Version classiqueVersion mobile

Batellerie gallo-romaine

 | 
Giulia Boetto
, 
Patrice Pomey
, 
André Tchernia

2 – Recent research on roman shipfinds from the Netherlands

Recherche récente sur les découvertes de navires romains aux pays-bas

André van Holk

Résumé

Cet article présente les nouvelles et anciennes découvertes de barges et d’un punt de l’époque romaine. Depuis 2003, quatre nouvelles épaves, datées du IIe au IIIe s. ap. J.-C., ont été retrouvées le long du limes aux Pays-Bas. Ces épaves ont été étudiées et publiées en néerlandais. Elles ont fourni des informations très intéressantes concernant la provenance du bois (De Meern 1 et 4, Woerden 7), la composition de l’équipage et la vie à bord (De Meern 1), la séquence de construction et l’usage de la technique méditerranéenne par tenon-et-mortaise pour l’assemblage des virures (De Meern 4), un punt, type de navire totalement nouveau (De Meern 6) et des détails sur la propulsion (Woerden 7). Pour ces raisons, de nouvelles recherches ont été entreprises sur les barges Zwammerdam 2 et 6 et Woerden 7. Cet article donne une vue d’ensemble de ces nouvelles découvertes et de ce nouveau projet de recherche afin d’en assurer une meilleure diffusion.

Texte intégral

Introduction

1Since 2003 four new wrecks dating to the second and the third century AD have been found along the limes in the Netherlands. The preliminary results of the De Meern 1 excavation have been published in English in the proceedings of the tenth ISBSA (van Holk 2006, p. 295-299), while three wrecks (De Meern 1 and 4 and Woerden 7) have been published in Dutch (Jansma, Morel 2007; de Groot, Morel 2007; Blom, Vorst, Vos 2008). The wreck De Meern 6 has been published as an internal report (Morel, n.d.). To make these investigations available to a wider audience the results are summarized in this paper in English.

2Furthermore, we would like to draw the attention to new research on the conserved barges 2 and 6 from Zwammerdam and the barge Woerden 7. The research questions of this project are directed towards the procurement and management of raw material, the building sequence and other aspects of the construction, as well as the function and disposal of barges in use on the lower Rhine (see infra).

De Meern 1

3The De Meern 1 barge was found in 1997 and excavated in 2003. The wreck was found in a side-branch of the River Rhine, the Heldammer stroom, at only six meters from the Roman road (fig. 1). Because the vessel was wrecked and not reused or sunken on purpose – as is the case with most of the Dutch wrecks from the Roman era – and quickly got covered by sediments, an extremely rich artefactual inventory was recovered.

Fig. 1. Map showing the limes and the findspots of the Roman wrecks (drawing E. Bolhuis, GIA).

Fig. 1. Map showing the limes and the findspots of the Roman wrecks (drawing E. Bolhuis, GIA).

1 = Zwammerdam 2 and 6 (Nigrum Pullum) ; 2 = Woerden 7 (Laurum) ; 3 = De Meern 1 ; 4 = De Meern 4 ; 5 = De Meern 6

4The building date of the vessel was established by dendrochronological research at 148 AD. The foundering must have taken place some forty years later, as could be made up from the composition of the artefactual inventory. But far more interesting was the result of the dendrochronological research concerning the provenance of the wood. It was a big surprise that the trees used for building the ship had grown in the middle part of the Netherlands. This means that the vessel was built in the nearby surroundings. So the hypothesis of Roman ships being built upstream the Rhine is untenable. Dendrochronological analyses led to some other interesting results. Certain elements of the ship were built out of the same tree: for example the L-shaped chines that form the transition between the bottom and the sides. The L-shaped chine is a common feature of the Zwammerdam-type barges.

5The hull-form of De Meern 1 is extremely simple: a box-like vessel with for-and-aft swimheads. A striking feature of the hull-form is the extreme L to B ratio of 9,1: 1. The ship was clearly adapted to narrow waterways (while still carrying as much cargo as possible).

