Viking-Age sailing routes of the western Baltic Sea – a matter of safety1 by Jens Ulriksen Included in the Old English Orosius, com- weather conditions, currents, shifting sand piled at the court of King Alfred the Great of bars on the sea floor and coastal morphol- Wessex around 890,2 are the descriptions of ogy. Being able to cope with the elements of two different late 9th-century Scandinavian nature is important for a safe journey, but sailing routes. These originate from Ohthere, equally important – not least when travelling who sailed from his home in Hålogaland in like Ohthere – is a guarantee of safety for northern Norway to Hedeby, and Wulfstan, ship and crew when coming ashore. Callmer probably an Englishman,3 who travelled suggests convoying as a form of self-protec- from Hedeby to Truso. The descriptions are tion, but at the end of the day it would be not detailed to any degree concerning way- vital to negotiate a safe passage with “supra- points or anchorages, and in spite of the fact regional or regional lords”.7 They controlled that lands passed are mentioned in both ac- the landing sites that punctuate Callmer’s counts, the information provided is some- route as stepping-stones. times unclear or confusing. For example, In consequence of the latter, Callmer departing from Hålogaland, Ohthere refers focuses on settlement patterns in order to to both Ireland and England on his starboard identify political and military centres – cen- side even though he obviously has been un- tres with lords who controlled certain areas able to glimpse these lands when sailing of land (and sea) and were able to guaran- along the Norwegian coast.4 The same pecu- tee safety within their ‘jurisdiction’. This ap- liarity applies to Wulfstan, who mentions the proach is important because it introduces an present-day Swedish landscapes Blekinge, obvious question regarding Ohthere’s travel- 1. This paper is the extended Möre, Öland and Gotland on his port side.5 ling pattern: how many lords would he have version of a comment on It is more likely that neither of the two were had to negotiate with on his trip? Would it Johan Callmer’s contribution describing sea routes, but rather describing have been 35, 15 or just two or three? at the seminar. 2. See Bately this volume: 20. the general geography to an audience with 3. See Jesch this volume: 30. limited knowledge of this area. 4. Lund 1983: 24. Most interesting in the accounts of Oht- Denmark and Danes – 5. See Bately this volume: 15. 6. Anton Englert argues that here and Wulfstan are the description of two the kingdom and people ca 890 Ohthere’s statement indicates very different ways of travelling. While Wulf- a way of measuring the dis- stan’s ship sailed for seven days and nights Callmer states that the level of political or- tance more than a description from Hedeby to Truso, Ohthere described ganisation and control must be considered of a specific voyage (Englert how a voyage from Hålogaland to Skiringes low and unstable in the (early) Viking Age.8 2007: 118, 122-125). He may healh (Kaupang) took at least a month in fair According to Callmer, a Danish kingdom be correct, but anchoring at night was apparently a wind and anchoring every evening.6 did exist, but not in a state comparable to the common sailing procedure, In his paper, Johan Callmer sketches the situation of the 11th and 12th centuries. otherwise Ohthere’s account assumed manner of sailing in the early Viking According to the Old English Orosius, would not make sense. Age, and the inspiration has obviously come Norway (Nordweg) stretched from the north 7. See Callmer this volume: 115. from Ohthere’s way of travelling. Callmer side of the Polar Circle to Vestfold. No Dan- 8. See Callmer this volume: sets out in a relatively small sailing ship ish supremacy is mentioned. Going south 114. with an insignificant draught; he is aware of from Skiringssal along the west coast of Viking-Age sailing routes of the western Baltic Sea 135 contemporary Sweden, Ohthere states that land) and then Blekinge, Møre, Øland and he had Denmark (Denamearc) to port (Figs 1- Gotland belonging to the Swedes (Sweon). 