151.3 100 100 100 71 Anura Ceratophryidae Telmatobius vellardi CR A2ace;B2ab(i,ii,iii,iv,v) 171.5 100 100 100 100 Anura Bufonidae Atelopus sernai CR A2ace 312.5 100 100 100 94 Anura Myobatrachidae Taudactylus pleione CR B1ab(v) + 2ab(v) 152.2 100 100 100 23 Anura Craugastoridae Craugastor escoces EX 1051.3 100 100 100 83 Anura Hylidae Isthmohyla angustilineata CR A2ae 1533.4 100 100 100 85 Anura Ceratophryidae Telmatobius niger CR A2ace 5961.2 100 100 100 97 Anura Bufonidae Atelopus glyphus CR A3ce 341.7 100 100 100 70 Anura Dendrobatidae Hyloxalus vertebralis CR A2ace 5956.1 100 100 100 99 Anura Ceratophryidae Telmatobius brevipes EN A3e 6333.7 100 100 99 47 Anura Amphignathodontidae Gastrotheca riobambae EN A2ac 7398.3 100 100 100 89 Anura Dendrobatidae Hyloxalus anthracinus CR A2ac 2254.1 100 100 98 92 Anura Bufonidae Atelopus senex CR A2ace 1181.9 100 100 99 73 Anura Bufonidae Atelopus nepiozomus CR A3ce 2271.8 100 100 97 69 Anura Bufonidae Atelopus quimbaya CR A3ce 254.4 100 100 97 77 Anura Ranidae Lithobates vibicarius CR A2ace 2773.5 100 100 100 96 Anura Amphignathodontidae Gastrotheca pseustes EN A2ace 12359.1 100 100 100 94 Anura Cycloramphidae Cycloramphus ohausi DD 1026.4 100 100 100 66 Anura Bufonidae Atelopus coynei CR A2ace 2955.3 100 100 100 92 Anura Bufonidae Atelopus bomolochos CR A2ace 11373.4 100 100 99 93 Anura Bufonidae Atelopus chiriquiensis CR A2ace 4600.3 100 100 100 98 Anura Centrolenidae Centrolene lynchi EN B2ab(iii,iv,v) 3388.8 100 100 97 82 Anura Bufonidae Atelopus galactogaster CR A3ce 65.2 100 100 90 0 Anura Dendrobatidae Hyloxalus pulchellus VU A2ace; B1ab(i,ii,iii, iv,v) 17380.8 100 100 96 44 Anura Craugastoridae Craugastor emcelae CR A3ce 1799.7 100 99 89 56 Anura Eleutherodactylidae Eleutherodactylus wightmanae EN A4ae; B1ab(v) 197.1 100 99 86 0 Anura Hylidae Scinax heyeri DD 1320.3 100 98 95 0 Anura Centrolenidae Centrolene heloderma CR A2ac 5143.5 100 99 88 31 Anura Bufonidae Atelopus peruensis CR A2ace 23124.3 100 98 91 48 Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a category range MaxEnt value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Anura Cycloramphidae Thoropa saxatilis NT 6374.4 100 100 100 0 Anura Hylidae Agalychnis annae EN A2abe 4044.1 100 98 86 42 Anura Bufonidae Leptophryne cruentata CR A2ac 712.0 100 96 77 30 Anura Eleutherodactylidae Eleutherodactylus orcutti CR A2ace 313.6 100 92 74 0 Anura Bufonidae Atelopus planispina CR A2ace 3242.9 100 98 76 3 Anura Dendrobatidae Hyloxalus elachyhistus EN A2ac; B2ab(iii,iv,v) 11103.3 100 97 77 51 Anura Hylidae Bromeliohyla bromeliacia EN A2ace 7264.6 100 94 78 33 Anura Bufonidae Atelopus pachydermus CR A2ace 9000.0 100 90 61 24 Anura Bufonidae Atelopus mindoensis CR A2e 3966.8 100 76 73 48 Anura Bufonidae Atelopus cruciger CR A2ace 7179.2 100 89 79 24 Anura Bufonidae Atelopus reticulatus CR A3ce 69.2 100 100 59 0 Anura Bufonidae Atelopus carrikeri CR A3ce 647.1 100 95 59 30 Anura Cycloramphidae Thoropa petropolitana VU B1ab(iii,v) + 2ab(ii,iii,iv,v) 19140.5 100 100 92 29 Anura Craugastoridae Craugastor fecundus CR A2ace 152.6 100 55 52 10 Anura Dendrobatidae Hyloxalus lehmanni NT 77330.3 100 95 82 48 Anura Hylidae Duellmanohyla uranochroa CR A2ace 14928.4 100 90 72 46 Anura Hylidae Bokermannohyla claresignata DD 9909.7 100 98 79 19 Anura Hylidae Plectrohyla dasypus CR A2ace;B1ab(iii,v) + 2ab(iii,v) 280.3 100 86 49 10 Anura Myobatrachidae Taudactylus eungellensis CR B2ab(v) 384.4 100 100 50 8 Anura Hylidae Plectrohyla thorectes CR A2ace 369.7 100 74 47 0 Anura Bufonidae Atelopus tricolor VU A3ce 42472.2 100 91 75 15 Anura Ceratophryidae Telmatobius marmoratus VU A3cde 193559.1 100 99 64 5 Anura Eleutherodactylidae Eleutherodactylus jasperi CR A2ae;B2ab(i,ii,iv,v) 158.3 100 100 46 0 Anura Bufonidae Atelopus varius CR A2ace 21998.4 100 94 70 38 Anura Bufonidae Atelopus famelicus CR A2ace 231.5 100 94 44 0 Anura Hylidae Charadrahyla altipotens CR A2ace 11.8 100 100 43 0 Anura Eleutherodactylidae Eleutherodactylus gryllus EN B1ab(v) 1416.8 100 77 51 0 Anura Bufonidae Atelopus balios CR A2ace 380.8 100 65 41 8 Anura Cycloramphidae Thoropa lutzi EN B1ab(iii,v) + 2ab(iii,v) 20655.2 100 98 78 16 Rapid Enigmatic Amphibian Decline of Chytrid Fungus Anura Hylidae Agalychnis moreletii CR A3e 38139.0 100 83 64 23 365  366 Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a category range MaxEnt value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Stefan Lo¨tters et al. Anura Eleutherodactylidae Eleutherodactylus eneidae CR A2ae 1947.4 100 86 48 0 Anura Myobatrachidae Taudactylus rheophilus CR A2ac; B2ab(v) 5140.6 100 86 58 0 Anura Bufonidae Atelopus limosus EN B1ab(iii) 453.3 100 100 37 5 Anura Hylidae Litoria nyakalensis CR D 12745.5 100 94 67 0 Anura Eleutherodactylidae Eleutherodactylus portoricensis EN A4ae 2335.7 100 82 43 0 Anura Myobatrachidae Mixophyes fleayi EN B2ab(ii,iii,iv,v) 21384.0 100 100 100 35 Anura Hylidae Plectrohyla guatemalensis CR A3e 58900.4 100 79 53 11 Anura Hylidae Hylomantis lemur CR A4ace 16873.9 99 75 51 17 Anura Hylidae Litoria rheocola EN A2ae 16602.7 100 88 55 0 Anura Hylidae Litoria pearsoniana NT 46975.0 100 100 100 30 Anura Bufonidae Atelopus longibrachius EN A3ce; B1ab(iii) 1018.4 72 44 24 0 Anura Eleutherodactylidae Eleutherodactylus richmondi CR A3ce 321.3 100 75 25 0 Anura Myobatrachidae Taudactylus acutirostris CR A2ace; B2ab(i,ii,iii,iv,v); C2a(i); D 15975.6 100 87 54 0 Anura Hylidae Litoria booroolongensis CR B2ab(i,ii,iii,iv,v) 189373.9 100 100 83 10 Anura Bufonidae Atelopus boulengeri CR A3ce 2389.0 100 78 23 0 Anura Bufonidae Atelopus halihelos CR A2ace; B1ab(iii) + 2ab(iii) 21.9 100 100 22 0 Anura Hylidae Litoria nannotis EN A2ae 20841.3 100 80 49 0 Anura Hylidae Litoria dayi EN A2ac 20670.5 100 80 47 0 Anura Hylidae Litoria aurea VU A2ace 263375.4 100 100 95 34 Anura Eleutherodactylidae Eleutherodactylus locustus CR A4ae 560.8 100 55 20 0 Anura Hylidae Litoria spenceri CR B2ab(ii,iii,iv,v) 25923.5 100 100 26 0 Anura Bufonidae Atelopus elegans CR A2ace 9228.5 100 60 37 4 Anura Cycloramphidae Rhinoderma darwinii VU A2ace 166596.5 100 100 61 6 Anura Eleutherodactylidae Eleutherodactylus hedricki EN B1ab(v) 1724.0 100 58 19 0 Anura Hylidae Litoria lorica CR D 1280.0 100 44 14 0 Anura Bufonidae Atelopus zeteki CR A2ace 1625.0 100 44 10 0 Anura Hylidae Litoria raniformis EN A2ae 1159458.8 100 96 63 18 Anura Bufonidae Atelopus andinus CR A3ce 2453.1 99 40 7 0 Anura Limnodynastidae Adelotus brevis NT 612412.2 100 91 56 13 Anura Bufonidae Atelopus chocoensis CR A3ce 345.5 100 69 6 0 Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a category range MaxEnt value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Anura Myobatrachidae Pseudophryne pengilleyi EN B1ab(ii,iv,v) + 2ab(ii,iv,v) 1668.4 100 100 5 0 Anura Hylidae Osteopilus pulchrilineatus EN B2ab(iii,v) 35131.2 92 26 15 5 Anura Ranidae Lithobates warszewitschii LC 162917.0 99 57 24 6 Anura Eleutherodactylidae Eleutherodactylus coqui LC 16045.2 100 26 10 0 Anura Aromobatidae Allobates olfersioides VU A2a; B2ab(iii) 373528.2 100 57 23 2 Anura Bufonidae Atelopus seminiferus CR A3ce 2285.7 100 69 3 0 Anura Bufonidae Atelopus pulcher CR A2ace 9373.0 63 27 4 0 Anura Ranidae Rana muscosa EN B2ab(ii,iv,v); C2a(i) 24198.3 97 65 3 0 Anura Bufonidae Atelopus spurrelli VU A3ce 24420.4 66 24 10 0 Anura Ranidae Lithobates tarahumarae VU A3e 125961.6 100 36 1 0 Anura Myobatrachidae Pseudophryne corroboree CR A2ace + 3ce; B2ab(ii,iii,iv,v); C1 1653.3 100 98 0 0 Anura Dicroglossidae Nanorana pleskei NT 197637.9 100 19 0 0 Anura Limnodynastidae Philoria frosti CR A2ace; B1ab(ii,iii,iv,v) + 2ab(ii,iii,iv,v); C2b 165.1 100 100 0 0 Anura Bufonidae Atelopus spumarius Excluded from analysis here, see Appendix 2 Anura Bufonidae Anaxyrus baxteri EW 185.9 100 0 0 0 Anura Bufonidae Anaxyrus boreas NT 6487261.0 82 10 2 0 Anura Bufonidae Anaxyrus canorus EN A2ae 16448.7 100 45 0 0 Anura Craugastoridae Craugastor saltuarius CR A2ace 314.4 100 89 88 22 Anura Hylidae Hyloscirtus colymba CR A4ace 16792.2 100 73 50 15 Anura Leiopelmatidae Leiopelma archeyi CR A2ae 1193.6 100 100 100 0 Anura Aromobatidae Mannophryne olmonae CR A2ae; B1ab(v) 104.0 100 16 0 0 Anura Bufonidae Nectophrynoides asperginis CR B1ab(ii,iii,v) + 2ab(ii,iii,v) 0.1 100 100 0 0 Caudata Plethodontidae Aneides aeneus NT 270921.7 100 38 0 0 Anura Bufonidae Atelopus arsyecue CR A3ce 75.8 100 100 87 24 Anura Bufonidae Atelopus carauta CR A3ce 142.5 100 99 0 0 Anura Bufonidae Atelopus certus EN A3ce 107.1 100 86 0 0 Anura Bufonidae Atelopus flavescens VU A3ce 488.8 100 75 0 0 Anura Bufonidae Atelopus franciscus VU A3ce 28755.6 96 35 0 0 Anura Bufonidae Atelopus ignescens EX 10189.7 100 100 99 78 Rapid Enigmatic Amphibian Decline of Chytrid Fungus 367  368 Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a category range MaxEnt value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Stefan Lo¨tters et al. Anura Bufonidae Atelopus longirostris EX 3620.5 100 80 77 59 Caudata Hynobiidae Batrachuperus pinchonii VU A2ad 183039.4 100 87 21 0 Caudata Plethodontidae Bolitoglossa pesrubra VU A2ace 799.1 100 100 100 98 Caudata Plethodontidae Bolitoglossa subpalmata EN B1ab(v) 1702.3 100 100 99 88 Anura Hylidae Bromeliohyla dendroscarta CR A2ace 2043.2 100 69 54 11 Caudata Plethodontidae Chiropterotriton cracens EN B1ab(v) 49.3 100 87 19 0 Anura Craugastoridae Craugastor angelicus CR A2ace 972.5 100 93 73 39 Anura Craugastoridae Craugastor berkenbuschii NT 62930.4 100 63 52 22 Anura Craugastoridae Craugastor catalinae CR A2ace 1306.7 100 100 100 100 Anura Craugastoridae Craugastor chrysozetetes EX 56.5 100 92 92 17 Anura Craugastoridae Craugastor emleni CR A2ace; B2ab(v) 844.4 100 97 83 27 Anura Craugastoridae Craugastor epochthidius CR A3ce 487.9 98 33 16 5 Anura Craugastoridae Craugastor fleischmanni CR A2ace 3362.9 100 100 97 73 Anura Craugastoridae Craugastor guerreroensis CR A2ace 7.3 100 0 0 0 Anura Craugastoridae Craugastor laevissimus EN A2ace 34170.1 99 65 41 3 Anura Craugastoridae Craugastor lineatus CR A3e 17200.0 100 89 71 23 Anura Craugastoridae Craugastor merendonensis CR A2ace; B1ab(v) + 2ab(v) 46.3 100 96 57 0 Anura Craugastoridae Craugastor milesi EX 445.6 100 70 38 7 Anura Craugastoridae Craugastor olanchano CR A2ace 636.0 100 68 3 0 Anura Craugastoridae Craugastor polymniae CR A2ace 331.2 100 100 59 9 Anura Craugastoridae Craugastor ranoides CR A2ace 41709.7 98 56 33 9 Anura Craugastoridae Craugastor sabrinus EN A2ace 7621.7 100 90 34 0 Anura Craugastoridae Craugastor sandersoni EN A2ace 6137.9 100 88 38 1 Anura Craugastoridae Craugastor stadelmani CR A2ace 846.6 100 92 80 13 Anura Hylidae Ecnomiohyla echinata CR A2ace 35.3 100 100 98 29 Anura Eleutherodactylidae Eleutherodactylus schmidti CR A2ace 13489.2 95 39 27 15 Anura Eleutherodactylidae Eleutherodactylus symingtoni CR A2ace; B2ab(iii,v); C1 276.2 100 17 0 0 Anura Hylidae Hyla bocourti CR A2ace 762.4 100 100 96 60 Anura Dendrobatidae Hyloxalus delatorreae CR B2ab(iii,iv,v) C2a(ii) 183.1 100 100 100 100 Anura Bufonidae Incilius fastidiosus CR A2ace 328.5 100 100 100 100 Table 1. continued Order Family Species Red List Red List criteria Geographic Portion of species geographic range exhibiting a MaxEnt category range value higher than threshold (%) (see footnote) (see footnote) (km2) Min training Min 10% training >0.5 >0.75 record (0.049) record (0.277) Anura Bufonidae Incilius holdridgei EX 235.5 100 100 100 100 Anura Bufonidae Incilius periglenes EX 147.8 100 100 100 82 Anura Bufonidae Incilius peripatetes CR A3ce 131.1 100 100 100 72 Anura Hylidae Isthmohyla calypsa CR A2ace 326.8 100 100 100 100 Anura Hylidae Isthmohyla debilis CR A2ace 2032.1 100 98 72 41 Anura Hylidae Isthmohyla graceae CR A2ace 1917.7 100 100 90 52 Anura Hylidae Isthmohyla rivularis CR A2ace 7137.4 100 100 94 70 Anura Hylidae Isthmohyla tica CR A2ace 6143.1 100 100 93 73 Anura Ranidae Lithobates omiltemanus CR B2ab(iii,v) 1189.3 86 33 25 1 Anura Hylidae Megastomatohyla pellita CR A2ace 1177.8 100 80 53 23 Anura Hylidae Plectrohyla hartwegi CR A3e 5891.1 100 90 78 44 Anura Hylidae Plectrohyla hazelae CR A2ace 545.5 100 77 75 28 Anura Strabomantidae Pristimantis urichi EN A2ae 3442.7 100 3 0 0 Caudata Plethodontidae Pseudoeurycea bellii VU A2ace 197983.0 100 88 70 10 Caudata Plethodontidae Pseudoeurycea smithi CR A2ace 265.0 100 100 100 56 Anura Myobatrachidae Rheobatrachus silus EX 1741.3 100 100 100 22 Anura Myobatrachidae Rheobatrachus vitellinus EX 149.7 100 100 67 14 Anura Myobatrachidae Taudactylus diurnus EX 4798.3 100 100 100 21 Caudata Plethodontidae Thorius narisovalis CR A2ab + 4ab; B1ab(ii,v) 84.2 100 100 100 41 Caudata Plethodontidae Thorius pennatulus CR A2ac + 4ac 811.2 100 100 100 71 Abbreviations used for IUCN Red List categories (for definitions and criteria see www.iucn.redlist.org). EX extinct; VU vulnerable; EW extinct in the wild; LC least concern; CR critically endangered; DD data deficient; EN endangered. Rapid Enigmatic Amphibian Decline of Chytrid Fungus 369 370 Stefan Lo¨tters et al. APPENDIX 2 gion have not (Lo¨tters et al., 2005; Rueda-Almonacid et al., 2005; Luger et al., 2008). Because there is no other Exclusion of one species due to insufficient taxonomic amphibian species from this remarkably large area, status. Under the GAA, one Atelopus population from the encompassed by A. spumarius (excluding the undescribed Amazon versant of the Andes in Ecuador was allocated to species), which has suffered from rapid decline at all, A. spumarius due to morphological resemblance. However, inclusion of the GAA shapefile for A. spumarius would have this Andean population represents an undescribed species misled our study. and is not conspecific with any of the populations from the Amazon basin and the adjacent Guyana Shield which ten- tatively have been referred to as A. spumarius (Lo¨tters et al., 2002; Rueda-Almonacid et al., 2005). Although the unde- APPENDIX 3 scribed species has experienced rapid enigmatic decline, those from the Amazon lowlands or adjacent Guyana re- See Figure 3. Figure 3. Worldwide potential distribution of the amphibian records for this pathogen used for SDM calculation, all located chytrid fungus (based on a MaxEnt SDM and six ‘‘bioclimate’’ outside areas of rapid enigmatic amphibian decline, are indicated by variables), with higher values indicating higher suitability. The 175 dots.  Rapid Enigmatic Amphibian Decline of Chytrid Fungus 371 REFERENCES Jeschke JM, Strayer DL (2008) Usefulness of bioclimatic models for studying climate change and invasive species. Annals of the New York Academy if Sciences 1134:1–24 Alford RA, Bradfield KS, Richards SJ (2007) Ecology: global warming and amphibian losses. Nature 447:E3–E4 Kielgast J, Ro¨dder D, Veith M, Lo¨tters S (2010) Widespread occurrence of the amphibian chytrid fungus in Kenya. Animal Anderson RM, May RM (1979) Population biology of infectious Conservation 13:1–8 diseases: part I. Nature 280:361–367 Kriger KM, Hero JM (2007) Large-scale seasonal variation in the Andre SE, Parker J, Briggs CJ (2008) Effect of temperature on host prevalence and severity of chytridiomycosis. Journal of Zoology response to Batrachochytrium dendrobatidis infection in the 271:352–359 mountain yellow-legged frog (Rana muscosa). Journal of Wildlife Diseases 44:716–720 Kriger KM, Pereoglou F, Hero J-M (2007) Latitudinal variation in the prevalence and intensity of chytrid (Batrachochytrium Bielby J, Cooper N, Cunningham AA, Garner TWJ, Purvis A dendrobatidis) infection in Eastern Australia. Conservation (2008) Predicting susceptibility to future declines in the world’s Biology 21:1280–1290 frogs. Conservation Letters 1:82–90 Lafferty KD, Gerber LR (2002) Good medicine for conservation Blehert DS, Hicks AC, Behr M, Meteyer CU, Berlowski-Zier BM, biology: the intersection of epidemiology and conservation Buckles EL, et al. (2009) Bat white-nose syndrome: an emerging theory. Conservation Biology 16:593–604 fungal pathogen? Science 323:227 Laurance WF (2008) Global warming and amphibian extinctions Bosch J, Martı´nez-Solano I, Garcı´a-Parı´s M (2001) Evidence of a in eastern Australia. Australian Ecology 33:1–9 chytrid fungus infection involved in the decline of the common midwife toad (Alytes obstetricans) in protected areas of central Lips KR, Brem F, Brenes R, Reeve JD, Alford RA, Voyles J, Carey Spain. Biological Conservation 97:331–337 C, Livo L, Pessier AP, Collins JP (2006) Emerging infectious disease and the loss of biodiversity in a Neotropical amphibian Bosch J, Carrascal LM, Duran L, Walker S, Fisher MC (2007) community. Proceedings of the National Academy of Science of Climate change and outbreaks of amphibian chytridiomycosis the USA 102:3165–3170 in a montane area of Central Spain: is there a link? Proceedings of the Royal Society B-Biological Sciences 274:253–260 Lips KR, Diffendorfer J, Mendelson JR III, Sears MW (2008) Riding the wave: reconciling the roles of disease and climate Cleaveland S, Hess GR, Dobson AP, Laurenson MK, McCallum change in amphibian declines. PLoS Biology 6:441–454 HI, Roberst MG, Woodroffe R (2002) The role of pathogens in biological conservation. In: The Ecology of Wildlife Diseases, Liu C, Berry PM, Dawson TP, Pearson RG (2005) Selecting Hudson PJ, Grenfell BT, Heesterbeek H, Dobson AP (editors), thresholds of occurrence in the prediction of species distribu- Oxford University Press: Oxford UK, pp 139–150 tions. Ecography 28:385–393 Cunningham AA, Daszak P, Rodriguez JP (2003) Pathogen pol- Lotters S, Haas W, Schick S, Bo¨hme W (2002) On the systematics ¨ lution: defining a parasitological threat to biodiversity conser- of the harlequin frogs (Amphibia: Bufonidae: Atelopus) from vation. Journal of Parasitology 89(Suppl):S78–S83 Amazonia. II: Redescription of Atelopus pulcher (Boulenger, 1882) from the eastern Andean versant in Peru. Salamandra Daszak P, Tabor GM, Kilpatrick AM, Epstein J, Plowright R 38:165–184 (2004) Conservation medicine and a new agenda for emerging diseases. Annals of the New York Academy of Sciences 1026:1–11 Lo¨tters S, Ro¨dder D, Bielby J, Bosch J, Garner TWJ, Kielgast J, et al. (2008) Meeting the challenge of conserving Madagascar’s Dormann CF, McPherson J, Arau´jo MB, Bivand R, Bollinger J, megadiverse amphibians: addition of a risk-assessment for the Carl G, et al. (2007) Methods to account for spatial autocor- chytrid fungus. PLoS Biology 6 relation in the analysis of species distributional data: a review. Ecography 30:609–628 Lo¨tters S, Schulte R, Co´rdova JH, Veith M (2005) Conservation priorities for harlequin frogs (Atelopus spp.) of Peru. Oryx Elith J, Graham CH, Anderson RP, Dudik M, Ferrier S, Guisan A, 39:343–346 et al. (2006) Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29:129–151 Luger M, Garner TJW, Ernst R, Ho¨dl W, Lo¨tters S (2008) No evidence for precipitous declines in harlequin frogs (Atelopus) in Fisher MC, Garner TWJ, Walker SF (2009) Global emergence of the Guyanas. Studies on Neotropical Fauna & Environment Batrachochytrium dendrobatidis and amphibian chytridiomyco- 43:177–180 sis in space, time and host. Annual Review of Microbiology 63:291–310 Manel S, Williams HC, Ormerod SJ (2001) Evaluating presence- absence models in ecology: the need to account for prevalence. Gascon C, Collins JP, Moore RD Church DR, McKay JE, Men- Journal of Applied Ecology 38:921–931 delson III JR (2007) Amphibian Conservation Action Plan. Gland, Switzerland, Cambridge, UK: IUCN, Conservation McCallum H (2008) Tasmanian devil facial tumour disease: les- International sons for conservation biology. Trends in Ecology & Evolution 23:631–637 Gaston KJ, Fuller RA (2008) The sizes of species’ geographic ranges. Journal of Applied Ecology 46:1–9 McCallum H, Dobson A (1995) Detecting disease and parasite threats to endangered species and ecosystems. Trends in Ecology Guisan A, Zimmermann NE (2000) Predictive habitat distribution & Evolution 10:190–194 models in ecology. Ecological Modeling 135:147–186 Phillips SJ, et al. (2007) Transferability, sample selection bias and Hijmans RJ, Cruz JM, Rojas E, Guarino L (2001) DIVA-GIS, background data in presence-only modelling: a response to Version 1.4, Data. Manual, Lima, Peru: International Potato Peterson et al. (2007). Ecography 31:272–278 Center and International Plant Genetic Resources Institute Phillips SJ, Dudı´k M (2008) Modeling of species distributions Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very with Maxent: new extensions and comprehensive evaluation. high resolution interpolated climate surfaces for global land Ecography 31:161–175 areas. International Journal of Climatology 25:1965–1978 372 Stefan Lo¨tters et al. Phillips SJ, Anderson RP, Schapire RE (2006) Maximum entropy Schloegel LM, Hero JM, Berger L, Speare R, McDonald K, Daszak modeling of species geographic distributions. Ecological Mod- P (2006) The decline of the sharp-snouted day frog (Taudactylus eling 190:231–259 acutirostris): The first documented case of extinction by infec- Piotrowski JS, Annis SL, Longcore JE (2004) Physiology of Ba- tion in a free-ranging wildlife species? EcoHealth 3:35–40 trachochytrium dendrobatidis, a chytrid pathogen of amphibians. Scott ME (1988) The impact of infection and disease on animal Mycologia 96:9–15 populations: implications for conservation biology. Conserva- Pounds JA, Bustamante MR, Coloma LA, Consuegra JA, Fogden tion Biology 2:40–56 MP, Foster PN, et al. (2006) Widespread amphibian extinctions Skerratt LF, Berger L, Speare R, Cashins S, McDonald KR, Phillott from epidemic disease driven by global warming. Nature AD, et al. (2007) Spread of chytridiomycosis has caused the 439:161–167 rapid global decline and extinction of frogs. EcoHealth 4:125– Puschendorf R, Carnaval AC, VanDerWal J, Zumbado-Ulate H, 134 Chaves G, Bolan˜os F, et al. (2009) Distribution models for the Smith KF, Acevedo-Whitehouse K, Pedersen AB (2009) The role amphibian chytrid Batrachochytrium dendrobatidis in Costa of infectious diseases in biological conservation. Animal Con- Rica: proposing climatic refuges as a conservation tool. Diversity servation 12:1–12 and Distributions 15:401–408 Stuart SN, Chanson JS, Cox NA, Young BE, Rodrigues ASL, Fis- Rachowicz LJ, Knapp RA, Morgan JAT, Stice MJ, Vredenburg VT, chman DL, Waller RW (2004) Status and trends of amphibian Parker JM, Briggs CJ (2006) Emerging infectious disease as a declines and extinctions worldwide. Science 306:1783–1786 proximate cause of amphibian mass mortality. Ecology 87:1671– Stuart SN, Hoffmann M, Chanson JS, Cox NA, Berridge RJ, Ra- 1683 mani P, et al. (2008) Threatened Amphibians of the World, Rahbek C (2007) Disease ecology: the silence of the robins. Nature Barcelona, Spain: Lynx Ed 447:652–653 Team RDC (2009) R: A Language and Environment for Statistical Ro¨dder D, Kielgast J, Bielby J, Schmidtlein S, Bosch J, Garner Computing. Vienna, Austria: R Foundation for Statistical TWJ, et al. (2009) Global amphibian extinction risk assessment Computing for the panzootic chytrid fungus. Diversity 1:52–66 Wake DB, Vredenburg VT (2008) Are we in the midst of the sixth Ro¨dder D, Veith M, Lo¨tters S (2008) Environmental gradients mass extinction? A view from the world of amphibians Pro- explaining prevalence and intensity of infection with the ceedings of the National Academy of Sciences USA 105:11466– amphibian chytrid fungus: the host’s perspective. Animal Con- 11473 servation 11:513–517 Warren DL, Glor RE, Turelli M (2008) Environmental niche Ron SR (2005) Predicting the distribution of the amphibian equivalency versus conservatism: quantitative approaches to pathogen Batrachochytrium dendrobatidis in the new world. niche evolution. Evolution 62:2868–2883 Biotropica 37:209–221 Woodhams DC, Alford RA (2005) Ecology of chytridiomycosis in Rueda-Almonacid JV, Rodrı´guez-Mahecha JV, La Marca E, Lo¨t- rainforest stream frog assemblages of tropical Queensland. ters S, Kahn T, Angulo A (2005) Ranas arlequines. Serie Libretas Conservation Biology 19:1449–1459 de Campo 5, Bogota´ DE, Colombia: Conservacio´n Internacional Colombia 

Vol. 85: 7–14, 2009 DISEASES OF AQUATIC ORGANISMS Published May 27 doi: 10.3354/dao02060 Dis Aquat Org Ranavirus in wild edible frogs Pelophylax kl. esculentus in Denmark Ellen Ariel1, 7,*, Jos Kielgast2, Hans Erik Svart3, Knud Larsen4, Hannele Tapiovaara5, Britt Bang Jensen1, 6, Riikka Holopainen5 1 National Veterinary Institute, Technical University of Denmark, Hangøvej 2, 8200 Århus N, Denmark 2 Department of Biology, Section for Microbiology and Evolution, University of Copenhagen, Universitetsparken 15, 2100 Copenhagen, Denmark 3 Danish Forest and Nature Agency, Haraldsgade 53, 2100 Copenhagen Ø, Denmark 4 Council of Slagelse, Dahlsvej 3, Korsør 4220, Denmark 5 Department of Veterinary Virology, Finnish Food Safety Authority Evira, Mustialankatu 3, Helsinki, Finland 6 Department of Large Animal Sciences, Faculty of Life Sciences, University of Copenhagen, Grønnegårdsvej 8, 1870 Frederiksberg C, Denmark 7 Present address: School of Veterinary and Biomedical Sciences, James Cook University, Douglas Campus, Townsville, Queensland 4811, Australia ABSTRACT: A survey for the amphibian pathogens ranavirus and Batrachochytrium dendrobatidis (Bd) was conducted in Denmark during August and September 2008. The public was encouraged via the media to register unusual mortalities in a web-based survey. All members of the public that reg- istered cases were interviewed by phone and 10 cases were examined on suspicion of disease- induced mortality. All samples were negative for Bd. Ranavirus was isolated from 2 samples of recently dead frogs collected during a mass mortality event in an artificial pond near Slagelse, Den- mark. The identity of the virus was confirmed by immunofluorescent antibody test. Sequencing of the major capsid protein gene showed the isolate had more than 97.3% nucleotide homology to 6 other ranaviruses. KEY WORDS: Ranavirus · Batrachochytrium dendrobatidis · Rana kl. esculenta · Pelophylax kl. esculentus · Amphibian declines · Survey · Frogs · Amphibians Resale or republication not permitted without written consent of the publisher INTRODUCTION viruses are part of the cause of amphibian declines across the world (Cunningham et al. 1996, Chinchar The recent listing of Batrachochytrium dendrobatidis 2002, Green et al. 2002, Docherty et al. 2003, Pearman (Bd) and ranavirus infection in amphibians by the et al. 2004, Greer et al. 2005, Jancovich et al. 2005). World Organisation for Animal Health (OIE) reflects a The type species of the ranavirus genus, frog virus 3 global concern for the health of farmed and declines of (FV3), was isolated for the first time in 1965 (Granoff et wild populations of amphibians (OIE 2008). Reports on al. 1965, Rafferty 1965). In Europe, ranaviruses were ranavirus disease in amphibians are listed in the litera- isolated from moribund edible frogs Pelophylax kl. ture (Speare & Smith 1992, Kanchanakhan et al. 2002 esculentus (formerly Rana kl. esculenta) in Croatia Zupanovic et al. 1998, Zhang et al. 2001, Green et al. (Fijan et al. 1991) and Italy (G. Bovo pers. comm.) and 2002, Weng et al. 2002, Greer et al. 2005, Fox et al. from wild populations of the common frog Rana tempo- 2006) and several publications indicate that rana- raria in the UK (Drury et al. 1995). The origin of these *Email: ellen.ariel@jcu.edu.au © Inter-Research 2009 · www.int-res.com 8 Dis Aquat Org 85: 7–14, 2009 viruses is unknown, as is the influence of ranaviral dis- (EPC) cells (Fijan et al. 1983) at 21°C. Titrations were ease on the wild populations of amphibians in Europe. carried out as 6 well replicates in 96 well plates with As part of the project Risk Assessment of New and 10-fold dilutions of the viral inoculum on EPC cells at Emerging Systemic Iridoviral Diseases for European 21°C. The progress of cytopathic effect (CPE) was Fish and Aquatic Ecosystems (RANA), a survey was observed by light microscope and recorded at regular conducted in several European countries to gain intervals with the final reading 7 d after inoculation. knowledge of the current presence of ranaviruses in The titre was calculated as TCID50 ml–1 according to wild amphibian populations. In Denmark, the public the Reed & Muench (1938) method for end-point was engaged in the survey to broaden the chances of titrations. identifying and registering unusual amphibian mortal- PCR for Bd. The diagnostic swabs were stored at ities in the wild populations. The results of the Danish –20°C until processed. DNA was extracted and survey are reported here. analysed via realtime PCR following Boyle et al. (2004) with minor modifications. The tip of each swap was cut off into a screw top Eppendorf tube with 60 µl of Prep- MATERIALS AND METHODS Man Ultra (Applied Biosystems) and 30 to 40 mg of zir- conium/silica beads (0.5 mm diameter, Biospec Prod- Survey of amphibian mortality. The purpose of the ucts) added. The samples were homogenised for 45 s survey was to identify amphibians that had died due and subsequently centrifuged (30 s at 13 000 × g in a to an infectious disease and the questionnaire was microfuge) to pellet the material. Homogenisation and therefore designed to filter out other causes. The centrifugation was repeated. The homogenised sample questionnaire covered the contact details of the was placed in a heating block at 100°C for 10 min, reporting person, the location and date of the cooled for 2 min and then centrifuged at 13 000 × g for observed amphibian(s), the perceived cause of death, 3 min in a microfuge. The supernatant was recovered type of amphibian, number of dead and/or affected and a subsample of this DNA extract diluted 10–1 in animals and details of previous similar incidences. water for further analysis. The questionnaire could be accessed via the RANA DNA samples were analysed by real-time PCR using project’s website and was active during August and primers ITS-1 Chytr and 5.8S Chytr with a minor September 2008. A statement was released to the groove binder probe ChytrMGB2 (GenBank accession media in early August 2008 to draw public attention no. AY598034). The specificity of the assay has been to the survey. If a case of mortality was registered as confirmed against 26 other Chytridiomycetes (Boyle et potentially induced by infectious disease, the respon- al. 2004). All samples and amplification standards of der to the survey was instructed to freeze the dead 0.1, 1, 10 and 100 genome equivalents were run in amphibian until contacted by a scientist. The scientist duplicate together with a positive extraction, a nega- conducted a secondary screening via a telephone tive extraction and a no template control (distilled interview to determine which cases should be submit- water). Twenty µl of TaqMan mastermix/primers/ ted to the National Veterinary Institute of Denmark probe/distilled water and 5 µl of sample, standard or for further investigation under permit no. SNS-441- control was pipetted into each well. The plate was then 00104. Follow-up investigation on the positive cases centrifuged at 2750 × g for 3 min and samples were was carried out as collaboration between the National screened using default amplification settings on an Veterinary Institute, the Council of Slagelse and the ABI prism 7700. Samples were identified as positive for Danish Forest and Nature Agency. Extensive testing Bd if a clear log-linear amplification was observed for for environmental pollutants in the water was carried both replicates and genomic equivalents quantified out (Højvang Environmental Laboratory, 4293 Diana- according to the standards. lund, Denmark) and weather patterns were consulted Immunofluorescent antibody test. Supernatant from for the period of the outbreak in the archives of the samples inducing CPE in the cell cultures were inocu- Danish Meteorological Institute (www.dmi.dk). lated onto EPC cells in 96 well plates and incubated for Laboratory examination. Thawed amphibian speci- 24 h at 24°C, using negative cell culture supernatant as mens were examined for gross lesions. Prior to dissec- negative control and epizootic hematopoietic necrosis tion, diagnostic skin swabs for Bd were taken using virus (EHNV; kindly supplied by R. Whittington, Uni- fine tip dry-swabs (MW100, Medical Wire and Equip- versity of Sydney, Australia) as positive control. They ment) according to the standard protocol by Hyatt et were then rinsed and fixed in 80% acetone prior to al. (2007). Spleen, liver, lung, kidney and heart tissues staining for immunofluorescence as described by Jør- were pooled for each animal and examined according gensen et al. (1989). Briefly, primary antibodies (poly- to standard virological techniques for fish tissues clonal rabbit sera) produced against European catfish (Anonymous 2001) in epithelioma papulosum cyprini virus (ECV) (Bovo et al. 1993; kindly provided by G.  