6The bottom of De Meern 1 consists of four wide bottom planks, in carvel position. The sides consist of an overlapping strake nailed to the standing part of the L-shaped chine. A narrow wash strake is nailed against the side strake. The framing system is regular and consists of double (paired) floor timbers with a floor timber knee against the starboard and port side.

7The presence of spike-plugs is an indication for the use of temporary timbers to keep the bottom planks together, before inserting the frames. The frames are nailed to the strakes from the inside and from the outside. This means that the ship had to be turned over during the building process. Above each seam between the planks, a limber hole is cut in the underside of the floors. According to Morel (Jansma, Morel 2007, p. 142) these were used for ventilating the seams.

8Slanting nails were used to connect the bottom planks for-and-aft, while long nails run through the thin part of the planks to connect the scarfs (fig. 2). This type of fastening could reflect a way of thinking about how a boat should be built. In this case by mortice-and-tenon (a mental template), which implies that planks are connected through the sides of the planks (van Holk 2006, p. 296). Other interesting details of the construction are the T-shaped stern and two ‘beam-cradles’ meant to support a removable beam. On top of the T-shaped stern probably a steering oar was mounted. It is possible that the removable beam supported a temporary deck to stand on while steering the vessel. Another curious detail are the zigzag lines cut into the strakes (both fore-and-aft). Were they made to absorb tension in the planking at the point of the swim-head where bending of the planks was at a maximum?

Fig. 2. Long nail running through scarf between two elements of the L-shaped chine (photo T. Penders, RCE).

Fig. 2. Long nail running through scarf between two elements of the L-shaped chine (photo T. Penders, RCE).

9For propulsion a mast step was cut into the thick part of one of the floor timbers (the zig-zag lines in the bow are situated just in front of the mast step). Beside the mast step two iron braces were mounted through which a wedge could be driven to secure the back of the mast foot.

10In the De Meern 1 wreck an extremely well preserved inside lining was present. The vessel was divided by three partitioning walls in two different living quarters: a cabin aft and in front of that a galley. The foremost compartment is interpreted as a galley because a roof-tile blackened by fire was found there, while the ceiling also showed traces of fire. The galley was only accessible through the front bulkhead, from the hold. The spatial separation of these two live-spheres might reflect a social segregation on board between the skipper and the other members of the crew (van Holk 2006, p. 297).

11A very exciting moment during the excavation was the finding of a cupboard and a box at a high level in the cabin (fig. 3). Probably the cupboard and the box had started drifting after the ship had sunk and got stuck against the roof of the cabin, which was obviously at that time, still in place.

Fig. 3. Cupboard and box found at a high level in the cabin (photo J.Koch).

Fig. 3. Cupboard and box found at a high level in the cabin (photo J.Koch).

12The artefactual inventory is dealt with in great detail in the comprehensive publication on the De Meern 1 vessel (van Holk 2007, p. 257-268). So only some conclusions based on the study of the material will be presented here. It could be concluded that in the cabin at least one grown-up male resided. The tools indicate that he was involved in carpentry and stonework (fig. 4). The idea is that the skipper carried out maintenance work along the limes. The military objects found in the vessel suggest that he was a veteran from the Roman army. Especially the fashionable shoes show that he was romanised, that is to say, integrated in Roman society and culture. He was used to move around in Roman company and to a certain extent well-to-do.

Fig. 4. Tools from the cabin (photo J.Koch).

Fig. 4. Tools from the cabin (photo J.Koch).

De Meern 4

13De Meern 4 was found just after the excavation of De Meern 1 was finished, at a distance of only 150 m from the De Meern 1 site (fig. 1). A big difference with the De Meern 1 shipfind is the fact that De Meern 4 was reused as a river embankment and not foundered like De Meern 1 ‘in action’. The ship was only partly excavated ; but nevertheless the results of the investigation are of great interest from a nautical archaeological point of view. The dimensions of the vessel had to be reconstructed. The estimated overall length is 27 m, while the greatest beam is 3,7 m. The stern was only preserved as a soilmark.

