(2006) Roman river barge from Sisak (Siscia), Croatia

An offprint from CONNECTED BY THE SEA Proceedings of the Tenth International Symposium on Boat and Ship Archaeology Roskilde 2003 Edited by Lucy Blue, Fred Hocker and Anton Englert ISBSA 10 Hosted by The Viking Ship Museum, Roskilde The National Museum of Denmark Centre for Maritime Archaeology The National Museum of Denmark Institute of Maritime Archaeology The Centre for Maritime Archaeology, University of Southampton With support from The National Research Foundation of Denmark The National Humanities Research Council of Denmark Queen Margrethe’s Fund © Oxbow Books 2006 ISBN 978 1 84217 228 5 1 84217 228 X Contents List of Contributors ........................................................................................................................................................ viii Preface ................................................................................................................................................................................ xi Keynote address: An international forum for nautical research 1976–2003 by Ole Crumlin-Pedersen .................. xiii Seán McGrail: Walking on water: Maritime archaeology by air, land and sea by Jonathan Adams ......................xvii A. EXPERIMENTAL ARCHAEOLOGY 1. Experimental archaeology and ships – principles, problems and examples Ole Crumlin-Pedersen ................................................................................................................................................ 1 2. Experimental boat archaeology: Has it a future? Seán McGrail .............................................................................................................................................................. 8 3. Experimental archaeology at the Viking Ship Museum in Roskilde Søren Nielsen ............................................................................................................................................................ 16 4. History written in tool marks Thomas Finderup ...................................................................................................................................................... 21 5. Reconstruction of rope for the copy of Skuldelev 2: Rope in the Viking Period Ole Magnus ............................................................................................................................................................... 27 6. Trial voyages as a method of experimental archaeology: The aspect of speed Anton Englert ............................................................................................................................................................ 35 7. An example of experimental archaeology and the construction of a full-scale research model of the Cavalière ship’s hull Sabrina Marlier ........................................................................................................................................................ 43 8. Reconstruction of the large Borobudur outrigger sailing craft Erik Petersen ............................................................................................................................................................ 50 9. The construction and trials of a half-scale model of the Early Bronze Age ship, Ferriby 1, to assess the capability of the full-size ship Edwin Gifford, Joyce Gifford and John Coates ..................................................................................................... 57 10. The value of experimental archaeology for reconstructing ancient seafaring Timm Weski ............................................................................................................................................................... 63 11. The Pacific migrations by canoe-form craft James Wharram and Hanneke Boon ....................................................................................................................... 68 B. THEORETICAL ISSUES IN THE CONSTRUCTION OF SHIPS 12. New light on the false clinkers in ancient Mediterranean shipbuilding Patrice Pomey ........................................................................................................................................................... 74 13. A preliminary report on the hull characteristics of the Gallo-Roman EP1-Taillebourg wreck (CharenteMaritime, France): archaeological evidence of regional practices of ancient flat-bottomed construction? Eric Rieth .................................................................................................................................................................. 78 14. The Dor 2001/1 wreck, Dor/Tantura Lagoon, Israel: Preliminary Report Yaacov Kahanov and Hadas Mor ........................................................................................................................... 84 15. A hypothesis on the development of Mediterranean ship construction from Antiquity to the Late Midde Ages Carlo Beltrame and Mauro Bondioli ...................................................................................................................... 89 16. Geometric rules in early medieval ships: Evidence from the Bozburun and Serçe Limanı vessels Matthew Harpster ..................................................................................................................................................... 