2). On the starboard side were Jutland (Got- All the way he had Wendland to starboard. land, Danish Jylland) at first, later Sillende Therefore, there was an area called Den- “and many islands”, and – Ohthere adds – amearc consisting of a rather substantial this was the ancient homeland of the An- region including the west coast of contem- gles before immigrating to England.9 Sil- porary Sweden, Scania, and the islands of lende is, therefore, thought to be the south- Langeland, Lolland and Falster. To this we ern part of the Jutland peninsula, where can surely add Zealand. West of here were the present-day region of Angeln is situated Gotland and, perhaps, Sillende. The lands between the Flensburg and Schlei fjords. An- are not described as an entity, and no eth- other identification of Sillende is put forward nic group (but the emigrated Angles) is con- by Bent Jørgensen, who connects the name nected with these areas. It is noteworthy that with Zealand (Danish Sjælland).10 While Ohthere describes Hedeby as situated be- passing “Sillende and many islands to star- tween Saxons, Angles and Wends, but be- board”, Ohthere had “the islands belonging longing to the Danes (hyrð in on Dene),12 to Denmark” to port. not to Denmark. Positioning Funen (Dan- Wulfstan contributes to the geo-po- ish Fyn) ethno-politically and geographically litical setting mentioning that he had the is- on Ohthere’s route is not straightforward. As lands Langeland, Lolland, Falster and Scania mentioned above, he noted that he had Sil- (Swedish Skåne) to port, and they were sub- lende and many islands to starboard and at the ject to Denmark.11 Still to port he passed an same time the islands belonging to Denmark independent island of Bornholm (Burgenda to port. Of course it is difficult to know ex- Fig. 1. Landscapes men- tioned by Ohthere and Wulfstan. Redrawn after Crumlin-Pedersen 1983. 9. Lund 1983; Bately this volume: 15. 10. Jørgensen 1994: 249-250. 11. See Bately this volume: 15. 12. Lund 1983; Bately this volume: 15. 136 II. The western and central Baltic Sea region Fig. 2. Sites mentioned in the text. 1. Avnslev Overby, 2. Fyns Hoved, 3. Gundslev, 4. Hedeby, 5. Hjulby, 6. Hoby, 7. Lejre, 8. Upp- åkra, 9. Nabbe-Kildegård, 10. Nr. Alslev, 11. Næs, 12. Odense, 13. Ribe, 14. Strandby-Gammeltoft, 15. Strøby-Toftegård, 16. Tissø, 17. Vejleby, 18. Vester Egesborg, 19. Vindinge, 20. Vålse. actly what he means by “many islands”. If the land from the mid 10th century.13 Neverthe- number five – or seven, including Samsø and less, contemporary written sources speak of Endelave – is “many” he could have entered “kings of the Danes” rather than a specific Little Belt between Jutland and Funen. Alter- Danish group or Denmark. However, both natively, his route passed through the Great Ohthere and Wulfstan refer to “Danes” and Belt separating Funen and Zealand. Conse- “Denmark”, but whether or not this indicates quently the “many islands” to starboard were the ethnic group of the Danes at this time, situated south of Funen, while “the islands having the same king is uncertain. Moreover, belonging to Denmark” must have been Zea- the people on Bornholm (the ‘Burgendan’) land and the islands to its south. However, are mentioned separately, a peculiarity un- this theory is not entirely in accordance with derlined by Wulfstan, who states that they Wulfstan’s mention of Langeland’s affiliation had their own king. The question is whether to Denmark. differentiation between ‘Danes’ and ‘Den- Several 9th-century Western European mark’ is important. In later documents, King annals and chronicles give us the impression Knud the Great calls himself “king of Eng- 13. Jacobsen & Moltke 1942: of a kingdom of Danes of some significance land and Denmark and the Norwegians and No. 217. that was strong enough to challenge Charle- a part of the Swedes”,14 “king of the English”15 14. Danmarks Riges Breve 1. magne and his successors in the border areas. and “king of the realm of England and the rk., 1. bd., nr. 422 (1975). In the Old English Orosius we learn that there Danes”.16 He randomly uses both the ethnic 15. Danmarks Riges Breve 1. rk., 1. bd., nr. 411 (1975). were two kinds of Danes: “North Danes” and name and the name of the kingdom. Dur- 16. Danmarks Riges Breve 1. “South Danes”, the latter also being men- ing the reign of the 11th-century Danish king rk., 1. bd., nr. 448 (1975). tioned on a rune stone from Sædinge on Lol- Svend Estridsen, he is called “king of the Viking-Age sailing routes of the western Baltic Sea 137 Danes”17 and Pope Alexander II mentions discussed in any detail here, it is paramount “the realm of the Danes”.18 Thus, there is no to elaborate further on the