Ariel et al.: Ranavirus in Danish frogs 9 Bovo, Instituto Zooprofilattico Sperimentale delle dition for testing (7 Bufo bufo, 2 Rana temporaria and 1 Venezie, Italy) was diluted 1:800 in phosphate Pelophylax esculentus). Most of these were single frog buffered saline (PBS) without Ca2+ and Mg2+ and used incidences, but in 2 cases mass mortality was reported: as the primary antibody. This antibody cross-reacts B. bufo near Ebeltoft and P. esculentus from a pond with other ranavirus isolates and is specific to near Slagelse, Denmark. The B. bufo mass mortality ranaviruses (G. Bovo pers. comm.). The plates were appeared to be caused by extreme arid conditions. then incubated for 30 min at 37°C. Rhodamine- Two individual samples (case nos. 2008-50-282-1 and conjugated swine-anti-rabbit IgG (F0205, Dako) was -2) from the P. esculentus mass mortality near Slagelse applied as secondary antisera in a dilution of 1:100 in tested positive for ranavirus. The landowner reported PBS for 30 min at 37°C. The samples were examined approximately 1200 dead frogs over the course of the for immunofluorescent staining by fluorescence micro- weekend of 3–4 August 2008. For 2 wk prior to this he scope. had occasionally noticed dead frogs in the area, but the PCR, DNA sequencing and DNA sequence analysis bulk of the population in the pond perished over those for ranavirus. The viral DNA was extracted from the few days and hardly any were left after the main die- infected EPC cells by a QiaAmp DNA Mini Kit (Qia- off. The frogs were found dead in resting position gen) according to the manufacturer’s protocol. The along the edge of a shallow (1 m deep) artificial pond. complete major capsid protein (MCP; 1392 bp) gene of No clinical signs were observed. The pond was dug out the 2 Pelophylax esculentus viral isolates, case nos. 9 yr previously and situated in the middle of a grazing 2008-50-283-1 and -2 (PEV-DK1 and PEV-DK2), was field with very little vegetation along the edge, no agri- amplified using PCR primers and conditions published cultural activity nearby and no water flowing through by Hyatt et al. (2000, modified from their Table 4) it. The invertebrate fauna (including introduced Asta- (Table 1). The PCR products were sequenced using a cus astacus) and the fish in the lake (Cyprinus carpio) Big Dye Terminator v1.1 Cycle Sequencing Kit appeared healthy and were not investigated. The air (Applied Biosystems) and an ABI PRISM 3100-Avant temperatures leading up to and during the period of Genetic Analyzer (Applied Biosystems). The sequence the outbreak were unusually high and the weather data was analyzed with Sequencing Analysis Software was described as a heat wave for Danish climate con- 5.1 (Applied Biosystems) and multiple sequence align- ditions: the maximum air temperature for that week ments were done with CLUSTAL_X 1.81 (Thompson et was registered as 30.7°C for the area (www.dmi.dk). al. 1997). The values for sequence pair percent identity All samples in the 10 cases were negative for Bd. were calculated by the MegAlign program from the Samples from 2 wild Pelophylax esculentus from the Lasergene 7.1 application package (DNASTAR). same location near Slagelse (case nos. 2008-50-283-1 and -2) produced CPE in cell culture. The titers of virus in the organs were 7 × 104 and 2.3 × 107 TCID50 ml–1 for RESULTS the 2 samples, respectively. Cultures infected with these isolates stained positive for ranavirus using the The press release created a good deal of interest immunofluorescent antibody technique. Pelophylax with the media in Denmark: 11 websites (4 news sites, esculentus viral isolates (PEV-DK1 and PEV-DK2) 6 environmental associations), 2 newspapers and 2 were 100% identical and showed more than 97.3% radio stations published the story. The web survey nucleotide similarity when compared to the previously returned 25 cases which were potentially induced by published MCP gene sequences of other ranaviruses: infectious disease. After telephone interviews, 13 cases frog virus 3 (FV3), AY548484; Bohle iridovirus (BIV), were submitted, out of which 10 were of a suitable con- AY187046; epizootic hematopoietic necrosis virus Table 1. PCR primers for amplifying the major capsid protein (MCP) gene of ranaviruses (Hyatt et al. 2000 adapted from their Table 4, with permission). Primer position is relative to the initiator nucleotide of the epizootic hematopoietic necrosis virus MCP gene Primer designation Primer position Nucleotide sequence (5’ to 3’) MCP-1 –27 to + 3 CAC CGT GTA TCT TAT AAT AAA AAG GAA ATG MCP-2R 516 to 498 GGC TCC GTC CTG GCC TGT G MCP-3 443 to 468 GAG GCC AAG CGC ACA GGC TAC MCP-4R 959 to 940 TTG GAG CCG ACG GAA GGG TG MCP-5 900 to 920 CGC AGT CAA GGC CTT GAT GT MCP-6R 1484 to 1463 AAA GAC CCG TTT TGC AGC AAA C 10 Dis Aquat Org 85: 7–14, 2009 (EHNV), AY187045; Ambystoma tigrinum virus (ATV), Denmark is on the northern border of the range of NC_ 005832; tiger frog virus (TFV), AF389451; and Pelophylax esculentus, and populations are frag- Rana esculenta virus (REV), REV 282/I02 (Holopainen mented and have decreased over the last decades (Fog et al. 2009) (Table 2). et al. 1997). These circumstances make this species The amino acid sequence alignment of PEV-DK1 particularly vulnerable to disease. and PEV-DK2 MCP genes with other ranaviruses is Interviews with the landowner at the site near presented in Fig. 1. The MCP gene sequence of PEV- Slagelse where the Pelophylax esculentus mortality DK1 was submitted to GenBank with accession no. was detected did not cast light on a possible pathway FJ515796. Laboratory tests (Højvang Environmental of introduction of this virus into the pond. However, Laboratory) for environmental pollutants in the water ranaviruses are causing disease outbreaks in frogs on (pesticides, solvents, phenols and components of oil all continents except for Africa and Antarctica and sulphur) were all negative. (Williams et al. 2005); therefore, the presence of ranavirus in the environment is not an unusual obser- vation (Green et al. 2002). Further epidemiological DISCUSSION studies may cast light on the origin of this isolate. A very high density of frogs was observed in and Amphibians are cryptic by nature and individuals near the aforementioned pond over the summer. In the dying of natural causes are rarely noticed before they weeks prior to the outbreak the weather was unusually are scavenged or decompose. Creating and utilising warm and the temperature of the water in the pond public awareness in surveys can increase the chances would have reflected this. There was no shade around of detecting disease-induced mortality in cryptic taxa or near the pond for the frogs to shelter. These excep- like amphibians. Our survey led to the discovery of one tional environmental conditions could affect the of the most severe ranavirus-associated mass die-offs immunological response of the frogs and induce condi- observed in a free-living amphibian population. Only tions for rapid spread and high pathogenicity of the one case of those reported in the survey was positive virus (Jancovich et al. 1997, Carey et al. 1999, Brunner for ranavirus and none for Bd. With all the public atten- et al. 2005). Water analysis did not reveal any known tion and only one case of ranavirus detected, it is likely pollutants and both frogs submitted from the site were that outbreaks caused by ranavirus or chytrid fungus negative for Bd and positive for ranavirus with high of this proportion were unusual in Denmark during this titres in the organs. Based on previous reports con- period. cerning the epidemiology of ranavirus disease in Frog farming is a rarity in northern Europe, and amphibians (Jancovich et al. 1997, Green et al. 2002, ranavirus infection in amphibians would therefore not Pearman et al. 2004, Rojas et al. 2005), it is highly likely have any commercial impact. However, the effect of that the infection caused or contributed to the deaths of ranavirus on wild populations of amphibians is poten- the frogs in this population. Challenge studies of frogs tially devastating. All amphibians are protected under with this isolate would further clarify its role in the epi- statutory order in Denmark due to the general threat of zootic, especially by enabling pathological investiga- extinction (Anonymous 2007). If the virus spreads in tion of fresh specimens. wild populations of amphibians it may have severe Repetitive isolations of ranavirus from moribund edi- consequences on the infected populations of amphib- ble frogs (11 positive out of 16 samples over an 11 yr ians, other susceptible species and their predators. period) collected from the wild and kept for commer- Table 2. Nucleotide sequence similarity (%) of complete major capsid protein gene (1392 bp) of Danish Pelophylax esculentus viral isolates (PEV-DK-2008-50-283-1 and -2) compared to other ranaviruses. REV 282/I02: Rana esculenta virus 282/I02; FV3: frog virus 3; BIV: Bohle iridovirus; EHNV: epizootic hematopoietic necrosis virus; ATV: Ambystoma tigrinum virus; TFV: tiger frog virus Isolate Nucleotide sequence similarity (%) Host species GenBank accession nos. PEV-DK1 100 Pelophylax esculentus FJ515796 PEV-DK2 100 Pelophylax esculentus FJ515796 REV 282/I02 99.6 Pelophylax esculentus Holopainen et al. (2009) FV3 98.3 Rana pipiens AY548484 BIV 98.4 Limnodynates ornatus AY187046 EHNV 98.8 Pelophylax fluviatilis AY187045 ATV 97.3 Ambystoma tigrinum NC_005832 TFV 98.4 Rana tigrina AF389451  Ariel et al.: Ranavirus in Danish frogs 11 * 20 * 40 * 60 FV3 : MSSVTGSGITSGFIDLATYDNLERAMYGGSDATTYFVKEHYPVGWFTKLPSLAAKMSGNP BIV : ......L..................................................... EHNV : ............................................................ ATV : .........................I.................................. TFV : ............................................................ PEV_DK1 : ............................................................ PEV_DK2 : ............................................................ * 80 * 100 * 120 FV3 : AFGQQFSVGVPRSGDYILNAWLVLKTPEVELLAANQLGDNGTIRWTKNPMHNIVESVTLS BIV : ............................................................ EHNV : .............................K.........................N.N.. ATV : .................I...........K........E................N.N.. TFV : ......................................E..................... PEV_DK1 : .............................K.........................N.N.. PEV_DK2 : .............................K.........................N.N.. * 140 * 160 * 180 FV3 : FNDISAQSFNTAYLDAWSEYTMPEAKRTGYYNMIGNTSDLINPAPATGQDGARVLPAKNL BIV : ...........................I................................ EHNV : ...........................I.....................N.......... ATV : ...........................I.....................NE......... TFV : ...........................I................................ PEV_DK1 : ...........................I.....................N.......... PEV_DK2 : ...........................I.....................N.......... * 200 * 220 * 240 FV3 : VLPLPFFFSRDSGLALPVVSLPYNEIRITVKLRAIQDLLILQHNTTGAISPIVASDLAGG BIV : ............................................................ EHNV : ......................................................A..E.. ATV : ...............................................V......A..E.. TFV : ............................................................ PEV_DK1 : ......................................................A..E.. PEV_DK2 : ......................................................A..E.. * 260 * 280 * 300 FV3 : LPDTVEANVYMTVALITGDERQAMSSTVRDMVVEQVQAAPVHMVNPRNATTFHTDMRFSH BIV : .................................................A.......... EHNV : .................................................A.......... ATV : .................................................A.......... TFV : .....................................V...........A.......... PEV_DK1 : .................................................A.......... PEV_DK2 : .................................................A.......... * 320 * 340 * 360 FV3 : AVKALMFMVQNVTHPSVGSNYTCVTPVVGVGNTVLEPALAVDPVKSASLVYENTTRLPDM BIV : ............................................................ EHNV : .......................A......D............................L ATV : .......................A...................I...............L TFV : .............................A.............................. PEV_DK1 : ............................................................ PEV_DK2 : ............................................................ Fig. 1. Amino acid sequence alignment of the major capsid protein gene of several ranaviruses. Previously published sequences: frog virus 3 (FV3; AY548484), Bohle iridovirus (BIV; AY187046), epizootic hematopoietic necrosis virus (EHNV; AY187045), Ambystoma tigrinum virus (ATV; NC_005832) and tiger frog virus (TFV; AF389451). Identical amino acids are marked with a dot, STOP-codons are marked with an asterisk 12 Dis Aquat Org 85: 7–14, 2009 * 380 * 400 * 420 FV3 : GVEYYSLVEPWYYATSIPVSTGHHLYSYALSLQDPHPSGSTNYGRLTNASLNVTLSAEAT BIV : ............................................................ EHNV : ........Q................................................... ATV : ........Q................................................... TFV : ...............................M............................ PEV_DK1 : ............................................................ PEV_DK2 : ............................................................ * 440 * 460 FV3 : TAAAGGGGNNSGYTTAQKYALIVLAINHNIIRIMNGSMGFPIL* BIV : ........D..................................* EHNV : ..S.....D..................................* ATV : A.......D..................................* TFV : ........D..................................* PEV_DK1 : ........D..................................* PEV_DK2 : ........D..................................* Fig. 1. (continued) cial purposes in Croatia spurred investigations into the the second closest relative to the Danish frog isolates virulence of the isolates to healthy edible frogs (Fijan was EHNV isolated from fish on another continent, et al. 1991). Experimental challenge via various routes with nucleotide identity of 98.8%. At the protein level did not produce disease in the frogs over 30 d, there were 5 amino acid changes between EHNV and although low titres of virus were present in frogs from PEV-DK (Fig. 1). all treatments apart from the negative control. This Despite the severity of amphibian declines due to questions the role of ranavirus as a primary pathogen infectious diseases, it is difficult to envisage how con- in edible frogs. It may be an opportunistic pathogen trol and containment of ranavirus in wild frogs can be which coexists at low levels within the host except achieved (Mendelson et al. 2006, Gascon et al. 2007). when the host is weakened and the environmental The nature of wild animals makes it very difficult to conditions are conducive to viral replication. Under ensure non-destructive containment of animals in an such conditions the virus may turn pathogenic and infected area. When frogs migrate from a pond in the cause disease in the host as reported here. autumn they may be in contact with other frog popula- The MCP gene is commonly used in defining taxon- tions and contaminate them. Viruses could also spread omy in the virus family Iridoviridae (Chinchar 2002) by vectors such as birds, fish or downstream via the and in differentiating ranavirus isolates (Mao et al. water. 1999, Hyatt et al. 2000, Marsh et al. 2002, Do et al. The risk of ranavirus outbreaks is not limited to 2005). The MCP sequences of PEV-DK1 and PEV-DK2 amphibians. Evidence for transmission of ranavirus were identical to each other. Our sequence compari- between frogs and fish has been recorded (Moody & son included 3 ranaviruses isolated in amphibian dis- Owens 1994, Ariel & Owens 1997, Mao et al. 1999) and ease outbreaks: BIV (Speare & Smith 1992), ATV one host species may act as reservoir for the other (Jancovich et al. 1997), TFV (Kanchanakhan et al. (Mao et al. 1999, Duffus et al. 2008). EHNV, ECV and 2002) and EHNV from disease outbreaks in perch in European sheatfish virus (ESV) are listed by the OIE Australia (Langdon et al. 1986). Based on sequence for fish and they are closely related to this new strain. comparison of the complete MCP gene, BIV, ATV, All the listed viruses were involved in natural out- TFV, EHNV, REV 282/I02 and FV3 were very closely breaks in wild populations of perch, catfish and sheat- related (97.3 to 99.6% nucleotide identity) to the Dan- fish (Langdon et al. 1986, Ahne et al. 1989, Bovo et al. ish frog virus isolates (Table 2). The comparison of the 1993, Bigarré et al. 2008) and several other species of MCP sequences indicates that PEV-DK belongs to the fish and frogs are experimentally susceptible to genus Ranavirus. These results are in accordance ranavirus infection (Langdon 1989, Moody & Owens with species demarcation criteria in the genus 1994, Ariel & Owens 1997, Cullen & Owens 2002, Bang Ranavirus (Chinchar et al. 2005). The closest relative Jensen et al. 2009). Potentially, this new Pelophylax to Danish frog isolates was PEV 282/I02, which was esculentus viral isolate could therefore pose a risk to also isolated from an edible frog, with nucleotide both amphibian and piscine fauna. To balance these identity of 99.6% and amino acid identity of 100% apocalyptic statements there are also many reports of (results not shown). Surprisingly, in the present study, ranavirus isolations from animals with no clinical  Ariel et al.: Ranavirus in Danish frogs 13 symptoms, where the virus apparently is not causing ➤ Cullen BR, Owens L (2002) Experimental challenge and clin- the host any distress (Jensen et al. 1979, Tapiovaara et ical cases of Bohle irodovirus (BIV) in native Australian anurans. Dis Aquat Org 49:83–92 al. 1998, Harp & Petranka 2006, Duffus et al. 2008). This is the first survey for viruses in Danish frogs and ➤ Cunningham AA, Langton TES, Bennett PM, Lewin JF, Drury SEN, Gough RE, McGregor SK (1996) Pathological and the first record of ranavirus in Danish amphibians. microbiological findings from incidents of unusual mortal- Although ranavirus could occur naturally as part of the ity of the common frog (Rana temporaria). Philos Trans R ecosystem, the finding of high titres of ranavirus asso- Soc Lond B Biol Sci 351:1539–1557 ciated with mass mortalities in frogs demonstrates the ➤ Do JW, Cha SJ, Kim JS, An EJ and others (2005) Phylogenetic analysis of the major capsid protein gene of iridovirus iso- possible consequences of infection given the right lates from cultured flounders Paralichthys olivaceus in environmental circumstances. Korea. Dis Aquat Org 64:193–200 ➤ Docherty DE, Meteyer CU, Wang J, Mao JH, Case ST, Chin- char VG (2003) Diagnostic and molecular evaluation of Acknowledgements. We thank the RANA consortium for three iridovirus-associated salamander mortality events. commenting on the questionnaire format. N. Nicolajsen, J. J Wildl Dis 39:556–566 Mølgaard, M. B. Christophersen and U. Tobler are acknowl- ➤ Drury SEN, Gough RE, Cunningham AA (1995) Isolation of edged for their technical assistance. This research was carried an iridovirus-like agent from common frogs (Rana tempo- out under the RANA project under the Sixth Framework Pro- raria). Vet Rec 137:72–73 gramme funded by the European Commission. ➤ Duffus ALJ, Pauli BD, Wozney K, Brunetti CR, Berrill M (2008) Frog virus 3-like infections in aquatic amphibian communities. J Wildl Dis 44:109–120 LITERATURE CITED ➤ Fijan N, Sulimanovic D, Bearzotti M, Muzinic D and others (1983) Some properties of the Epithelioma papulosum ➤ Ahne W, Schlotfeldt HJ, Thomsen I (1989) Fish viruses: isola- cyprini (EPC) cell line from carp Cyprinus carpio. Ann Inst tion of an icosahedral cytoplasmic deoxyribovirus from Pasteur Virol 134:207–220 sheatfish (Siluris glanis). J Vet Med Ser B 36:333–336 Fijan N, Matasin Z, Petrinec Z, Valpotic I, Zillenberg LO Anonymous (2001) Commission Decision of 22 February 2001 (1991) Isolation of an iridovirus-like agent from the green laying down the sampling plans and diagnostic methods frog (Rana esculenta L.). Vet Arh 61:151–158 for the detection and confirmation of certain fish diseases Fog K, Schmedes A, Rosenørn de Lasson D (1997) Nordens and repealing Decision 92/532/EEC. Off J Eur Union padder og krybdyr. Gad, Copenhagen 31:436–442 ➤ Fox SF, Greer AL, Torres-Cervantes R, Collins JP (2006) First Anonymous (2007) Bekendtgørelse om fredning af visse case of ranavirus-associated morbidity and mortality in dyre- og plantearter mv., indfangning af og handel med natural populations of the South American frog Atelog- vildt og pleje af tilskadekomment vildt; Bekendtgørelse nathus patagonicus. Dis Aquat Org 72:87–92 nr. 901 af 11. juli 2007 (Statutory order no. 901, July 2007). Gascon C, Collins JP, Moore RD, Church DR, McKay JE, Danish Ministry of the Environment, Copenhagen Mendelson JR III (eds) (2007) Amphibian Conservation ➤ Ariel E, Owens L (1997) Epizootic mortalities in tilapia Ore- Action Plan. IUCN/SSC Amphibian Specialist Group, ochromis mossambicus. Dis Aquat Org 29:1–6 IUCN, Gland and Cambridge ➤ Bang Jensen B, Kjær Ersbøll A, Ariel E (2009) Susceptibility of ➤ Granoff A, Came PE, Rafferty KA (1965) The isolation and pike Esox lucius to a panel of Ranavirus isolates. Dis properties of viruses from Rana pipiens: their possible Aquat Org 83:169–179 relationship to the renal adenocarcinoma of the leopard Bigarré L, Cabon J, Baud M, Pozet F, Castric J (2008) frog. Ann NY Acad Sci 126:237–255 Ranaviruses associated with high mortalities in catfish in ➤ Green DE, Converse KA, Schrader AK (2002) Epizootology of France. Eur Assoc Fish Pathol Bull 28:163–168 sixty-four amphibian morbidity and mortality events in Bovo G, Comuzi M, DeMas S, Ceschia G, Giorgetti G, Gia- the USA, 1996–2001. Ann NY Acad Sci 969:323–339 cometti P, Cappellozza E (1993) Isolamento di un agente ➤ Greer AL, Berrill M, Wilson PJ (2005) Five amphibian mortal- virale irido-like da pesce gatto (Ictalurus melas) dalleva- ity events associated with ranavirus infection in south cen- mento. Boll Soc Ital Patol Ittica 11:3–10 tral Ontario, Canada. Dis Aquat Org 67:9–14 ➤ Boyle DG, Boyle DB, Olsen V, Morgan JAT, Hyatt AD ➤ Harp EM, Petranka JW (2006) Ranavirus in wood frogs (Rana (2004) Rapid quantitative detection of chytridiomycosis sylvatica): potential sources of transmission within and (Batrachochytrium dendrobatidis) in amphibian samples between ponds. J Wildl Dis 42:307–318 using real-time Taqman PCR assay. Dis Aquat Org 60: ➤ Holopainen R, Ohlemeyer S, Schütze H, Bergmann SM, Tapi- 141–148 ovaara H (2009) Ranavirus phylogeny and differentiation ➤ Brunner JL, Richards K, Collins JP (2005) Dose and host char- based on major capsid protein, DNA polymerase and acteristics influence virulence of ranavirus infections. neurofilament triplet H1-like protein genes. Dis Aquat Oecologia 144:399–406 Org (in press) ➤ Carey C, Cohen N, Rollins-Smith L (1999) Amphibian de- ➤ Hyatt AD, Gould AR, Zupanovic Z, Cunningham AA and oth- clines: an immunological perspective. Dev Comp Immu- ers (2000) Comparative studies of piscine and amphibian nol 23:459–472 iridoviruses. Arch Virol 145:301–331 ➤ Chinchar VG (2002) Ranaviruses (family Iridoviridae): ➤ Hyatt AD, Boyle DG, Olsen V, Boyle DB and others (2007) Diag- emerging cold-blooded killers. Arch Virol 147:447–470 nostic assays and sampling protocols for the detection of Ba- Chinchar VG, Essbauer S, He JG, Hyatt A, Miyaki T, Seligy trachochytrium dendrobatidis. Dis Aquat Org 73:175–192 V, Williams T (2005) Iridoviridae. In: Fauquet CM, Mayo ➤ Jancovich JK, Davidson EW, Morado JF, Jacobs BL, Collins MA, Maniloff J, Desselberger U, Gall LA (eds) Virus tax- JP (1997) Isolation of a lethal virus from the endangered onomy: 8th report of the International Committee on the tiger salamander Ambystoma tigrinum stebbinsi. Dis Taxonomy of Viruses. Elsevier, London, p 163–175 Aquat Org 31:161–167  14 Dis Aquat Org 85: 7–14, 2009 ➤ Jancovich JK, Davidson EW, Parameswaran N, Mao J and OIE (World Organisation for Animal Health) (2008) Aquatic others (2005) Evidence for emergence of an amphibian iri- animal health code. Section 2.4. Diseases of amphibians. doviral disease because of human-enhanced spread. Mol OIE, Paris, available at www.oie.int/eng/normes/fcode/ Ecol 14:213–224 en_titre_2.4.htm Jensen NJ, Bloch B, Larsen JL (1979) The ulcus syndrome in ➤ Pearman PB, Garner TWJ, Straub M, Greber UF (2004) cod (Gadus morhua). III. A preliminary virological report. Response of the Italian agile frog (Rana latastei) to a Nord Vet Med 31:436–442 Ranavirus, frog virus 3: a model for viral emergence in Jørgensen PEV, Olesen NJ, Ahne W, Lorenzen N (1989) naïve populations. J Wildl Dis 40:660–669 SVCV and PFR viruses: serological examination of 22 iso- ➤ Rafferty KA Jr (1965) The cultivation of inclusion-associated lates indicates close relationship between the two fish viruses from Lucké tumor frogs. Ann NY Acad Sci 126: rhabdoviruses. In: Ahne W, Kurstak E (eds) Viruses of 3–21 lower vertebrates. Springer-Verlag, Berlin, p 349–366 Reed LJ, Muench H (1938) A simple method of estimating Kanchanakhan S, Saduakdee U, Kreethachat A, Chinabut S fifty percent endpoints. Am J Hygiene 27:493–497 (2002) Isolation of FV3-like iridovirus from a cutaneous ➤ Rojas S, Richards K, Jancovich JK, Davidson EW (2005) Influ- ulceration disease of cultured frog, Rana tigrina Cantor, in ence of temperature on Ranavirus infection in larval sala- Thailand. In: Lavilla-Pitogo CR, Cruz-Lacierda ER (eds) manders Ambystoma tigrinum. Dis Aquat Org 63:95–100 Diseases in Asian aquaculture IV. Fish Health Section, ➤ Speare R, Smith JR (1992) An iridovirus-like agent isolated Asian Fisheries Society, Manila, p 417–425 from the ornate burrowing frog Limnodynastes ornatus in Langdon JS (1989) Experimental transmission and patho- northern Australia. Dis Aquat Org 14:51–57 genicity of epizootic haematopoietic necrosis virus ➤ Tapiovaara H, Olesen NJ, Lindén J, Rimaila-Pärnänen E, von (EHNV) in redfin perch, Perca fluviatilis L., and 11 other Bonsdorff CH (1998) Isolation of an iridovirus from pike- teleosts. J Fish Dis 12:295–310 perch Stizostedion lucioperca. Dis Aquat Org 32:185–193 ➤ Langdon JS, Humphrey JD, Williams LM, Hyatt AD, West- ➤ Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Hig- bury HA (1986) First virus isolation from Australian fish: gins DG (1997) The CLUSTAL_X windows interface: flex- an iridovirus-like pathogen from redfin perch, Perca fluvi- ible strategies for multible sequence alignment aided by atilis L. J Fish Dis 9:263–268 quality analysis tools. Nucleic Acids Res 25:4876–4882 ➤ Mao J, Green DE, Fellers G, Chinchar VG (1999) Molecular ➤ Weng SP, He JG, Wang XH, Lu L, Deng M, Chan SM (2002) characterisation of iridoviruses isolated from sympatric Outbreaks of an iridovirus disease in cultured tiger frog, amphibians and fish. Virus Res 63:45–52 Rana tigrina rugulosa, in southern China. J Fish Dis 25: ➤ Marsh IB, Whittington RJ, O’Rourke B, Hyatt AD, Chisholm 423–427 O (2002) Rapid differentiation of Australian, European Williams T, Barbosa-Solomieu V, Chinchar VG (2005) A and American ranaviruses based on variation in major decade of advances in iridovirus research. Adv Virol capsid protein gene sequence. Mol Cell Probes 16: 53:905–912 137–151 ➤ Zhang QY, Xiao F, Li ZQ, Gui JF, Mao J, Chinchar VG (2001) ➤ Mendelson JR III, Lips KR, Gagliardo RW, Rabb GB and oth- Characterization of an iridovirus from the cultured pig ers (2006) Confronting amphibian declines and extinc- frog Rana grylio with lethal syndrome. Dis Aquat Org tions. Science 313:48 48:27–36 ➤ Moody NJG, Owens L (1994) Experimental demonstration of ➤ Zupanovic Z, Musso C, Lopez G, Louriero CL, Hyatt AD, the pathogenicity of a frog virus, Bohle iridovirus, for a Hengstberger S, Robinson AJ (1998) Isolation and charac- fish species, barramundi Lates calcarifer. Dis Aquat Org terization of iridoviruses from the giant toad Bufo marinus 18:95–102 in Venezuela. Dis Aquat Org 33:1–9 Editorial responsibility: Alex Hyatt, Submitted: December 18, 2008; Accepted: March 17, 2009 Geelong, Victoria, Australia Proofs received from author(s): May 7, 2009 

Chapter 14 Hotspots, Conservation, and Diseases: Madagascar’s Megadiverse Amphibians and the Potential Impact of Chytridiomycosis Stefan L€ otters, Dennis R€ odder, Jos Kielgast, and Frank Glaw Abstract Worldwide amphibian diversity is threatened through the emergence of the disease chytridiomycosis, caused by the amphibian chytrid fungus. This patho- gen apparently is absent from the amphibian hotspot Madagascar. However, an extinction risk assessment based on environmental niche modelling suggests that a major portion of this island is climatically highly suitable to the fungus. This includes regions of high amphibian species richness. Many species have their entire geographic range in such areas and are at the same time predicted to suffer potentially from chytridiomycosis due to their life history traits. Human-mediated dissemination of the chytrid fungus to Madagascar is considered likely. In particu- lar, there may be a high risk of accidental cointroduction via the animal trade. Severe decline and possibly extinction are expected in a postemergence scenario on Madagascar with more than 270 described and numerous undescribed anuran amphibian species under threat. Effective responses to this potential threat might include (1) an increased attention to ‘biosecurity’, including the consequent imple- mentation of measures to avoid the introduction of the chytrid fungus, (2) the development of breeding procedures for representatives of all major clades of Madagascan amphibians as a ‘pre-emergency prophylaxis’ and (3) the development of plans for ‘emergency response’. S. L€otters (*) Biogeography Department, Trier University, 54286 Trier, Germany e-mail: loetters@uni-trier.de D. R€odder Zoologisches Forschungsmuseum Alexander Koenig, Adenauerallee 160, 53113 Bonn, Germany J. Kielgast Department of Biology, University of Copenhagen, Sølvgade 83H, 1307 København K, Denmark F. Glaw Zoologische Staatssammlung M€ unchen, M€ unchhausenstr. 21, 81247 Munich, Germany F.E. Zachos and J.C. Habel (eds.), Biodiversity Hotspots, 255 DOI 10.1007/978-3-642-20992-5_14, # Springer-Verlag Berlin Heidelberg 2011 256 S. L€ otters et al. 14.1 Biodiversity Loss Through Diseases Biodiversity is unevenly distributed across the globe. Several regions have been identified in which both species richness and endemism is particularly high. The persistence of these “hotspots” comprises an essential element in maintaining global biodiversity by focused conservation strategies and natural resource man- agement (e.g., Mittermeier et al. 2004). Although there are tools to minimize direct anthropogenic impact on hotspots (e.g., by designation of protected areas), a range of indirect impacts remains with few mitigation measures available. Emerging infectious diseases (EID) fall within this class of biodiversity threats and, although their effects on wildlife population dynamics and conservation are not new, there has been an alarming pathogen-related increase in species declines during the last decades (e.g., Smith et al. 2009a). Human impact is linked to this phenomenon and the term “pathogen pollution” has been introduced specifically covering anthropo- genic dissemination of pathogens across evolutionary or ecological boundaries (Daszak et al. 2000). Recent examples of dramatic biodiversity loss due to EIDs include virus infections in North American birds (Rahbek 2007), infectious cancer in the Tasmanian Devil (McCallum 2008), or fungus-associated mass mortality in North American bats (Blehert et al. 2009). At the global scale, the emergence of a fungal disease in amphibians (chytridiomycosis) has within the last decade been made responsible for the most severe case of disease-induced biodiversity loss ever observed (Gascon et al. 2007; Fisher et al. 2009). It has been documented that amphibian species have become extinct and will become extinct through chytridiomycosis (e.g., L€otters et al. 2010; R€odder et al. 2010). Currently, amphibians belong to the most threatened of all vertebrates, with approximately one third of the over 6,500 (according to: http:// www.amphibiaweb.org) assessed species threatened with extinction (Stuart et al. 2008; http://www.iucnredlist.org/initiatives/amphibians). Noteworthy, it is sug- gested that diversity in several of the amphibian hotspots has been affected by chytridiomycosis (e.g., Lips et al. 2008). As emphasized in the IUCN Amphibian Conservation Action Plan, new conservation strategies need to be rapidly implemented to prevent large-scale amphibian diversity loss due to chytridio- mycosis (Gascon et al. 2007). 