95 17. Oak growing, hull design and framing style. The Cavalaire-sur-Mer wreck, c. 1479 Brad Loewen and Marion Delhaye ......................................................................................................................... 99 18. Ship design in Holland in the eighteenth century Ab Hoving ............................................................................................................................................................... 105 19. Archaeobotanical characterisation of three ancient, sewn, Mediterranean shipwrecks Stéphanie Wicha and Michel Girard .................................................................................................................... 111 20. Coating, sheathing, caulking and luting in ancient shipbuilding Ronald Bockius ....................................................................................................................................................... 117 C. BETWEEN LAND AND SEA 21. Roman techniques for the transport and conservation of fish: the case of the Fiumicino 5 wreck Giulia Boetto .......................................................................................................................................................... 123 22. Land and sea connections: the Kastro rock-cut site (Lemnos Island, Aegean Sea, Greece) Christina Marangou ............................................................................................................................................... 130 23. Local boat-building traditions in the Bristol region Anthony J. Parker ................................................................................................................................................... 137 24. The harbour of Haiðaby Sven Kalmring ........................................................................................................................................................ 143 25. Money, port and ships from a Schleswig point of view Christian Radtke ..................................................................................................................................................... 147 26. Inland water transport in the Pre-Roman and Roman Iron Age in Northern Germany and its role in intra- and intercultural communication Ulrike Teigelake ..................................................................................................................................................... 152 27. Staraya Ladoga: a seaport in medieval Russia Petr Sorokin ............................................................................................................................................................ 157 28. The APES Archaeological Study: The North Carolina Sounds, an interface between land and sea Lawrence E. Babits, Frank Cantelas and Keith Meverden ................................................................................. 163 D. LONG DISTANCE SEAFARING AND THE CONNECTIONS BETWEEN CULTURES 29. The ends of the earth: maritime technology transfer in remote maritime communities Valerie Fenwick ...................................................................................................................................................... 171 30. The ships that connected people and the people that commuted by ships: The western Baltic case-study George Indruszewski, Marcus Nilsson and Tomasz Ważny ................................................................................. 177 31. Early cogs, Jutland boatbuilders, and the connection between East and West before AD 1250. Fred Hocker and Aoife Daly ................................................................................................................................. 187 32. Couronian ship building, navigation and contacts with Scandinavia Inese Karlina .......................................................................................................................................................... 195 E. HISTORICAL, ICONOGRAPHIC AND ETHNOGRAPHIC SOURCES AND APPROACHES 33. From Carl Reinhold Berch to Nils Månsson Mandelgren: On the concept of maritime history, (Sw. sjöhistoria), and its meanings in Sweden since the latter 18th century Carl Olof Cederlund .............................................................................................................................................. 199 34. Ships and subsidies David A. Hinton ...................................................................................................................................................... 205 35. Sea-lanes of communication: Language as a tool for nautical archaeology Katrin Thier ............................................................................................................................................................ 210 36. Medieval shipping in the estuary of the Vistula River. Written sources in the interpretation of archaeological finds Robert Domzal ........................................................................................................................................................ 217 37. Linking boats and rock carvings – Hjortspring and the North John Coles ............................................................................................................................................................... 223 38. Aeneas’ Sail: the iconography of seafaring in the central Mediterranean region during the Italian Final Bronze Age Claire Calcagno ..................................................................................................................................................... 