use of the term absolute contradiction between the “areas “regional” when pointing out two or even belonging to the Danes” and “Denmark” in three “regional centres” situated very close this period, and the same might have been (5 km) to each other on the northern part of the case in Ohthere’s and Wulfstan’s time. Falster (Fig. 3).19 In the Annals of Fulda it is noted for Sites with extra-ordinary buildings and 873 that the peace between King Louis and other constructions as well as lay-out and the Danish kings Halfdan and Sigurd was artefacts like Tissø,20 Lejre on Zealand21 confirmed so that trade between the king- and Uppåkra in Scania22 seem insignificant doms could continue unhindered. This ‘trade in Callmer’s presentation while other sites pact’ could solely have been aimed at the traf- on northern Funen are described. But how fic across the border in Schleswig. On the shall we consider the relations between the other hand it might also have meant that the proposed centres Odense, labelled “the most kings could guarantee safe passage and trade important centre of the island of Funen” and within their realm. If we assume that Half- Hjulby, referred to as a “small centre”, with dan and Sigurd controlled ‘Denmark’ as regard to safe passage at sea? Odense, situated defined by Ohthere and Wulfstan, Ohthere 9-10 km upstream on the Odense river, has could travel safely most of the way from Oslo an indicative name,23 while the archaeologi- Fjord to Hedeby if he was granted the protec- cal finds are not older than the 10th century. tion of the Danish kings. But how did Oht- The artefacts – dominated by local pottery – here gain safe conduct in the first place? Was are primarily retrieved from scattered pits it a time-honoured privilege handed down and odd pit houses, and at least one of the through generations and centuries to travel archaeological sites is connected with a vil- along established routes? Was it necessary to lage structure called “Hetby”.24 negotiate with each and every magnate with At Hjulby, situated 4-5 km from the a landing site along the route? Or, did the Great Belt coast, 15 pit houses have been king decentralise the power of issuing safe excavated, along with a cultural layer with conduct to his trusted magnates, who con- workshop refuse and a variety of metal ob- trolled larger or smaller areas of their own, jects dating from around the 6th to 12th cen- 17. Danmarks Riges Breve 1. and could Ohthere then fly a banner from turies, all found within a 150 m x 250 m rk., 2. bd., nr. 5 (1975). his mast, showing his acquired status to area.25 Some 4 km to the south, at the vil- 18. Danmarks Riges Breve 1. everyone? lage of Vindinge, post-built longhouses and rk., 2. bd., nr. 5 (1975). pit houses from the late Viking Age have re- 19. Vålse, Gundslev and cently been found, as has a nearby workshop Nørre Alslev; see Callmer this volume. Centres area with artefacts dating from the 8th to 10th 20. Jørgensen 2003. centuries.26 Approximately 3 km north-west 21. Christensen 1993. Callmer’s proposed sailing route focuses on of Hjulby, at Avnslev Overby, yet another site 22. Helgesson 2002: 45-62; settled areas in order to make the presence with a long house, pit houses and a workshop Larsson 2002. 23. Odense means “the of a protected landing site probable. Ac- area has been partly excavated.27 The artefacts sacred place (‘vi’) of Odin” cording to Callmer, protection relies on a date from the 7th to 12th centuries, and a rune (Jørgensen 1994: 218). “centre”, and he adds the adjectives “most stone has been found by the church. Which 24. Jacobsen 2001: 72-73; important” (Odense), “small” (Hjulby), “re- of these sites is the “centre”? Hjulby is by far Vestergade 70-74 (OBM j. gional” (Vålse, Gundslev and Nørre Alslev). the largest at this point, both in regards to the nr. 8236). 25. Henriksen 2000: 35-55; However, it is not very clear which criteria area of activity and the number of artefacts. Henriksen 2002. lie behind the terms used in describing the But it has also been a site of focused archaeo- 26. Henriksen 2002: 156-157. centres. Although the term “centre” is not logical research for a decade, while Vindinge 27. Henriksen 2002: 174. 