14Dendrochronological analysis of the wood samples gave a felling date of 100 AD for the trees used to built the ship. Again the most surprising result was the provenance of the wood: the mid-Netherlands. This result indicates that the wharf where the vessel had been built was probably situated in that area (fig. 5). This is a confirmation of the hypothesis concerning De Meern 1 that at least some of the Roman barges found in the Netherlands are of local origin.

Fig. 5. Map showing the provenance of some of the wood samples from De Meern 4 (after Jansma, Morel 2007).

Fig. 5. Map showing the provenance of some of the wood samples from De Meern 4 (after Jansma, Morel 2007).

15The De Meern 4 is, not surprisingly, constructed with an L-shaped chine. The seven bottom planks are somewhat narrower than the ones of De Meern 1. The side consists of a carvel side plank. On the outside a whale is nailed against the side strake. Against the inside of the whale a stringer was fastened. The frames are – in contrast to De Meern 1 – single with alternating floor timber knees against the port and starboard side. Another feature of the construction, yet a complete surprise and uncommon in Zwammerdam-type barges, was the presence of mortice-and-tenon joints to connect the planks mutually. So a genuine Mediterranean technique was used. Frames and strakes are connected by nails from the inside and outside. Again like De Meern 1, above each seam between planks a ‘limber hole’ is cut, probably for ventilating the seams (de Groot, Morel 2007, p. 49). The mortice-and-tenon fastenings are not placed in a regular pattern. The slots are remarkable in their size and form: very wide at the seam between the planking. The fit between slots in adjacent planks is not exact. Some tenons are even placed beside the mortices (fig. 6). This mortice-and-tenon technique is used in a rather rough and clumsy way. As to the sequence of construction, the presence of spike-plugs is of relevance (de Groot, Morel 2007, p. 39). They showed up in two athwartship rows, 1,25 m apart. They were placed centrally in the planks, with only one spike-plug for each plank. In one row the spike-plugs were found only in the two outer bottom planks while the other row has them only in the five central planks (fig. 7). Morel (de Groot, Morel 2007, p. 39-41, 54) states that the bottom planks were temporary connected to a beamframe, on which the bottom planks were nailed. After inserting and fastening the frames with short nails from the inside to the bottom planks, the ship was tilted and long nails were hammered in from the outside and turned over on the floors on the inside of the hull. After this the temporary beamframe could be removed, the ship was put back into a horizontal position and the nails connecting the bottom planks with the temporary beamframe were pulled out. The nailholes were closed with spike-plugs, only from the inside. The nail holes in the outside of the bottom planks probably closed automatically, after the ship was launched (J. Morel, written communication 2010).

Fig. 6. Mortice-and-tenon connection between bottom planks of De Meern 4 (drawing J.M.A.W. Morel, RCE).

Fig. 6. Mortice-and-tenon connection between bottom planks of De Meern 4 (drawing J.M.A.W. Morel, RCE).

Fig. 7. Different rows of spike-plugs in the bottom planks of De Meern 4 (drawing J.M.A.W. Morel, RCE).

Fig. 7. Different rows of spike-plugs in the bottom planks of De Meern 4 (drawing J.M.A.W. Morel, RCE).

Woerden 7

  • 1 In the excavation report, the wreck is erroneously situated to the southwest of the castellum (Blom (...)