226 39. Western European design boat building in Buton (Sulawesi, Indonesia): a “sequence of operations” approach (SOA) Daniel Vermonden .................................................................................................................................................. 234 40. Balagarhi Dingi: An anthropological approach to traditional technology Swarup Bhattacharyya ........................................................................................................................................... 243 F. NEWS FROM THE BALTIC 41. The Roskilde ships Morten Gøthche ...................................................................................................................................................... 252 42. Two double-planked wrecks from Poland Waldemar Ossowski ................................................................................................................................................ 259 43. Mynden. A small Danish frigate of the 18th century Jens Auer ................................................................................................................................................................. 266 44. The wreck of a 16th/17th-century sailing ship near the Hel Peninsula, Poland Tomasz Bednarz ...................................................................................................................................................... 273 G. NEWS FROM AROUND THE WORLD 45. Sewn boat timbers from the medieval Islamic port of Quseir al-Qadim on the Red Sea coast of Egypt Lucy Blue ................................................................................................................................................................ 277 46. A Roman river barge from Sisak (Siscia), Croatia Andrej Gaspari, Miran Erič and Marija Šmalcelj ............................................................................................... 284 47. Contributions of maritime archaeology to the study of an Atlantic port: Bordeaux and its reused boat timbers Patricia Sibella, John Atkin and Béatrice Szepertyski ........................................................................................ 290 48. A Roman barge with an artefactual inventory from De Meern (the Netherlands) André F.L. Van Holk .............................................................................................................................................. 295 49. The Arade 1 shipwreck. A small ship at the mouth of the Arade River, Portugal Filipe Castro ........................................................................................................................................................... 300 50. A Black Sea merchantman Kroum N. Batchvarov ............................................................................................................................................. 306 51. Medieval boats from the port of Olbia, Sardinia, Italy Edoardo Riccardi ................................................................................................................................................... 312 284 Andrej Gaspari, Miran Eri¹ and Marija Šmalcelj 46 Roman river barge from Sisak (Siscia), Croatia Andrej Gaspari, Miran Eri¹ and Marija Šmalcelj Introduction Prehistoric Segestica and the subsequent Roman settlement of Siscia are located on the site of modern Sisak (100 m above sea-level), 57 km south of Zagreb, Croatia. They commanded a geographically strategic position on the intersection of waterways and roads between the Danube and the Adriatic, and between Italy and the Balkans, respectively, but also at a naturally defendable position on the confluence of the Kupa (Colapis) and the Sava (Savus) Rivers. Good waterways undoubtedly played a decisive role in the formation of the prehistoric settlement, as is discernible from the role of Segestica as a river port during Octavian’s siege (Appian, Illyr. 22–24; Šašel Kos 2005: 437–442). Siscia, on the other hand, is rightly supposed to have been the base of the Pannonian fleet (classis Pannonica). The flourishing economy of the city, which received the status of a colony (Colonia Flavia Siscia) during the Flavian period, was also significantly influenced by the waterways since they provided a passage to the rich iron ore deposits in western Bosnia (Loliº 2003: 133–134). The importance of the navigable waterways and the epigraphically attested river port (CIL III 11382) are best illustrated by the remains of loading platforms, a shipwreck, the focus of this paper, and a number of logboats, as well as numerous archaeological finds from the Kupa Riverbed. The latter represents one of the most extensive collections of objects obtained from European rivers. The majority of the several thousand finds, dating from the period between the Copper Age and the High Middle Ages, were initially discovered during dredging or amateur explorations, and only a small number through archaeological research, which took place during the 1980s and 1990s. The remains of three extensive groups of wooden piles from the Early and Late Iron Ages, as well as the Roman period, were detected and partially researched at several places along the eastern bank of the last meander of the River Kupa, in the region of Pogorelec, the site of the prehistoric settlement (Durman 1992). This article discusses the wreck of a Roman river barge discovered in 1985 in the former swimming area of the city of Sisak, in a part of the riverbed long known among the locals as “the Mint”, opposite the city centre. When water levels are low, the tops of wooden piles are visible in an extensive gravel sandbank, an area known to be rich in archaeological remains from the second half of the 19th century onwards (Fig. 46.1). The ship’s remains were documented during a rescue excavation conducted by the Department of Archaeology of the Faculty of Philosophy in Zagreb, headed by “Kovnica/ The Mint” Pogorelec Fig. 46.1. The location of the site in the Kupa Riverbed next to the centre of modern Sisak (Executed by A. Gaspari). 