14.1.1 Chytridiomycosis and Amphibian Declines This EID is an epidermal infection by the parasitic amphibian chytrid fungus, Batrachochytrium dendrobatidis (Bd). This pathogen spreads via motile infectious zoospores in aqueous environments and encysts and grows to reproductive zoosporangia on amphibian skin. Severe infections cause osmoregulatory imbal- ance in physiologically essential ions and may lead to a breakdown of neurological function causing cardiac arrest and death (Voyles et al. 2009). The origin of Bd and its natural host–pathogen system remain unknown but low global molecular 14 Hotspots, Conservation, and Diseases 257 diversity indicates that it has rapidly and only recently spread to its present pan- demic state (James et al. 2009). Currently, Bd is known from more than 1,700 localities in the wild within all continents where amphibians occur and has been detected in over 400 different host species (http://www.spatialepidemiology.net/ Bd-maps). It has been suggested that anthropogenic dissemination of Bd is likely to play an important role in the panzootic chytridiomycosis. In particular, the interna- tional animal trade has been identified as a plausible pathway of dissemination with hundreds of amphibian species being traded annually (Garner et al. 2009). In recognition of its impact on wild amphibian populations, Bd in 2008 was implemented on the list of notifiable diseases of the aquatic health code under the World Organisation for Animal Health (http://www.oie.int/eng/en_index.htm). Pos- sible actions and policy changes have been discussed. However, Australia is so far the only country strongly enforcing specific restrictions on Bd (Garner et al. 2009). Chytridiomycosis has been identified as the key causal mechanism in several focal studies of declines and extinctions on a large scale both geographically and taxonomically (e.g., Schloegel et al. 2006; Lips et al. 2008) including so-called enigmatic declines (L€ otters et al. 2010) and there is evidence that many more species have been or will be affected (R€ odder et al. 2009; L€otters et al. 2010). However, it is not possible to generalize on consequences of Bd emergence as substantial interspecific variation in susceptibility has been observed both in nature and under controlled exposure trials (e.g., Blaustein et al. 2005; Smith et al. 2009b). The host response span from highly susceptible species, which have undergone extraordinarily rapid populations declines, to highly resistant species, which live with high prevalence and pathogen load and may serve as Bd reservoirs and vectors (e.g., Rollins-Smith et al. 2009; Schloegel et al. 2010). Recently, Bielby et al. (2008) defined a biotic index and suggested 837 out of 3,976 worldwide anuran amphibians to exhibit the highest risk of Bd-related decline or extinction. This was based on biological and life history information. Particularly susceptible species were pointed out to be living at high altitude with an aquatic life stage, having a restricted range and low fecundity. Subsequently, developing an ecological niche model (e.g., Guisan and Zimmermann 2000), R€odder et al. (2009) estimated the worldwide potential distribution of the amphib- ian chytrid fungus. The authors used the Maxent algorithm to develop a map indicating the occurrence probability of Bd based on 365 presence records of this pathogen (taken from http://www.spatialepidemiology.net/bd) and six bioclimatic variables with spatial resolution 2.5 arc min (taken from http://www.worldclim.org) i.e. ‘annual mean temperature’, ‘maximum temperature of the warmest month’, ‘minimum temperature of the coldest month’, ‘annual precipitation’, ‘precipitation of the wettest month’ and ‘precipitation of the driest month’. These variables were identified as biologically important for Bd in previous studies (e.g. Kielgast et al. 2010). Model evaluation via the area under the receiver operating characteristic curve (AUC) revealed a high precision of the modelling efforts (mean training AUC = 0.937, mean test AUC = 0.910 in 100 models each trained with 70% of the records and the remaining 30% used for model testing). The model uncovered regions of different suitability to the pathogen. Applying a rule set of predictions 258 S. L€ otters et al. from their spatial analysis and key host life history variables inferred by Bielby et al. (2008), R€ odder et al. (2009) prioritized anuran species according to the risk of decline and extinction due to the developing pandemic chytridiomycosis (see Table 14.1). For this purpose, R€ odder et al. (2009) derived a risk factor balancing the relative environmental suitability for the chytrid fungus within the range of a species and its susceptibility according to the biotic index derived by Bielby et al. (2008) (Table 14.1). In this risk assessment, the combined index ranges from 0 (no risk) to 1 (high risk) (R€ odder et al. 2009). Most of the 837 anuran species, which by their biology and life history show a high predicted susceptibility to Bd, occur in regions which at the same time are characterized by high bioclimatic suitability to the chytrid fungus. In total, 379 of them, with the entire geographic range, fall into areas highly suitable to Bd. R€ odder et al. (2009) considered these amphibians to be the most threatened by the emergence of chytridiomycosis. 14.1.2 Mitigating the Problem: Ex Situ Conservation Although some susceptible species may survive severe epizootics of chytridiomycosis on long term (Retallick et al. 2004; Murray et al. 2009), it has been stressed that new rapid and highly comprehensive conservation actions are necessary to avoid catastrophic biodiversity loss. Useful measures include revision of animal trade regulations, development and implementation of pathogen hygiene protocols, increasing public awareness at airports and ports, development of risk factors, monitoring and further research (Andreone and Randriamahazo 2008; Weldon and Du Preez 2008). As there is no certain measure to mitigate Bd dissemination and transmission in the wild, ex situ conservation has been promoted in the IUCN Amphibian Conservation Action Plan both in nonrange countries and in-range countries but outside the nature (Gascon et al. 2007). Ex situ conservation includes cryobanking of viable biomaterials and short-term conservation breeding. Although the former has had limited success with regard to vertebrates except fishes, there is some experience and potential for amphibian captive breeding in zoos and aquariums (Gascon et al. 2007; Lermen et al. 2009). Conservation breeding has been successfully performed in a few species, e.g., the Mallorcan Midwife Toad (Alytes muletensis), Kihansi Spray Toad (Nectophrynoides asperginis), or the Panamanian Golden Frog (Atelopus zeteki). Numerous programs in zoos, on- or off-exhibit, have been implemented, leaving conservationists optimistic that some amphibian species may survive through short-term conserva- tion breeding (e.g., McGregor Reid and Zippel 2008). Most of them are coordinated by the “Amphibian Ark” of IUCN Amphibian Specialist Group, IUCN Captive Breeding Specialist Group, and the World Association of Zoos and Aquariums. Species prioritization for several countries including Madagascar have been conducted based on species’ status (e.g., IUCN Red List category, occurrence in protected habitat, biological distinctness), availability of specimens and general feasibility of ex situ conservation efforts (http://portal.isis.org/partners/AARK/ Lists/Prioritization%20workshop%20results/AllItems.aspx). Table 14.1 List of 234 Madagascan anuran amphibian species as recorded in the IUCN Red List of Threatened Species on 7 July 2009, their IUCN Red List 14 category and geographic range size (taken from http://www.natureserve.org/getdata/amphibianmaps.jsp; accessed 7 July 2009), each followed by its biotic Index (after Bielby et al. 2008), risk factor for threat through extinction due to chytridiomycosis (after R€ odder et al. (2009) and its conservation priority according to ‘Amphibian Ark’ (with 10 species suggested for conservation breeding each indicated by an asterisk) (http://portal.isis.org/partners/AARK/Lists/ Prioritization%20workshop%20results/AllItems.aspx; accessed 5 July 2010). Species are sorted by risk factor; in 48 taxa the biotic index and risk factor were not determined Species IUCN Red List Geographic Biotic index (Bielby Risk factor (Ro¨dder ‘Amphibian Ark’ category range (km2) et al. 2008) et al. 2009) priority value Aglyptodactylus laticeps EN 31.9 0.8 0 44 Boophis jaegeri VU 413.95 0.36 0 12 Boophis xerophilus DD 195.58 0.35 0 8 Heterixalus tricolor LC 28.15 0.99 0 0 Heterixalus variabilis LC 3356.71 0.22 0 0 Hotspots, Conservation, and Diseases Mantidactylus cowanii NT 34854.52 0.18 0 0 Mantidactylus melanopleura LC 99953.85 0.3 0 8 Mantidactylus noralottae VU 3.62 0.35 0 0 Rhombophryne testudo VU 570.4 0.51 0 15 Stumpffia psologlossa DD 1919.09 0.58 0 8 Stumpffia pygmaea VU 328.57 0.46 0 20 Heterixalus luteostriatus LC 49184.02 0.1 0 0 Aglyptodactylus securifer LC 6372.37 0.27 0 11 Wakea madinika DD 93.47 1 0 11 Hoplobatrachus tigerinus LC 4202408.01 0.05 0 5 Dyscophus insularis LC 152722.46 0.09 0 3 Scaphiophryne calcarata LC 165446.88 0.09 0 3 Mantella betsileo LC 89439.46 0.15 0 10 Blommersia wittei LC 74120.37 0.11 0 0 Boophis doulioti LC 279388.71 0.12 0 0 Scaphiophryne brevis LC 90888.33 0.16 0 3 Ptychadena mascareniensis LC 9151425.38 0.06 0.01 0 259 (continued) Table 14.1 (continued) 260 Species IUCN Red List Geographic Biotic index (Bielby Risk factor (Ro¨dder ‘Amphibian Ark’ category range (km2) et al. 2008) et al. 2009) priority value Heterixalus carbonei NT 4858 0.28 0.01 4 Heterixalus boettgeri LC 4938.35 0.25 0.01 0 Mantella viridis EN 839 1 0.01 47* Boophis opisthodon LC 49586.7 0.12 0.03 0 Cophyla phyllodactyla LC 10255.36 0.43 0.06 3 Gephyromantis luteus LC 72241.32 0.17 0.07 0 Stumpffia gimmeli LC 9512.22 0.5 0.07 0 Mantidactylus zipperi LC 76195.45 0.51 0.08 0 Mantidactylus ulcerosus LC 54277.51 0.63 0.1 0 Heterixalus punctatus LC 36614.74 0.28 0.13 0 Mantidactylus ambreensis LC 8198.09 0.61 0.13 0 Boophis tephraeomystax LC 123576.94 0.23 0.14 0 Mantidactylus curtus LC 254685.51 0.27 0.14 8 Stumpffia tetradactyla DD 495.35 0.54 0.14 16 Scaphiophryne spinosa LC 103192.65 0.23 0.14 13 Mantella expectata EN 1290.53 1 0.15 47* Mantella laevigata NT 23090.51 0.33 0.15 22 Boophis haematopus VU 4109.44 0.44 0.15 12 Platypelis grandis LC 145481.81 0.25 0.17 8 Heterixalus alboguttatus LC 35515.74 0.26 0.17 0 Mantella manery DD 37.13 0.93 0.17 18 Mantidactylus majori LC 40087.57 0.98 0.18 0 Gephyromantis plicifer NT 20275.76 0.3 0.18 4 Gephyromantis boulengeri LC 30019.63 0.33 0.19 0 S. L€ Scaphiophryne gottlebei EN 724.96 1 0.19 60* Plethodontohyla ocellata LC 78481.67 0.28 0.19 0 Mantidactylus femoralis LC 210805 0.29 0.21 0 otters et al.  14 Gephyromantis malagasius LC 69119.48 0.3 0.22 0 Gephyromantis redimitus LC 55563.98 0.3 0.22 0 Mantidactylus betsileanus LC 96904.14 0.3 0.24 8 Guibemantis liber LC 159146 0.31 0.25 0 Gephyromantis corvus EN 1239.26 0.76 0.25 13 Mantidactylus lugubris LC 115484.08 0.36 0.26 0 Gephyromantis silvanus EN 892.25 0.7 0.27 44 Gephyromantis pseudoasper LC 22586.44 0.51 0.27 0 Mantidactylus guttulatus LC 60561.53 0.37 0.27 13 Platypelis milloti EN 3529.49 0.83 0.28 44 Platypelis pollicaris DD 12442.21 0.57 0.29 8 Gephyromantis webbi EN 1649.04 0.64 0.29 36 Mantidactylus biporus LC 96111.76 0.35 0.29 0 Hotspots, Conservation, and Diseases Platypelis barbouri LC 41782.08 0.4 0.3 0 Guibemantis tornieri LC 69844.24 0.34 0.31 0 Stumpffia grandis DD 25270.95 0.43 0.31 8 Mantidactylus opiparis LC 82533.99 0.41 0.31 0 Mantidactylus mocquardi LC 139555.73 0.35 0.31 0 Mantella nigricans LC 31432.53 0.45 0.31 10 Boophis albilabris LC 90431.13 0.36 0.32 0 Mantidactylus grandidieri LC 85656.99 0.29 0.32 13 Gephyromantis granulatus LC 15283.02 0.57 0.32 0 Gephyromantis zavona DD 745.06 0.98 0.32 8 Boophis tasymena LC 69295.9 0.37 0.32 0 Boophis occidentalis NT 652.75 0.9 0.33 4 Guibemantis pulcher LC 67174.54 0.36 0.33 0 Gephyromantis horridus EN 3243.21 0.85 0.33 44 Blommersia grandisonae LC 68237.09 0.37 0.34 0 Boophis madagascariensis LC 99482.06 0.38 0.34 8 Paradoxophyla palmata LC 35398.42 0.42 0.34 11 261 (continued) Table 14.1 (continued) 262 Species IUCN Red List Geographic Biotic index (Bielby Risk factor (Ro¨dder ‘Amphibian Ark’ category range (km2) et al. 2008) et al. 2009) priority value Gephyromantis moseri LC 32336.51 0.42 0.34 0 Spinomantis aglavei LC 100698.02 0.38 0.34 0 Boophis hillenii DD 15039.61 0.53 0.35 8 Heterixalus betsileo LC 119509.18 0.4 0.35 0 Boophis albipunctatus LC 38429.16 0.42 0.36 0 Gephyromantis rivicola VU 9008.09 0.54 0.36 12 Boophis boehmei LC 69297.28 0.39 0.36 0 Spinomantis peraccae LC 112472.12 0.39 0.37 0 Boophis viridis LC 67939.83 0.39 0.37 0 Boophis pauliani LC 52880.96 0.42 0.37 0* Boophis andreonei VU 1426.18 0.92 0.38 12 Boophis reticulatus LC 67718.81 0.42 0.38 8 Mantidactylus tricinctus DD 2625.07 1 0.38 8 Heterixalus andrakata LC 1136.93 0.76 0.38 0 Guibemantis punctatus DD 11984.08 0.55 0.38 8 Dyscophus antongilii NT 5264.81 0.57 0.39 12 Boophis brachychir DD 6523.13 0.65 0.39 8 Gephyromantis asper LC 45983.06 0.45 0.4 0 Boophis goudotii LC 91462.49 0.45 0.4 0 Mantella madagascariensis VU 13914.61 0.44 0.4 22 Spinomantis fimbriatus LC 38131.65 0.47 0.41 0 Boophis miniatus LC 31016.03 0.47 0.42 0 Gephyromantis spinifer NT 28829.74 0.52 0.44 4 Dyscophus guineti LC 7135.13 0.64 0.44 8 S. L€ Mantidactylus aerumnalis LC 42295.77 0.46 0.45 0 Gephyromantis decaryi NT 18617.09 0.5 0.45 4 Mantella pulchra VU 17751.99 0.55 0.46 22 otters et al.  14 Stumpffia tridactyla DD 1213.56 0.84 0.46 8 Spinomantis bertini NT 25632.84 0.5 0.46 4 Boophis idae LC 37094.62 0.47 0.46 0 Boophis rappiodes LC 36815.13 0.47 0.46 0 Boophis marojezensis LC 46017.99 0.48 0.46 0 Mantidactylus zipperi LC 48425.56 0.28 0.47 0 Boophis pyrrhus LC 33584.44 0.52 0.47 0 Boophis luteus LC 42584.25 0.5 0.48 8 Boophis picturatus LC 32307.51 0.48 0.48 0 Platypelis tetra EN 6756.66 0.63 0.49 44 Mantidactylus argenteus LC 31887.43 0.51 0.5 0 Mantella baroni LC 37510.05 0.51 0.51 10 Boophis vittatus LC 12464.45 0.66 0.51 0 Hotspots, Conservation, and Diseases Gephyromantis striatus VU 4886.48 0.67 0.52 12 Spinomantis elegans VU 20489.63 0.55 0.52 12 Blommersia domerguei LC 58378.34 0.53 0.52 0 Boophis guibei LC 22625.61 0.52 0.52 0 Gephyromantis tandroka VU 10762.58 0.68 0.53 12 Heterixalus rutenbergi NT 52372.63 0.6 0.53 4 Plethodontohyla mihanika LC 37795.36 0.57 0.54 8 Mantella haraldmeieri VU 2385.73 0.8 0.56 22 Boophis bottae LC 19258.93 0.57 0.57 0 Guibemantis albolineatus DD 11994.87 0.68 0.57 8 Gephyromantis blanci NT 21342.83 0.6 0.58 4 Blommersia blommersae LC 20046.08 0.6 0.6 0 Gephyromantis salegy VU 2411.23 0.78 0.61 12 Boophis erythrodactylus LC 18211.35 0.66 0.62 0 Boophis rhodoscelis NT 22207.75 0.66 0.62 12 Boophis blommersae VU 1857.27 0.95 0.63 12 Boophis microtympanum LC 34738.96 0.63 0.63 0 263 (continued) Table 14.1 (continued) 264 Species IUCN Red List Geographic Biotic index (Bielby Risk factor (Ro¨dder ‘Amphibian Ark’ category range (km2) et al. 2008) et al. 2009) priority value Scaphiophryne marmorata VU 13186.32 0.64 0.64 25 Boophis majori NT 19767.4 0.65 0.65 4 Boophis ankaratra LC 35360.77 0.66 0.65 0 Gephyromantis ambohitra VU 1244.13 0.97 0.65 12 Boophis rufioculis NT 17871.18 0.65 0.65 4 Mantidactylus alutus LC 38272.38 0.65 0.65 0 Anodonthyla nigrigularis DD 2268.21 0.87 0.68 11 Gephyromantis klemmeri VU 3351.33 0.84 0.68 12 Spinomantis guibei EN 5984.76 0.8 0.7 44 Boophis englaenderi DD 636.04 0.95 0.72 8 Gephyromantis cornutus DD 10700.35 0.77 0.75 8 Mantella cowanii CR 287.38 0.76 0.76 51* Spinomantis brunae EN 1488.94 0.86 0.78 44 Boophis elenae DD 5154.56 0.78 0.78 8 Blommersia sarotra DD 6765.31 0.8 0.8 8 Madecassophryne truebae EN 4366.38 0.83 0.82 44 Gephyromantis tschenki DD 4583.14 0.83 0.83 8 Blommersia kely LC 10432.06 0.86 0.86 0 Gephyromantis eiselti DD 2225.89 0.91 0.91 8 Gephyromantis enki DD 1486.97 0.93 0.93 8 Boophis periegetes DD 1339.56 0.94 0.94 8 Plethodontohyla brevipes EN 580.73 0.95 0.95 44 Plethodontohyla tuberata VU 3895.6 0.96 0.96 20 Boophis laurenti DD 1076.87 0.96 0.96 8 S. L€ Boophis liami DD 1181.38 0.96 0.96 8 Gephyromantis thelenae DD 951.19 0.96 0.96 8 Boophis schuboeae DD 563.94 0.97 0.97 8 otters et al.  14 Boophis anjanaharibeensis DD 373.27 0.98 0.98 8 Mantidactylus madecassus EN 1431.53 0.32 0.98 44 Boophis solomaso DD 793.19 0.98 0.98 8 Platypelis alticola EN 1035.38 0.99 0.98 44 Boophis sibilans DD 697.25 0.98 0.98 16 Platypelis mavomavo EN 531.66 0.98 0.98 44 Boophis andohahela DD 345.61 0.99 0.98 8 Mantella aurantiaca CR 601.34 0.99 0.99 51* Spinomantis microtis EN 430.89 0.99 0.99 44 Guibemantis kathrinae DD 491.74 0.99 0.99 8 Boophis feonnyala DD 360.83 0.99 0.99 8 Anodonthyla montana VU 644.86 1 1 15 Anodonthyla rouxae EN 124.06 1 1 47 Hotspots, Conservation, and Diseases Boophis burgeri DD 163.84 1 1 8 Boophis mandraka DD 260.07 1 1 8 Boophis williamsi CR 407.99 1 1 56* Gephyromantis schilfi VU 27.03 1 1 12 Mantella crocea EN 259.96 1 1 26 Mantella milotympanum CR 56.19 1 1 46* Mantidactylus pauliani CR 286.38 0.41 1 56* Mantidactylus zolitschka DD 311.83 1 1 8 Scaphiophryne boribory EN 401.65 1 1 29 Stumpffia helenae CR 322.78 1 1 46* Mantidactylus delormei VU 2287.11 not dermined not dermined 0 Spinomantis massi VU 10054.3 not dermined not dermined 12 Spinomantis phantasticus LC 33914.54 not dermined not dermined 0 Aglyptodactylus madagascariensis LC 170824.75 not dermined not dermined 8 Anodonthyla moramora DD 12.19 not dermined not dermined 11 Boehmantis microtympanum EN 5701.03 not dermined not dermined 52 Boophis axelmeyeri NT 375.46 not dermined not dermined no data 265 (continued) Table 14.1 (continued) 266 Species IUCN Red List Geographic Biotic index (Bielby Risk factor (Ro¨dder ‘Amphibian Ark’ category range (km2) et al. 2008) et al. 2009) priority value Boophis lichenoides LC 64929.79 not dermined not dermined 0 Boophis sambirano VU 1275.25 not dermined not dermined 8 Boophis septentrionalis DD 1275.12 not dermined not dermined 8 Boophis tampoka EN 47.81 not dermined not dermined 0 Cophyla berara CR 10.24 not dermined not dermined 11 Cophyla occultans DD 3264.5 not dermined not dermined 8 Gephyromantis azzurrae EN 9.1 not dermined not dermined 0 Gephyromantis leucocephalus NT 12978.4 not dermined not dermined 4 Gephyromantis leucomaculatus NT 20951.6 not dermined not dermined 4 Gephyromantis runewsweeki EN 29.08 not dermined not dermined no data Gephyromantis sculpturatus LC 16525.55 not dermined not dermined 0 Gephyromantis ventrimaculatus LC 37318.97 not dermined not dermined 0 Guibemantis bicalcaratus LC 130407.16 not dermined not dermined 0 Guibemantis depressiceps LC 152182 not dermined not dermined 0 Guibemantis flavobrunneus LC 70833.65 not dermined not dermined 0 Guibemantis timidus LC 28279.98 not dermined not dermined 0 Heterixalus madagascariensis LC 33716.12 not dermined not dermined 0 Laliostoma labrosum LC 338436.67 not dermined not dermined 3 Mantella bernhardi EN 13903.13 not dermined not dermined 42* Mantella ebenaui LC 46818.91 not dermined not dermined 10 Mantidactylus albofrenatus DD 2070.17 not dermined not dermined 8 Mantidactylus ambohimitombi DD 295.14 not dermined not dermined 8 Mantidactylus bellyi LC 14334.86 not dermined not dermined no data Mantidactylus bourgati DD 1512.35 not dermined not dermined 0 S. L€ Mantidactylus brevipalmatus LC 43516.61 not dermined not dermined 0 Paradoxophyla tiarano DD 9.89 not dermined not dermined 11 Platypelis cowanii DD no data not dermined not dermined 8 otters et al.  14 Platypelis tsaratananaensis VU 25.52 not dermined not dermined 12 Platypelis tuberifera LC 66461.9 not dermined not dermined 0 Plethodontohyla angulifera DD no data not dermined not dermined 8 Plethodontohyla bipunctata LC 52478.29 not dermined not dermined 0 Plethodontohyla fonetana VU 16.89 not dermined not dermined 0 Plethodontohyla guentheri DD 3.34 not dermined not dermined 8 Plethodontohyla inguinalis LC 80354.97 not dermined not dermined 0 Plethodontohyla notosticta LC 903.56 not dermined not dermined 8 Scaphiophryne madagascariensis NT 23694.27 not dermined not dermined 7 Scaphiophryne menabensis NT 31618.63 not dermined not dermined 11 Scaphiophryne obscura DD no data not dermined not dermined 11 Scaphiophryne verrucosa DD 16.63 not dermined not dermined 11 Stumpffia roseifemoralis DD 1367.67 not dermined not dermined 8 Hotspots, Conservation, and Diseases Tsingymantis antitra VU 11.52 not dermined not dermined 8 267 268 S. L€ otters et al. 14.2 Amphibian Hotspot Madagascar: A Special Case 14.2.1 Anuran Megadiversity The amphibian fauna of Madagascar is highly exceptional both in terms of species diversity and endemism and is represented only by frogs (order Anura), whereas the two other orders (Caudata and Gymnophiona) are absent. With currently more than 270 described species and probably more than 200 still undescribed species (Vieites et al. 2009), Madagascar has the highest amphibian species diversity of all African countries and clearly ranks among the global amphibian hotspots. Diversity is concentrated in rainforests along the East coast mainly between sea level and 1,000 m above sea level and can locally reach over 100 species (Andreone et al. 2008), whereas the largely deforested central high plateau and the relatively dry western slopes harbor much fewer species (Glaw and Vences 2007). Impressively, 100% of the autochthonous species and 88% of the genera are naturally endemic to Madagascar and its inshore islands. Recent studies indicate that the degree of microendemism is much higher than formerly expected (e.g., Gehring et al. 2011), indicating that the currently recognized geographic range size of many species (Table 14.1) might be an overestimate that will strongly decrease when taxonomic progress (e.g., by increasing use of integrative taxonomy) will have deciphered the relationships within all species complexes. Furthermore, the con- tinuing high level of deforestation and fragmentation of primary forests will result in a further decrease of available habitats, population size, and density (Vallan 2000). All these factors increase the potential impact that can be expected from an outbreak of chytridiomycosis on this “micro-continent”. 14.2.2 Potential Impact of Chytridiomycosis Bd has so far not been detected in Madagascar despite reasonably thorough survey activities. At least, eight localities have been surveyed covering the three major biogeographical regions and a wide range of altitudes with more than 500 samples from 74 different species (Weldon et al. 2008). This implies that the pathogen may currently be absent on the island. How likely is Bd to enter Madagascar? The dispersal ability of this pathogen is indisputably high, even though mechanism and pathways are areas of ongoing debate and discovery (e.g., Fisher et al. 2009). Temporal patterns of spread have been studied in the Neotropics and show a possible annual range expansion of 25–282 km where the lower end of the spectrum is more prominent in well-studied areas (Lips et al. 2008). Importantly, the island status of Madagascar does not exclude the possibility of Bd invasion. There have been reports of Bd emergence on more than 20 islands all over the World including remote oceanic systems like Hawaii (http://www.spatialepidemiology.net/Bd-maps). This underlines the imminent danger 14 Hotspots, Conservation, and Diseases 269 that Bd may enter Madagascar. Because of the remoteness of this island, human- mediated spread may be the most likely pathway of introduction (Wollenberg et al. 2010). Possible modes of import include infected live or dead amphibians, fishes or contaminated water, moist substrates, or other fomites (i.e., via animal trade or accidental cointroduction with other imported products). Interestingly, on the Afri- can mainland, Bd has been detected widespread with high prevalence (Kielgast et al. 2010) and may hence potentially serve as a source of introduction. R€odder et al. (2009), in their ecological niche model, identified a major portion of Madagascar to be highly climate-suitable to the chytrid fungus. When comparing the areas within the island that are most suitable to the fungus with amphibian species richness, a remarkable spatial impact is evident (Fig. 14.1). When applying Fig. 14.1 Potential distribution of the amphibian chytrid fungus in Madagascar following the risk assessment of R€odder et al. (2009), re-projected from a resolution of 2.5 arc min to 30 arc sec. In this ecological niche model, warmer colors indicate a higher climatic suitability. Amphibian species richness as proposed by Kremen et al. (2008) is highlighted from lighter to darker stippling: >18, >36, >63 species per grid cell of 0.1 270 S. L€ otters et al. the rule set of predictions from the ecological niche and key host life history traits (R€odder et al. 2009), 40 of 186 considered Madagascan anurans (i.e. 234 recorded species less 48 for which no biotic index is available) exhibit risk factor 0.75 or higher (Table 14.1), including many species with small ranges from higher altitudes. We expect that the inclusion and exclusion of species has to be taken with care in some cases, as the risk factor of R€ odder et al. (2009) highly depends on the geographic range encompassed by a species. If this is remarkably small, this may merely reflect limited collections efforts in remote areas. If a species’ distribution is extraordinarily large, this can be the result of an unsolved taxonomy. However, with the current state of knowledge, it remains unanswered if emer- gence of chytridiomycosis on Madagascar will in fact lead to amphibian decline or extinctions. The only available empirical evidence of susceptibility comes from an outbreak in captive Tomato Frogs (Dyscophus antongilii) with high mortality rate (Oevermann et al. 2005) and, more recently, infection of captive Plethodontohyla tuberata (Une et al. 2008). Due to the long history of evolutionary isolation it may be expected that Madagascan amphibians naı¨ve to Bd will respond with drastic population declines. Comparing the potential risk imposed by Bd with the prioritization for Madagascan amphibians as proposed by the “Amphibian Ark”, there is little overlap (Fig. 14.2, Table 14.1). Particularly noteworthy is that “Amphibian Ark” 70 LC NT VU EN CR DD 60 'Amphibian Ark' priority value 50 40 30 20 10 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 risk factor Fig. 14.2 Relationship between IUCN Red List status, conservation priorities according to “Amphibian Ark” and corresponding risk factor, indicating threat of extinction by the emergence of chytridiomycosis, as proposed by R€odder et al. (2009). Data are taken from Table 14.1. IUCN Red List categories are: DD Data Deficient, LC Least Concern, NT Near Threatened, VU Vulnerable, EN Endangered, CR Critically Endangered (http://www.redlist.org) 14 Hotspots, Conservation, and Diseases 271 prioritized only 10 out of 242 species assessed for ex situ conservation measures. As indicated in Table 14.1, five of these are potentially exposed to a high risk of extinction through Bd following R€ odder et al. (2009), while for one (Mantella bernhardi) no risk factor was assessed. The four remaining are expected to be little affected by potential Bd introduction (Table 14.1). In conclusion, Madagascar’s unique and rich amphibian diversity may be expected to potentially suffer heavy consequences post Bd introduction and running ex situ conservation measures may not be sufficient to cushion the threat. 14.2.3 Recommendations for Conservation Strategies Even though the Madagascan species prioritized by “Amphibian Ark” represent a minute fraction of the island’s amphibian diversity at stake, the proportion of Madagascan species that are considered for conservation breeding, compared to other regions, e.g., African mainland, is considerably high (L€otters 2008). This illustrates a strong recognition of the importance of the distinctive Madagascan amphibian fauna (e.g., Andreone et al. 2008). Therefore, species prioritization by “Amphibian Ark”, in principle, has to be seen as a first step forward. In a subsequent step, we recommend a revision of the ‘Amphibian Ark’ prioritization considering the particular threat to species through chytridiomycosis (e.g., risk factor of R€ odder et al. 2009). Furthermore, additional short-term conservation breeding programs should be implemented in advance to develop breeding procedures and conditions for representatives of all major clades of Madagascan amphibians and representatives of the different reproductive modes (Buley et al. 2008). These “breeding manuals,” which might be obtained in close cooperation with qualified hobbyist breeders, would allow to establish captive breeding programs without delay in the expected case of Bd’s arrival in Madagascar. This preemergency prophylaxis might be a crucial measure, as Bd spread and population breakdowns may undergo within a couple of months only after Bd arrival (e.g., Lips et al. 2008), and it remains uncertain which species or taxonomic groups will be most heavily affected. For the same reason, following the interventions advocated through the IUCN Amphibian Conservation Action Plan (Gascon et al. 2007), there is a strong need to develop plans for “emergency response”. That is collecting, treating, quarantining, and subsequently using for breeding efforts as many as possible specimens per species and perhaps as many as possible localities where populations are observed to decline due to chytridiomycosis (Gascon et al. 2007). This may especially aim on the provision of funding and ex situ capacities and ad hoc coordination, permission, and transport facilitation. This may give reason to consider the establishment of an in-country conservation breeding center. Notewor- thy, emergency response can only be effectively established when species are monitored, i.e., both population and Bd status. As a preventive and certainly the “best” measure, we here stress the need for an increased attention to “biosecurity.” The implementation of quarantine measures 272 S. L€ otters et al. related to commercial trade in aquarium fishes and plants to prevent the accidental introduction of the amphibian chytrid fungus into Madagascar is in progress (Andreone et al. 2008). However, a more rigorous import risk assessment is still needed to create a basis for further strategies and possibly specific restrictions to mitigate the threat posed by Bd to Madagascar’s megadiverse anuran fauna. References Andreone F, Carpenter AI, Cox N, du Preez L, Freeman K, Furrer S et al (2008) The challenge of conserving amphibian megadiversity in Madagascar. PLoS Biol 6:943–946 Andreone F, Randriamahazo H (2008) Sahonagasy Action Plan. Conservation Strategies for the Amphibians of Madagascar. Mus Reg Sci Nat Torino, Torino, Italy Bielby J, Cooper N, Cunningham AA, Garner TWJ, Purvis A (2008) Predicting susceptibility to future declines in the World’s frogs. Conserv Lett 1:82–90 Blaustein AR, Romansic JM, Scheessele EA, Han BA, Pessier AP, Longcore JE (2005) Interspe- cific variation in susceptibility of frog tadpoles to the pathogenic fungus Batrachochytrium dendrobatidis. Conserv Biol 19:1460–1468 Blehert DS, Hicks AC, Behr M, Meteyer CU, Berlowski-Zier BM, Buckles EL et al (2009) Bat white-nose syndrome: an emerging fungal pathogen? Science 323:227 Buley K, Furrer SC, Garcia G, Gibson RC, Gili C, Mattioli F, Andreone F (2008) Captive breeding and zoo actions. In: Andreone F, Randriamahazo H (eds) Sahonagasy Action Plan. Conserva- tion Strategies for the Amphibians of Madagascar. Mus Reg Sci Nat Torino, Torino, Italy: 67–74 Daszak P, Cunningham AA, Hyatt AD (2000) Emerging infectious diseases of wildlife – threats to biodiversity and human health. Science 287:443–449 Fisher MC, Garner TWJ, Walker SF (2009) Global emergence of Batrachochytrium dendrobatidis and amphibian chytridiomycosis in space, time and host. Ann Rev Microbiol 63:291–310 Garner T, Stephen I, Wombwell E, Fisher M (2009) The amphibian trade: bans or best practice? EcoHealth 6:148–151 Gascon C, Collins JP, Moore RD, Church DR, McKay JE, Mendelson JR III (2007) Amphibian Conservation Action Plan. IUCN, Conservation International, Gland, Switzerland, Cambridge, UK Gehring PS, K€ohler J, Strauß A, Randrianiaina RD, Glos J, Glaw F, Vences M (2011) The kingdom of the frogs: Amphibian radiations in Madagascar. In: Zachos FE, Habel JC (eds) Biodiversity hotspots. Springer, Heidelberg Glaw F, Vences M (2007) A field guide to the amphibians and reptiles of Madagascar, 3rd edn. Vences and Glaw Verlag, Cologne, Germany Guisan A, Zimmermann N (2000) Predictive habitat distribution models in ecology. Ecol Modell 135:147–186 James TY, Litvintseva AP, Vilgalys R, Morgan JAT, Taylor JW, Fisher MC et al (2009) Rapid global expansion of the fungal disease chytridiomycosis into declining and healthy amphibian populations. PLoS Pathog 5:e1000458 Kielgast J, R€odder D, Veith M, L€ otters S (2010) Widespread occurrence of the amphibian chytrid fungus in Kenya. Anim Conserv 13:1–8 Kremen C, Cameron A, Moilanen A, Phillips S, Thomas C, Beentje H et al (2008) Aligning conservation priorities across taxa in Madagascar with high-resolution planning tools. Science 320:222 Lermen D, Bl€omeke B, Browne R, Clarke A, Dyce P, Fixemer T et al (2009) Cryobanking of viable biomaterials: necessities of new strategies for conservation purposes. Mol Ecol 18:1030–1033 14 Hotspots, Conservation, and Diseases 273 Lips KR, Diffendorfer J, Mendelson JR III, Sears MW (2008) Riding the wave: reconciling the roles of disease and climate change in amphibian declines. PLoS Biol 6:441–454 L€otters S (2008) African amphibians in zoos and aquariums: will they be on the ark? Int Zoo Yb 42:136–142 L€otters S, Kielgast J, Bielby J, Schmidtlein S, Bosch J, Veith M et al (2010) The link between rapid enigmatic amphibian decline and the worldwide emerging chytrid fungus. EcoHealth 6:358–372 McCallum H (2008) Tasmanian devil facial tumour disease: lessons for conservation biology. Trends Ecol Evol 23:631–637 McGregor Reid G, Zippel KC (2008) Can zoos and aquariums ensure the survival of amphibians in the 21st century? Int Zoo Yb 42:1–6 Mittermeier RA, Robles-Gil P, Hoffmann M, Pilgrim JD, Brooks TM, Mittermeier CG et al (2004) Hotspots revisited: Earth’s biologically richest and most endangered terrestrial ecoregions. CEMEX, Mexico City Murray K, Skerratt L, Speare R, McCallum H (2009) Impact and dynamics of disease in species threatened by the amphibian chytrid fungus, Batrachochytrium dendrobatidis. Conserv Biol 23:1242–52 Oevermann A, Schildger B, Feldman S, Robert N (2005) Chytridiomykose bei Tomatenfr€ oschen (Dyscophus antongilii) in der Schweiz. Tier€arztl Umsch 60:211–217 Rahbek C (2007) Disease ecology: the silence of the robins. Nature 447:652–653 Retallick RWR, McCallum H, Speare R (2004) Endemic infection of the amphibian chytrid fungus in a frog community post-decline. PLoS Biol 2:e351 R€odder D, Kielgast J, Bielby J, Schmidtlein S, Bosch J, Garner TWJ et al (2009) Global amphibian extinction risk assessment for the panzootic chytrid fungus. Diversity 1:52–66 R€odder D, Kielgast J, L€ otters S (2010) The potential of the emerging amphibian chytrid fungus under anthropogenic future climate change. Dis Aquat Org 92:201–207 Rollins-Smith L, Ramsey J, Reinert L, Woodhams D, Livo L, Carey C (2009) Immune defenses of Xenopus laevis against Batrachochytrium dendrobatidis. Front Biosci 1:68–91 Schloegel LM, Hero JM, Berger L, Speare R, McDonald K, Daszak P (2006) The decline of the sharp-snouted day frog (Taudactylus acutirostris): The first documented case of extinction by infection in a free-ranging wildlife species? EcoHealth 3:35–40 Schloegel LM, Ferreira CM, James TY, Hipolitos M, Longcore JE, Hyatt AD et al (2010) The North American bullfrog as a reservoir for the spread of Batrachochytrium dendrobatidis in Brazil. Anim Conserv 13:53–61 Smith K, Lips K, Chase J (2009a) Selecting for extinction: nonrandom disease-associated extinc- tion homogenizes amphibian biotas. Ecol Lett 13:1069–1078 Smith KF, Acevedo-Whitehouse K, Pedersen AB (2009b) The role of infectious diseases in biological conservation. Anim Conserv 12:1–12 Stuart SN, Hoffmann M, Chanson JS, Cox NA, Berridge RJ, Ramani P et al (2008) Threatened amphibians of the world. Lynx Ed, Barcelona, Spain Une Y, Kadekaru S, Tamukai K, Goka K, Kuroki T (2008) First report of spontaneous chytridiomycosis in frogs in Asia. Dis Aquat Org 82:157–160 Vallan D (2000) Influence of forest fragmentation on amphibian diversity in the nature reserve of Ambohitantely, highland Madagascar. Biol Conserv 96:31–43 Vieites DR, Wollenberg KC, Andreone F, K€ ohler J, Glaw F, Vences M (2009) Vast underestimation of Madagascar’s biodiversity evidenced by an integrative amphibian inven- tory. Proc Nat Acad Sci USA 106:8267–8272 Voyles J, Young S, Berger L, Campbell C, Voyles W, Dinudom A et al (2009) Pathogenesis of chytridiomycosis, a cause of catastrophic amphibian declines. Science 326:582 Weldon C, Du Preez L (2008) Managing emerging amphibian diseases. In: Andreone F, Randriamahazo H (eds) Sahonagasy Action Plan. Conservation Strategies for the Amphibians of Madagascar. Mus Reg Sci Nat Torino, Torino, Italy:34–41 274 S. L€ otters et al. Weldon C, du Preez L, Vences M (2008) Lack of detection of the amphibian chytrid fungus (Batrachochytrium dendrobatidis) in Madagascar. Monogr Mus Reg Sci Nat Torino 45:95–106 Wollenberg KC, Jenkins RKB, Randrianavelona R, Ralisata M, Rampilamanana R, Ramanandraibe A et al. (2010) Raises awareness of amphibian chytridiomycosis will not alienate ecotourists visiting Madagascar. EcoHealth 7:248–251 