16The year 2003 was an extremely rich ‘Roman barge year’ in the Netherlands. For in that same year not only De Meern 1 and 4 were found and (partly) excavated, but also another Zwammerdam-type barge, Woerden 7, was unearthed. The barge was found to the northwest1 of the castellum Laurium (fig. 1). In the vicinity of the vessel no traces of a sheetpiling or a wharf were present. Although the excavators state that the ship had been foundered, it seems more plausible the ship was put aside, because it was found on a southern riverbank along the river Rhine, on the transition of the bank to the river. Moreover there was no artefactual inventory or cargo of any significance left inside the wreck. After the barge was left aside the riverbed moved within a short period of time 50 m to the north, so the wreck got covered up quickly by sediments. What is quite clear is that the vessel has to be placed in the military context of the Roman frontier, the limes.

17The Woerden 7 is a barge of Zwammerdam-type, with a box-like shape and parallel sides. The close resemblance to Zwammerdam 6 may indicate that the two were sister-ships (Blom, Vorst, Vos 2008, p. 391). To name but a few similarities: a long keelson with mast step, cross-bond frames (probably both fore-and-aft) and a single framing system. The reconstructed length is 29,6 m and the beam is 4,7 m (fig. 8). Again the transition of the bottom to the sides consists of an L-shaped chine. The chine consists of five pieces: a long part over most of the length of the vessel and two shorter pieces at the ends where the swimhead is situated. A triangular piece of wood connects the middle and end parts of the chine. The bottom is flat and made of ten flush-laid strakes. The seams between the strakes were caulked with reed mace (Typha latifolia), which was held in place by thousands of small nails and laths at the swimheads. The frames are single with alternating floor timber knees at port and starboard. Again, as with the barges De Meern 1 and 4, limber holes were cut in the underside of the floors above all plank seams.

Fig. 8. Woerden 7 during excavation (after Blom, Vos 2008).

Fig. 8. Woerden 7 during excavation (after Blom, Vos 2008).

18The vessel shows some very interesting details. The side plank is carvel-built and fastened to the L-shaped chine with iron cramps. Also interesting is the L-shaped covering board on top of the side plank. Finally the rowing fittings are an interesting element of the construction of the vessel (Blom, Vorst, Vos 2008, p. 366). In the gunwale holes are made to secure the oars (with a rope or leather strap) and on top of the gunwale oarlocks are visible (fig. 9). Furthermore, in the covering board dovetail joints are cut, probably for thwarts. The thwarts themselves have not been found. The rowing fittings might indicate that the vessel was not only used for transport downstream but also was capable of rowing upstream.

Fig. 9. Rowing fittings of Woerden 7 : gunwale with holes and oarlocks. (after Blom, Vos 2008).

Fig. 9. Rowing fittings of Woerden 7 : gunwale with holes and oarlocks. (after Blom, Vos 2008).

19The trees used for building the vessel have been felled in the winter of 162/163 AD. The trees grew in Mid-Germany and again – surprisingly – in the Mid-Netherlands. The wood provenance makes a building place for the barge in the Mid-Netherlands very likely. Like the wood provenance of De Meern 1 shows, the idea that Roman barges were built upstream in Germany has also to be revised in the light of the results of the dendrochronological research of the barge Woerden 7. One of the other interesting results of this research is the arrangement of the planking (Blom, Vorst, Vos 2008, p. 383). Planks from the same tree were placed sytematically around strake 5 of which the left side actually forms the centre line of the ship (fig. 10).

Fig. 10. Planks from the same tree are placed systematically around strake 5 (drawing Y. E. Vorst).

Fig. 10. Planks from the same tree are placed systematically around strake 5 (drawing Y. E. Vorst).

De Meern 6

20In March 2008 fragments of wood were discovered near De Meern (fig. 1). Investigation by the Cultural Heritage Agency (former RACM, former NISA) showed the timbers to be from a highly interesting shipwreck: a punt-like craft from the Roman era. At this moment the date is uncertain, but on the basis of the stratigraphy a date in the third century AD seems likely. If the date is correct this is the first punt dating in Roman times in North-western Europe (Morel n.d., p. 3).