46 Roman river barge from Sisak (Siscia), Croatia 285 Fig. 46.2. The view from the right bank of the Kupa River towards the site (Photo by K. Kiš). Fig. 46.3. The axonometric view of the piles and the remains of the barge (Executed by M. Eri¹). Professor Marija Šmalcelj. The excavation took place from the 9th to the 17th October 1985 (Šariº 1986). They were enabled by the severe droughts of the summer of 1985 that dried out a large section of the Kupa Riverbed (Fig. 46.2), facilitating an almost dry land excavation of the site, including photographic documentation, a geodetic record of the piles, and several plans of the ship’s remains. The latter were, unfortunately, not available during the preparation of this article. During the excavation, the piles and the ship parts were sampled by Aleksander Durman for subsequent dendrochronological and radiocarbon analyses. The excavation proceeded smoothly until unauthorised diggers also took advantage of the low water level, and, over one weekend in the absence of the archaeologists, they ransacked the site and the uncovered part of the ship was completely destroyed. Official excavation was subsequently suspended. There were two large pile concentrations separated by a 2 m wide and 9 m long gap that extend along the direction of the river. The shipwreck discovered a few meters upstream from both groups of piles, might be related to the function of this gap (Fig. 46.3). The excavation showed that the upper parts of the piles were covered by an alluvial layer with pieces of Roman brick, tegulae, as well as large limestone plates and worked stones. Approximately 50 cm underneath the pile tops there was a layer of fine sand with small sized stones mixed with archaeological material that dates, for the most part, from the 1st–4th centuries AD. Beneath was another alluvial layer that contained somewhat larger sized stones and numerous coins, mostly from the second half of the 4th century. Underneath, there was a layer of wooden shavings, interpreted as the possible surface of the foundation layer into which the piles were driven. Excavation did not extend to this layer. However, a coin of Gordianus III (238–244 AD) and a small, excellently preserved wooden chest with a metal lock, were found on top of this context. The natural colouring of the wood indicates that the material in this layer was not exposed to the river flow subsequent to its deposition. However, in contrast, the alluvial layers above were rather mixed. This is demonstrated by fragments of individual ceramic vessels found scattered at different ends of the groups of piles, as well as at different heights, even up to a meter apart (Wiewegh 2001: 92). Dendrochronological and radiocarbon analyses of over 100 samples, revealed episodes of repeated renovation of the wooden structure, with continual reuse of old piles that were gradually replaced by new ones. The precise data derived from the dendrochronological assessment of the boundary between the heartwood and sapwood, allowed us to infer that the piles were installed in series, approximately every seven years, over a period spanning two or three centuries i.e. between the 1st–3rd century (Durman 1992: 120). A good absolute chronological indicator is provided by the coin of Tiberius, found in a shoe of the extracted pile No. 153. The pile structure was only partially researched which The site The site lies on the eastern bank of the Kupa River in a region known as Pogorelec. The excavation covered an area of 20 x 15 m in total and comprised the easternmost part of the sandbank that extended from the right bank sloping gently down to the middle of the 100-m-wide riverbed. An alluvial layer covered the tops of innumerable wooden piles, vertically driven into the gravel bed. More or less densely positioned piles lay at least 2 m beneath the water (around 90 m above sea-level) during normal water levels and covered the entire width of the site (Fig. 46.1). Over 200 piles were documented during the 1985 excavations, made from round (20–45 cm in diameter) and square (10–20 cm in thickness) hewn oak trunks, with varying heights of up to 4 m. An extracted pile showed that the points of at least some of the piles were tipped with iron shoes and numerous pile tops still bore traces of rust patina left by iron pots. In some places there were also remains of wooden horizontal beams found, fastened with iron nails and positioned approximately 30–40 cm underneath the pile tops. 286 Andrej Gaspari, Miran Eri¹ and Marija Šmalcelj Fig. 46.4. The barge, not excavated in its entirety (Photo by K. Kiš). unfortunately hinders any interpretation of its function. However, the hypotheses that it represented the remains of a Roman pile dwelling (Brunšmid 1909: 24) or a bridge, that lead towards the western gate of Siscia (Koš¹eviº 1995: 10), seem less likely. The distribution and the number