138 II. The western and central Baltic Sea region Fig. 3. Sites on Lolland and Falster. and Avnslev Overby are ‘virgin’ sites in com- minde” mentioned by Callmer is yet to be parison. Further excavations could change found. The most convincing traces of a set- the picture completely. The truth is that these tlement of this period have been found in sites could be nothing more than production Gloslunde Parish at Hoby on the banks of sites tied to the true centre of power: a mag- the reclaimed fjord.30 The excavations have nate’s manor yet to be revealed. been limited, and it is too early to conclude The southern Danish island of Lolland whether Hoby was a magnate’s residence, an is another point of focus in Callmer’s paper ordinary farm or a landing site during the Vi- (see Fig. 3). In the southern part of the is- king Age. Another potentially interesting site land is the reclaimed Rødby Fjord, originally is situated between Vejleby church and the comprised of several fine natural harbours reclaimed shore of Rødby Fjord, where metal in shallow and protected waters. From the detection has identified several artefacts of hinterland of the fjord, there are five rune bronze dating from the Viking Age.31 This stones dating from the second half of the situation is paralleled at a number of landing 10th century, and one from the 11th century.28 sites throughout southern Scandinavia. Callmer mentions the island as a “mini-re- gion” suggesting a centre of a 9th-century “re- 28. Jacobsen & Moltke 1942. 29. Schilling 2003. gional lord”, approximately 1 km north of the Landing sites 30. Archaeological excavations present-day village of Vejleby. From fields at in Denmark 2001: nos 185 Duesminde, silver hoards were retrieved dur- In his paper, Callmer defines maritime sites and 186. ing the 1960s and in 2002,29 and 12 poorly- as “primarily functioning as ports, shipyards 31. ‘Vejleby Kirke’, no. 07.03.10 - 10 Vejleby parish, furnished Viking-Age graves were revealed and contact zones between natives and visi- Fuglse herred, Maribo amt. 1.5 km to the west in 1923. The hoards are tors”. This must be seen in contrast to what See www.dkconline.dk. indeed spectacular, but the “manor of Dues- Callmer calls coastal sites – “settlements of Viking-Age sailing routes of the western Baltic Sea 139 normal agrarian type situated close to the The difficulty of making early conclu- sea”. sions can be illustrated by a locality from He suggests a typology of five categories southern Zealand. Vester Egesborg is a land- of maritime sites, but it is difficult to appreci- ing site situated on the bank of the Dybsø ate the definitions as useful tools in real-life Fjord some 10 km south of Næstved.37 It was archaeology. As mentioned previously, the found in the early 1960s and a trial excava- concept of the term “regional” is somewhat tion was carried out in 1965.38 Surveying with obscure. Therefore it is not very helpful to metal detector in the 1990s produced an ex- label maritime sites “local”, “regional” or “su- traordinary amount of high-quality metal pra-regional” as long as these terms remain objects from the 6th to 10th centuries. A large- undefined. There is also no estimate of size scale excavation has been conducted on the or physical presence. What is “small” com- site during recent years, and every year has pared with “middle-sized” and “extensive”? If delivered surprising results, especially con- a “semi-permanent” element is a house (type cerning the layout of the site. If excavations C),32 which kind of construction is “perma- had stopped in 1999, the interpretation of nent”? The presence of houses does not make the site would not have been adequate and a landing site more “regional”. as a consequence the conclusions would have Although there is some (modern) reason- been incorrect. ing to Callmer’s site typology,33 the archaeo- Indeed, there are different types of Vi- logical documentation is as yet too flimsy king-Age landing sites.39 Classifying settle- to support it. The number of documented ments connected in a network founded on sites is fairly limited and the size of excava- a theory of extensive and systematic trade tions often inadequate to determine chronol- relations in the early Viking Age is another ogy, layout and function. The studies of land- matter. There is no convincing argument ing sites in Denmark and southern Sweden that the level of 9th-century economics were have shown that the majority date from the in need of a fine-meshed trading network late Iron and Viking Ages, demonstrating a with numerous sites of sizes “small”, “me- large variation in physical presence and arte- dium” and “extensive”, widespread along the facts.34 Only large-scale excavations can reveal coasts and rivers of Scandinavia, frequently their true nature. Callmer suggests Strandby- visited by independent merchants during Gammeltoft and Nabbe-Kildegård on Funen the sailing season. The number of landing as examples of type C sites, but at this point sites in Roskilde Fjord and the Limfjord in they are not comparable at all. At Strandby- mind40 – perhaps paralleled by the Schlei Gammeltoft more than an acre has been fjord41 – would imply a rather abundant ‘class’ excavated and amongst the features were 29 of persons who produced a surplus of their 32. See definitions in Callmer pit houses dating from the 7th/8th to 10th cen- own large enough to put on the market. This this volume. turies.35 The site has produced artefacts con- market should, then, also take place on their 33. See Callmer this volume. sisting of pottery, metal objects and workshop own beach, having a standard and goods of 34. Callmer 1991; Carlsson refuse. At Nabbe-Kildegård, a single pit house such a quality that it attracted merchants. 1991; Ulriksen 1990; Ulriksen 1998; Ulriksen 2004. has been found in a sewer trench, containing The increase in the number of archaeo- 35. Henriksen 1997. uncharacteristic pottery of Iron-Age type and logically-known landing sites of the 8th and 36. Ulriksen 1998: 153. animal bones.36 Stray finds from the vicinity 9th centuries is evident, but only a few were 37. Gärtner & Ulriksen 1997; consist of four spindle whirls, two glass beads participants in fixed trading networks. The Ulriksen 1998: 169-178; Ulrik- and a gold arm-ring (dating to the 9th cen- long-distance trading sites differ from the sen 2006. 38. Ørsnes 1966: 262. tury). The true size, structure, function and majority in their layout, their plots and the 39. Ulriksen 1998: 189-194. chronology of Nabbe-Kildegård can only be amount of refuse from production. It is also 40. Ulriksen 1998. revealed through further excavations. these sites that are mentioned in the contem- 41. Dobat 2002; Dobat 2004. 140 II. The western and central Baltic Sea region  porary written sources: i.e., Ribe, Hedeby, breaking.45 Vester Egesborg possesses 18 rath- Skiringssal and Birka. They were legs in a er short three-aisled post-built houses as well Northern European trading network, and as 98 pit houses. Wells for retting have yet to connected Scandinavian magnates and their be found, but a couple of pits similar to those surplus with the rest of the world. The ma- for drying flax at Næs are documented. The jority of landing sites only occasionally were evidence from these two recently-excavated in contact with this sphere of trading. They sites might indicate the phenomenon of spe- were multifunctional42 and their maritime cialised coastal agrarian production. connection was mostly based on their situ- The archaeological evidence demon- ation on the coast/beach, not by ship-finds strates that the majority of the 9th-century or imported goods. Generally, they contrast landing sites were involved in small-scale with the average agrarian settlement site by production, refining of agricultural produce having no farmsteads or fenced plots, and if and fishing and also functioned as harbours there are buildings, pit houses are dominant. for local ships. In relation to trade, they were A number of these landing sites are character- points of departure for a minor group of ised by stray finds of women’s jewellery and people controlling the agricultural surplus, refuse from small-scale craft production, typ- which could be exchanged at the long-dis- ically iron, bone and antler working, bronze tance trading sites, with contacts in an ‘in- casting and cloth weaving. An average land- ternational’ sphere of trade. Here they could ing site has between three and four types of acquire luxurious items important for main- crafts documented through the archaeologi- taining the social order and political connec- cal material.43 At the rural settlements there tions. The restricted number of long-distance are rarely more than two types of handicrafts trading sites in 9th-century Scandinavia is no represented and these are dominated by cloth coincidence. Society had no need for com- weaving followed by iron working.44 Rural petitive sites in the modern sense of the word, settlements that have the same spectrum of and these emporia were situated in