21Only one – of the probably two - bottom planks was found, together with one strake of the sides (fig. 11). Of the inner timbers only a fragment of a frame survived that was nailed to the bottom planks. Nails and nail holes indicate the position of the other frames. The nails used to connect the frames to the planking are quite large: this is another indication for a Roman vessel.

Fig. 11. Plank remains of De Meern 6 in situ (drawing J.-M.A.W. Morel and F. Dallmeijer, RCE).

Fig. 11. Plank remains of De Meern 6 in situ (drawing J.-M.A.W. Morel and F. Dallmeijer, RCE).

22One end of the bottom strake shows a rabbet for a stem or stern post.

23A preliminary reconstruction of the vessel was made, which led to the following dimensions (Morel n.d., p. 9): length of the bottom 9 m, width of the bottom 1,05 m. The bottom consisted of two planks and had a lancet form at the ends; the sides were probably made out of one plank. The side plank fitted into a rabbet in the edge of the bottom plank. From the oblique underside of the end of the one preserved frame the rake of the sides could be deduced at about 125°. The mean framing distance is 59,75 cm, which is close to two Roman feet of 29,6 cm.

24The reconstructed overall length of the vessel is between 9,5 to 10 m, the width 1,5 to 1,6 m and the depth 0,35 to 0,40 m.

25Punt-like vessels are still being built in the Netherlands and they are often referred to as typically Dutch. This find shows that the punt-like vessel might be of ‘Italic’ design (Morel n.d., p. 13)!

Research project concerning ships of the Zwammerdam type

  • 2 As indicated on the Peutinger map, a medieval copy of a Roman travel map.

26Between 1968 and 1971 the Roman auxiliary fort Nigrum Pullum2 (fig. 1) situated on the southern riverbank of the Old Rhine near the modern town of Zwammerdam was excavated by the University of Amsterdam (former Institute for Pre- and Protohistory IPP) (Haalebos 1977). Nigrum Pullum was founded in the middle of the 1st century AD. The military settlement was burnt down during the Batavian revolt in 69 AD when most Roman forts along the lower Rhine were destroyed. A wooden fort was built shortly thereafter, which survived until c. 175 AD. A third occupation phase in which the main building (the principia) and the walls surrounding the fort were rebuilt in stone, lasted until c. 275 AD. During the occupation a wooden revetment of some 500 m in length was constructed along the riverbank just outside of the fort (De Weerd 1988, p. 26-27).

27Along this revetment a seven meter oak dugout was found in 1971, followed by two other dugouts and three barges of 20, 23 and 34 m in length all in the same area. The barges showed strong similarities and are known as the Zwammerdam type barges. These ships are long and flat-bottomed with no keel. They have long, low vertical sides, with ribs mostly comprised of single pieces of L-shaped timber, with the long sides nailed to the bottom and the shorter vertical pieces supporting the sides. Most distinctive is the L-shaped plank, cut from a single piece of timber, which forms the right angled junction between the side and the bottom of the craft (Marsden 1976, p. 44-46).

28All vessels from Zwammerdam were excavated and put in conservation which, with some interruptions, took 30 years to complete.

  • 3 As described above; published in Jansma, Morel 2007 ; Blom, Vos 2008.
  • 4 As described above; published in De Groot, Morel 2007.

29During that time, in the Netherlands new barges of the Zwammerdam type were discovered and excavated in 2003 (De Meern 1 and Woerden 7)3 and 2005 (De Meern 4)4. De Meern 1 was lifted and put in conservation taking up much of the capacity to conserve large ‘wet’ wooden artefacts in the Netherlands. Therefore, the discovery of Woerden 7 shortly after the excavation of De Meern 1 made the excavators of this barge decide to record this vessel as best as possible by taking it apart while recording, retaining only some characteristic parts for conservation. This approach allowed for extensive dendrochronological sampling.

  • 5 Barges of the Zwammerdam type: chronology, provenance, construction and wood technology’.