Qatar Islamic Archaeology and Heritage Project End of Season Report Environmental Studies 2011-2012 Qatar Islamic Archaeology and Heritage Project End of Season Report, Environmental Studies 2011-2012  The executive and staff of the Qatar Islamic Archaeology and Heritage Project acknowledge and appreciate the perceptive leadership and considerable support of: Her Excellency Sheikha Al Mayassa Bint Hamad Bin Khalifa Al Thani Chairperson of the QMA His Excellency Sheikh Hassan Bin Mohammad Bin Ali Al Thani Vice-Chairperson of the QMA  Acknowledgements Mr. Mansoor Al Khater Board of Trustees and Former CEO of Qatar Museums Authority Engineer Abdullah Al Najjar Former CEO of Qatar Museums Authority Professor Sultan Muhesen Director, Department of Archaeology and Heritage, Qatar Museums Authority Mr. Faysal Al Na‘imi Head of Archaeology Section, Qatar Museums Authority Mr. Sayf Al Na‘imi Archaeology Section, Qatar Museums Authority Mr. Søren Frank & Mr. Steffen Sandvig Bach Maersk Oil Qatar Mr. Mohammed Y. Al Jaidah & Dr. Nayla Beyrouthi Ministry of Environment, Qatar Professor Friedhelm Krupp Director, Qatar Natural History Museum Mr. Khalid Hassan Al Jaber Head of Natural History, Qatar Museums Authority The QIAH 2011-2012 Team Members, Environment: Pernille Bangsgaard, Jakob Walløe Hansen, Anne Majgaard Jensen, Aslak Jørgensen, Jos Kielgast, Reinhardt Møbjerg Kristensen, Jeppe Møhl, Peter Rask Møller, Phillipe Provencal, Ken Puliafico. Contributing Authors: Pernille Bangsgaard, Jakob Walløe Hansen, Anne Majgaard Jensen, Aslak Jørgensen, Jos Kielgast, Reinhardt Møbjerg Kristensen, Nadja Møbjerg, Jeppe Møhl, Peter Rask Møller, Dennis K. Persson, Ken Puliafico, Gitte Petersen, Phillipe Provencal, Ole Seberg. Editors: Pernille Bangsgaard, Reinhardt Møbjerg Kristensen, Peter Rask Møller & Hanne Nymann Cover: Alexis Pantos First Revised Edition University of Copenhagen and Qatar Museum Authority 2012 CONTENTS 1. BIOLOGICAL SURVEY OF THE AL ZUBARAH, FIELD WORK SEASON 2012 .... 3 1.1 Introduction ................................................................................................................. 3 1.2 The marine environment ............................................................................................. 3 1.3 Findings within the buffer zone .................................................................................. 4 Meiofauna .............................................................................................................................. 4 Fish ......................................................................................................................................... 4 Molluscs ................................................................................................................................. 5 Other taxa ............................................................................................................................... 5 1.4 The terrestrial environment ......................................................................................... 5 Herpetology............................................................................................................................ 5 Entomology ............................................................................................................................ 6 1.5 Other areas visited (Coral Heads, Düvel Rock, Mangrove) ........................................ 6 Coral Heads ............................................................................................................................ 6 The small mangrove at Al Jumail .......................................................................................... 7 1.6 Conclusions and recommendations ............................................................................. 7 2. THE SEAGRASSES OF AL ZUBARAH ....................................................................... 11 2.1 Introduction ............................................................................................................... 11 2.2 Seagrasses of the Arabian Gulf ................................................................................. 11 2.3 Al Zubarah................................................................................................................. 12 3. THE MARINE TARDIGRADES AT AL ZUBARAH ................................................... 13 3.1 Introduction ............................................................................................................... 13 3.2 Material and Methods................................................................................................ 14 3.3 Results ....................................................................................................................... 14 3.3.1 Taxonomic account ............................................................................................ 15 3.4 Findings within the buffer zone ................................................................................ 18 3.5 Conclusions and recommendations ........................................................................... 18 4. THE RECENT MARINE MOLLUCS AT AL ZUBARAH ........................................... 25 4.1 Introduction ............................................................................................................... 25 4.2 Findings within the buffer zone ................................................................................ 25 4.3 Conclusions and recommendations ........................................................................... 28 1 5. ENTOMOLOGY ............................................................................................................. 31 5.1. Introduction ................................................................................................................... 31 5.2 Methods.......................................................................................................................... 31 6. FISHES ............................................................................................................................ 46 6.1 Introduction ............................................................................................................... 46 6.2 Methods ..................................................................................................................... 46 6.3 Findings within the buffer zone ................................................................................ 47 6.4 Species List ............................................................................................................... 47 6.5 Findings, other locations ........................................................................................... 52 6.6 Conclusions and Recommendations.......................................................................... 53 7. NAMES OF FISH AND OTHER MARINE ANIMALS IN QATAR ............................... 62 7.1 Introduction .................................................................................................................... 62 7.3 Main conclusions from the work undertaken this season .............................................. 64 7.4 Future research ............................................................................................................... 64 8. REPTILES .......................................................................................................................... 68 8.2 Findings within the buffer zone ..................................................................................... 68 8.2.1 species list ............................................................................................................... 71 8.3 Conclusions and recommendations................................................................................ 72 9. PALEONTOLOGY ............................................................................................................ 79 9.1 Introduction .................................................................................................................... 79 9.2 Findings within the buffer zone at Al Zubarah .............................................................. 79 9.2.1 Species list .............................................................................................................. 80 9.3 Findings at Umm Bab, Dukhan Region, Southern Qatar .......................................... 80 9.4 Conclusions and recommendations ........................................................................... 81 10. THE COMPARATIVE COLLECTION OF ANIMAL SKELETONS ........................... 87 10.1 Introduction .................................................................................................................. 87 10.2 The process of cleaning a skeleton .............................................................................. 87 10.3 The collection............................................................................................................... 88 10.4 Other work processes ................................................................................................... 89 10.5 Conclusion and recommendations ............................................................................... 89 2 1. BIOLOGICAL SURVEY OF THE AL ZUBARAH, FIELD WORK SEASON 2012 Pernille Bangsgaard, Reinhardt Møbjerg Kristensen & Peter Rask Møller. (The Natural History Museum of Denmark, University of Copenhagen, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark) 1.1 INTRODUCTION Al Zubarah Archaeological Site is a rare intact example of a pearl fishing and trading town from the 18th-19th centuries. Moreover, the area around the site in NW Qatar has seen little modern urban development and as such showcases the flora and fauna typical of the Arabian Gulf desert and semi-desert ecoregion. As part of the on-going efforts to inscribe Al Zubarah Archaeological Site on the UNESCO World Heritage List, research into the natural environment was initiated in 2011. In March 2012 three weeks of field work was carried out by a team of researchers from the Natural History Museum in Copenhagen, in cooperation with the QMA and the Qatar Ministry of Environment. The aim was to investigate the natural environment around the archaeological site and surrounding areas as well as explore the potential for future research project. This report presents the findings from the field work, summarizing the results and highlighting the potential for further research in a number of areas based on the natural environment of NW Qatar and the country as a whole. 1.2 THE MARINE ENVIRONMENT Several marine expeditions have worked in the Arabian Gulf. Most famous is the Danish Arabian Voyage to the Arabian Peninsula in 1761-67, which included zoologist Peter Forsskål, whose descriptions of many marine plants, invertebrates and fishes were published posthumously in 1775 (Niebuhr 1775). The book contains the first descriptions of the marine environment and the marine fauna of the Arabian Seas. In the following two hundred and fifty years, many scientific expeditions have worked in the Arabian Gulf including the Danish fishery investigations of 1937-38 (Blegvad & Løbbentin 1944) and the German oceanographic expedition with the vessel Meteor in 1965. A comprehensive FAO guide to the marine living resources of the southern gulf was published in 1997 (Carpenter et al. 1997). But very little scientific biological sampling has been done in the marine waters of Qatar, and no information was previously available on the biology of the Al Zubarah buffer zone. The present end of season report, therefore presents the very first data from the area. The main marine faunas investigated in 2012 were: Meiofauna (mainly Tardigrades), Ichthyofauna (fishes), Malacofauna (molluscs), Botany (sea grasses). The buffer zone contains several habitats, including sea grass beds, sand, mud, rocks, sea weed and biogenous reefs of e.g. peal oysters (Figure 1.1). Most distinctively the sea water 3 around Qatar has an extremely high salinity - more than 45‰ compared with the 35‰ at the entrance of Hormuz Strait (Sheppard et al. 2010, Uchupi et al, 1999). Such hypersaline environments has previously been described from Shark Bay; Western Australia. Closer to Qatar, several scientific teams have worked on sedimentary processes (Basson et al. 1981) and the oil company ARAMCO, operating in Eastern Saudi Arabia have investigated the marine environment and particularly the meiofauna (Hansen et al. 2006). The materials contained a very rich tardigrade fauna with a large proportion of currently undescribed species. The most surprising results were the hyper saline environments attested by the sediment samples. Therefore, the salinity as an ecological factor was a priority during the field work season at Al Zubarah. The results clearly show that the salinity varies strongly depending on the tidal rhythm. The maximum tide differentials were seen on the 11 March 2012, at about 1.2 meters. The highest salinity was measured in a small lagoon close to the Old Pier (Figure 1.2). The salinity in the sediment was 58.0 ‰ at low tide – at high tide the salinity was 48.5 ‰. The temperature at the sea in the buffer zone was only 16°C, but in the sediment at low tides the temperature (11 March) was easily rising to 22°C. The buffer zone appears to be in a healthy environmental shape and only few signs of human impact were documented. No oil spills were observed in the water, in contrast to some areas on shore. A few lost gill nets were seen, as well as an unmarked fish trap. A transect of sediment samplings was set up (5 March to 11 March 2012) at the Old Pier at Al Zubarah to investigate the meiofauna and to obtain data on the high salinity tolerance of different invertebrate groups in the area. The sediment consists of very fine coral sand with a lot of sulphur bacteria (strong smell of sulphur). At the end of the Old Pier a dense grass bed of sea grasses of the species Halophila stipulacea and Halodule uninervis exists. The buffer zone fish fauna was investigated from 6-23 March 2012, by multi-mesh gill-net (one night), baited traps (3 nights), and snorkelling with camera and spear gun (3 nights, 2 days) and scuba diving (3 day dives). 1.3 FINDINGS WITHIN THE BUFFER ZONE Meiofauna In the sediment of the sea grasses a rich tardigrade fauna was found. Furthermore, lot of barnacles, polychaetes and molluscs were collected in the tidal zone (Kristensen et al. 2012, Chapter 3). Together with sediment and sea grass samples the tardigrade fauna was investigated. Finally the tardigrade fauna related to the invertebrates was sorted by fresh water shocking. Four new species of tardigrades were found at Al Zubarah. Two new species (Pseudostygarctus nov. sp. and Echiniscoides nov. sp.1) of tardigrades were found in the tubes of the polychaete Pomatoleios kraussii and two new species of tardigrades (Echiniscoides nov. sp. 2 and 3) were found in the bysus threads of the mytilid mussel Brachidontes variabilis. Fish A total of ca. 48 identified fish species belonging to 33 families were observed, photographed and/or caught within the buffer zone (Møller et al. 2012, Chapter 6). The most effective method was night snorkelling and most species were observed near Ras Ushayriq. 4 The most diverse families were seabreams (Sparidae) with 5 species, followed by stingrays (Dasyatidae), Blennies (Blennidae) and emperors (Lethrinidae) all with 3 species. Two species of stingray (Himantura gerrardi and Himantura uarnak are considered Vulnerable (VU), a shark Chiloscyllium arabicum and a labrid Halichoeres leptotaenia are considered Near threatened (NT), a butterflyfish Chaetodon nigropunctatus and an angelfish Pomacanthus maculosus are placed in the Least Concern (LC) category, whereas another stingray Pastinachus sephen, a flathead Platycephalus indicus and a grouper Epinephelus tauvina are currently Data Deficient (DD). Molluscs In total 40 species of molluscs were reported from within the UNESCO Buffer Zone (Jørgensen 2012, Chapter 4): Gastropoda 31, Cephalopoda (1), Scaphopoda (1), Bivalvia (7), Polyplacopora (1) The registration of four genera of ellobiid snails previously not reported from Qatar is the single most important finding in the 2012 field work. Other taxa A dense population of the blue swimming crab (Portunus pelagicus, Figure 1.3) was observed on the sandy shores at high tide and several were caught in our traps. The Common Bottlenose Dolphin (Tursiops truncatus) was seen at several occasions, hunting in the zone, close to Al Zubarah. A single Indian humpback dolphin Sousa plumbea was also observed in the buffer zone. Sea turtles, a single young Hawksbill (Eretmochelys imbricate) was observed in the water, but not on shore, except for one large dead Green turtle (Chelonia mydas) at Al Jumail village and several of the same species at Al Zubarah Beach. Dugongs or sea cows (Dugong dugon) were not seen alive, however, dead dugongs were found and the skulls (Figure 1.4) were prepared for the museum in Doha. The Greater Flamingo (Phoenicopterus roseus) was observed close to The Old Pier, Al Zubarah in nearly every day in flock up till 30 individuals. This bird is an abundant winter visitor in Qatar. 1.4 THE TERRESTRIAL ENVIRONMENT The main habitat types in the area are the sabkha and the stony desert. There is a well vegetated transition between the two, but patches of shrub and grass vegetation are also scattered around other parts of the area, although it is generally sparsely vegetated. Moreover there are a few natural stony ridges and several places with scattered piles of recent or historical building debris. The archaeological site itself makes up an important habitat for many species. A preliminary terrestrial survey of the Buffer Zone was conducted in 2010 (Kielgast & Hansen 2010). Herpetology A herpetological survey was conducted in the Al Zubarah archaeological site and buffer zone during 5th-23rd of March 2012. Activities were targeted at supplying a checklist as complete as possible for the local reptile fauna of the area. Due to the limited size of the buffer zone and the relatively long time frame of the survey it was possible to investigate all areas at a fine spatial scale with at least one visit during day and night. The survey activities were thereafter targeted at the most promising localities of the respective habitat types in the area. 5 A total of 17 reptile species belonging to eight different families were recorded during the survey in March 2012 (Kielgast 2012, Chapter 8). All species are widespread in the region and their occurrence in the buffer zone of Al Zubarah archaeological site is neither surprising nor unique. Entomology The entomological survey found about 170 species of terrestrial arthropods (Puliafico and Jensen 2012, Chapter 5). The most species were insects; however, species of scorpions (Androctonus crassicauda) and wind scorpions (Solifugids) were recorded. The beetle fauna was a dominating element, and several darling beetles (Tenebrionids) were observed running on the hot sands during daylight hours at Al Zubarah. Several night-active moths (Lepidoptera) were observed at the end of the season. The timing was probably due to the night temperature, which were too cold for all kind of flying insects in March 2012. 1.5 OTHER AREAS VISITED (CORAL HEADS, DÜVEL ROCK, MANGROVE) Coral Heads The maritime location (26°02.447’N and 50°53.254’E) east of the Al Zubarah Buffer Zone were visited by speedboat on the 12th of March 2012. The area is a very large low water coral reef. Unfortunately most of the corals are dead today, and very few fish were seen. A part of the reef is not submerged at low tide. Here many resting Socotra Cormorant (Phalacrocorax nigrogularis) were observed. Between the dead corals a dense population of sea grass was present. Samples of coral sand were taken from 2-4 m depths by SCUBA-diving and from 6 to 10 m depths with a mini van Veen grab. The salinity was relatively low 45‰ in the water and 48‰ in the sediment. A dead coral bloc was taken back to Al Zubarah and crushed. Lots of boring mussels and crustaceans were found inside the dead coral. The tardigrade fauna was rich. The sediment samples are not sorted out yet; however, the samples contain at least five species of tardigrades and a species of kinorhynchs (Mud dragon). Düvel Rock The maritime location known as Düvel Rock (26°16.717’N and 50°58.754’E) northeast of Al Zubarah was visited by speedboat on the 12th of March 2012. The area is a true coral reef with a Blue Lagoon in the middle. The corals were alive, but were largely overgrown by algae. The crew members told us that in the old days ships would visit the locality because of the upwelling fresh water, similar to the fresh water springs around Bahrain. We could not locate the wells in the lagoon; however, the salinity was low in the sea water (45 ‰) and only 44‰ in the sediment from the lagoon. The current was extremely strong at the edge of the reef and very weak in the “Blue lagoon”. The sediment in the lagoon was very fine and smelled strongly of sulphur. A small sea turtle, the Hawksbill (Eretmochelys imbricate) was caught by the divers. The turtle was entirely covered by large barnacles (Figure 1.5). These are currently being investigated for tardigrades and for molecular studies. 6 The small mangrove at Al Jumail The area near Al Jumail village (26°05.717’N and 50°09.401’E) was visited on the 5th of March 2012. The salinity was extremely high (58‰) at low tide. The mangrove has a muddy sediment, but also a lot of stones. The stones were nearly covered with barnacles. Furthermore a dead Green turtle with barnacles was found at this locality. The barnacles from both localities are still being investigated for tardigrades. 1.6 CONCLUSIONS AND RECOMMENDATIONS The recommendations for the preservation of the natural environment of the buffer zone are to continue the marine biological research in the buffer zone with investigations of Molluscs, Fishes, Polychaetes and Crustaceans in more seasons in order to get a more complete overview of the fauna. Furthermore, a full investigation of The Canal of Al Zubarah is very relevant (Figure 1.6). This fantastic man-made canal has a very high salinity (>60 ‰). Lots of crustacean (crab) made holes are seen in the walls and perhaps it is also functioning as a nursery area for juvenile mullets (Mugilidae) The carbonated sediment dug out from the canal still exists as very large mounds close to it. This sediment consists of sub fossil marine snails (Figure 1.6). The marine meiofauna in the canal was not investigated in 2012; however, this investigation has first priority in 2013. The full investigations of the terrestrial mammals and birds in the Al Zubarah Archaeological Site are also recommended. 7 References Basson, P.W., Burchard, J.E., Hardy, J. T., & Price, R. G. 1981. Biotopes of the Western Arabian Gulf, Marine Life and Environments of Saudi Arabia. ARAMCO, Second Edition. Shenval “80”, Harlow, Great Britain. 284 pp. Blegvad, H. & Løppenthin, B. 1944. Fishes of the Iranian Gulf. Einar Munksgaard. Carpenter, K.E., Krupp, F., Jones, D.A. & Zajons, U. 1997. The living marine resources of Kuwait, Eastern Saudi Arabia, Bahrain, Qatar & the United Arab Emirates. Rome : FAO. Glennie, K.W. 1998. The desert of southeast Arabia: A product of Quaternary climate change. In Alsharhan, A.S., Glennie, K.W., Whittle, G.L., Kendall, C. St.G., (Eds.) Quaternary Deserts and Climatic Change. (pp. 279-291). A. A. Balkema, Rotterdam. Niebuhr, C. (Ed.) 1775. Descriptiones Animalium. Haunia. Sheppard, C., Al-Husiani, M., Al-Jamali, F., Al-Yamani, F., Baldwin, R., Bishop, J., Benzoni, F., Dutrieux, E., Dulvy, N. K., Durvasula, S. R. V., Jones, D. A., Loughland, R., Medio, D., Nithyanandan, M., Pilling, G. M., Polikarpov, I., Price, A. R. G., Purkis, S., Riegl, B., Saburova, M., Namin, K. S., Taylor, O., Wilson, S., Zadinal, K. (2010). The Gulf: A young sea in decline. Marine Pollution Bulletin 60: 13-38. Uchupi, E., S.A. Swift & D.A. Ross. 1999. Late Quaternary stratigraphy, palaeoclimate and neotectonism of the Persian (Arabian) Gulf region. Marine Geology, 160, 1–23. Walmsley, A., Barnes, H., & Macumber, P. 2010. Al-Zubārah and its hinterland, north Qatar: excavations and survey, spring 2009. Proceedings of the Seminar for Arabian Studies 40: p.55–68. 8 Figures Figure 1.1: Pearl oysters at Ras Ushayriq Figure 1.2: Old Pier and the coral beach, 11 March, 2012. Al Zubarah. Figure 1.3: The blue swimming crab (Portunus pelagicus). 9 Figure 1.4: Skulls from Green turtle (Chelonia mydas) and skull from dugong (Dygon dugon) found on the beach. Figure 1.5: Düvel Rock. Collecting barnacles on the cement piles and on the Hawksbill, 12 March, 2012, North of Qatar. Figure 1.6: The Canal of Al Zubarah and the carbonate sediment dug out from it. 10 2. THE SEAGRASSES OF AL ZUBARAH Gitte Petersen and Ole Seberg (Natural History Museum of Denmark, gittep@snm.ku.dk and oles@snm.ku.dk) 2.1 INTRODUCTION Worldwide only some 60 species of flowering plants, collectively named the seagrasses, have reinvaded the marine environment, occuring along the coastline, on reef tops and in estuaries. They all belong to the same order of monocotyledons, the Alismatales, a group which also including many well-known fresh water plants. The seagrasses belong to five different families, which do not form a monophyletic group. Thus, the ability to tolerate the harsh salinity of the ocean has evolved more than once. Seagrasses usually form beds of vegetation in coastal, shallow watered areas, where they have an immense ecological importance. They provide the habitat and breeding ground for fish and shellfish and serve as feeding grounds for marine mammals and turtles. Thus, in the Arabian Gulf the presence of seagrasses is essential to the populations of green sea turtles (Chelonia mydas) and to the endangered dugong (Dugong dugon). Seagrasses are also important for stabilizing the ocean bottom, which would otherwise be vulnerable to intense wave action from currents and storms. Additionally, seagrasses trap the finer sediments and thereby filter the water increasing its clarity and filters nutrients that are washed out to sea and might otherwise be a treat to sensitive habitats such as coral reefs. 2.2 SEAGRASSES OF THE ARABIAN GULF Compared to the Arabian Sea, the seagrass communities of the Gulf are species poor – probably due to high salinity of the waters within the Gulf area. But a very high temperature variation (from 10°C to 39°C) may be another limiting factor for some species. Oil-related pollution has so far not been recorded as having negative effects on seagrass populations in the Gulf. Thus, dredging and land reclamation seem the major threat to seagrasses in the Gulf. In the Arabian Gulf, the seagrass communities are composed of three species: Halophila ovalis, Halophila stipulacea (both Hydrocharitaceae), and Halodule uninervis (Cymodoceaceae). The two former, Halophila ovalis and Halodule uninervis, are both rather common and widely distributed along the coastline from East Africa to the Pacific. Halophila stipulacea on the other hand has a more restricted distribution along the coast line of East Africa, the Arabian Peninsula, the eastern Mediterranean, and isolated occurrence along the coast of Southeast India. 11 2.3 AL ZUBARAH Along the coast of Al Zubarah all three species have been recorded and collected, though species delimitation in Halophila is uncertain. The collected material include DNA samples, which are useful both for studies of phylogenetic relationships of the seagrasses and other Alismatales and for the purpose of DNA barcoding seagrasses worldwide. The latter will enable rapid identification of all seagrass species even for non-specialists. Figure 2.1: Halophila stipulacea (with broad leaves) mixed with Halodule uninervis (with narrow leaves). Near Ras Ushayriq, immediately south of Al Zubarah Figure 2.2: Halodule uninervis. Near Ras Ushayriq, immediately south of Al Zubarah 12 3. THE MARINE TARDIGRADES AT AL ZUBARAH Reinhardt Møbjerg Kristensen¹, Dennis K. Persson¹,², Nadja Møbjerg² & Aslak Jørgensen3. (¹Invertebrate Department, The Natural History Museum of Denmark, University of Copenhagen, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark, rmkristensen@snm.ku.dk, dpersson@snm.ku.dk. ²Department of Biology, August Krogh Centre, University of Copenhagen, Universitetsparken 13, DK-2100 Copenhagen Ø, Denmark, nmobjerg@bio.ku.dk. 3Laboratory of Molecular Systematics, The Natural History Museum of Denmark, University of Copenhagen, Sølvgade 83, DK-1307 Copenhagen K, Denmark, aslak@snm.ku.dk). 