of piles lead most researchers to interpret the structure as a part of a platform, by way of which the gently sloping bank of the Pogorelec was extended, at an appropriate height towards the landing area, thereby bridging the silted flood area of the right bank (Šašel 1974: 722). Pile dimensions indicate a high load capacity and it is therefore possible that the above mentioned brick and stone building material actually represent the remains of a massive superstructure, most likely workshops or warehouses, that formed part of the port’s infrastructure (Šariº 1986: 28; Wiewegh 2001: 90). The numerous finds of iron clamps, iron nails and wood working tools support this proposal. The excavators believe that the gap between the two pile groups provided the waterpower to drive a metallurgic plant, which is suggested by the large amounts of uniform metal objects, found in different contexts and within different concentrations (for example, by products of fibulae and 600 keys, collective find of 19 axes, found together with coins from the 4th century). The location, however, was given its name by finds of blanks and even a coin die which indicated the possible location of the imperial mint, active in Siscia from the reign of Gallienus to the beginning of the 5th century (Loliº 2003: 144). Geological research and the remains of Roman architecture on the left bank indicate that the navigable part of the river and the wooden construction were both located on the same side of the river as they are today (Durman 1992: 120). Due to considerable differences in water levels that extended up to 10 m, the Kupa River is nowadays navigable only in the port of Sisak. Nevertheless, the presumed elevation of the platform with wooden foundations allows us to suppose that these oscillations were much smaller in the Roman Period. Fig. 46.5. The joints and scarfs of the planks were joined by simple iron clamps (Photo by K. Kiš). The remains of the barge The remains of a large box-shaped vessel with a flat bottom and low sides were documented within a 7 m long and 4 m wide stretch of the riverbed (Fig. 46.4). The excavated remains included a steep side of the vessel and a substantial part of the width of the bottom of the craft with associated planking, floor timbers and frames. They were found positioned so that the bottom timbers jutted out at an oblique angle, and the side planking, on the other hand, was angled against the current. The opposite side of the vessel that protruded above the sediment was not discovered and was probably broken off by the river current. The eastern part of the vessel, that is the part that extended towards the deeper part of the river, was also missing. The western part of the ship, however, still lies underneath the sediment. The barge lay on a layer of fine sand that begins about 50 cm under the pile tops and was covered by a 30–50 cm thick sediment. South of the vessel and partly within it, although not sitting directly on the planks, lay several large unworked stones. It is presumed that they did not represent cargo remains but rather ruins that originate from the topmost alluvial layer. Three wooden piles, driven through the bottom of the shipwreck, indicate that the vessel was probably intentionally sunk. The planks of the vessel are 25 to 40 cm wide and around 6 cm thick. The planks on the inner side of the vessel, between which luting material of an undetermined plant species was preserved, were fastened with numerous iron clamps that were driven in perpendicular or slightly obliquely, across the joints every 4 to 7 cm (Fig. 46.5). 46 Roman river barge from Sisak (Siscia), Croatia The clamps are simple crosspieces with perpendicularly bent ends that were driven up to a centimeter into adjoining planks. They do not vary greatly in size and do not exceed 5 cm in length. The clamps held together floor and side planks, but were also used to secure the scarfs of individual strakes. Identical clamps were found in great numbers throughout the area of the excavation. Only two planks (SIS-3, SIS-6) underwent a xylotomic analysis. They were radially cut out of the same oak trunk. Ten transverse oak timbers and 14 knee-shaped oak frames were uncovered. Exceptionally massive floor timbers, rectangular in section (height 20–25 cm; thickness 10 cm), were positioned 55–60 cm apart, when measured from timber centre. Small, up to 2 cm high, triangular notches were cut in the lower faces of the timbers above the plank joints. The timbers were then fastened to the planks by one or two cylindrical treenails; no iron nails were detected within the vessel remains. The knee-like frames extended from the top of the vessel’s sides and across several floor planks. Naturally curved wood was used and was not always worked to a precise fit prior to installation. There were two knees, set slightly apart, positioned between two transverse timbers. The transition from the bottom to the sides was formed by chine-blocks with a near L-shaped cross-section, whose inner faces were hewn out in a semi-circle. The angle between the sides and the bottom of the vessel was estimated at 110 degrees. Dendrochronological analysis of the planks revealed that they were made of wood contemporaneous with the piles from one of the renovations of the platform, dating most probably between the second half of the 2nd and the first half of the 3rd century AD. The Middle Imperial date of the ship was confirmed by radiocarbon analysis of the rib “K”, performed at