30The dendrochronological research of Woerden 7 was directed at answering questions about provenance (i.e. to locate the source area of the trees used for building the ship), but also yielded information on the construction date, the building sequence, and the choice of timber (Vorst 2005). The results still need to be integrated in a report on the construction of the barge Woerden 7. This will be part of a PhD project carried out by Yardeni Vorst, which is currently investigating the conserved ships Zwammerdam 2 and 6 in a similar way5. Zwammerdam 2 will be rerecorded and reassembled after dendrochronological study. Zwammerdam 6 will be dendrochronologicaly examined, but not reassembled.

  • 6 NWO-research programme (Humanities) ‘A sustainable frontier: The establishment of the Roman frontie (...)

31When investigating regional influences and Mediterranean characteristics of fluvial ships from the Roman period, it is important to look at the ships themselves, but also at the region where they were found and functioned. What did these vessels carry, i.e. what was their function and where did they navigate? Current research6 informs us that the barges in the western part of the lower Rhine limes functioned in an environment that was highly structured towards shipping by the Roman army (Van Dinter 2008, p. 30). The Zwammerdam ships were found right outside a military fort, as were other ships of this type, for instance the barges of Woerden. Hence the military connection should be further explored.

  • 7 NWO-research programme (Humanities) ‘Arts and crafts in Roman ship building: raw material managemen (...)

32The research is part of a larger project7, which looks at the Roman wood and timber economy in the lower Rhine delta and will also focus on barges from the transalpine region in general.

Conclusions

33The artefactual inventory of the barge De Meern 1 sheds some light on the organisation of shipping along the limes. The skipper might have been a veteran of the Roman army, not necessarily Roman as the composition of the artefactual inventory shows, but rather of indigenous Batavian origin.

34Dendrochronological research of barges De Meern 1 and 4 and Woerden 7 indicates that at least some of these barges were built in the Netherlands. This means that the Roman tradition of ship-building was maybe more firmly established along the northern part of the limes than until now has been assumed. The barge De Meern 4 makes clear that the Mediterranean technique of mortice-and-tenon to connect the strakes was in use along this part of the limes around 100 AD.

35From the find De Meern 6 it can be assumed that punt-like vessels are certainly not typically Dutch, but are of ‘Italic’ design.

36It is clear that lots of questions are unsolved and more research has to be done on this interesting group of ships. That is exactly the reason why a PhD program has been started in the Netherlands in which the barges Zwammerdam 2 and 6 and Woerden 7 will be re-examined, with regard to questions concerning the chronology, provenance, construction and wood technology.

Bibliographie

Blom, Vos 2008 : BLOM (E.), VOS (W.K.) eds., Woerden-Hoochwoert. De opgravingen 2002-2004 in het Romeinse castellum Laurium, de vicus en van het schip de ‘Woerden 7’, Amersfoort, ADC ArcheoProjecten/Hazenberg Archeologie Leiden, 2008 (ADC Rapport, 910).

Blom, Vorst, Vos 2008 : BLOM (E), VORST (Y.), VOS (W.K.) – De ‘Woerden 7’: Een Romeinse platbodem. In : Blom (E.), Vos (W.K.) eds., Woerden-Hoochwoert. De opgravingen 2002-2004 in het Romeinse castellum Laurium, de vicus en van het schip de ‘Woerden 7’, Amersfoort, ADC ArcheoProjecten/Hazenberg Archeologie Leiden, , 2008, p. 349-401 (ADC Rapport, 910).

van Dinter 2008 : DINTER VAN (M.) – Transformatieprocessen in een Romeins landschap. In : Erdich (M.), Dinter van (M.), Rijn van (P.), Polak (M.), Kooistra (L.I.), Cavallo (C.), Dütting (M.K.), Chorus (J.), Een duurzaame Romeinse grens. Archeobrief, 12-1, 2008, p. 30.

de Groot, Morel 2007 : GROOT DE (T.), MOREL (J.-M. A. W.) eds. – Het schip uit de Romeinse tijd De Meern 4 nabij boerderij de Balije, Leidsche Rijn, gemeente Utrecht: waardestellend onderzoek naar de kwaliteit van het schip en het conserverend vermogen van het bodemmilieu, Amersfoort, Rijksdienst voor archeologie, cultuurlandschap en monumenten, 2007 (Rapportage archeologische Monumentenzorg, 147).