3.1 INTRODUCTION The microscopic water bears (Tardigrada) have a global distribution with terrestrial and limnic species primarily inhabiting mosses and lichens, whereas marine species inhabit various sediments. Only about 1100 tardigrade species have been described (Guidetti & Bertolani 2005, Degma et al. 2010); however, it is expected that many more yet undescribed species exist. These small animals are well-known for their ability to survive large fluctuations in abiotic environmental conditions (Møbjerg et al. 2011). As a response to environmental stress some tardigrades enter cryptobiosis; a state in which metabolism is reversibly switched off. Some of these tardigrades found within the marine genus Echiniscoides, live on barnacles, mussels and polychaete tubes in the intertidal zone throughout the world. Echiniscoides could be among the toughest animals on this planet; however, detailed information on their tolerance to environmental extremes is lacking. The four species of Echiniscoides in Al Zubarah represent the highest diversity reported from any locality (Kristensen & Hallas 1980, Hallas & Kristensen 1982, Miller & Kristensen 1999). The high diversity may reflect that this investigation comprised three different substrates in Al Zubarah, whereas most other published investigations of Echiniscoides usually have focused solely on barnacles. Detailed information is also lacking regarding the biodiversity of marine tardigrades. Currently only app. 150 species of marine tardigrades have been described (Degma et al. 2010; Guidetti & Bertolani; Hansen 2011). However, many species and genera new to science await description in various museum collections particularly at the Natural History Museum of Denmark (SNM). The major reasons for this lack in detailed information on biodiversity are that the marine tardigrades are small, relatively rare, difficult to extract from sediment samples and they require expert knowledge to identify to species (Jørgensen et al. 2010). The ARAMCO Northern Area Intertidal Sampling Program in 1982 included 22 sediment samples collected in the Arabian Gulf of Saudi Arabia. These samples revealed a rich tardigrade fauna. In total, 12 species of heterotardigrades were found from four intertidal sediment localities containing three new interstitial species of Arthrotardigrada. The three new species comprise two species of Megastygarctides (Hansen & Kristensen 2006) and one species of Pseudostygarctus (Hansen et al. in press). Our preliminary investigations at Al Zubarah have revealed four species of tardigrades new to science. However, every identified species from Qatar will be a new record from the 13 country and add to the growing knowledge of the biodiversity of Qatar. Once the biodiversity of the marine tardigrade fauna has been determined it will be possible to make comparisons to the fauna of Saudi Arabia (Hansen & Kristensen 2006, Hansen et al. in press), the Mediterranean Sea (Accogli et al. 2011, D’Addabbo Gallo et al. 1987,2000, 2001, Grimaldi de Zio et al. 1998), and the Faroe Bank, North Atlantic Ocean (Hansen et al. 2001). Only a few localities have been thoroughly investigated with regard to the biodiversity of marine tardigrades. Of these the faunae of the Mediterranean Sea and the Faroe Bank (North Atlantic Ocean) are the best known. Hence a thorough investigation of the marine tardigrade fauna of Qatar will greatly increase the current knowledge regarding marine tardigrade faunae. The new investigation in Qatar should focus on the following 3 questions: How many marine tardigrade species live in Qatar? Description of the marine tardigrade species new to science from Qatar. Is the marine tardigrade fauna of Qatar similar to faunae known from other places in the world, especially Saudi Arabia? 3.2 MATERIAL AND METHODS A transect of sediment samplings was set up at the Old Pier at Al Zubarah (Figure 3.1) and samples were collected at Coral Head and Düvel Rock using a mini van Veen grab handled by SCUBA divers. The sediment samples were freshwater shocked at the camp in Al Zubarah and plankton nets (30-60 µm) were used to collect the animals from the samples. Large clumps of annelid tubes (figure 3.3), byssi of mussels (Figure 3.2) and several species of barnacles were collected and sun-dried in coffee-filters. Later on, these samples were freshwater shocked to extract the marine tardigrades from the samples. Good stereo- and compound microscopes were needed for sorting of the sediment samples and identification/description of species. Finally, two species of Echiniscoides were critical point dried, transferred to aluminum stubs, coated with palladium and examined using a JEOL JSM-6335-F scanning electron microscope. A trained and skillful scientific illustrator is still needed for preparing illustrations of the marine tardigrade species new to science. 3.3 RESULTS Currently 11 species of marine Heterotardigrada have been sorted out of the samples from Qatar. These species are: 1). Pseudostygarctus nov. sp. (Old Pier, Polychaete tubes of Pomatoleios kraussii) 2). Batillipes similis (Old Pier, coral sand mixed with sea grasses from 2 meter water depth) 3). Batillipes pennaki (Old Pier, coral sand, low tide) 14 4). Halechiniscus nov. sp. (Coral Head) 5). Wingstrandarctus sp. (Coral Head) 6). Florarctus sp. 1. (Old Pier, 2 meter water depth) 7). Florarctus sp. 2. (Düvel Rock, 1 meter water depth) 8). Echiniscoides sigismundi cfr. (Barnacles, Ras Ushayriq, high tide) 9). Echiniscoides nov. sp. 1 (Old Pier, Polychaete tubes of Pomatoleios kraussii) 10). Echiniscoides nov. sp. 2 (Beach Rock of Al Zubarah, from the mussel Brachidontes variabilis, Mytilidae) 11). Echiniscoides nov. sp. 3 (Beach Rock of Al Zubarah, from the mussel Brachidontes variabilis, Mytilidae). 3.3.1 Taxonomic account Order: Arthrotardigrada Diagnosis: With a median cirrus located on the head. Family Stygarctidae Schulz, 1951 Genus Pseudostygarctus McKirdy, Schmidt & McGinty-Bayly, 1976 Diagnosis: Marine tardigrades always with plates. Stygarctidae with dorsal cuticle thickened to form semicircular head plate, three body plates and two intersegmental plates. A caudal plate devoid the spike-like processes. Cephalic appendages complete; however, the secondary clavae modified as semi-globular structures; cirrus E attached to the body by means of a “ball and socket”-type articulation. Each leg terminated by two or three claws. The seminal receptacles with internal cuticular bars and external cuticular pockets containing spermatozoa. Type species: Pseudostygarctus triungulatus McKirdy, Schmidt & McGinty-Bayly, 1976 from Galapagos Islands. Pseudostygarctus nov. sp. from Al Zubarah, Qatar. Diagnosis: A medium sized Pseudostygarctus with two beautiful crown-shaped intersegmental plates. The lateral processes of the three segmental plates only with weakly developed sheets. A single clover-shaped ventral plate between the head and the first trunk segment. The four pairs of strongly telescopic legs with three claws on each tarsus. The bucco-pharyngeal apparatus with very long stylets, the placoides in the pharyngeal bulb with so-called internal stylet supports. True stylet supports are lacking. 15 Type material: Holotype, an adult female and 3 paratypes (all females) collected inside the polychaete tubes (Pomatoleios kraussii), the 11.of March, 2012 at the Old Pier, Al Zubarah, Qatar (25°58.347’ N and 51°01.121’ E). The polychaete tubes were strongly adhered to the stones of the pier at high tide level. Temperature: 18.3° C. Salinity: 55‰. Description: The holotypic female (Figures 3.4A and 3.4B) is 165.2 µm long from the anterior margin of the head to the base of the caudal plate. The dorsal cuticle forms seven plates: the semi-circular head plate, three body plates each with a pair of flexible spine-like appendages, the caudal plate with the cirrus E, and the two crown-shaped intersegmental plates (Figure 3.4A). The head plate is semi-circular and with four deep indentions, dividing the head into five lobes. A complete set of well-developed cephalic sense organs is present on the head. Primary clava and lateral cirrus are inserted separately. The primary clava is club- shaped, and the secondary clava is semi-globular and inserted ventrally. All the cephalic cirri consist of well-developed cirrophore, a long scapus and shorter flagellum. The cirrus E on the caudal plate is inserted at a small lateral process and has the so-called “ball and socket” articulation – only found in Pseudostygarctus. The sense organ on the fourth leg is a pedunculate papilla without spine. No sense organs are present on the first three pairs of legs. There are three claws on each leg and each claw has a very tiny spur. All three claws originate from small pedestals insetting on the tarsus. The ventral mouth cone consists of a basal part and three telescopic segments. The sclerified/carbonated structures of the bucco-phangyngeal apparatus are for the first time observed in the genus Pseudostygarctus. The stylets are very long (49 µm) with small furcae. The buccal canal (47 µm) is carbonated all the way down to the placoids in the pharyngeal bulb. The placoids in the bulb have a very large anterior extension – called the internal stylet support. These three structures terminate in a ball-shaped structure, where the furcae of the two stylets may articulate. The true stylet supports are lacking. The small pharyngeal bulb (11 µm) is nearly filled up with three placoids and from the bulb a short oesophagus leads to the white diverticulated mid-gut. The reproductive system of the holotype consists of a single ovary with one very large and several small oocytes. The gonopore system consists of six rosette cells and very complex seminal receptacles. The ventral spheroid vesicles with looped ducts open close to the rosette- formed gonopore. Two internal bars are present at the openings of the seminal receptacles, and two external cuticlar pockets with spermatozoa overlap the female gonopore. The anus is sub-terminal and is closed by a three-lobed cuticular system. Remarks: Based on these morphological data, the new species of Pseudostygarctus from Qatar is very closely related to the type species of the genus Pseudostygarctus triungulatus from the Galapagos Islands (McKirdy et al. 1976), and not to the new species Pseudostygarctus galloae currently under publication from sandy sediments in Saudi Arabia (Hansen et al. in press). The only other stygarctid from polychate tubes is Mesostygarctus spiralis from Sidney, Australia; however, this may be due to the fact that this habitat has not been investigated for tardigrades before. Family Batillipedidae Ramazzotti, 1962 Type genus: Batillipes Richters, 1909 16 Diagnosis: Marine tardigrades without dorsal or ventral plates. Legs with 6 toes of equal or differing lengths. Each toe expanding distally into a sucking or adhesive round or shovel- shape disk. Claws are never present. Sense organs on all legs. Cirrus E always present. Complete set of cephalic sensory organs; however, the secondary clavae may be indistinct or only slightly dome-shaped. Cuticular seminal receptacles always lacking. Type species: Batillipes mirus Richters, 1909 Two species of Batillipes were found at the Old Pier, Al Zubarah; Batillipes pennaki Marcus, 1946 tidal in coral sand and Batillipes similis Schulz, 1955 subtidal in coral with roots of sea grass. Family Halechiniscidae Ramazzotti, 1962 Type genus: Halechiniscus Richters, 1908 Diagnosis: Marine tardigrades without dorsal plates. Telescopic legs with 4 toes in adults, 2 toes in larvae. Claws simple without spurs. Sense organs on all legs. Two cuticular seminal receptacles always present. Only one specimen of Halechiniscus was found at Coral Head in the coral sand. The specimen may be a new species. Genus: Florarctus Delamare-Deboutteville & Renaud-Mornant, 1965 Diagnosis: Marine tardigrades with large wing-shaped rostral, lateral and caudal expansions (alae) with procuticular support (caestus). In adults the two external toes are shorter than the two internal ones. The external toe has a hook-shaped cuticular structure (peduncle) inside the base, and the claw ending with an articulate portion (avicularia). The two cuticular seminal receptacles have an S-shaped duct. Two species of Florarctus were found in the Qatar-investigation. One subtidal at Old Pier and another at Düvel Rock. None of the species have been classified. Genus: Wingstrandarctus Kristensen, 1984 Diagnosis: Marine tardigrades with wing-shaped expansions without procuticular support. Symbiotic bacteria in cephalic vesicles. The avicularia on the external claws only a small notch. A species (one specimen) similar to Wingstrandarctus intermedius (Renaud-Mornant, 1967) was found in coral sand at Coral Head. Order: Echiniscoidea Diagnosis: Without a median cirrus on the head. Family: Echiniscoididae Kristensen & Hallas, 1980 Type genus: Echiniscoides Plate, 1889 17 Diagnosis: Trunk without dorsal plates. Legs without toes and not telescopic. Claws from 5 to 13 on each leg. Most species are found intertidal on algae, barnacles or other marine invertebrates. A few species are found in subtidal sediments. Four species of Echiniscoides were found inside the buffer zone. Three of them are new to science. One of the species look very similar to the nominate Echiniscoides sigismundi sigismundi. All four species were found on invertebrates. Echiniscoides sigismundi sigimundi cfr. Found on barnacles, Ras Ushayriq, Qatar at high tide (Figure 3.5A). Dorsal cuticle smooth. 7-9 claws on each leg. Echiniscoides nov. sp. 1. Found inside polychaete tubes of Pomatoleios kraussii at Old Pier, Al Zubarah, Qatar. High tide. Dorsal cuticle with a fine dorsal sculpture. 7-10 claws on each leg. Echiniscoides nov. sp. 2. Beach Rock (Q4) of Al Zubarah, Qatar. From the mussel Brachidontes variabilis, (Mytilidae) (Figure 3.2A). Dorsal cuticle with mammalate dorsal sculpture. 7-8 claws on each leg. Echiniscoides nov. sp. 3. Beach Rock of Al Zubarah, Qatar from the mussel Brachidontes variabilis, (Mytilidae) (Figures 3.5B and 3.6B-3.6D). High tide. Strongly sculptured dorsal cuticle with small plate-like areas. Head very aberrant with large secondary clavae (looks like a toad). Surprisng this species has some similarities with Echiniscoides horningi from Macquarie Island, Subantarctica (Miller & Kristensen1999) 3.4 FINDINGS WITHIN THE BUFFER ZONE The preliminary “taxonomic account” presented above shows a rich intertidal tardigrade fauna with many heterotardigrade species; several species new to science. This zoogeographical finding is surprising, as the Arabian Gulf is very young – not more than 15,000 years old (Glennie 1998; Sheppard et al. 2010; Uchupi et al. 1999). Some of the species identifications are still tentative as only a single specimen was found. Furthermore, the use of DNA sequences could help clarify the identification of some specimens of Echiniscoides that differ slightly from closely related species, i.e. are they genetically different (another species) or are they genetically identical (same species with another phenotype). 3.5 CONCLUSIONS AND RECOMMENDATIONS Several aspects of the tardigrade fauna at Al Zubarah deserve further investigation and this report is only preliminary. Especially the sediment samples are still not sorted out and these samples might reveal further new species if thoroughly investigated. A number of identifications (e.g. four species of Echiniscoides) should be validated by the use of molecular methods to investigate if the morphological variation is reflected in the DNA (Kristensen & Hallas 1980, Hallas & Kristensen 1982, Faurby et al. 2011, 2012). The relationship between various Echiniscoides species could be investigated through 18 phylogenetic analysis of the fast evolving gene cytochrome c oxidase subunit I (COI). This gene has been suggested as a universal DNA barcode for numerous groups of animals and has recently been used to illustrate a very large genetic differentiation within Echiniscoides (Faurby et al. 2012). Furthermore the mangrove habitats within the UNESCO Buffer Zone should be more thoroughly investigated for the class Eutardigrada. Eutardigrades may be found on the trunk of the trees. Not a single eutardigrade was found in the current investigation. 19 References Accogli, G., Gallo, M., D’addabbo, R. & Hansen, J. G. 2011. Diversity and ecology of the marine tardigrades along the Apulian Coast. Journal of Zoological Systematic Research 49(Suppl. 1): p.53–57. Degma, P., Bertolani, R. & Guidetti, R. 2010. Actual checklist of Tardigrada species (2009- 2012, Ver. 21: 30-06-2012). Faurby, S., Jørgensen, A., Kristensen, R. M. & Funch, P. 2011. Phylogeography of North Atlantic intertidal tardigrades: refugia, cryptic speciation and the history of the Mid- Atlantic Islands. Journal of Biogeography 38(8): p.1613–1624. Faurby, S., Jørgensen, A., Kristensen, R. M. & Funch, P. 2012. Distribution and speciation in marine intertidal tardigrades: testing the roles of climatic and geographical isolation. Journal of Biogeography 39(9): p.1596–1607. Gallo D’Addabbo M., Morone De Lucia M. R. & Grimaldi de Zio, S. 1987. Heterotardigrada of the Amendolara Shoal, high Ionian Sea. In: Bertolani R (ed.), Biology of Tardigrades. Selected Symposia and Monographs UZI I Mucchi, Modena, Italy, pp 103–110. Gallo D’Addabbo, M., S. de Zio Grimaldi & R. D’Addabbo, R. 2000. Pseudostygarctus apuliae (Tardigrada, Heterotardigrada): a new species from the lower Adriatic Sea. Italian Journal of Zoology 67: p.125–128. Gallo D’Addabbo, M., S. de Zio Grimaldi & Sandulli, R. 2001. Heterotardigrada of two submarine caves in S. Domino Island (Tremiti Islands) in the Mediterranean Sea. Zoologischer Anzeiger 240: p.361–369. Glennie, K.W. 1998. The desert of southeast Arabia: A product of Quaternary climate change. In Alsharhan, A.S., Glennie, K.W., Whittle, G.L., Kendall, C. St.G., (Eds.) Quaternary Deserts and Climatic Change. (pp. 279-291). A. A. Balkema, Rotterdam. Grimaldi de Zio, S., M. D’Addabbo Gallo & Morone De Lucia, M. R. 1998. A new Stygarctidae from South Tyrrhenian Sea (Tardigrada, Heterotardigrada). Cahiers de Biologie Marine 39: p.85–91. Guidetti, R. & Bertolani, R. 2005. Tardigrade taxonomy: an updated checklist of the taxa and a list of characters for their identification. Zootaxa 845: p.1–46. Hallas, T. E. & Kristensen, R. M. 1982. Two new species of the tidal genus Echiniscoides from Rhode Island, U.S.A. (Echiniscoididae, Heterotardigrada). Proc. Third Intern. Symp. Tardigrada, Aug. 3-6, Johnson City, Tennessee, U.S.A., ETSU Press: 179–192. Hansen, J. G., Jørgensen A. & Kristensen R. M. 2001. Preliminary studies of the tardigrade fauna of the Faroe Bank. Zool Anz 240: p.385–393I. 20 Hansen, J. G. & Kristensen, R. M. 2006. The ”hyena female” of tardigrades and descriptions of two new species of Megastygarctides (Arthrotardigrada: Stygarctidae) from Saudi Arabia. Hydrobiologia 558: p.81–101. Hansen J. G. 2011. The phylogeny of Arthrotardigrada. Unpublished PhD thesis, University of Copenhagen. Hansen, J.G., Kristensen, R.M. & Jørgensen, A. in press. The armoured marine tardigrades (Arthrotardigrada, Tardigrada). Scientia Danica, Series B, Biologica, Vol. ? The Royal Danish Academy of Sciences and Letters. Hiruta, S. 1985. A new species of marine interstitial Tardigrada of the genus Stygarctus Schulz from Hokkaido, Japan. Special Publication of the Mukaishima Marine Biological Station 245: p.127–129. Jørgensen, A., Faurby, S., Hansen, J.G., Møbjerg, N. & Kristensen, R.M. 2010. Molecular phylogeny of Arthrotardigrada (Tardigrada). Molecular Phylogenetics and Evolution 54: p.1006–1015. Kristensen, R. M. & Hallas, T. E. 1980. The tidal genus Echiniscoides and its variability with erection of Echiniscoididae fam.n. (Tardigrada). Zoologica Scripta 9: p.113–127. McKirdy, D., Schmidt, P. & McGinty-Bayly, M. 1976. Interstitielle fauna von Galapagos XVI. Tardigrada. Mikrofauna des Meeresbodens 58: 410-449. Miller, W. R. & Kristensen, R. M. 1999. Tardigrades of the Australian Antarctic: A new species of the marine genus Echiniscoides from Macquarie Island, Subantarctica. In H. Greven (ed.), Proceedings of the seventh international symposium on the Tardigrada, August 1997, Düsseldorf, Germany. Zoologischer Anzeiger 238: p.289– 294. Møbjerg, N., Halberg, K. A., Jørgensen, A., Persson, D., Bjørn, M., Ramløv, H. & Kristensen, R. M. 2011. Survival in extreme environments – on the current knowledge of adaptations in tardigrades. Acta Physiologica 202(3): p.409–420. Sheppard, C., Al-Husiani, M., Al-Jamali, F., Al-Yamani, F., Baldwin, R., Bishop, J., Benzoni, F., Dutrieux, E., Dulvy, N. K., Durvasula, S. R. V., Jones, D. A., Loughland, R., Medio, D., Nithyanandan, M., Pilling, G. M., Polikarpov, I., Price, A. R. G., Purkis, S., Riegl, B., Saburova, M., Namin, K. S., Taylor, O., Wilson, S., Zadinal, K. 2010. The Gulf: A young sea in decline. Marine Pollution Bulletin 60: p.13–38. Uchupi, E., S.A. Swift & Ross, D. A. 1999. Late Quaternary stratigraphy, palaeoclimate and neotectonism of the Persian (Arabian) Gulf region. Marine Geology 160: p.1–23. 21 Figures Figure 3.1: The “Old Pier” at Al Zubarah, Qatar provides easy access to the different substrates in the tidal zone, in particular the coral sand with many species of tardigrades. Figure 3.2: A) Brachidontes variabilis, Mytilidae from Al Zubarah, Qatar. B) Close-up of the byssi threads, a good place to find tardigrades. 22 Figure 3.3: A) Polychaete tubes of Pomatoleios kraussii from Old Pier, Al Zubarah, Qatar. B) The average size of these habitat-structures. Several species of tardigrades were found in the polychaete tubes at a salinity of 50‰. This is also the type locality of the new species of Pseudostygarctus. Figure 3.4: A new species of Pseudostygarctus with three claws. The species was found inside the polychaete tubes of Pomatoleios kraussii. A) Notice the two crown-shaped intersegmental plates. B) Notice the shape of the stylets and placoides. 23 Figure 3.5: Light microscopy of two tardigrade species from Qatar. A) This could possibly be a subspecies of Echiniscoides sigismundi. B) Echiniscoides nov. sp. 3 from mussels. Figure 3.6: A) The dorsal cuticle of the Echiniscoides nov. sp. 2. B) The dorsal cuticle of the Echiniscoides nov. sp. 3, notice the cuticle differences between A and B. C) Ventral cuticle and claws of Echiniscoides nov. sp. 3, notice the shape of the head. D) Close-up of the head of Echiniscoides nov. sp. 3. Scale bars = 10 µm. 24 4. THE RECENT MARINE MOLLUCS AT AL ZUBARAH Aslak Jørgensen (The Natural History Museum of Denmark, Sølvgade 83, DK-1307 Copenhagen K, Denmark, aslak@snm.ku.dk). 4.1 INTRODUCTION The molluscs are an important marine faunal component within the Al Zubarah buffer zone both with regard to the current marine biodiversity and the socioeconomic impact of the pearl fishing trade on Al Zubarah. Many of the mollusc classes are present within the buffer zone with representatives of Gastropoda (snails), Bivalvia (bivalves), Cephalopoda (squids), Polyplacophora (chitons) and Scaphopoda (tusk shells). Of these the fisheries of Pinctada (pearl oyster) and Sepia (squid) have an economic impact on the Arabian Gulf communities. Several studies have focused on the Qatari marine molluscan fauna resulting in the reported occurrence of 129 species of gastropods, 2 cephalopods, 5 scaphopods, 144 bivalves and 6 polyplacophorans (Al-Ansi & Al-Khayat, 1999; Al-Khayat, 1997; Al-Khayat, 2007; Al- Khayat, 2008; Al-Khayat & Al-Ansi (2008), Al-Khayat & Al-Khayat, 2000; Al-Khayat & Jones, 1999; Mohammed & Al-Khayat, 1994). Furthermore, five species of recently introduced terrestrial gastropods have been reported (Al-Khayat, 2010). 4.2 FINDINGS WITHIN THE BUFFER ZONE The preliminary species list presented below show a rich intertidal fauna with many gastropod species and a relatively poor bivalve fauna due to the lack of extensive soft bottom (mud) habitats within the buffer zone. Many of the species identifications are tentative as a good reference collection illustrating the conchological variation is still not established. Furthermore, the use of DNA sequences could help clarify the identification of some specimens that differs slightly from closely related species, i.e. are they genetically different (another species) or are the genetically identical (same species with another phenotype). Species list Gastropoda Vetigastropoda Chilodontidae Euchelus asper Fissurellidae Diodora sp. Phasianellidae Phasianella solida 25 Trochidae Osilinus kotschyi Priotrochus obscurus Osilinus kotschyi/Priotrochus obscurus-like but with fine, more subtle nodules. Trochus cf. scabrosus (three teeth in aperture). Trochus sp. (very triangular appearance; 1 cm; T. erithreus or T. radiatus?). Turbinidae Lunella coronata Mesogastropoda Cerithiidae Cerithium caeruleum Cerithium rueppelli Clypeomorus bifasciatus persicus Littorinidae Echinolittorina arabica Nodilittorina cf. millegrana (“very” large 10+ mm) Littorina sp. (smooth shells) Peasiella isseli Planaxidae Planaxis sulcatus Potamididae Cerithidea cingulata Potamides conicus Neogastropoda Columbellidae Mitrella blanda Muricidae Thais savignyi Thais cf. tissoti Nassariidae Nassarius fissilabris Nassarius cf. stolatus Nudibranchia Chromodorididae Goniobranchus obsoletus 26 Panpulmonata Siphonariidae Siphonaria belcheri Ellobiidae Allochroa bronnii Laemodonta monilifera Laemodonta rapax Melampus sp. Pedipes sp. Cephalopoda Sepiidae Sepia sp. Remarks: The specimens closely resemble S. pharaonis; however S. pharaonis has a split anterior spine on the internal shell according to Bosch et al., (1995) and the collected specimens have an internal shell with a single unsplit anterior spine. Anderson et al., (2007) suggest that “S. pharaonis” is a cryptic species complex consisting of at least five species within its geographical range. Scaphopoda Laevidentaliidae Only dead shells of Laevidentalium longitrosum. Bivalvia Arcidae Barbatia setigera Chamidae Chama sp. Glycymerididae Glycymeris pectunculus Mytilidae Brachidontes variabilis Mytilidae sp. on Pier (perhaps Brachidontes variabilis with frayed shell edges). Malleidae Malvufundus sp. (juvenile specimens). Pinnidae Pinna muricata (or Pinna bicolor) 27 Pteriidae Pinctada radiata Polyplacophora Chitonidae Chiton sp. 4.3 CONCLUSIONS AND RECOMMENDATIONS Several aspects of the molluscan fauna at Al Zubarah deserve further investigation. Especially the nudibranchs, ellobiids and the polyplacophorans might reveal further species if thoroughly sampled. A number of identifications should be validated by the use of molecular methods to investigate if the morphological variation is reflected in the DNA. Furthermore, the mangrove habitats within the UNESCO Buffer Zone should be more thoroughly investigated as minute species of assiminid gastropods have been reported from mangroves in the United Arab Emirates (Feulner & Hornby, 2006). Finally, a more thorough sampling effort in the few and small subtidal mud habitats present within the buffer zone and more extensive digging into the tidal mud flats will most likely increase the currently recorded number of bivalves. The findings of the 2012 field work have been summarized in the table below. Molluscan class Number of families Number of species Gastropoda 15 31 Cephalopoda 1 1 Scaphopoda 1 1 (only dead shells) Bivalvia 7 7 Polyplacopora 1 1 Total 25 40 28 References Al-Ansi, M.A., & Al-Khayat, J.A. 1999. A preliminary study of coral reef and its associated biota in the Qatari waters, Arabian Gulf. Qatar University Science Bulletin 19: p.294– 311. Al-Khayat, J.A. 1997. The marine mollusc of the Qatari waters, Arabian Gulf. Qatar University Science Bulletin 17: p.479–491. Al-Khayat, J.A. 2007. Macrofauna abundance in seagrass of Qatari Water, Arabian Gulf. Egyptian Journal of Aquatic Research 33: p.257–276. Al-Khayat, J.A. 2008. Molluscs of the State of Qatar. Qatar Biodiversity Newsletter 2: p.1–5. Al-Khayat, J.A. 2010. First record of five terrestrial snails in the State of Qatar. Turkish Journal of Zoology 34: p.539–545. Al-Khayat, J.A., & Al-Ansi, M.A. 2008. Ecological features of oyster beds distribution in Qatari waters, Arabian Gulf. Asian Journal of Scientific Research p.1–18. ISSN 1992- 1454. Al-Khayat, J.A., & Al-Khayat, F.A. 2000. Study of macrobenthic invertebrates in the Qatari waters, Arabian Gulf. Bulletin of the National Institute of Oceanography and Fisheries 26: p.125–148. Al-Khayat, J.A., & Jones, D.A. 1999. A comparison of the macrofauna of natural and replanted mangroves in Qatar. Estuarine, Coastal and Shelf Science 49: p.55–63. Anderson, F.E., Valinassab, T., Ho, C.-W., Mohamed, K.S. & Asokan, P.K. 2007. Phylogeography of the Pharaoh Cuttle Sepia pharaonis based on partial mitochondrial 16S sequence data. Reviews in Fish Biology and Fisheries 17: p.345–352. Bosch, D.T., Dance, S.P., Moolenbeek, R.G., & Oliver, P.G. 1995. Seashells of Eastern Arabia, Motivate Publishing, 296 pp. Feulner, G.R., & Hornby, R.J. 2006. Intertidal molluscs in UAE lagoons. Tribulus 16: p.17– 23. Mohammed, S.Z., & Al-Khayat, J.A. 1994. A preliminary checklist of benthic mollusca on the Qatari coasts, Arabian Gulf. Qatar University Science Bulletin 14: p.201–206. 29 Figures Laemodonta sp. Melampus sp. Pedipes sp. Figure 4.1: Some ellobiid gastropods from the Old Pier at Al Zubarah. The shells are app. 10 mm in height. The ellobiids are a common, albeit often hidden, component of mangrove fauna. The specimens from Al Zubarah were collected by hand picking from rock crevices and washed up sea weed at the Old Pier. They represent new faunal records for the Qatari animal inventory. 30 5. ENTOMOLOGY Kenneth P. Puliafico (Entomology Department, Zoological Museum, The Natural History Museum of Denmark, puliafico@gmail.com) Anne Majgaard Jensen (University of Copenhagen. anne-majgaard@hotmail.com). 5.1. INTRODUCTION Qatar’s native terrestrial environment is characterized by desert and semi-desert ecological zones, yet even under these harsh conditions there is a rich diversity of arthropod species to be found. Arthropods; invertebrate animals with legs and an exoskeleton; are the most abundant and diverse animals in terrestrial habitats and make up over 80% of all animals known throughout the world. The native land-living arthropods of Al Zubarah are well adapted to the hot, dry conditions of the Qatari Desert and have several behavioural and physiological adaptations to tolerate extreme temperatures and to prevent water loss. Some of these adaptations also mean that these creatures are not easy to find during the daylight hours when most visitors come to Al Zubarah. Our survey used a variety of different trapping and hunting techniques to find these animals when they are most active, so that we can give a better assessment of the true diversity of this fascinating region. It has been estimated that over 500 species of the phylum Arthropoda occur in Qatar and the vast majority of these are insects. To date only a preliminary check list of Qatar’s insects has been published (Table 1, Abdu & Shaumar 1985). An extensive review of the literature indicated that there are very few scientific publications covering the arthropod fauna of this region (>20 papers mentioning Qatar), and many of these journal articles are primarily ecological studies that focus on a subset of Qatar’s insects (e.g. Abushama 1999). As a result the taxonomic literature specific to Qatar are extremely limited and usually only covers a specific group at the family level (e.g. Soldati 2009). Furthermore, most of these taxonomic treatments do not include species identification keys (e.g. Háva & Pierre 2008, Keith & Bordat 2011). This lack of basic taxonomic information for the majority of arthropod groups occurring within the region has made it particularly difficult to put exact names on the animals we collected. Despite these obstacles we found several new taxa in the designated area that have not been previously recorded for the nation of Qatar (Table 5.1). 5.2 METHODS The area within the Al Zubarah Archaeological Site exclusion and buffer zones can be divided into several local habitat types, including beachfront, sabkha (salt flats), and rocky desert. Our sampling focused mainly on the native habitats outside of those heavily influenced by activities relating to the preservation and restoration of the Fort and old city of Al Zubarah (archaeological sites and the modern camp). In the surrounding natural habitat plant features that are significant for animal life such as the presence of acacia trees (Acacia ehrenbergiana and A. tortillis), shrubs (e.g. Alhagi maurorum, and Limonium axillare) and grasslands were a particular focus for collecting. Therefore we recognized the following five natural habitats to direct our sampling: beach, lower sabkha, upper sabkha, acacia trees, and grassy depressions in the rocky desert. 