conven- handicrafts as the landing sites, including ient contact zones. It is noteworthy that Oht- comb making and glass bead production, are here obviously was very focused in his travel typically magnate’s sites like Lejre, Tissø and behaviour. He did not refer to any incidents Strøby-Toftegård on Zealand. of trade or exchange in connections with However, there are two sites situated on his numerous stops from northern Norway the coast of southern Zealand that are differ- to the western Baltic. Only Skiringssal and ent. Separated by only 10 km, Vester Eges- Hedeby were important. They were the focal borg on the bank of Dybsø Fjord and Næs points in the communication system of the on the bank of the Avnø Fjord do not fit into time, and where Ohthere could exchange his the description of the average coastal land- cargo of raw materials for what he required. ing site. At Næs, dating from the second half of the 8th century to the 10th century, there have been excavated 20 three-aisled post- Sailing routes built houses, representing a single farm in four phases, and 79 pit houses. In addition, The matter of safety is paramount on a sail- 58 wells and a 150 m-long canal connecting ing route. This implies specific knowledge some of the wells have been identified. The of the route, as well as the sailor’s ability 42. Ulriksen 1998: 184; Ulrik- wells were largely used for retting flax in con- to cope with capricious nature and avoid at- sen 2004: 11. 43. Ulriksen 2002: 9-11. nection with linen production, and pits and tacks. If we assume that a skipper was experi- 44. Ulriksen 2002: 12-13. ditches with charcoal and burnt stones might enced and familiar with the way-points and 45. Hansen & Høyer 2000. have been used for drying the flax before was granted safe conduct, proposing sailing Viking-Age sailing routes of the western Baltic Sea 141  Fig. 4. Supposed sailing route from the mouth of the Schlei fjord to Kalmar Sound. routes is straight forward. He could choose prefer Wulfstan’s way of travelling. A route any sailing route and established or uninhab- from the Schlei fjord to the Kalmar Sound ited landing site he liked. Close to the coast would head for the southern spit of Falster, he would have numerous opportunities to where the Bøtø Nor provides a landmark and find a proper resting place. Natural sheltered a sheltered anchorage (Fig. 4). From here a anchorages and resting places are abundant, northeasterly course towards the Bornholm especially on the coasts of southern Denmark Gap would bring a ship between the promi- (see Fig. 3). nent landmarks of Cape Arkona on the island However, it is not a necessity only to of Rügen and the white cliffs of Møn. A pos- look for settled and inhabited landing sites. sible anchorage could be the northeast coast A skipper could choose to anchor at deso- of Bornholm, and from here to Utlängen the late islands and spits of land. A partly-exca- course is northbound. This trip would take vated site of the latter type is known at Fyns around five or six days and nights. Hoved.46 This might very well have been a Even though the winds and currents of commonly used and relatively safe way of the Baltic Sea can be difficult or even treach- spending the night. In this way the risk of erous for a sailing ship, it is not the natural sudden attacks from local residents looking conditions that constitute the primary ob- for easy plunder could be reduced. stacle to navigation. It is more a question of A skipper unfamiliar with the waters, re- security for ship and crew: a skipper or pilot sponsible for his and his crew’s safety, would familiar with the area, the guarantee of a safe have to avoid the coastal routes as well as in- journey and landing or showing the ability habited landing sites and densely-populated to defend oneself in an unmistakable way areas. He would either choose desolate and would have ensured the risk of attack would isolated camping sites like Fyns Hoved or be kept to a minimum. 46. Henriksen 1994. 142 II. The western and central Baltic Sea region References Englert, A. 2007: Ohthere’s voyages seen from a nautical angle. In J. Bately & Archaeological excavations in Denmark 2001, A. Englert (eds), Ohthere’s Voyages. A late 2002: Kulturarvsstyrelsen. Copenhagen. 