Haalebos 1977 : HAALEBOS (J.K.) – Zwammerdam-Nigrum Pullum; ein Auxiliarkastell am Nieder-Germanischen Limes, Amsterdam, 1977 (Cingula, 3).

van Holk 2006 : HOLK VAN (A.F.L.) – A Roman barge with an artefactual inventory from De Meern (the Netherlands). In : Blue (L.), Hocker (F.) And Englert (A.) eds., Connected by the sea. Proceedings of the 10th International Symposium on Boat and Ship Archaeology (ISBSA 10), Roskilde (2003), Oxford, Oxbow Books, Oxford, 2006, p. 295-299.

van Holk 2007 : HOLK VAN (A.F.L.) – Analyse van de inventaris. In : Jansma (E.), Morel (J.-M. A. W.) eds., Een Romeinse Rijnaak, gevonden in Utrecht-De Meern. Resultaten van het onderzoek naar de platbodem ‘De Meern 1’, Amersfoort, Rijksdienst voor archeologie, cultuurlandschap en monumenten, 2007, p. 257-268 (Rapportage archeologische Monumentenzorg, 144).

Jansma, Morel 2007 : JANSMA (E.), MOREL (J.-M. A. W.) eds. – Een Romeinse Rijnaak, gevonden in Utrecht-De Meern. Resultaten van het onderzoek naar de platbodem ‘De Meern 1’, Amersfoort, Rijksdienst voor archeologie, cultuurlandschap en monumenten, 2007 (Rapportage archeologische Monumentenzorg, 144).

Marsden 1976 : MARSDEN (P.) – A boat from the Roman period, found at Bruges, Belgium, in 1899, and related types. International Journal of Nautical Archaeology, 5.1, 1976, p. 23-55.

Morel n. d. : MOREL (J.-M.A.W.) – Standaardrapport inventarisatie punter De Meern 6, Utrecht/Lelystad, Gemeente Utrecht, Sectie Cultuurhistorie, Rijksdienst voor Archeologie, Cultuurlandschap en Monumenten (RACM).

Vorst 2005 : VORST (Y. E.) De constructie en herkomst van de Romeinse platbodem ‘Woerden 7’; een studie van jaarringpatronen en bewerkingssporen, MA thesis, University of Amsterdam, 2005.

de Weerd 1988 : WEERD DE (M.D.) – Schepen voor Zwammerdam, PhD thesis, University of Amsterdam, 1988.

Notes

1 In the excavation report, the wreck is erroneously situated to the southwest of the castellum (Blom, Vorst, Vos 2008, p. 352).

2 As indicated on the Peutinger map, a medieval copy of a Roman travel map.

3 As described above; published in Jansma, Morel 2007 ; Blom, Vos 2008.

4 As described above; published in De Groot, Morel 2007.

5 Barges of the Zwammerdam type: chronology, provenance, construction and wood technology’.

6 NWO-research programme (Humanities) ‘A sustainable frontier: The establishment of the Roman frontier in the Rhine Delta’.

7 NWO-research programme (Humanities) ‘Arts and crafts in Roman ship building: raw material management, Construction, technology, use and disposal of barges in the lower Rhine region in the Roman period’.

Auteur

Professor Maritime Archeology
State University of Groningen
NL – 9712 CP Groningen (The Netherlands)
a.vanholk@nieuwlanderfgoedcentrum.nl

© Publications du Centre Camille Jullian, 2011

Conditions d’utilisation : http://www.openedition.org/6540

Lire

Open access

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search