31 For three weeks in the spring of 2012 (5. – 24. March) we collected terrestrial arthropods within the exclusion and buffer zones of the Al Zubarah Archaeological Site. Using a number of different collecting methods (pitfall traps, yellow pan traps, light traps, and hand collecting) we captured a wide variety of insects and other arthropods within the five local habitat types. Specimens were preserved in propylene glycol or ethanol until their return to the Zoological Museum, University of Copenhagen, Denmark. Specimens were sorted to family and assigned to a morpho-species based on their external morphological characters. Taxonomic experts for particular groups have been contacted for further identification of species whenever possible. A preliminary list of the species and morpho-species identified thus far are shown in the tables below (Tables 5.2, 5.3, and 5.4). In addition to the general species survey of the entire arthropod fauna we also used standardized sampling protocols to compare the habitats for their diversity and abundance of four major taxonomic groups. Beetles, spiders, ants, bees and wasps (Coleoptera, Arachnida, Formicidae and Hymenoptera, respectively) are commonly used as indicator species in broader ecological studies throughout the world. Three independent trapping sites were placed in each habitat type. Five plastic cups (95mm diameter x 120 mm depth) were placed approximately 5 meters apart, buried so that they were level with the ground and filled with propylene glycol. After 5 days the traps were emptied, specimens were rinsed using ethanol and sorted for identification. For each of these taxonomic groups we counted the number of individuals captured within a series of pitfall traps. To display the results so that comparisons can be made between taxa the count for each habitat type was divided by the total number of specimens collected and presented as a percentage using weighted averages. The use of standardized pitfall trapping allowed us to compare the five habitat types identified within the Al Zubarah exclusion and buffer zones. Although each animal type had a unique distribution within these habitats we can see an overall pattern which suggests that places with more stable environments tend to have greater arthropod abundance (Figure 5.1). The beach and lower Sabkha, which tend to be extremely unstable environments had the lowest trap catch while the areas with less disturbance had greater abundance and diversity (not shown). Also important was the presence of vegetation resources such as shrubs in the upper sabkha, grasses and herbs within the moist depressions of the rocky desert, or the acacia trees. Plants provide food for herbivores, nectar resources, prey items, increased humidity, shade and shelter. Interestingly the Acacia tree species appear to have lower levels of both spiders and winged Hymenoptera (bees and wasps), but this is most likely because these animals have moved off of the ground and are utilizing the flowers, leaves and branches of the tree. In contrast, the high abundance of ground nesting ants near the trees is probably due to the presence of aphids and other Hemipterans that produce honeydew to feed the ants in exchange for protection from predators. The unique ecological interactions that occur because of these plant resources make the preservation of these habitats essential for maintaining a diverse fauna in the harsh desert landscape. 5.3 CONCLUSIONS AND RECOMMENDATIONS These results are a first step to better understanding Qatari fauna, but we are only beginning to get a picture of the complete terrestrial Arthropod community for Qatar. This study should still be considered very preliminary since some major taxonomic groups including Diptera, Hymenoptera, Lepidoptera, Orthoptera and Hemiptera have not been fully examined from 32 our samples. Several of these taxonomic groups required identification by experts who have not had time to return their results. Therefore we expect the species list to greatly increase as more accurate names are added. The current study also should be viewed as a snap-shot of the entire community rather than a comprehensive picture of all the insects, spiders and other arthropods that inhabit the Al Zubarah area. Our collecting trip occurred in March before the end of winter temperatures. Since all of the Arthropoda are ectothermic (“cold blooded”) most of these animals were not particularly active during this time of year, therefore trap catches were much smaller than what could be expected in other seasons. Warmer temperatures later in the year also allow the new generation of insects to complete their development and emerge as adults. Several species known from the other areas in the eastern Arabian Peninsula have only been collected during the summer and autumn. One study of the insects in Doha showed that there are two peaks of insect abundance, one in April-May and a second much larger peak in September- October (Abushama, 2006). Therefore we recommend later spring and autumn collecting trips be initiated as soon as possible to take advantage of these peaks of insect activity. Every insect collection has some level of bias, and this study is no exception. Due to the failure of some equipment arriving from an overseas supplier we were not able utilize all the trapping methods originally planned for. As a result many of the collecting methods deployed tended to favor ground dwelling animals such as beetles, ants, spiders, and scorpions. The fact that we did collect so many bees and wasps (Hymenoptera) and flies (Diptera) in the pitfall traps is an indication of the scarcity of vegetative cover and the attractiveness of the white cups used to make these traps. Future workers should utilize the UV light traps, Malaise traps, flight intercept traps and other equipment that will open the door to collecting a wider variety of day and night flying insect species. Our final recommendation is to encourage the preservation of the habitats that support Qatar’s unique plant and animal life. Studies such as our standardized pitfall sampling should be conducted to quantify the species richness and assess the quality of these habitats. Our results suggest that areas that support the deserts sparse plants also encourage wildlife diversity. Features like the shrub covered upper sabkha, acacia trees and the moist grassy depressions of the desert that have traditionally provided fodder for domestic animals also provide important resources for the wild animals of Qatar. Beyond the arthropod community surveyed in our study we also observed the spiny-tailed lizard or dhub (Uromastyx aegyptia), several rodents, and a variety of birds utilizing these habitats. The placement of future buildings should avoid disruption of vegetated habitats as much as possible. That would leave room for interpretive pathways which might take advantage of these unique parts of the ecosystem, especially the transition zone between the sabkha and the slightly higher elevations where the rocky desert plants thrive. 33 References R.M. Abdu and N.F. Shaumar, 1985. A preliminary list of the insect fauna of Qatar. Qatar University Science Bulletin. 5: 215-232. F.T. Abushama, 1999. Dominance value and community production of desert Arthropoda in Qatar. Qatar University Science Journal. 18: 137-153. F.T. Abushama, 2006. Night and early morning flying insects in a residence backyard in Doha City, Qatar. Qatar University Science Journal. 26: 83-90 List of Lepidoptera of Qatar. Wikipedia, accessed 8/24/2012 at 11:32, http://en.wikipedia.org/wiki/List_of_Lepidoptera_of_Qatar Carabidae Checklist of the Qatar. Carabidae of the World, accessed 8/25/2012 at 14:57, http://carabidae.pro/carabidae/pcid_83.html J. Háva and E. Pierre, 2008. Contribution to the Family Dermestidae (Coleoptera) from Qatar. Journal of the Entomological Research Society. 10(2): 37-41. D. Keith and P. Bordat, 2011. Sur la faune des Scarabaeoidea du Qatar (Coleoptera). Nouvelle revue d'entomologie (N.S.). 27 (2): 183-189. L. Soldati, 2009. Darkling Beetles (Coleoptera: Tenebrionidae) of Qatar. Natura optima dux Foundation, Warszawa (Poland). pp. 101. 34 Tables Table 5.1. Summary of the terrestrial Arthropoda species known from Qatar. All Insecta Class Family Genera Species Date References Orders Orders* Qatar Checklist 1 15 (15) 63 154 170 1985 1. 1., 2., 4., 5. Literature review 4 20 (15) 86 208 262 25 Sep. 2012 and others Al Zubarah current 5 30 (16) 82 80 171 25 Sep. 2012 Morpho-species study New taxonomic current 1 12 (3) 27 23 3 25 Sep. 2012 records† study *Columns present totals for all Arthropods except for “Insecta Orders” which are a subset of Arthropod Orders. † Taxa found in this survey not previously reported from the literature for Qatar. 35 Table 5.2. Minor Arthropod Classes collected from proposed UNESCO Exclusion and Buffer Zone, Al Zubarah, Qatar, in March 2012. Morpho- Class Order Family Genus Species Author Date species New record References Malacostraca Amphipoda sp. 1 O Isopoda Oniscidia Porcellio evansi Omer-Cooper 1923 1? 2. Chilopoda Geophilomorpha Geophilidae sp. 1 O/F Scalopendramorpha Scolopendridae Scolopendra mirabilis Porat 1876 1 2. Entognatha Entomobryomorpha sp. 1 C/O Poduromorpha sp. 1 C/O Symphypleona sp. 1 C/O The numbers of Morpho-species are distinct specimens types thought to represent undetermined species; a “?” indicates uncertainty at the lowest taxonomic level indicated. New records are indicated for Qatar for taxonomic Class (C), Order (O), and Family (F). Notice that none of these taxa have been previously reported from Al Zubarah. 36 Table 5.3. Arachnids (Arthropoda: Arachnida) collected from proposed UNESCO Exclusion and Buffer Zone, Al Zubarah Archaeological Site, Qatar, in March 2012. Morpho- Class Order Family Genus Species Author Date species New record References Arachnida Ixodida Ixodidae sp. 2 Trombidiformes sp. 1 O Oribatida sp. 1 O Araneaea Araneidae sp. 1 F Clubionidae sp. 1? F Gnaphosidae sp. 1 Linyphiidae Prinerigone cf vagans (Audouin) 1826 1 F/G/S Lycosidae Lycosa sp. 1 (Fig. 5.2) 2. Oonopidae sp. 1 F Oxyopidae sp. 1? F Philodromidae Phillodromus sp. 2 2. Pholcidae Artema sp. 1 F/G Salticidae sp. 1 F Scytodidae sp. 1 F Sicariidae Loxosceles sp. 1 F/G Sparasidae sp. 1 F Tetragnathidae sp. 1? F Theridiidae Enoplagnatha sp. 1 F/G Theridiidae Latrodectus sp. 1 F/G Zodariidae sp. 1 F Pseudoscorpiones Withiidae sp. 1? O/F Scorpiones Buthidae Androctonus crassicauda (Olivier) 1807 1? (Fig 5.3) 2. Scorpiones sp. 2 Solifugae Daesiidae sp. 1 O/F Rhagodidae sp. 1 O/F The numbers of Morpho-species are distinct specimens types thought to represent undetermined species; a “?” indicates uncertainty at the lowest taxonomic level indicated. New records are indicated for Qatar for taxonomic Order (O), Family (F), Genus (G) and Species (S). Notice that none of these taxa have been previously reported from Al Zubarah. 37 Table 5.4. Insects (Arthropoda: Insecta) collected from proposed UNESCO Exclusion and Buffer Zone, Al Zubarah, Qatar, in March 2012. Lit. Al Morpho- New Order Family Subfamily Genus Species Author Zubarah species record References Coleoptera Anobiidae Ptilinus sp. 1 G Anobiidae sp. 1 Anthicidae Anthicus sp. 1? F/G Anthicidae sp. 4 F Carabidae Anthiinae Anthia duodecimguttata Bonelli x 1., 2., 5. Carabidae Carabinae Calosoma sp. 1 (Latreilla & Carabidae Cicindelinae Megacephala euphratica Dejean) 1 G Carabidae Cicindelinae Nebria sp. 1 G Carabidae Trechinae Bembidion sp. 3? G Carabidae sp. 7 Cerambycidae Apomecyna lameerei Pic x 1. Chrysomelidae Bruchinae sp. 1 F/SF Chrysomelidae sp. 1 F Cleridae Necrobia sp. 2 Coccinellidae Coccinella undecimpunctata Linnaeus x 1? 1. Coccinellidae sp. 2 Cryptophagidae Cryptophagus sp. 1 F/G Curculionidae Tychius sp. 1 G Curculionidae sp. 5 Dermestidae Dermestinae Dermestes sp. 1 Dermestidae sp. 5 Elateridae sp. 1 Endomychidae Holoparamecus sp. 1 F/G Histeridae Saprinus sp. 1 F/G Histeridae sp. 4 F Latridiidae Migneauxia sp. 1 F/G Ptiliidae Acrostrichis sp. 1 F/G Ptiliidae Actidium sp. 1 F/G Scarabaeidae Aphodiidae Aphodius sp. 1 G Scarabaeidae Dynastinae Pentodon bispinosus Kuster 1? 1. Scarabaeidae sp. 3 Staphylinidae Aleocharinae Aleochara sp. 1 38  Staphylinidae Aleocharinae sp. 1 Lit. Al Morpho- New Order Family Subfamily Genus Species Author Zubarah species record References Coleoptera Staphylinidae Oxytelinae Bledius sp. 3 SF/G Staphylinidae Staphylininae sp. 3 SF Tenebrionidae Adesmia sp. x 1 (Fig. 5.4) 1. Tenebrionidae Akis sp. 1 Tenebrionidae Blaps mortisaga Linnaeus 1? (Fig 5.5) 2. Tenebrionidae Erodius sp. 2 2. Tenebrionidae Gonocephalum sp. 4 Tenebrionidae Mesostina sp. 3 Tenebrionidae Micipsa sp. 3? 1. Tenebrionidae Pimelia sp. 1? Tenebrionidae Trachyderma sp. 1 Tenebrionidae Trachyderma hispida (Forsskal) 1? 2. Tenebrionidae Zophosis sp. 1 1., 2. Tenebrionidae sp. 10 Dermaptera Labiduridae Labidura confusa Capra x 1. Dictyoptera Blattidae Blattella germanica Linnaeus x 1. Blattidae sp. 2 Mantidae Blepharopsis mendica (Fabricius) x 1. Mantidae sp. 1 Diptera Bombyliidae sp. 1 Calliphoridae sp. 1 Culicidae sp. 1 Muscidae Musca domestica Linnaeus 1? 1. Sarcophagidae Blaesoxipha setose (Salem) 1 G/S Sarcophagidae Sarcophaga sp. 1 Sarcophagidae Wohlfahrtia sp. 1 Syrphidae sp. 1 Tachinidae sp. 1 Embioptera sp. x 1 1. 39  Lit. Al Morpho- New Order Family Subfamily Genus Species Author Zubarah species record References Hemiptera Aphididae sp. 1 F Coccoidea sp. 1 Cydnidae Cydnus hispidulus Klug x 1. Cydnidae Cydnus macrophthalmus E. Wagner x 1. Cydnidae sp. 1 Lygaedae sp. 1 Pentatomidae sp. 1 Pyrrhocoreidae Scantius aegyptius (Linnaeus) 1? 1. Reduviidae Reduvius sp. 1 Rhopalidae Agraphopus lethierryi Stål 1? 1. Hymenoptera Apidae sp. 1 (Fig. 5.6) Dryinidae sp. 1 F Formicidae Camponotus maculatus Emery x 1. Formicidae Formicinae sp. 1 Formicidae Myrmicinae Cardiocondyla sp. 1 G Formicidae Myrmicinae Crematogaster sp. 1 G Formicidae Myrmicinae Monomorium sp. 1 2. Formicidae Myrmicinae sp. 2 Formicidae Ponerinae Ponera sennaarensis (Mayr) 1 G/S Formicidae Formicinae Cataglyphis niger (Linnaeus) 1? 2. Sphecidae Liris haemorrhoidalis Fabricius x 1. Isoptera Hodotermitidae Anacanthotermes ochraceus (Burmeister) 1? 1., 2. Rhinotermitidae Psammotermes hybostoma Desneux 1? 1., 2. Lepidoptera Acrtiidae Utetheisa pulchella (Linnaeus) 1? (Fig. 5.7) 1., 4. Crambidae Herpetogramma licarsisalis (Walker) x 1., 4. Noctuidae Clytie haifae (Habich) x 1., 4. Noctuidae Trichoplusia ni (Hubner) x 1., 4. Papilionidae Papilio demoleus Linnaeus x 1., 4. Pieridae Madais fausta Oliver x 1., 4. Sphingidae Hippotion celerio (Linnaeus) x 1., 4. 40  Lit. Al Morpho- New Order Family Subfamily Genus Species Author Zubarah species record References Neuroptera Chrysopidae sp. 1 Coniopterygidae sp. 1 F Myrmeleontidae Morter hyalinus Oliver x 1. Myrmeleontidae sp. 1 Odonata Aeshnidae Anax parthenope Selys 1? 1. Orthoptera Acrididae Anacridium aegyptium Linnaeus x 1. Acrididae Cyclopternacris etbaica Ramme x 1. Acrididae Locusta migratoria (Linnaeus) 1? 1., 2. Gryllidae Oecanthus sp. x 1. Gryllidae Acheta domesticus Linnaeus 1? 1., 2. Gryllotalpidae Gryllotalpa gryllotalpa Linnaeus x 1. Tettigonidae Homorocryphus nitidulus Scopoli x 1. Psocoptera sp. 1 O Siphonaptera sp. 1 O Thysanoptera Thripidae sp. 1 O Thysanura Lepismatidae Thermobia domestica (Packard) 1? 1. “Lit. Al Zubarah” column shows if the species has been previously recorded for the locality in the literature. The numbers indicated in the Morpho-species are distinct specimens types thought to represent undetermined species; a “?” shows uncertainty at the lowest taxonomic level indicated. New records are indicated for Qatar for taxonomic Order (O), Family (F), Subfamily (SF), Genus (G) and Species (S). 41 Figures 70% 60% 50% % ot total catch 40% 30% 20% 10% 0% Beach Lower Sabkha Upper Sabkha Meadow & Acacia Trees Desert Habitat types Beetles Spiders Ants Bees & Wasps Figure 5.1: Terrestrial arthropod collections made in each habitat type using standardized pitfall sampling. Percentages are calculated from average values of each habitat type divided by a weighted average of the total catch. Figure 5.2: A large wolf spider (Lycosidae). These spiders are solitary hunters that usually come out at night to feed on other arachnids and small animals. Although they can give a poisonous bite when continuously harassed, they prefer running away from people. (photo by J. Kielgast). 42 Figure 5.3: Arabian fat-tailed scorpion, Androctonus crassicauda (Olivier 1807), is one of the most lethal scorpion species in the world. This mildly aggressive species can grow up to 10 cm long and is the largest of the three species we found in Qatar. (photo by K.P. Puliafico). Figure 5.4: A pitted darkling beetle in the genus Adesmia is one of the few beetles commonly seen during daylight hours at Al Zubarah. These long-legged beetles are well adapted to running on the hot sands of the desert. Their angular shape and hard exoskeletons help to protect them from daytime predators like birds. (photo by K.P. Puliafico). 43 Figure 5.5: Blaps cf. mortisaga Linnaeus, 1758, is the largest beetle species (over 4 cm) collected in the Al Zubarah Archaeological Site during our study. Like most of the beetles found here, it is most active at night. Despite its large size this species only feeds on plant material (photo by J. Kielgast). Figure 5.6: A bee pollinating flowers near the grassy depressions found in the rocky desert near Al Zubarah Fort. Possibly a member of the family Apidae, pollinators like this are most active during the day because they search out the flowers by sight (photo by K.P. Puliafico). 44 Figure 5.7: Other pollinators such as these moths (Utetheisa sp.) are active at night. This pair is preparing to lay eggs for the next generation (photo by J. Kielgast). 45 6. FISHES Peter Rask Møller1, Philippe Provencál1, Jos Kielgast2 (Vertebrate department1 and Center of GeoGenetics2, Natural History Museum of Denmark, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark, Email: pdrmoller@snm.ku.dk). 6.1 INTRODUCTION The Arabian Gulf including the waters of Qatar is characterized by shallow depths, high salinity and dramatic seasonal temperature changes (Sheppard et al. 1992). Due to recent glacial extinctions, the fish fauna is relatively young and comparatively species poor for a tropical sea, but characterized by high productivity and abundance. This extreme marine setting makes the Gulf interesting for conventional marine biodiversity research, but even more so as a study ground for the development of novel bio-monitoring approaches. For the present study we tested the use of night snorkeling as a tool for fish biodiversity surveys. Additionally environmental DNA (eDNA) was sampled, in filtered seawater for future biodiversity projects (see Thomsen et al. 2011, 2012). Danish investigations of the marine fauna of Arabian seas goes back to the famous Arabian Voyage 1761-67, where Peter Forsskål collected and described more than 150 common fish species. The type specimens of 60 species are still stored at the Natural History Museum of Denmark, available for modern studies (see http://www.zmuc.dk/VerWeb/Peter_ Forsskaal/Peter_Forsskaal.html). A very significant contribution to the knowledge about the fish fauna of the Arabian Gulf was the survey (1937-38) and publication (1944) of H. Blegvad and B. Løppenthin. In fact there was no comprehensive scientific study of the ichthyofauna of the region until this Danish systematic survey. Despite of many more recent studies in the region (e.g. Randall 1995, Carpenter et al. 1997, Al-Baharra 1986), the fish fauna in the waters of Qatar is still not well known. Available literature does not represent complete check-lists (e.g. Sivasubramaniam 1982) and no detailed information exists on the fish fauna of the Al Zubarah buffer zone. The purpose of the present study is to provide an overview of the local fish fauna in the Al Zubarah buffer zone. With only 3 weeks of field work all we can expect is a snapshot of the fauna and more seasons is needed before we can expect a full species list. We hope, however, that this small contribution will be relevant for the future management of the zone and that it will represent at new beginning and continuation of the fine Danish tradition for exploration in the Arabian Seas 6.2 METHODS The buffer zone was investigated from the 6 March to 23March 2012, by multi-mesh gill-net (one night), baited traps (3 nights), and snorkelling with camera (Cannon D7) and spear gun (3 nights, 2 days) and scuba diving (3 day divers). Gil-netting was abandoned due to problems with floating seaweed. The baited traps were not successful and caught swimming crabs only. 46 6.3 FINDINGS WITHIN THE BUFFER ZONE A total of ca. 48 identified fish species in 33 families were observed, photographed and/or caught within the buffer zone. A total of 33 species were collected for museum collections and tissue was sampled for DNA sequencing. A total of 35 species were UV-photographed. Most species were observed along the Ras Ushayriq area, where 38 species were found. At the old pier 13 species were recorded, followed by 7 at the outer zone and 1 in the canal, respectively. The most diverse families were seabreams (Sparidae) with 5 species, followed by stingrays (Dasyatidae), Blennies (Blennidae) and emperors (Lethrinidae) all with 3 species. 6.4 SPECIES LIST Species recorded inside the Buffer zone. Families listed according to Nelson (2006). Abundance symbols: - absent, + 0-5 specimens, ++ 6-50 specimens, +++ >50 specimens observed. Localities Ras Ushayriq Outer zone UV photo Collected Old Pier Canal IUCN Species Hemiscylliidae Chiloscyllium arabicum Gubanov, 1980 Arabian carpetshark NT - + - - + - Dasyatidae Himantura gerrardi (Gray, 1851)* Sharpnose stingray VU - + - - - + Himantura uarnak (Gmelin, 1789) Honeycomb stingray VU - + - - + + Pastinachus sephen (Forsskål, 1775)* Cowtail stingray DD + - - - + + Chanidae Chanos chanos (Forsskål, 1775)* Milkfish NE - + - - - + Plotosidae Plotosus lineatus (Thunberg, 1787)* Striped eel catfish NE - + - - + + Batrachoididae Allenbatrachus grunniens (Linnaeus, 1758)* Grunting toadfish NE - + - - + + Mugilidae Moolgarda seheli (Forsskål, 1775)* Bluespot mullet NE - +++ - - + - Mugilidae sp. (juv) - - - + - + Atherinidae Atherinomorus lacunosus (Forster, 1810)* Hardyhead silverside NE + +++ - - + + Hemiramphidae Hemiramphus far (Forsskål, 1775)* Black-barred halfbeak NE - ++ - - + + Belonidae Ablennes hians (Valenciennes, 1846)* Flat needlefish NE + ++ - - - + Tylosurus crocodilus (Péron & Lesueur, 1821)* Hound needlefish NE - ++ - - - + Platycephalidae Platycephalus indicus (Linnaeus, 1758)* Bartail flathead DD - ++ - - + + Serranidae Epinephelus tauvina (Forsskål, 1775)* Greasy grouper DD - ++ - - + + Pseudochromidae Pseudochromis percicus Murray, 1887* Bluespotted dottyback NE - - + - + - Teraponidae Terapon puta (Cuvier and Valenciennes, 1829)* Smallscaled terapon NE ++ - - - + - 47 Apogonidae Apogonichthyoides nigripinnis (Cuvier, 1828)* Bullseye NE + +++ - - + + Carangidae Carangoides bajad (Forsskål, 1775) Orangespotted trevally NE - + - - + + Gnathanodon speciosus (Forsskål, 1775)* Golden trevally NE - + - - - - Lutjanidae Lutjanus argentimaculatus (Forsskål, 1775) Mangrove red snapper NE - + - - - + Lutjanus fulviflamma (Forsskål, 1775) Dory snapper NE +++ +++ - - + + Gerreidae Gerres oyena (Forsskål, 1775)* Common silver-biddy NE +++ +++ - - + + Haemulidae Plectorhinchus sordidus (Klunzinger, 1870)* Sordid rubberlip NE - ++ - - + + Nemipteridae Scolopsis ghanam (Forsskål, 1775) Arabian monocle bream NE - + + - + - Black-streaked monocle NE Scolopsis taeniata (Cuvier, 1830) - + - - + - bream Lethrinidae Lethrinus lentjan (Lacepède, 1802) Pink ear emperor NE - ++ - - + + Lethrinus microdon (Valenciennes, 1830) Smalltooth emperor NE - ++ - - - + Lethrinus nebulosus (Forsskål, 1775) Spangled emperor NE - ++ - - - + Sparidae Acanthopagrus bifasciatus (Forsskål, 1775) Twobar seabream NE + ++ - - + + Diplodus sargus kotschyi (Steindachner, 1876) One spot seabream NE - ++ - - + + Rhabdosargus haffara (Forsskål, 1775) Haffara seabream NE - ++ - - + + Rhabdosargus sarba (Forsskål, 1775) Goldlined seabream NE + - - - + - Sparidentex hasta (Valenciennes, 1830) Sobaity seabream NE - + - - + + Mullidae Upeneus tragula Richardson, 1846* Freckled goatfish NE + + + - + + Monodactylidae Monodactylus argenteus (Linnaeus, 1758)* Silver moony NE - ++ - - - + Chaetodontidae Black-spotted LC Chaetodon nigropunctatus Sauvage, 1880* - + - - - - butterflyfish Pomacanthidae Pomacanthus maculosus (Forsskål, 1775) Yellowbar angelfish LC - + - - + + Labridae Halichoeres leptotaenia Randall & Earle, 1994* A labrid NT - - + - + - Blennidae Ecsenius pulcher (Murray, 1887)* A combtooth blennie NE - + - - - - Istiblennius lineatus (Valenciennes, 1836)* Lined rockskipper NE + - - - + - Petroscirtes ancylodon Rüppell, 1835* Arabian fangblenny NE - - + - + - Gobiidae Istigobius ornatus (Rüppell, 1830)* Ornate goby NE - + - - + - Cryptocentrus lutheri (Klausewitz, 1960)* Luther's shrimpgoby NE - - + - + - Gobidae sp. - + - - + - Siganidae Siganus canaliculatus (Park, 1779)* White-spotted spinefoot NE + +++ - - + + Acanthuridae Acanthurus sohal (Forsskål, 1775)* Sohal surgeonfish NE - + - - - + Sphyraenidae Sphyraena obtusata Cuvier, 1829* Obtuse barracuda NE - ++ - - + + Soleidae Solea stanalandi Randall & McCarthy, 1989* Stanaland's sole NE + - - - - + Monacanthidae Paramonacanthus oblongus (Temminck & Schlegel, NE Hair-finned filefish - - + - + - 1850)* 48 Hemiscylliidae (Bamboo sharks) A single specimen of Chiloscyllium arabicum was photographed at night near Ra`s ´Ushayriq. Dasyatidae (Whiptail stingrays) Three species were collected from the zone. Most abundant seems to be Pastinachus sephen often observed at daytime around the Old Pier. Himantura gerrardi has not previously been reported from Qatari waters. The collected specimens are currently being studied by Peter Last, CSIRO, Hobart. Chanidae Chanos chanos was collected from the Ras Ushayriq area, where it appeared to be common. Plotosidae (Eeltail catfishes) A few specimens of Plotosus lineatus was found on rocky bottoms at Ras Ushayriq. Batrachoididae (Toadfishes) A few specimens of Allenbatrachus grunniens was found at night time on rocky bottoms at Ras Ushayriq. Mugilidae (Mullets) Huge schools of mullets were seen at night time in shallow water near Ras Ushayriq. Based on the dark spot in the pectoral basis they were identified as Moolgarda seheli, but more species might be present. A juvenile specimen (12 mm) was collected in the old water canal by the entomology team (Anne and Ken). Mullidae (Goatfishes) A single species, Upeneus tragula, was observed both day and night in low densities in the zone. The taxonomy of the genus is currently being investigated by Franz Uiblein, IMR, Bergen. Schools of another species, Parupeneus margaritatus Randall & Guézé, 1984 were photographed at Düvel Rock. Atherinidae (Old world silversides) At least one species, Atherinomorus lacunosus, was very common in the zone. They were observed near the surface at night. Hemiramphidae (Halfbeaks) A single species Hemiramphus far was common near Ras Ushayriq. 49 Belonidae (Needlefishes) At least two species seen and collected at night time near Ras Ushayriq. Platycephalidae (Flatheads) A single species Platycephalus indicus was common on sandy bottoms near Ras Ushayriq. Serranidae (Sea basses) A single species, Epinephelus tauvina, was common near Ras Ushayriq. Some of the archeologists had also seen it from the Old Pier. It was found in rocky crevices, but they were also observed digging holes in the sand in very shallow waters, typically under rocks. This is one of the most popular and valuable species at local fish markets. Specimens up to 69 cm were collected. Pseudochromidae (Dottybacks) A few specimens of Pseudochromis percicus was photographed at daytime during diving in the outer zone. Teraponidae (Terapons) A small school of Terapon puta was photographed in the sea weed near the “Old Pier”. Apogonidae (Cardinal fishes) At least one species, Apogonichthyoides nigripinnis, was very common in the zone. A bright red specimen seen in a cave at Ras Ushayriq is likely to represent another species. Carangidae (Jacks and pompanos) Carangoides bajad was caught and UV photographed, whereas Gnathanodon speciosus was observed only, and needs further confirmation. Lutjanidae (Snappers) Two species was found in the zone. Lutjanus fulviflamma was very common including juvenile specimens. Gerridae (Mojarras) At least one species was very common on the shallow, sandy habitats. More species is likely to be present. Haemulidae (Grunts) A single species, Plectorhinchus sordidus, was common at Ras Ushayriq. 50 Nemipteridae (Thredfin breams) Scolopsis ghanam and Scolopsis taeniata were observed and photographed in low numbers in the zone. Lethrinidae (emperors) Three species were caught near Ras Ushayriq. Most common was Lethrinus lentjan, a well know fish at local fish markets. All were recorded by night snorkelling, except for one specimen found in a trap set by locals from Ras Ushayriq. Sparidae (Porgies) Five species were recorded in the buffer zone. Most abundant was Acanthopagrus bifasciatus, whereas Rhabdosargus sarba was represented by a single specimen. The largest collected specimen was a Sparidentex hasta, 66 cm in total length. Monodactylidae (Moonfishes) A school of ca. 10 specimens of Monodactylus argenteus was observed at night time at the tip of the Ras Ushayriq area. Chaetodontidae (Butterflyfishes) A single species was observed at Ras Ushayriq. Furthermore, the species was collected from the pier at Al Dhabiyah and photographed at Düvel Rock. Pomacanthidae (Angelfishes) A single species, Pomacanthus maculosus, was observed at Ras Ushayriq. Two specimens were collected from a trap set by local people from Ras Ushayriq. Labridae (Wrasses) A few specimens of Halichoeres leptotaenia, was photographed in the outer zone at ca. 5 m. Blennidae (Combtooth blennies) Three species were observed in the buffer zone, all as single specimens. None of these were collected and further confirmation of the identification is needed. One of the species, Petroscirtes ancylodon, was collected outside the zone at the pier at Al Dhabiyah. Gobiidae (Gobies) Two species were identified on the basis of photos, whereas one additional photographed species was left unidentified. Only few specimens of gobies were observed. 51 Siganidae (Rabbitfishes) A single species, Siganus canaliculatus, was very abundant near Ras Ushayriq. Rabbitfishes can change colour in seconds, so more species might be present. The identification of S. canaliculatus was based on the dark spot behind the head. Acanthuridae (Surgeon fishes) A single specimen of Acanthurus sohal was observed and collected at the tip of Ras Ushayriq. It was found in the rocky pier at night time. Sphyraenidae (Barracudas) One species, Sphyraena obtusata, was common in small groups at Ras Ushayriq. Soleidae (Soles) A single, juvenile specimen of Solea stanalandi was collected by a push net near the Old Pier. Monacanthidae (Filefishes) A single juvenile specimen of Paramonacanthus oblongus was photographed hiding in sea weed, at the outer buffer zone at about 5 meters depth 6.5 FINDINGS, OTHER LOCATIONS Pier at Al Dhabiyah This abandoned harbor seemed to be severely polluted by oil. During one hour night snorkeling 13 March 2012, the following species were collected: Petroscirtes ancylodon, Chaetodon nigropunctatus, Plectorhinchus sordidus, Lutjanus argentimaculatus and Epinephelus tauvina. Coral heads The coral there were all dead, and only two fish species were observed during day snorkeling 12 March 2012: Acanthopagrus bifasciatus and Scolopsis taeniata. Düvel Rock Many living corals were forming reef. Most, however, were covered by algae. Several fish species were photographed during day, snorkeling on the 12 March 2012: e.g. Acanthopagrus bifasciatus, Scolopsis ghanam, Parupeneus margaritatus, Chaetodon nigropunctatus, Abudefduf vaigiensis (Quoy & Gaimard, 1825), Pseudochromis percicus, Lutjanus fulviflamma and Ecsenius pulcher. 52 6.6 CONCLUSIONS AND RECOMMENDATIONS A total of 48 species in 33 families were found in the buffer zone. No less than 31 of the species are not mentioned as present in Qatari waters in FishBase (Frose and Pauly 2012), a global internet facility much used by ichthyologists and managers worldwide. This large proportion is most likely caused by the fact that no study of the Qatari fish fauna has been done, and that few entries of Qatari fishes have been made into the database. Most of the species (39) have not been evaluated for the international IUCN Red list of threatened species. Two species of stingray (Himantura gerrardi and Himantura uarnak (Figure 6.3) are considered Vulnerable (VU), a shark Chiloscyllium arabicum (Figure 6.2) and a labrid Halichoeres leptotaenia are considered Near threatened (NT), a butterflyfish Chaetodon nigropunctatus and an angelfish Pomacanthus maculosus are placed in the Least Concern (LC) category, whereas another stingray Pastinachus sephen (Figure 6.4), a flathead Platycephalus indicus (Figure 6.5) and a grouper Epinephelus tauvina (Figure 6.6) are currently Data Deficient (DD). Although the current study does not have a quantitative approach, some species were clearly more common than others. Based on the observations the following 20 species can be considered common in the zone: Ablennes hians, Acanthopagrus bifasciatus, Apogonichthyoides nigripinnis, Atherinomorus lacunosus, Diplodus sargus kotschyi, Epinephelus tauvina, Gerres oyena, Hemiramphus far, Lethrinus lentjan, Lethrinus microdon, Lethrinus nebulosus, Lutjanus fulviflamma, Monodactylus argenteus, Moolgarda seheli, Platycephalus indicus, Plectorhinchus sordidus, Rhabdosargus haffara, Siganus canaliculatus, Sphyraena obtusata, and Tylosurus crocodiles. Recommendations for future studies The buffer zone contains a wide range of habitats, from sea grass, sand, mud, rocks, sea weed and biogenous reefs of peal oysters. The zone seemed to be in a healthy environmental shape; with a few signs of human impact were notes. No oil spills were observed in the water, in sharp contrast to the situation on shore, but a few lost fishing gill nets were seen, as well as an unmarked fish trap. In order to preserve the natural environment in the zone, it is essential that fishing and collection of animals and plants are regulated. The zone should probably be a No-Take zone, so that snorkelers and divers get to see as mush of the fauna as possible. Alternatively – a small bag-limit could be introduced – so that recreational fishermen and visitors can catch a few fish or blue-crabs per day. Gill-nets, trawls and traps should not be allowed, whereas fishing rods, spear guns, and hand nets should be accepted, but with strict bag-limits. As many as 35 % (17 of 48) of the species recorded from the zone are described by Forsskål (1758). The types of most of these species were collected in the Red Sea during the Arabian voyage (1761- 67), most likely obtained from local fishermen. Future studies based partly on the collections made in the Buffer zone may focus on the small morphological (and genetic) differences that are found between Red Sea and Arabian Gulf populations. Several fish species were heavily infected with external parasites. Some of the worst examples were seen on e.g. Upeneus tragula, Platycephalus indicus, Rhabdosargus sarba, Siganus canaliculatus, Plectorhinchus sordidus, Apogonichthyoides nigripinnis, Gerres oyena and Scolopsis taeniata were 53 infected by unknown fish lice (Argulidae) (Figure 6.5). The high parasite density might be caused by osmoregulatory stress as a result of the high salinity. The parasite fauna is another obvious topic for future studies in the buffer zone. 54 References Al-Baharra , W.S. 1986: Fishes of Bahrein, Ministry of Commerce and Agriculture, Directorate of Fisheries, Bahrein. Blegvad, H. & Løppenthin, B. 1944. Fishes of the Iranian Gulf. Einar Munksgaard. Carpenter, K.E., Krupp, F., Jones, D.A. & Zajons, U. 1997. The living marine resources of Kuwait, Eastern Saudi Arabia, Bahrain, Qatar & the United Arab Emirates. Rome : FAO. Forsskål, P. 1775. Piscium. In: Niebbuhr, C. (Ed.) Descriptiones Animalium. Haunia. Froese, R. and D. Pauly. Editors. 