9th-century account of voyages along the Callmer, J. 1991: Platser med anknytning coasts of Norway and Denmark and till Handel och Hantverk i yngre järn- its cultural context. Maritime Culture ålder. Eksempel från södre Sverige. In P. of the North 1, 117-129. Roskilde. Mortensen & B.M. Rasmussen (eds), Gärtner, B. & Ulriksen, J. 1997: Vester Eges- Fra Stamme til Stat 2: Høvdingsam- borg-pladsen – en anløbsplads fra vi- fund og kongemagt. Jysk Arkæologisk Sel- kinge tid ved Dybsø Fjord. Liv og skabs Skrifter 32:2, 29-48. Levn 11, 11-14. Carlsson, D. 1991: Harbours and trading Hansen, K.M. & Høyer, H. 2000: Næs – en places on Gotland. In O. Crumlin-Ped- vikingetidsbebyggelse med hørproduk- ersen (ed.), Aspects of Maritime Scandi- tion. Kuml 2000, 59-89. navia AD 200 - 1200. Proceedings of the Helgesson, B. 2002: Järnålderns Skåne: sam- Nordic Seminar on Maritime Aspects of hälle, centra och regioner. Uppåkrastudier Archaeology, Roskilde, 13th-15th March, 5, Stockholm. 1989, 145-158. Roskilde. Henriksen, M.B. 1994: Vikinger på Fyns Christensen, T. 1993: Lejre beyond the Myth. Hoved. Fynske Minder 1994, 181-191. Journal of Danish Archaeology 10, 163-185. Odense. Crumlin-Pedersen, O. 1983: Skibe, sejlads og Henriksen, M.B. 1997: Vikinger ved Helnæs- ruter hos Ottar og Wulfstan. In Lund bugten. Fynske Minder 1997, 25-58. 1983, 32-44. Odense. Danmarks Riges Breve, 1. række, 1. og 2. bind, Henriksen, M.B. 2000: Lundsgård, Seden 789-1169, 1975: Det danske Sprog- og Lit- Syd og Hjulby. Tre fynske boplads- teraturselskab. områder med detektorfund. In M.B. Dobat, A.S. 2002: Snekke-namen an der Henriksen (ed.), Detektorfund – hvad Schlei. Zeugnisse einer Infrastruktur im skal vi med dem? Dokumentation og Umfeld von Haithabu und Schleswig? registrering af bopladser med detektor- Archäologisches Korrespondenzblatt 32 fund fra jernalder og middelalder. Rap- Heft 4, 607-620. Mainz. port fra et bebyggelseshistorisk seminar Dobat, A.S. 2004: Hedeby and its Maritime på Hollufgård den 26. oktober 1998, 17-60. Hinterland. The Schlei Fjord as an Early Odense. Medieval Communication Route. In Henriksen, M.B. 2002: Er Hjulby Nyborgs H. Jöns & F. Lüth (eds), 53. Internation- forgænger? Fynske Minder 2002, 155-185. ales Sachsensymposium vom 31. August Odense. bis 4. September 2002 in Schwerin. Über- Jacobsen, J.A. 2001: Fynske jernalderbopladser regionaler Kulturaustausch im Nordsee- bind 2. Odense herred. Skrifter fra Odense und Ostseeraum von 200 bis 800 AD. Bys Museer. Odense. Verkehrswege – Marktplätze – Handels- Jacobsen, L. & Moltke, E. 1942: Danmarks waren. Bodendenkmalpflege in Mecklen- runeindskrifter. Copenhagen. burg-Vorpommern, Jahrbuch 2003 – 51, Jørgensen, B. 1994: Stednavneordbog. Copen- 419-435. Lübstorf. hagen. Viking-Age sailing routes of the western Baltic Sea 143 Jørgensen, L. 2003: Manor and market at lake Ulriksen, J. 1990: Teorier og virkelighed i for- Tissø in the 6th to 11th centuries: the bindelse med lokalisering af anløbspla- Danish ‘productive’ sites. In T. Pestell dser fra germanertid og vikingetid i Dan- & K. Ulmschneider (eds), Markets in mark. Aarbøger for Nordisk Oldkyndighed Early Medieval Europe. Trading and og Historie 1990, 69-101. Copenhagen. ‘Productive’ Sites, 650-850, 175-207. Mac- Ulriksen, J. 1998: Anløbspladser. Besejling og be- clesfield. byggelse i Danmark mellem 200-1100 e.Kr. Larsson, L. 2002: Uppåkra – research on En studie af søfartens pladser på baggrund a central place. Recent excavations af undersøgelser i Roskilde Fjord. Roskilde. and results. In B. Hårdh & L. Larsson Ulriksen, J. 2002: Håndværksspor på yngre (eds), Central places in the Migration jernalders anløbspladser In M.B. Hen- and Merovingian Periods. Papers from riksen (ed.), Rapport fra seminar om the 52nd Sachsensymposium, Lund, August håndværksspor på bopladser fra yngre 2001.Uppåkrastudier 6, 19-30. Stock- jernalder, 7-16. Odense. holm. Ulriksen, J. 2004: Danish coastal landing Lund, N. 1983: Ottar og Wulfstan – to places and their relation to navigation rejsebeskrivelser fra vikingetiden. Ros- and trade. In J. Hines, A. Lane & M. Red- kilde. knap (eds), Land, sea and home: proceed- Ørsnes, M. 1966: Form og stil i Sydskand- ings of a conference on Viking-period set- inaviens yngre germanske jernalder. tlement at Cardiff, July 2001, 7-26. Maney. Arkæologisk-historisk række XI. Copen- Ulriksen, J. 2006: Vester Egesborg – a coastal hagen. settlement from the late Iron Age on Schilling, H. 2003: Duesmindeskatten. Skalk Zealand. Journal of Danish Archaeology 6, 5-12. 14, 187-199. 144 II. The western and central Baltic Sea region