2012.FishBase. World Wide Web electronic publication. www.fishbase.org, version (08/2012). Randall, J. E. 1995. Coastal Fishes of Oman. University of Hawaii Press. Sheppard, C.R.C., Price, A.R.G. & Roberts, C.M. 1992. Marine Ecology of the Arabian Region. Academic Press, New York. Thomsen, P. F., Kielgast, J., Iversen, L. L., Wiuf, C., Rasmussen, M., Gilbert M. T. P., Orlando, L. & Willerslev, E. 2011. Monitoring endangered freshwater biodiversity by environmental DNA. Molecular Ecology 21: 2565–2573). Thomsen, P. F., Kielgast, J., Iversen, L. L., Møller, P.R., Rasmussen, M. & Willerslev, E. 2012. Detection of a diverse marine fish fauna using environmental DNA from seawater samples. PLoS ONE 7(8): e41732. Doi: 10.1371/journal.pone.0041732. Sivasubramaniam, K. 1982: Common Fishes of Qatar, marine Science Department, Faculty of Science, University of Qatar, Doha. 55 Figures Figure 6.1: Map of collection sites. Four subareas studied: 1. “Old Pier”, 2. “Ras Ushayriq”, 3. “Outer zone” and 4. “Old Canal”. 56 Figure 6.2: Arabian carpetshark , Chiloscyllium arabicum, phographed near Ras Ushayriq, at night 12 March 2012. Photos PRM. 57 Figure 6.3: Honeycomb stingray Himantura uarnak. Upper photo: Specimen photographed near Ras Ushayriq, at night 12 March 2012. Lower four photos of a specimen ZMUC uncat. (10114), collected near Ras Ushayriq, at night 10 March 2012. 58 Figure 6.4: Cowtail stingray Pastinachus sephen. Upper photo of an un-collected specimen from the Old Pier, 9. March 2012. Lower five photos of ZMUC uncat. (10225) collected 9 March 2012 near the Old Pier. 59 Figure 6.5: Bartail flathead Platycephalus indicus phographed near Ras Ushayriq, at night 12 March 2012. Mid-left and lower photo shows a specimen infected with an unknown fish lice (Argulidae). Photos PRM. 60 Figure 6.6: Greasy grouper, Epinephelus tauvina photographed and/ or collected from near Ras Ushayriq, at night 8 -12 March 2012. 61 7. NAMES OF FISH AND OTHER MARINE ANIMALS IN QATAR Philippe Provençal (Vertebrate department, Natural History Museum of Denmark, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark), 7.1 INTRODUCTION My area of study was two folds for this field season. Firstly, the goal was to gather local fish names in Qatar, in the NW region and in Doha. Secondly, the aim was to compare these names to fish names gathered from the Red Sea and the Mediterranean region as well as to names known from Classical Arabic texts. 7.2 LOCAL FISH NAMES The study was effectuated by visits to a smaller fish shop in the town of Al Shamāl and to a larger fish market in Doha. The sellers were asked for the names of the fishes displayed. As all sellers were from the Indian subcontinent, the names provided were subsequently compared with the names provided by K. Siwasubramaniam & M.A. Ibrahim (1982) and Wajeeh S. Al-Baharna (1986) whose publications were at the Al Zubarah camp. Below are the lists showing the scientific names of the investigated species and the corresponding local Arabic names in both Latin scientific transcriptions and in Arabic letters. The scientific names are written in italics. Table 7.1. Names gathered from a fish shop in Al Shamāl 18 March 2012 1 Siganus javus ṣāfī ‫صافِي‬ َ 2 Rhabdosargus sarba qurqufān ‫قُرْ قُفَان‬ 3 Carangoides chrysophys suwaydī ‫ُس َو ْي ِدي‬ 4 Coral groupers Serranidae spp. (common designation) hāmūr ‫هَا ُمور‬ 5 Alepes mate jī ‫ِجيش‬ 6 Lethrinidae (common designation) a rī ِ ‫َشع‬ ‫ْري‬ ( 7 Queenfish Scomberoides commersionanus laḫlā (laḥlāḥ) ( 8 Crab qabqab ‫قَ ْبقَب‬ 9 Cuttlefish Sepia sp. ḫatāq ‫َخثَاق‬ Table 7.2. Names gathered from a poster in Doha on 21 March 2012 10 Queenfish Scomberoides commersionanus basār ‫بسار‬ qī (vocalisation according to the 11 Parrot fish Scaridae sp. ‫قيش‬ pronunciation provided at the fish market) 62 Table 7.3. Names gathered at the fish market in Doha on 21 March. 12 Doublebar bream Acanthopagrus bifasciatus fuskar ‫فُ ْسكَر‬ 13 Scomberomorus commersoni kanʿad ‫َك ْن َعد‬ 14 Gnathanodon speciosus rubīb ‫ُربِيب‬ 15 Mylio berda uʿm ‫ُشعْم‬ 16 Barracuda Sphyraena sp. jidd ّ‫ِجد‬ 17 Arius thalassinus kim (Classical Arabic) im (Qa ar dialect) ‫ِكم‬ 18 Scaridae qī ‫قِيش‬ 19 Lethrinus spp. (common designation) uʿrī ِ ‫ُشع‬ ‫ْري‬ 20 Rhonciscus stridens ḫīrah ‫ِخيرة‬ 21 Nemipterus sp. bāsī ‫بَا ِسي‬ 22 Trevallly Carangidae sp. zubaydī ‫ُزبَ ْي ِدي‬ 23 Blue Crab kukūb ‫ُك ُكوب‬ 24 Shark Pleurotremata sp. nawr ‫نَوْ ر‬ ّ،ّ‫هَا ُمور‬ 25 Coral groupers Serranidae spp. (common designation) hāmūr, hamūr ‫هَ ُمور‬ 26 Platax orbicularis ḥammūd, hammūd ‫ّهَ ُّمود‬،ّ‫َح ُّمود‬ 27 Euthynnus affinis (identity not certain) ḍabān, dabbān ‫ّ َدبَّان‬،ّ‫ضبَّان‬ َ 28 Coral hind Cephalopholis miniatus ridāmūr ‫ِردَا ُمور‬ 29 Sole Soleidae sp. mūsā ‫ُمو َسا‬ 30 Black tail bream Diplodus sargus kirkafān ‫ِكرْ َكفَان‬ ʿāqūl (correct pronunciation persumably ّ‫عَاقُول‬ 31 Needlefish Tylosurus leiurus, Ablennes hians ḥākūl) ‫( َحا ُكول)ـ‬ 32 Pony fish Leiognathus sp. zubaydī ‫ُزبَ ْي ِدي‬ sikkīn (correct pronunciation persumably ّ‫ّ ِس ِّكين‬ 33 Rachycentron canadus sikin) ‫( ِس ِكن)ـ‬ 34 Greater amberjack Seriola dumerili ḥamām ‫َح َمام‬ 35 mullet Mugilidae sp. būrī ‫ُوري‬ِ ‫ب‬ ّ‫س ُْلطـَان‬ 36 Red goatfish Parupeneus sp. sul ān Ibrāhīm ‫إِ ْب َرا ِهم‬ 37 Halfbeack Hemiramphidae sp. surūs ‫ُسرُوس‬ 38 Flathead Thysanophrys sp. baḥr ‫بَحْ ر‬ 63 7.3 MAIN CONCLUSIONS FROM THE WORK UNDERTAKEN THIS SEASON The list of fish names from Qatar bears witness to the elaborate knowledge of the sea and sea creatures found in the coastal populations around the Arabian Peninsula. Such knowledge manifest itself in a wide vocabulary regarding all sea animals and the phenomenon was documented by the research of Peter Forsskål (1775) and earlier by the description of fishes and sea creatures in the Arabic geographical and scientific texts from the Classical period, ie. the period from the VIIIth to the XVth century AD. The fish names gathered during the Qatar field work season show some interesting features. When the names are compared to names for the same species from the Red Sea, it is clear that not only is the notion of species the same on both shores of the Arabian Peninsula, but the notion of genera also remain the same. Thus Coral Groupers Serranidae spp. are designated by a common name both in the Red Sea and in the Arabic Gulf, but this designation varies with the locality. In Egypt and in other locations around the Red Sea the groupers as a group are called u ar while their common designation in the Qatar is hamūr. In Egypt the members of the gender Lethrinus have aᶜūl as common designation while in Qatar they have uᶜrı as common designation (personal observations). This pattern illustrates that the Red Sea and Qatar represent two different linguistic entities when it comes to the lexical contents of the fish fauna, but the notion of which species belong together in families or genera remain the same across the Arabian Peninsula. More interesting is the fact, that some species, e.g. mullets, mugilidae, have the same name as the one used in Arabic in the Mediterranean region. In Egypt the fact that mullets are called by the same name both on the Mediterranean and on the Red Sea Coasts could be explained by the geographical proximity of the Red Sea to the Mediterranean. But the fact that mullets are called by the name būrī in both the Mediterranean region, the Red Sea and in Qatar shows an interesting continuity. Such continuity is also visible in the historical record, as this fish name is well known from Classical Arabic texts. For instance in the list of 79 fishes and aquatic animals included in the descriptions of the island of Tinnīs in Lake Manzalah in the Delta of the Nile by Yāqūt ibn ᶜAbdullāh al-Ḥamāwī (1179-1229) in his book Muᶜjam al-Buldān and by Zakariāʾ b. Muḥammad b. Maḥmūd al-Qazwīnī (1203-1283) in his Cosmography, Kitāb ʿAjāʾib al-Maḫlūqāt wa Gharāʾib al-Mawjūdāt. A similar observation could be made on the name of the red goatfish. The Mediterranean species of red goatfish Mullus barbatus and Mullus surmuletus are called sul ān Ibrāhīm in Egypt and the Syro-Palestinian region (Oman 1966 and personal observations), and a red species of mullet is known by this name at the fish market in Dawḥah. This name has also been recorded for the Red Sea area, but it is not very common there (cf. Provençal 1997, Oman 1992, personal observations). 7.4 FUTURE RESEARCH This study is part of the project The Arabic Animal-Names of Forsskål’s Descriptiones Animalium by myself and B. Skaarup, housed at the Zoological Museum in Copenhagen. The aim is to systematize the Arabic names of primarily fishes and other marine animals noted by Peter Forsskål during the expedition The Arabian Voyage 1761-1767. This material has not yet been adequately studied. This systematisation includes investigations in contemporary ichtiological nomenclature around the Arabic Peninsula and in the Middle East as well as studies of fish 64 nomenclature in Classical Arabic geographical and scientific texts. In this respect material from Qa ar is of great interest as it provides data from a part of the Arab World, where the sea and fisheries play a crucial role from antiquity to the present day and thus gives an indispensable contribution to investigations on the importance of the sea and its resources and on the knowledge of marine life the Arabic culture in both Classical and modern context. 65 References Al-Baharna, W. S. 1986. Fishes of Bahrein, Ministry of Commerce and Agriculture, Directorate of Fisheries, Bahrein. Forsskål, P. 1775. Descriptiones Animalium, post mortem auctoris edidit Carsten Niebuhr. Möller, Copenhagen. Oman, G. 1966. L'ittionimia nei Paesi Arabi del Mediterraneo. Quaderni dell'archivio linguistico veneto. Casa editrice Leo S. Olschiki. Firenze. Oman, G. 1992. L’Ittionimia nei paesi arabi dei Mari Rosso, Arabico e del Golfo Persico (o Arabico), Instituto Universitario Orientale, Napoli. Provençal, P. 1995. Enquête lexicographique sur les noms d'animaux en arabe /Alexicographic survey of arabic animal names. Ph. D. afhandling ved Carsten Niebuhr Instituttet for Nærorientalske Studier, Københavns Universitet. Provençal, P. 1997. Animal names gathered by interviews with members of the Muzın tribe in Sinai. Acta orientalia vol. 58 pp. 35-46. Siwasubramaniam, K. & M.A. Ibrahim 1982. Common Fishes of Qatar, Marine Science Department, Faculty of Science, University of Qatar, Doha. 66 Figures Figure 7.1: A specimen of Gerres acinaces gathered during the field work 2012. Specimen and photo: P. Provençal. 67 8. REPTILES Jos Kielgast (Natural History Museum of Denmark, Universitetsparken 15, 2100 Copenhagen, Denmark. joskielgast@snm.ku.dk). 8.1. INTRODUCTION Reptiles are among the most conspicuous terrestrial animals in the arid landscapes of the Arabian Peninsula and account for a substantial part of its vertebrate biodiversity. However, there are substantial knowledge gaps in our understanding of the herpetofauna in the region and Qatar features among comparatively poorly studied areas. Although reptiles have been covered in recent national check-lists (Mohammed 1988; El-Sherif and Al-Thani 2000) very little scientific work has been published on this sector of Qatar’s biodiversity. The country comprises characteristic habitats of the Persian Gulf desert and semi-desert terrestrial ecoregion (Olson et al. 2001) and is hence expected to harbour a fauna similar to the xeric lowlands of Western Saudi Arabia and The United Arab Emirates. Currently a total of 29 species have been recorded in Qatar (El-Sherif and Al-Thani 2000). In comparison WWF lists 53 species to occur in the ecoregion and adjacent marine areas (WWF 2006). This is some indication that the herpetological diversity in the country may potentially be underestimated. The reference work for the entire Gulf region (Leviton et al. 1992) provide an overview of all occurring species and subspecies including a total of 150 taxa from South Eastern most Iraq, Iran, Kuwait, Western Saudi Arabia, Bahrain, Qatar and the United Arab Emirates. Obviously this large area covers a number of habitats and zoogeographic zones which are not present in Qatar. However, the neighbouring and more thoroughly investigated United Arab Emirates has 72 nominal species which have been covered in a number of accessible popular and scientific publications (Baha el Din 1996; Hornby 1996, Jongbloed 2000; Gardner 2005, 2008; Baldwin et al. 2008; Soorae et al. 2010; Uetz et al. 2012). Such a substantial discrepancy in the known diversity of these two neighbouring states should spur curiosity to the exploration of Qatar. Intriguingly the most notable contribution to the knowledge on the herpetofauna of Qatar comes from Saudi Arabia in the form of the book series “Fauna of Saudi Arabia” which provide the most important broad herpetofaunistic references for the region (e.g. Arnold 1986; Gasperetti 1988, 1993; Schätti et al. 1994) as well as comprehensive treatment of specific groups (e.g. Arnold 1980, 1994; Hillenius 1984). 8.2 FINDINGS WITHIN THE BUFFER ZONE Methods A herpetological survey was conducted in the Al Zubarah archaeological site and buffer zone during 5 March to 23 March 2012. Activities were targeted at supplying a checklist as complete as possible for the local reptile fauna of the area. In arid landscapes the trade-offs between quantitative and qualitative survey methods are noticeable due to the low abundance animals and patchy nature of suitable micro habitat - in particular shelter from sun and predators. These features make quantitative survey approaches very inefficient and opportunistic transect walks much more feasible with regards to both number of records and taxonomic coverage. For similar reasons collection was 68 opportunistic by visual encounter and hand catching rather than installing time consuming contraptions such as pitfalls and drift fences. During the survey the entire area of the archaeological site and buffer zone was systematically surveyed by transect walks both during day and night. Due to the limited size of the buffer zone and the relatively long time frame of the survey it was possible to systematically investigate all areas at fine spatial scale with at least one visit during day and night. The survey activities were thereafter targeted at the most promising localities of the respective habitat types in the area. Day transect walk surveys in desert areas are normally focussed at mornings and late afternoons due to the basking and foraging behaviour of diurnal species. However, cold weather especially during the beginning of the survey made it more productive to focus on the hottest hours of the day. In spite of this unconventional strategy many findings of diurnal species were made, while the species were clearly inactive, by turning stones, vegetation and debris they hide under during night and hibernation. Nocturnal surveys were carried out from 19:00-05:00. The main habitat types in the area are sabkha and stony desert. There is a well vegetated transition between the two, but also patches of shrub and grass vegetation scattered around other parts of the area although it is generally sparsely vegetated. Moreover, there are a few natural stony ridges and several places with scattered piles of recent or historical building debris. The archaeological site itself makes up such important habitat for many species. Shelter in all forms, be it natural or man-made, is a decisive micro habitat feature to most of the species in the area. The localities with the highest abundance and diversity of reptiles in the buffer zone area also include two graveyards. Reptiles recorded: The Al Zubarah area hosts a relatively rich herpetofauna of lizards, amphisbaenids, snakes and sea turtles including at least 18 reptile species belonging to eight different families. Lizards are by far the most abundant and diverse group in the area with geckoes representing the most specious family. Lizards The gulf short-fingered gecko (Pseudoceramodactylus khobarensis) (Figure 8.1A) is perhaps the most widespread species in the area occurring both in the sabkha and the stony desert hinterland. It was found even in quite bare habitat of sparsely vegetated gravel plain, but more abundantly on moist soils. It is particular numerous around the area of the camel racetrack in the east of the buffer zone. The former congeneric slevin’s sand gecko (Stenodactylus slevini) (Figure 8.1B) also occur in the area, but it was only found where there were scattered large rocks or other shelter. The arabian rough-tailed gecko (Cyrtopodion scabrum) (Figure 8.1D) and baluch ground gecko (Bunopus tuberculatus) (Figure 8.1C) were very numerous in the same sheltered areas and especially around the stony ridges in the northern sector of the buffer zone and among the archaeological building debris. The latter is the more common of the two. The diurnal dwarf rock gecko (Pristurus rupestris) (Figure 8.1E) can be found in decent numbers on the permanent buildings of the archaeological site camp and scattered around the buffer zone were large rocks and permanent piles of historical or recent building debris occur. Only a single specimen of persian leaf-toed gecko (Hemidactylus persicus) (Figure 8.1F) was recorded in an old well north of the camp. The most prominent reptile species in the area is the large diurnal spiny-tailed lizard (Uromastyx aegyptia microlepis) (Figure 8.2A). This primarily herbivorous animal grows up to 80 cm in length and can be observed in vegetated areas sunbathing next to its burrows. During the survey only four 69 specimens were recorded inside the buffer zone. However, two of these were found very close to camp on the particularly hot last day of the survey while the third one was found hiding under debris during broad daylight. Many burrows were recorded scattered around the area during the survey and the finding of a juvenile documents that reproduction occur in the area. These findings all indicate that the temperature during the survey period was too low for the species to be active and that the density was hence most likely under estimated. Only one other agamid lizards, the yellow-spotted agama (Trapelus flavimaculatus) (Figure 8.2B), was observed a single time at the graveyard south of the camp, while short-nosed lizards (Mesalina brevirostris) (Figure 8.2C) were present at moderate density scattered throughout all habitats in the area - from the shore to the hinterlands. The peculiar fossorial amphisbaenid zarudnyi’s worm lizard (Diplometopon zarudnyi) (Figure 8.3E) was recorded with a single specimen under debris in an un-restored sector of the Al Zubarah archaeological site. Snakes A single specimen of each of the regions harmless rear-fanged colubrids, shokari sand snake (Psammophis shokarii) (Figure 8.3A) and hooded malpolon (Malpolon moilensis) (Figure 8.3D), were recorded at the old peer and along the road from camp to Al Shamal, respectively. The terrestrial fauna also include a single venomous species, the sand viper (Cerastes gasperetti) (Figure 8.3B). It was not recorded during the survey, but has been documented from the archaeological site where it is occasionally found during restoration activities. A number of sea snakes potentially occur in the marine area of the buffer zone (Gasperetti 1988, Baldwin & Gardner 2005) including the blue-banded sea snake (Hydrophis cyanocinctus), shaw’s sea snake (Lapemis curtus), yellow-bellied sea snake (Pelamis platurus), beaked sea snake (Enhydrina schistosa), common small headed sea snake (Microcephalophis gracilis), Arabian Gulf sea snake (Hydrophis lapemoides), and the ornate sea snake (Hydrophis ornatus). Only the latter two have been documented at Al Zubarah by collected dead specimens or photographs (e.g. Hydrophis ornatus Figure 8.3C), but none were observed during the actual survey. Turtles In terms of species conservation the presence of marine turtles in the waters off Al Zubarah is the most important component of the herpetofauna. Dead specimens of the endangered green turtle (Chelonia mydas) and the critically endangered hawksbill turtle (Eretmochelys imbricata) (Figure 8.4) are recurrently found washed up on the shore in the buffer zone, indicating that these species occur in the area on a regular basis. A single live adult green turtle and a sub-adult hawksbill turtle were observed by snorkelling during the survey. Skeletal remains of a leatherback turtle (Dermochelys coriacea) have also been recorded on the shore, but this species most likely only occur vagrantly in the area. The marine section of the buffer zone may hence constitute a foraging ground for sea turtles, but the shores are not known to be used as nesting sites. 70 8.2.1 species list Reptile species recorded within the Al Zubarah archaeological site buffer zone. Species which were not recorded during the survey, but documented by photos or specimens collected during the archaeological excavations are marked with * Agamidae Spiny-tailed lizard (Uromastyx aegyptia microlepis) Yellow-spotted agama (Trapelus flavimaculatus) Lacertidae Short-nosed lizard (Mesalina brevirostris) Gekkonidae Slevin’s sand gecko (Stenodactylus slevini) Gulf short-fingered gecko (Pseudoceramodactylus khobarensis) Persian leaf-toed gecko (Hemidactylus persicus) Rough-tailed gecko (Cyrtopodion scabrum) Baluch ground gecko (Bunopus tuberculatus) Dwarf rock gecko (Pristurus rupestris) Trogonophidae Zarudnyi’s worm lizard (Diplometopon zarudnyi) Viperidae Arabian horned viper (Cerastes gasperetti)* Colubridae Shokari sand snake (Psammophis shokarii) Hooded malpolon (Malpolon moilensis). Hydrophiidae Ornate sea snake (Hydrophis ornatus)* Arabian Gulf sea snake (Hydrophis lapemoides)* Cheloniidae Green turtle (Chelonia mydas) Hawksbill turtle (Eretmochelys imbricata) Leatherback turtle (Dermochelys coriacea)* 71 8.3 CONCLUSIONS AND RECOMMENDATIONS A total of 18 reptile species belonging to eight different families were recorded within the buffer zone of Al Zubarah archaeological site during the survey in March 2012. This equals almost two thirds of all species previously known from the entire country. However, all recorded species are widespread in the gulf region and their occurrence in the area is neither surprising nor unique. Notably the marine turtles recorded in the area are of global conservation concern. The buffer zone does not include coral reef structures or sea grass beds substantial enough for the area to be regarded as important key habitat for these species. However, they do occur and forage here and may hence constitute the most important biological feature of the area to global biodiversity conservation. The generally cold weather conditions during the survey and consequently inactive behaviour of the reptile fauna likely caused an underestimate of species abundance and possibly even species richness. To ensure that the list of species from the area is complete would necessitate further survey activity during a later part of the spring or early fall, when animals are at their peak activity. This would certainly also be necessary to ensure a more complete spatial mapping e.g. of the occurrence of Uromastyx in the area. A few recommendations can be made regarding the conservation of the reptile diversity recorded in the buffer zone of Al Zubarah archaeological site. All grazing by domestic animals should be avoided to protect the vegetation in the area both as important micro habitat structure and as a source of food. This is particularly important for Uromastyx, but also for other species in this reptile assemblage. Compared to surrounding areas the buffer zone of Al Zubarah archaeological site is notably rich in vegetation due to hydrology and may therefore be considered important to local reptile conservation. Conversely little has been done to document the fauna in surrounding areas so it is premature to conclude anything on a larger spatial scale. Human disturbance in the area should be kept to a minimum and in particular sheltered areas of rocks, debris and vegetation should be left as is. Any bulldozing, re-location or removal of rocks, vegetation or building debris (new or historical) will likely be harmful to the resident reptile fauna, as these landscape features provide essential micro habitat. Similarly restoration practises on buildings at the archaeological site which include filling of crevices and polishing surfaces with cement like materials make the area less suitable reptile habitat. On the other hand excavation activities that expose parts of the ruins that are currently completely covered by sand can provide new suitable habitat for some species. 72 References Arnold, E. N. 1980. Reptiles of Saudi Arabia. A review of the lizard genus Stenodactylus (Reptilia: Gekkonidae). Fauna of Saudi Arabia 2: 368-404 Arnold, E. 1986. A key and annotated check list to the lizards and amphisbaenians of Arabia. Fauna of Saudi Arabia 8:385-435. Arnold, E. N. & Gardner, A. S. 1994. A review of the middle eastern leaf-toed geckoes (Gekkonidae: Asaccus) with descriptions of two new species from Oman. Fauna of Saudi Arabia 14: 424-441. Baldwin, R., & Gardner, A.S. 2005. Marine Reptiles. Pp. 242–251. In The Emirates: A Natural History. Hellyer, P, and S. Aspinall (Eds.). Trident Press, London, U.K Baha El Din, S. 1996. Terrestrial Reptiles of Abu Dhabi. Pp. 124-147 In Desert Ecology of Abu Dhabi - a Review and Recent Studies. Osborne, P.E. (Ed.). Pisces Publications, Newbury, U.K. El-Sherif, G. & Al-Thani, A. S. 2000. Record, histological and enzyme histochemical demonstration of Qatari reptiles in relation to seasonal and environment variations. Report for First Part HE/45/95 Project SARC, University of Qatar: 3. Gardner, A.S. 2005. Terrestrial reptiles. Pp. 229–241. In The Emirates: A Natural History. Hellyer, P, and S. Aspinall (Eds.). Trident Press, London, U.K Gardner, A.S. 2008. The Terrestrial Reptiles of the United Arab Emirates: Herpetological History, Zoogeography and Conservation. Pp. 281–307. In Terrestrial Environment of Abu Dhabi Emirate Gasperetti, J. 1988. The snakes of Arabia. Fauna of Saudi Arabia 9:169–450. Gasperetti, J., Stimson, A.F., Miller, J.D., Ross, J.P. & Gasperetti, P.R.. 1993. Turtles of Arabia. Fauna of Saudi Arabia 13:170–367. Hillenius, D. & Gasperetti, J. 1984. Reptiles of Saudi Arabia: the chameleons of Saudi Arabia. Fauna of Saudi Arabia 6: 513-527. Hornby, R. 1996. A checklist of the amphibians and reptiles of the UAE. Tribulus, 6: 9-13. Jongbloed, M. 2000. Field Guide to the Reptiles and Amphibians of the UAE. Atlas Printing Press. Levinton, A. E., Anderson, S. C., Adler, K. & Minton, S. A. 1992. Handbook to the Middle East Amphibians and Reptiles. Society for the Study of Amphibians and Reptiles. Mohammed M.B.H. 1988. Survey of the reptiles of Qatar. Proceedings of the Zoological Society A.R. Egypt, 15: 17-26. 73 Olson D, Dinerstein E, Wikramanayake E, Burgess N, Powell G, et al. (2001) Terrestrial ecoregions of the world: a new map of life on earth. Bioscience 51: 933-938. Pritpal S. Soorae, Myyas Al Quarqaz, and Andrew S. Gardner. 2010. An Overview and Checklist of the Native and Alien Herpetofauna of the United Arab Emirates. Herpetological conservation and biology, 5:529-536. Schätti, B. & Gasperetti, J. 1994. A contribution to the herpetofauna of Southwest Arabia. Fauna of Saudi Arabia 14: 348-423. Uetz, P. (ed.), The Reptile Database, http://www.reptile-database.org, accessed July 2012 WWF 2006. WildFinder: Online database of species distributions, ver. Jan-06. www.worldwildlife.org/WildFinder 74 Figures Figure 8.1: Lizards 75 Figure 8.2: Lizards 76 Figure 8.3: Snakes 77 Figure 8.4: Turtle Figure 8.5: Rocky outcrops in the buffer zone represent hotspots for reptile diversity at Al Zubarah 78 9. PALEONTOLOGY Jakob Walløe Hansen (Natural History Museum of Denmark, Øster Voldgade 5-7, DK-1350 Copenhagen K, Denmark, jwhansen@snm.ku.dk). 9.1 INTRODUCTION During three seasons from 2009 through to 2012 geomorphology and hydrology studies was carried out in the Al Zubarah Region and NW Qatar (Macumber, 2009, 2010, and 2011). Both are two important factors when studying the relationship between earlier human occupation and the natural environment. As a part of the UNESCO application in 2011, a brief survey and following assessment report on the biosphere of the Al Zubarah Region was initiated. This programme was extended in 2012 with a month of field work covering all main areas of the zoology of the area, including the paleontology. Preliminary paleontological sampling was carried out during the 2012 season in order to shed light on this unstudied part of the natural history. This involved smaller scale sampling in and around the ruins of the old town of Al Zubarah, where sub-fossils have been used as construction material. But fossils were also collected in the entire buffer zone along the coastline, on the mud plains, in the sabkha (and paleo-sabkha), and in the old canal and its ridges, as well as in the arid stony desert area towards the Al Zubarah Fort. An extended part of the preliminary survey, took place in the Dukhan Region, especially at Umm Bab (80 km south Al Zubarah) where a prolific Eocene formation, yielding a high quantity of fossil shark teeth, had a certain interest. This report covers the results of the surveys and tries to provide new information on the natural history of Al Zubarah and the Dukhan Region. 9.2 FINDINGS WITHIN THE BUFFER ZONE AT AL ZUBARAH The Al Zubarah Region stretches over a vast area, and exhibits many different habitats and rock types. Holocene deposits dominate the sabkha and mud plain areas, around the city ruins and close to the sea. Further inland, the prevalent habitat is arid stony desert, dominated by Eocene limestone. A certain habitat, labelled paleo-sabkha, covers the vast area in between the city ruins and the project compound. This is an area with low relief and only sparse vegetation. But huge quantities of Holocene fossils are present, and of which some were collected for reference (see species list). The excavation at Al Zubarah testifies to the use of fossiliferous limestone for construction material. These building blocks consist predominantly of smaller gastropods and are used for both floors as well as for walls. These limestone building blocks originate from the Holocene deposits in the sabkha and mud plain areas. 79 The Eocene limestone (Dammam Formation) yields almost no fossil evidence other than the random gastropod and bivalve. These are of considerable size though; the gastropods can easily grow to be more than 25 cm long. But they are only exposed in very few and poor outcrops. The lithology is a very hard and dolomitized limestone and it appears that the population only occasionally used this rock type for construction material. 9.2.1 Species list The fossils listed beneath are some of the most common among the sub-fossils found in the paleo- sabkha and on the present sabkha, as well as within the old city ruins. Furthermore, the fossil assemblage, include a large amount of especially gastropods. For studies on these, see end of season report by Dr. Aslak Jørgensen. Anodontia edentula (Linnaeus, 1758). Toothless clam, normally found buried in mudflats in the intertidal and subtidal zones. Asaphis violascens (Forsskål, 1775). Marine bivalve found in the intertidal mudflats. Barbatia sp. (Arcoidea). Ark clam, normally stuck under stones and rocks on the shore, and normally covered by periostracum (Figure 9.1). Hexaplex sp. Large marine gastropod sometimes referred to as rock snail (Figre 9.2). Pinnidae sp. Large saltwater/marine clam sometimes called pen shell. Lives anchored in sediment using a byssus (Figure 9.3) Strombus (Conomurex) persicus. Marine gastropod belonging to the true conchs (Figure 9.4). Tellinoidea sp. Marine bivalve living fairly deep in soft sediments in shallow seas. They respire using long siphons that reach up to the surface of the sediment. Turbo sp. Marine gastropod/large sea snail with gills and an operculum (Figure 9.5) Veneroidea sp. Venus clam with short siphon and long foot (Figure 9.6). 9.3 FINDINGS AT UMM BAB, DUKHAN REGION, SOUTHERN QATAR Smaller scale operations took place during the field season of 2012. Amongst other places, the Eocene formations of the Dukhan Region were examined and a number of samples were brought home to the University of Copenhagen for processing and further studies. The preliminary results of these examinations are quite spectacular. One small sample (approx. 250 grams) contained large quantities of fossil fragments from many animal groups, primarily from marine invertebrates such as echinoderms, foraminifera, and bivalves. Shark teeth were also present in large numbers (Figures 9.7 – 9.11). Species are not listed, but pictures of the relevant fossils are presented). 80 The geological surface of Qatar consists mainly of Eocene deposits (Al-Saad, 2005) and may so retain important information about the latest “greenhouse climate” of the Earth, since the Late Cretaceous. During this warm period, Qatar was situated centrally in the Tethyan Sea, which stretched from North Africa to East Asia. This area separated the warmer low latitudes from the cooler European Boreal Realm, making it an important location for environmental and faunal information. Shark teeth are an interesting source for faunistic and climatic information. They take top positions in the marine food web and due to rapid tooth replacement; their teeth are fairly common in the geological record. Furthermore, the enameloid of the teeth are very resistant to diagenesis (Kolodny & Luz, 1991; Pucéat et al., 2003) and as such, an almost perfect agent for preserving isotopic information, which can be utilized for very precise climatic information. Shark teeth are well known from the Midra Member of the Dammam Formation and may also occur in other formations. The last published paper on this important shark fauna was published by Casier (1971) and contained only larger taxa and thereby missed the important smaller species. This year’s results show that many of the smaller species are also present, and many of these are new to science. The before mentioned sample did contain more than 25 teeth; surprisingly so it also contained many fragments of marine invertebrates. Particularly the uncommon echinoderm morphological features attracted attention (Figure 9.11). These will be determined by experts from the Natural History Museum in London in order to properly determine the paleoecological characteristics. 9.4 CONCLUSIONS AND RECOMMENDATIONS At present the fossils from the buffer zone can be attributed to the Holocene age but not further. The geomorphology provides strong evidence for the sea level increasing and decreasing several times during the same period. But at present there is no strong difference in the fossil assemblage of the paleo-sabkha and the present sabkha, and therefore a true fossil assemblage analysis would be beneficial. Further studies of the geology of the buffer zone (and surrounding areas) can reveal a more complex solution to the natural history of the area. However, this will only be possible if large scale sampling is initiated. Furthermore it would require on shore and probably also off shore drilling to be initiated. Such a study should reveal structures within the sediments on which Al Zubarah rests; information than could enhance our understanding of the area. Finally, it would be interesting to investigate the geological formations of the Dukhan area in order to obtain further knowledge of the fossil assemblage of especially the Dammam Formation. Such a project could be linked with a newly initiated project at the Natural History Museum of Denmark, on the selachian faunas from the Eocene of Denmark at Trelde Næs. Thus the first initial step for comparing the fossil faunas of Qatar and Denmark could be taken and may include advanced isotopic research utilizing the latest technology within oxygen studies in phosphates. 81 References Al-Saad, H. 2005. Lithostratigraphy of the Middle Eocene Dammam Formation in Qatar, Arabian Gulf: effects of sea-level fluctuations along a tidal environment. Journal of Asian Earth Sciences 25: 781-789. Casier, E. 1971. Sur un materiel ichthyologique des Midra (and Saila) shales du Qatar (Golfe Persique). Bulletin de l'Institut Royal des Sciences Naturelles de Belgique 47 (2): 1-9. Kolodny, Y. & Luz, B. 1991. Oxygen isotopes in phosphate of fossil fish - Devonian to recent. In Taylor, H.P., O’Neil, J. R. and Kaplan, I. R. eds. Stable Isotope Geochemistry: A Tribute to Samuel Epstein. University Park, Geochemical Society 3: 105-119. Macumber, P. 2009. Preliminary Report on the Geomorphology and Hydrology of the Al Zubarah Region, Northern Qatar. Macumber, P. 2010. Geomorphology and Geoarchaeology. In Season 2 Report (2010). University of Copenhagen. Macumber, P. 2011. Geomorphology, Hydrology and Occupation across North-Eastern Qatar – Geomorphological and Geoarchaeological Results from the third Season of the Copenhagen University Study in Northern Qatar. Pucéat, E., Lécuyer, C., Sheppard, S.M., Dromart, G., Reboulet, S., & Grandjean, P. 2003. Thermal evolution of Cretaceous Tethyan marine waters inferred from oxygen isotope composition of fish tooth enamels. Paleoceanography 18 (2): 1029. 82 Figures Figure 9.1: Barbatia sp. (Arcoidea). (Photo by Sten Lennart Jakobsen). Figure 9.2: Hexaplex sp. (Photo by Sten Lennart Jakobsen) Figure 9.3: Pinnidae sp. (Photo by Sten Lennart Jakobsen) 83 Figure 9.4: Strombus (Conomurex) persicus. (Photo by Sten Lennart Jakobsen) Figure 9.5: Turbo sp. (Photo by Sten Lennart Jakobsen) Figure 9.6:. Veneroidea sp. (Photo by Sten Lennart Jakobsen) 84 Figures 9.7 and 9.8: Not yet described fossil shark teeth from Umm Bab, Dukhan Region. (SEM- photos by Jan Schulz Adolfssen) Figures 9.9 and 9.10: Not yet described fossil shark teeth from Umm Bab, Dukhan Region. (SEM- photos by Jan Schulz Adolfssen) 85 Figure 9.11: Echinoderm spines revealing special morphology features, yet to be described by experts. (Photo by Sten Lennart Jakobsen) 86 10. THE COMPARATIVE COLLECTION OF ANIMAL SKELETONS Jeppe Møhl and Pernille Bangsgaard (the Natural History Museum of Denmark, Quaternary Zoology, Department of vertebrate, jeppemhl99@gmail.com and bangsgaardp@hotmail.com). 10.1 INTRODUCTION The first biological field season of the QIAH project included the creation of a comparative collection due to the efforts of conservator Jeppe Møhl with some assistance from Pernille Bangsgaard. The aim of this collection is two-folds. Firstly, the comparative collection is a necessary tool for the Zooarchaeologists working with the excavated faunal remains from the QIAH excavations at Al Zubarah and other sites. These analyses were also initiated during the 2012 season. Due to this use the collection will be stored and used for scientific research at Al Zubarah camp in the coming years. Secondly, for the long term the comparative collection is intended to remain in Qatar as a tool for biological and zooarchaeological research. It is envisioned that the collection will become part of the planned Qatar Natural History Museum, where the collection can be maintained and remain accessible to other researchers. During the previous archaeological field season in 2011 to 2012, it became clear that opportunities for collecting skeletons in NW Qatar was abundant, from marine animals and birds to all terrestrial mammals, wild and domesticated. The situation is probably mainly due to the dry climate, the flat landscape and sparse vegetation, thus creating optimal conditions for spotting cadavers. This combined with a large team of biologists, archaeological surveyors and archaeologists, who are constant moving within the area, create optimal conditions for retrieving multiple dead animals. For few of the domesticated species it was possible to buy a fresh animal, but in most cases a cadaver was the only option, due to cost concerns. Cadavers were also the only option for many of the wild animals, particularly those which are currently featured on the IUCN red list of threatened species, such as the green turtles, Chelonia mydas. 10.2 THE PROCESS OF CLEANING A SKELETON The ideal procedure for skeletonising an animal is often hard to apply in the field, with less than ideal equipment. Furthermore the best starting point is a newly dead animal, which most of the found specimen in Qatar were not. But when possible the following procedure was followed: 1. Ideally the following data was registered for all collected animals: secure species identification, measurements of body size, total body weight, in situ photograph of the animal and time and place of the find. In many instances, however, measuring and weighing the animal was not possible or useful due to the unset of putrefaction. 2. Skinning the animal and roughly skeletonising it, removing most meat and sinew. A wire was pushed through the spine and hand and foot bones were put in separate bags. 87  3. The roughly skeletonised animal was left to soak in plenty of water overnight to pull out any remaining blood. 4. The skeleton was then transferred to clean water, preferably kept at around 30 degrees. It was left for about 10 to 14 days depending on age and size of the animal. Enzymes or washing power was in some cases added at the beginning, to kick start the process. At the end of the putrefaction process all meat, tendons and sinew should have dissolved or at least be completely soft and loose. 5. In order to stop the process of putrefaction the skeleton was briefly boiled. Some soda or chlorine was added to the mixture to slightly bleach the bones or were subsequently left over night in cold water with 2% chlorine added. Any residue of meat, fat or soft tissues should loosen during the final boiling. 6. The skeleton was then rinsed in running water over a fine sieve, so none of the small bones were lost. The bones were then gently dried. In order to avoid cracking, particularly of the teeth, direct sunlight was avoided. The skeletal material we found in Qatar varied significantly in species and state. In most cases the animals were far from fresh, more rotting or downright mummified. Due to the cold nights we found that step 4 often did not lead to proper putrefaction even though the highest day temperature was around or above 30 degrees. Therefore step 4 was most often followed by a longer more intensive boiling with extra Chlorine added, in order to loosen the remaining tissues that could then be removed manually. 10.3 THE COLLECTION All the cleaned skeletons were subsequently packed in boxes with tags contained information on the collection number, species, find location and date. All are currently stored in one of the containers on site, at Al Zubarah Camp. In total 40 more or less complete skeletons have so far been processed (table 10.1). These include at least 18 different species and some are therefore clearly represented multiple times, such as domesticated sheep (8 complete skeletons) and green sea turtle (3 skulls and 4 complete skeletons). The remaining species include mainly mammals, marine and terrestrial, but also birds and reptiles are represented. Additionally, work on a collection of fish skeletons have also begun, but this is not included in the current report. A comparative collection intended purely for field identification purposes does not necessarily have to include multiple specimens from each species, although it may prove useful for sex differentiation and age distinction. But due to the intended destination in the Qatar Natural History Museum it is well worth the effort to collect as many skeletons as possible. Such a large scale collection may serve as the base for much research in the future. 88 10.4 OTHER WORK PROCESSES A small part of the collection has been temporarily deposited with the school service program. These include three skulls and the lower leg of a horse (table 10.2). They are intended to be used in the coming season for the many school visits where the story of the wild life and nature of the area is included. An experimental treatment was initiated this season with two small cadavers, a house mouse and a cape hare (table 10.3). These were skinned and emptied for intestine and then deposited in baskets of fine mesh. Due to the substantial amount of dermestid beetles that were attracted by the cadavers we decided to test if smaller animals left out in the open could be process to cleaned skeletons by the insects during the summer. The fine mesh baskets were then deposited inside an iron frame with a heavy-duty mesh for safekeeping from potential scavengers. Due to time constraints a small number of dried-out or in some instances almost mummified animals were wrapped in plastic and packed up without further treatment. These will be processed next season when more time is available (table 10.4). Finally, three partially cleaned skeletons were buried near the camp (table 10.5) also to be processed next season. In total this group include a further 7 species of animals, that can be added to the current comparative collection of 40 skeletons from 18 species. 10.5 CONCLUSION AND RECOMMENDATIONS During 2012 season the creation of a comparative collection of modern skeletons was initiated. 40 more or less complete skeletons from at least 18 different species were processed during this time. They represent a substantial start to what will hopefully become a large collection of animal skeletons that may serve as the base for future zooarchaeological and biological research in Qatar. A placement with the future Qatar Natural History Museum has therefore been arranged. The results from the current season clearly show the potential for building such a collection as part of the QIAH field seasons and with fairly limited funds due to the large amount of animals that can be found in NW Qatar. It is therefore the recommendation of this report that a least two to three seasons of field work are to include a trained conservator, who can work together with the zooarchaeologists and biologists, developing the comparative collection. Such initiative would also be useful for the future school service program and the planned visitors centre in order to prepare materials for exhibits on the modern biology of the area. 89 Tables Table 10.1 Comparative Collection of animal skeletons. No Species Skeletal Information 1 Dugong, Dugong dugon Skull, unknown age and sex, very large specimen 2 Green Sea Turtle, Chelonia mydas Skull, unknown age and sex, fusing 7 Ethiopian hedgehog, Paraechinus Partial skeleton, unknown age and sex aethiopicus 8 Socotra cormorant, Phalacrocorax Complete skeleton, unknown sex, juvenile, PIC nigrogularis 9 Dromedary, Camelus dromedarus Near complete skeleton, unknown age and sex, PIC 10 Lesser Crested Tern, Sterna bengalensis Partial skeleton – unknown age and sex 11 Sheep, Ovis aries Complete skeletons, big female, adult, black and a little white coat, PIC 12 Sheep, Ovis aries Almost complete skeleton, smaller female, adult, brown and white coat, PIC 13 Green Sea Turtle, Chelonia mydas Complete skeleton, unfused skull, unknown sex and age, PIC 14 Cat, Felis sp. Mandible, unknown age and sex 15 Spiny tailed lizard, Uromastyx aegyptia Partial skeleton, most limb bones are missing, PIC 16 Dugong, Dugong dugon Skull, unknown age and sex, adult 17 Green Sea Turtle, Chelonia mydas Skull, unknown sex and age, adult, fusing 18 Dromedary, Camelus dromedarus Skull, unknown age and sex, 19 Dugong, Dugong dugon Partial skeleton, unknown age and sex 20 Dolphin, Delphinidae sp. Partial skeleton – maxilla missing, unknown age and sex, PIC 22 Sea snake, Hydrophis lapemoides Complete skeleton, unknown age and sex, PIC 23 Socotra Cormorant, Phalacrocorax Complete skeleton, unknown age and sex, PIC nigrogularis 25 Green Sea Turtle, Chelonia mydas Complete skeleton, juvenile, unknown sex, PIC, carapace length 24cm, width 22 cm, weight: 1120 g 26 Green Sea Turtle, Chelonia mydas Complete skeleton, juvenile, unknown sex, PIC, carapace length 30cm, width 28cm, weight: 1860 g 27 Horse, Equus caballus Near complete skeleton, unknown age and sex 29 Domesticated cat, Felis catus Complete skeleton, adult, unkown sex, PIC 30 Chicken, Gallus gallus domesticus Partial skeleton, adult and male 32 Dog, Canis familiaris Complete skeleton, juvenile male of what looks like a golden retriever/German Shepard mix 33 Green Sea Turtle, Chelonia mydas Skull, unknown sex and age, adult, fusing 34 Green Sea Turtle, Chelonia mydas Complete skeleton, adult, unknown sex, PIC carapace length 109cm, width 93cm 36 Medium size gull, Larus sp. Complete skeleton, unknown age and sex, PIC 38 Grey Heron, Ardea cinerea Complete skeleton, unknown age and sex, both wings broken, PIC 39 Domesticated cat, Felis catus Complete skeleton, adult, probably male 41 Sundevall´s jird, Meriones crassus Complete skeleton, unknown age and sex, PIC 42 Sheep, Ovis aries Partial skeleton, unknown age and sex – probable male 43 Spiny tailed lizard, Uromastyx aegyptia Partial skeleton, juvenile unknown sex, 45 Sheep, Ovis aries Almost complete skeleton, pullus, unkown sex 46 Sheep, Ovis aries Almost complete skeleton, juvenile, unkown sex 90 47 Sheep, Ovis aries Almost complete skeleton, pullus, unkown sex 48 Sheep, Ovis aries Almost complete skeleton, juvenile, unkown sex 49 Sheep, Ovis aries Almost complete skeleton, juvenile, unkown sex 50 Sundevall´s jird, Meriones crassus Complete skeleton, unknown age, male 51 Brown rat, Rattus norvegicus Complete skeleton, unknown age and sex 56 Dugong, Dugong dugon Skull, adult, unknown sex, very large specimen Table 10.2 Skeletons currently on loan to the school service program. No Species Skeletal Information 3 Dromedary, Camelus dromedarus Skull, unknown age and sex 53 Ethiopian hedgehog, Paraechinus skull, adult, unknown sex aethiopicus 54 Domesticated cat, Felis catus skull, adult, unknown sex 55 Horse, Equus domesticus Front leg, unknown age and sex Table 10.3 Collected specimens that are partially processed, now in experimental treatment. No Species Skeletal Information 44 House mouse, Mus musculus Complete skeleton, unknown age and sex. 52 Cape hare, Lepus carpensis Complete skeleton, unknown age and sex. Table 10.4 Collected animal skeletons that has not been processed No Species Skeletal Information 21 Tortoise, testudinidae sp. Complete skeleton, unknown age and sex Complete skeleton, adult, unknown sex, PIC, Probably 24 Large white-headed gull, Larus sp. Caspian (L. cachinnans) 28 Dove, Columba sp. Complete skeleton, pullus and unknown sex Ethiopian hedgehog, Paraechinus 31 aethiopicus Complete skeleton, unknown age and sex, PIC 35 Lilith Owl, Athene (noctua)lilith Complete skeleton unknown age and sex, PIC Socotra Cormorant, Phalacrocorax 37 nigrogularis Complete skeleton, unknown age and sex, PIC 40 Common Redshank, Tringa tatatus Complete skeleton, unknown age and sex, PIC Table 10.5 Collected animal skeletons that has been buried No Species Skeletal Information 4 Dugong, Dugong dugon Partial skeleton, juvenile 5 Dolphin, Delphinidae sp. Partial skeleton, unknown age and sex 6 Sheep, Ovis aries Partial skeleton, unknown age and sex 91 Figures Figure 10.1: Collecting and processing animals Figure 10.2: Some of the processed animal skeletons in the comparative collection 92 

1 Supporting Information for: “Monitoring Endangered Freshwater Biodiversity by Environmental DNA” 100 Accumulated probability of detection (%) 90 80 70 60 50 40 1 2 3 Number of samples Fig. S1. Accumulated probabilities of detecting the targeted species in the field studies when taking 1, 2 or 3 samples. The number of ponds positive for the respective species when considering all 3 samples is set to 100%. P. fuscus (blue, n=14), T. cristatus (red, n=10), L. pectoralis (black, n=9), M. fossilis in stagnant water (green, n=11), M. fossilis in running water (purple, n=8), L. apus (yellow, n=10) and L. lutra (brown, n=4).  2 Fig. S2. Model fit on the DNA concentration in individual containers of the mesocosm experiment through time. PF1.1 denoting density of one individual of P. fuscus in replicate 1, TC4.2 denoting 4 individuals of T. cristatus in replicate 2 etc.  3 Fig. S3. Modeled parameter estimates for A: DNA excretion and B: DNA degradation by the two species T. cristatus (TC) and P. fuscus (PF) in mesocosm experiment at densities of 1, 2 and 4 specimens pr. 80 l.  4 Fig. S4. QQ-plot for mesocosm experiments demonstrating fulfillment of the assumption of normality after log10 transformation. TC: T. cristatus; PF: P. fuscus.  5 Field experiment Success qPCR Species Locality Conventional DNA rate replicates Coordinates N55.98225, GRIB29 + + 1/3 4 E12.24362 N55.98246, GRIB30 + + 3/3 4,4,4 E12.24107 N49.81814, PFDE1 + + 2/3 3,3 E8.378260 N49.82278, Pelobates fuscus PFDE2 + + 3/3 3,3,3 E8.374980 N55.98929, HEL56 + + 3/3 4,3,3 E12.20933 N55.99674, HEL70 + + 3/3 3,3,3 E12.15944 N56.00489, HEL76 + + 3/3 3,3,3 E12.14852 N55.88863, STRO1 + + 3/3 3,5,3 E12.12061 LILYNG1 + + 1/3 4 N55.94363,  6 E12.13673 N55.98916, HEL60 - + 2/3 3,3 E12.20613 N55.89581, STRO2 - + 2/3 3,3 E12.12089 N55.95206, LILYNG2 - + 1/3 3 E12.14250 N55.93526, MEL1 - + 1/3 4 E12.17440 N55.98831, GRIB21 - + 2/3 4,4 E12.18037 N55.95175, LILYNG3 - - 0/3 - E12.14628 N55.99076, HEL57 - - 0/3 - E12.21112 N55.96758, HIL28 - - 0/3 - E12.23795 N55.80773, NAR1 - - 0/3 - E12.53963 N55.70701, controls BRO1 - - 0/3 - E12.50857 N55.83233, MAL1 - - 0/3 - E12.52472  7 N55.78913, JD1 + + 2/3 3,3 E12.57943 N55.78860, JD2 + + 2/3 4,4 E12.57722 N55.78705, JD3 + + 3/3 3,4,4 E12.56818 N55.78702, JD4 + + 3/3 4,4,4 E12.56808 N55.79373, JD5 + + 2/3 3,5 E12.56872 Triturus cristatus N55.79307, JD6 + + 2/3 3,3 E12.58687 N55.80070, JD7 + + 3/3 4,4,4 E12.57713 N55.79685, JD8 + + 3/3 4,4,4 E12.56059 N55.79574, JD9 + - 0/3 - E12.55170 N55.79847, JD10 + + 3/3 4,4,4 E12.54997 N55.79799, JD11 + + 2/3 4,3 E12.58399 N55.68651, controls BOT1 - - 0/3 - E12.57432  8 N55.70701, BRO1 - - 0/3 - E12.50857 N55.83233, MAL1 - - 0/3 - E12.52472 N57.69979, KAR1 + + 3/3 3,3,3 E26.45889 N53.22960, MFP1 + + 3/3 3,3,3 E21.87083 N53.22980, MFP2 + + 3/3 3,3,3 E21.87074 N53.22970, MFP3 + + 3/3 3,3,3 E21.87130 Misgurnus fossilis (ponds) N53.24053, MFP4 + + 2/3 3,3 E21.96304 N53.24008, MFP5 + + 2/3 3,3 E21.96298 N53.24576, MFP6 + + 3/3 3,3,3 E22.00651 N53.24590, MFP7 + + 3/3 3,3,3 E22.01087 N53.24710, MFP8 + + 2/3 3,3 E22.00922 MFP9 + + 1/3 3 N53.24729,  9 E22.01024 N53.24295, MFP10 + + 3/3 3,3,3 E22.01768 N53.80837, SH2 ? + 3/3 3,3,3 E9.54235 N53.80003, SH3 ? + 1/3 3 E9.49517 N53.81951, SH4 ? + 1/3 3 E9.51973 N53.82235, SH5 ? - 0/3 - E9.49389 Misgurnus fossilis (streams) N53.84867, SH6 ? + 1/3 3 E9.49438 N53.84240, SH7 ? - 0/3 - E9.50423 N53.78150, SH8 ? + 1/3 3 E9.49425 N53.76588, SH10 ? - 0/3 - E9.51403 N53.76135, SH11 ? - 0/3 - E9.51103 N53.76235, SH12 ? - 0/3 - E9.46593 SH13 ? - 0/3 - N53.75945,  10 E9.47942 N53.74716, SH15 ? - 0/3 - E9.55456 N53.74588, SH16 ? + 3/3 3,3,3 E9.58558 N53.75279, SH17 ? + 1/3 1 E9.56611 N53.79403, SH18 ? + 2/3 3,3 E9.55748 N55.98831, GRIB21 - - 0/3 - E12.18037 N55.98225, controls GRIB29 - - 0/3 - E12.24362 N55.98246, GRIB30 - - 0/3 - E12.24107 N55.69851, SKA1 + - 0/3 - E13.44245 Leucorrhinia pectoralis N55.69856, SKA2 + + 1/3 4 E13.44263 N55.69872, SKA3 + + 1/3 3 E13.44227 N55.69898, SKA4 + + 1/3 3 E13.44236  11 N55.69886, SKA5 + + 3/3 3,3,3 E13.44214 N55.76740, SKA7 + - 0/3 - E13.83848 N55.76749, SKA8 + + 1/3 3 E13.83607 N55.76749, SKA9 + + 1/3 5 E13.83607 N55.69911, SKA10 + + 1/3 4 E13.36750 N55.55074, SKA11 + + 2/3 4,4 E13.35331 N55.53414, SKA12 + + 3/3 4,4,4 E13.26749 N55.98831, GRIB21 - - 0/3 - E12.18037 N55.98225, controls GRIB29 - - 0/3 - E12.24362 N55.98246, GRIB30 - - 0/3 - E12.24107 N53.24848, Lepidurus apus LEP1 + + 3/3 3,3,3 E21.96302 LEP2 + + 1/3 3 N53.24889,  12 E21.96283 N53.24904, LEP3 + + 3/3 3,3,3 E21.96311 N53.23977, LEP4 + + 2/3 3,3 E21.91604 N53.23867, LEP5 + + 2/3 3,3 E21.99447 N53.22165, LEP6 + + 3/3 3,3,3 E22.05579 N53.23973, LEP7 + + 3/3 3,3,3 E21.99547 N53.24710, MFP8 + + 2/3 3,3 E22.00922 N53.24729, MFP9 + + 3/3 3,3,3 E22.01024 N53.24295, MFP10 + + 3/3 3,3,3 E22.01768 N55.69999, DM1 - - 0/3 - E12.50484 N55.56671, control HER1 - - 0/3 - E9.57437 N55.55949, HER6 - - 0/3 - E9.58360  13 N57.04169, OD10 ? - 0/3 - E8.76599 N57.04839, OD12 ? + 2/3 3,3 E8.71714 N57.04626, OD13 ? - 0/3 - E8.71756 N57.04613, OD14 ? - 0/3 - E8.71738 N57.04854, OD15 ? - 0/3 - E8.71666 N57.03394, Lutra lutra OD18 ? - 0/3 - E8.58624 N57.10795, OD2 ? - 0/3 - E8.91368 N57.03943, OD22 ? + 2/3 3,3 E8.75691 N57.04726, OD23 ? + 1/3 3 E8.72327 N56.88909, OD27 ? - 0/3 - E8.40157 N56.81362, OD28 ? + 1/3 3 E8.34860 N57.07795, OD6 ? - 0/3 - E8.81398  14 N57.04404, OD7 ? - 0/3 - E8.84741 N57.04312, OD8 ? - 0/3 - E8.84879 N57.04180, OD9 ? - 0/3 - E8.76587 N55.98831, GRIB21 - - 0/3 - E12.18037 N55.98225, control GRIB29 - - 0/3 - E12.24362 N55.98246, GRIB30 - - 0/3 - E12.24107 Table S1. Overview of the results of field experiments. Localities, presence of species confirmed by conventional or environmental DNA sampling, success rate and number of positive qPCR replicates for each sample as well as the coordinates (Datum: WGS84) for the sampling sites. Controls were taken from localities with certain absence of the targeted species. Success rate denotes the number of positive 15 ml samples out of three for each locality. qPCR replicates denotes how many replicates were obtained from each 15 ml sample and included in final DNA concentration estimation.  15 Controlled mesocosm experiment Species Sampling day Density in 80 l Success rate qPCR replicates 1 0/3 - Before 0 2 0/3 - 4 0/3 - 1 3/3 4,3,4 2 2 2/21 4,4 4 3/3 4,4,4 1 3/3 4,4,4 9 2 2/21 4,4 4 3/3 4,4,4 Pelobates fuscus 1 3/3 4,4,4 23 2 2/21 4,4 4 3/3 4,4,4 1 3/3 4,4,4 44 2 2/21 4,4 4 1/12 8 1 3/3 4,4,4 64 2 2/21 4,4 4 1/12 10 1 1/3 3 After 66 2 2/21 4,3 4 1/12 4  16 1 0/3 - 73 2 0/21 - 4 0/12 - 1 0/3 - 79 2 0/21 - 4 0/12 - 1 0/3 - 112 2 0/21 - 4 0/12 - 1 0/3 - Before 0 2 0/3 - 4 0/3 - 1 2/3 3,3 2 2 2/3 6,4 4 2/3 3,3 1 2/3 4,3 Triturus cristatus 9 2 3/3 3,6,4 4 2/3 4,4 1 3/3 3,4,3 23 2 3/3 3,3,3 4 3/3 3,3,3 1 3/3 4,4,4 44 2 3/3 3,4,4  17 4 1/13 6 1 3/3 3,3,3 64 2 3/3 3,3,3 4 2/24 3,3 1 2/3 3,3 66 2 1/3 5 4 2/24 4,3 1 0/3 - 73 2 0/3 - 4 1/24 3 After 1 0/3 - 79 2 0/3 - 4 0/24 - 1 0/3 - 112 2 0/3 - 4 0/24 - 1 One sample was discarded due to tadpole mortality at set-up of experiment 2 Two samples were discarded due to mortality in the container at sampling 3 Two samples were lost in DNA extraction centrifugation step 4 One sample was discarded due to metamorphosis of experimental animals Table S2. Overview of the results of controlled mesocosm experiments. Sampling day and density of individuals for both species in each container. Day 0 is June 24th 2009. Success rate  18 denotes the number of positive 15 ml samples out of three for each animal density. qPCR replicates denotes how many replicates were obtained from each 15 ml sample and included in final DNA concentration estimation.  19 taxon primer/probe sequence (5'-'3) fragment gene PFCBL GGCTTTTTCCATCGCAATTCT 72 cyt-b Pelobates fuscus PFCBR CGAAATATCATGCTCCGTTGTTT PFCB.probe CCCTTATGCCCATTCTTCACACCGC TCCBL CGTAAACTACGGCTGACTAGTACGAA 81 cyt-b Triturus cristatus TCCBR CCGATGTGTATGTAGATGCAAACA TCCB.probe CATCCACGCTAACGGAGCCTCGC MFCBL AGGTGGAGTCCTGGCCCTAT 70 cyt-b Misgurnus fossilis MFCBR TTTTGAGGTGTGTAAGATGGGAACT MFCB.probe TTCTCTATCCTGGTCTTAATAG LPCOIbL GCTTTCCCACGATTAAATAA 90 COI Leucorrhinia pectoralis LPCOIbR TGCACCTCTTTCAACTATACTT LPCOIb.probe ATAAGATTTTGACTTTTGCCTCC LACOIL TCGGATCCCCTTGTTTGTATG 85 COI Lepidurus apus LACOIR AGTGATGGCTCCTGCAAGGA LACOI.probe TCTGTAGCAATCACAGCAT LLCBL AAAGCCACCCTGACACGATT 80 cyt-b Lutra lutra LLCBR AGCAGGTGGATTGTTGCTAGTG  20 LLCB.probe TTCGCTTTCCACTTTAT AmpCBLb AGTCCTGTTGGGTTGGTTGACCCNGTTT 119 cyt-b Amphibians AmpCBR AATGCAACTCTCACCCGATTCTT PFcytbL TCGCCCTTTCTCACAATTTC 80 cyt-b Amphibians/Pelobates PFcytbR CAACAGGTTGACCTCCGATT TCcytbL TGATGGAACTTCGGCTCTCT 105 cyt-b Amphibians/Triturus TCcytbR TGAAAATGCCGATTGTGTGT FishCBL TCCTTTTGAGGCGCTACAGT 130 cyt-b Fish FishCBR GGAATGCGAAGAATCGTGTT TincaCBL CACACCTCCACACATTCAGC 100 cyt-b Fish/Tinca TincaCBR GAATAATAGTGCAAGAACACCTCCT PercaCBL ACGCTCGATTCCAAACAAAC 119 cyt-b Fish/Perca PercaCBR AGAGCGGTCGGAATGTAATG Table S3. Primers and probes designed and used in this study. Fragment lengths are given in base-pairs including primers. The amplified genes are Cyt-b: Cytochrome b, COI: Cytochrome Oxidase I. Probes are Minor Groove Binding (MGB) probes and have the modifications; 5’: 6-Fam (D-L-Probe), 3’: BHQ-1.  21 taxon species location pcr # sequence (5'-'3) primers Pelobates fuscus TATTCTGATCCCTTATTGCTAACACACTAATT Amphibians (mesocosm) PF1.1, day2 37 CTTACTTG PFcytbL/R PF4.1, day2 37 - - PF1.1, day9 37 - - PF4.1, day9 37 - - PF4.1, day64 37 - - PF2.1, day2 166 CGCCCTTATGCCCATTCTTCACACCGCC PFCBL/R PF1.3, day64 167 - - PF2.2, day64 167 - - PF4.2, day2 166 CGCCCTTATGCCCATTCTTCACACCGCT - PF4.3, day64 167 CGCCCTTATGCCCACTCTTCACACCGCT - Pelobates fuscus (field) HEL56 84 same as PF2.1, day2 - PFDE1.2 77 CGCCCTTATGCCCATTCTTCACACCGCT - TATTCTGATCCCTTATTGCTAACACACTAATT STRO1.1 41 CTTACTTG PFcytbL/R STRO2.1 41 - - LILYNG2.1 41 - - PDFE2.1 41 - - Triturus cristatus (mesocosm) TC1.2, day9 149 AAAAATAGCGAGGCTCCGTTAGCGTGGATGT TCCBL/R TC2.2, day9 149 - - TC4.1, day9 149 - - TC1.1, day44 184 - - TC2.2, day44 184 - - TC2.3, day44 184 - - CTGCTGTATAGTGTATCGCTAGAAATAGTCCT GTGAGGATTTGTGTAATTAGGCACACTCCTA TC1.2, day64 43 GA TCcytbL/R TC2.3, day64 43 - - TC4.2, day64 43 - - Triturus cristatus (field) JD5.1 92 same as TC1.2, day9 TCCBL/R JD6.1 92 - - JD4.2 45 same as TC1.2, day64 TCcytbL/R JD7.2 45 - - JD8.1 45 - - JD10.3 45 - - GGTGTAAGAAGAGTAGGTGAACAATGCTAGT CCCTGCAATTAAGAATGGGAACAGGAAGTGG JD11 207 AATGCA AmpCBLb/R GGTGTAAGAAGAGTAGGTGAACAATGCTAGC CCCTGCAATTAAGAATGGGAACAGGAAGTGG HEL56 207 AATGCA - GGTGAAGGAACAGTAAGTGAACAATGCTTGC Lissotriton CCCGGCAATAAGAAAGGGGAATAAGAAGTG vulgaris JD11 207 GAATGCG - HEL56 207 - -  22 GGTGAAGAAACAGGAGATGAATTATGCTTGC Pelophylax kl. GGCTGTGATAATAAAGGGGAGAATAAAGTG esculentus JD11 207 GAATGTA - GGTGAAGAAACAGGAGATGAATTATGCTTGC GGCTGCGATAATAAAGGGGAGAATAAAGTG HEL56 207 GAATGTA - GGTGTAGAAATAGAAGATGAATTATGCTCAC AGCAGCGATAATAAACGGAAGAATAAAGTG Rana temporaria HEL56 207 GAATGTG - GATGTAGGAACAGAAGGTGAATTATGCTCAT AGCAGCGATGATAAATGGGAGGATGAAGTG Rana arvalis HEL56 207 GAACG - Fish Misgurnus fossilis SH2.3 159 TATTCTCTATCCTGGTCTTAATAGT MFCBL/R KAR9 161 - - SH8.1 163 - - SH16.1 163 - - SH18.1 164 - - MFP1.3 240 - - AATTACAAACCTCCTATCCGCTGTGCCATATA Carassius TAGGAGATATATTAGTTCAATGAACTTGAGG carassius ELL1 209 AGGCTTCTCCGTAGACAACGCAACATT FishCBL/R AATTACAAACCTCCTATCCGCTGTGCCATATA TAGGAGATATATTAGTTCAATGAATTTGAGG BOT1 209 AGGCTTCTCCGTAGACAACGCAACATT - AATCACAAACCTTCTATCCGCCGTGCCATAT ATAGGAGATATATTAGTTCAATGAATTTGAG Carassius auratus ELL1 209 GAGGCTTCTCCGTAGACAATGCAACATT - BOT1 209 - - AATCACAAACCTCCTATCTGCCGTACCATAC ATGGGAGACATGTTAGTCCAATGAATCTGAG Cyprinus carpio ELL1 209 GTGGGTTCTCAGTAGACAATGCAACACT - BOT1 209 - - AATTACAAACCTCCTCTCAGCAGTCCCCTAC Scardinius ATGGGTGATACCCTTGTTCAATGAATCTGAG erythrophthalmus ELL1 209 GCGGTTTCTCAGTAGACAACGCGACCCT - BOT1 209 - - AGTTTGTTTGGGATTGATCGTAAAATGGCGT Tinca tinca ELL1 238 AGGCAAATAAGAAATATCATTCTG TincaCBL/R AGCTTGTTTGGGATTGATCGTAAAATGGCGT BOT1 289 AGGCAAATAAGAAATATCATTCTG - Leucaspius AGTTTATTAGGAATAGACCGGAGGATGGCAT delineatus ELL1 238 ACGCAAATAAGAAGTATCATTCTG - CCACGTTGTTTAGAAGTGTGAAGGATGGGGA CAACTATAAGAACCAGGATGGAGGCAAGTA Perca fluviatilis BOT1 289 AGGCTAAGACCCCTCCTA PercaCBL/R Leucorrhinia CTAGCTAATAGTAGGGTAAATGAAGGAGGCA Insects pectoralis SKA5.2 182 AAAGTCAAAATCTTATA LPCOIbL/R SKA9.1 131 - - SKA10.2 157 - - SKA12.1 157 - - CTAGCTAATAGTAAGGTAAATGAAGGAGGCA - - ATAGTCAAAATCTTATA - CTAGCTAATAGTAGGGTAAATGAAGGGGGCA - - AAAGTCAAAATCTTATA - Crustaceans Lepidurus apus LEP2.3 248 ATCTGTAGCGATCACAGCATTACTTTTACTAC LACOIL/R  23 TTTCTTTACCTG ATCTGTAGCAATCACAGCATTACTTTTACTAC LEP3.3 248 TTTCTTTACCTG - LEP7.3 248 - - MFP8.2 246 - - CTAAGATGATGAATGGCAGGATAAAGTGGA Mammals Lutra lutra OD12.2 254 AAGCGAAG LLCBL/R CGTGAAGGAAGAGTAAGTGTACTATAGCGAG TGCTGCGATGATAAATGGGAGAATAAAGTGG Cervus elaphus JD11 207 AAAGCG AmpCBLb/R CATTACCAATCTATTCTCAGCCGTCCCATACA Columba TCGGTCAAACTCTCGTCGAATGAGCCTGAGG Birds palumbus ELL1 209 CGGATTCTCAGTAGATAACCCCACATT FishCBL/R CATGGAGGAAGGTGAGATGGATTGTGGTTAT GCCTGCAATAATAAATGGGAGGAGGAAGTG Fulica atra HEL56 207 GAGAGC AmpCBLb/R CGTGTAGTAAGGTCAGATGTACTAGGGTCAG Acrocephalus GCCTGCAATAATGAATGGTAGGAGGAAGTGG palustris HEL56 207 AGGGCG - Table S4. Summary of DNA sequences recovered in this study. Sequences derived from pcr # 207 and 209 are generated by pyrosequencing using Roche GS FLX 454 platform, the remaining are generated by cloning and subsequent Sanger sequencing. Symbol (-) denotes that the information is the same as above. For full details on locations see table S1 and S2, for full primer details see table S3.