Leibniz AMS Laboratory in Kiel (sample KIA22918 K) which indicates the age of 1749 ± 36 BP or 241-264 cal AD at One Sigma Range. A bronze coin of Theodosius (388 AD) was found on one of the floor planks between timbers L and K. However, due to the disturbed stratigraphy any chronological relevance is not reliable. The position of the vessel lying perpendicular to the gap in the wooden construction, the three piles driven through the floor planks and the ruins inside the vessel, all contribute to the opinion that we are dealing with an out-of-service vessel that was intentionally sunk during one of the platform renovations; in its secondary function it then served either to regulate the river flow in the gap or to relieve the pressure of water on the construction. The analysis The Roman barge from Sisak, despite the fact that some of its crucial characteristics, like the overall dimensions and the bow and stern sections, are not known, can nevertheless be compared to the Gallo-Roman barges from the northern provinces of the Empire dated between 1st – 287 3rd centuries AD. It can also be compared with some examples of Roman Age river vessels found in the area between the hinterland of the northern Adriatic and the lower Danube Basin. A geometric building concept, with a flat bottom and steeply inclined or vertical sides, the use of near L-shaped chine strakes, and the manner in which floor timbers and frames were fastened to the planks, all link this vessel to a group of barges from the Rhine and the Rhône region. The relatively small spacing between the floor timbers that corresponds to two Roman feet (dupondius), usual for this type of cargo vessel, indicates a large vessel of the type described above. With the preserved width of the wreck exceeding 4 m, the original length of the ship should have surpassed 25 m (cfr. Bockius 1998: 515). The setting of the planks in parallel strakes with diagonal scarfs, represents an ancient technical solution, probably of Mediterranean origin. This method of hull assemblage occurs on Greek and Roman ships, both of sewn and mortise-and-tenon construction (Steffy 1994: 41, 48, 59 and 65). Floor planks with more or less parallel edges and diagonal scarfs are common also in the barges of Gallo-Roman type (for example the Zwammerdam 2 and 6 wrecks; Bockius 2000: Abb. 4, 6), as well as in river barges of later periods. An exceptional constructional feature of the Sisak barge is the fastening of planks with tightly spaced iron clamps that held the joints between the floor and the side elements. The only known analogy for this type of fastening can be found in the wrecks of two other smaller vessels (possibly one vessel in two parts?) from the Danube near Kušjak in the Iron Gate area, of which the second is dated to the 2nd century AD (Bockius 2003). In the Kušjak 2 wreck, iron clamps were recorded on the inside along some of the joints between the floor and also between the side planks, spaced at 5 to 10 cm. In contrast to the Sisak barge, the sequence of clamps is broken in places and clamps in the Kušjak wreck were also used to secure the cracks. Elsewhere, Roman shipbuilders used typologically similar clamps only sporadically. In the barge from Yverdon-les-Bains they were used to repair cracks (Arnold 1992: 35), while in the Zwammerdam 2 wreck they reinforced the transition from the amidship section to the swim-heads (De Weerd 1988: 103, Figs. 51, 54, 59). In the vessels from Oberstimm, built using mortise-andtenon technology, clamps were used during the construction itself and for repairing cracks (Bockius 2002: 53–54). Similar clamps were also used in medieval and later vessels, exclusively to reinforce the luting and not to fasten the planks (Bockius 2003). Bockius (2003) noticed that the tight distribution of metal clamps on the Kušjak wreck resembles the intervals between the stitch holes in planks of sewn ships, a similarity that is all the more obvious with the barge from Sisak. It seems that by using the metal clamps to fasten the planks they were looking to simplify a complex building process as well as to facilitate the expensive maintenance of sewn vessels. 288 Andrej Gaspari, Miran Eri¹ and Marija Šmalcelj The Sisak barge has another constructional feature in common with ancient ships of sewn construction, namely the limber holes that are cut into the lower parts of the floor timbers above the plank seams. These appear in the same places in wrecks from Mainz 1, Woerden 1, and Zwammerdam 2, while in Zwammerdam 4 and 6 they are cut above the centres of the planks (De Weerd 1988: 295– 300). Both mortise-and-tenon joints and luting appear only sporadically in Gallo-Roman type barges. However, this arrangement was noted in addition to the Sisak and Kušjak vessels, in the vessels from Chalon-sur-Saône and Lyon, where a resin-soaked cloth was used for the purpose (Bockius 2000: 122). Another similarity with the sewn boats of the Roman period exists in the fact that treenails were used, almost exclusively, to fasten the frame to the shell (cfr. Beltrame 2000: 93 – 94). This assumed technological link between the characteristics of the ships from Sisak and Kušjak and those of the Mediterranean shipbuilding construction technique, is further confirmed by floor timbers and pairs of kneeshaped frames that were installed in the same sequence as on Greek and Roman seagoing vessels (Steffy 1994: 49–50, 65 and 67). Conclusion The constructional features of the Sisak and the Kušjak vessels point to a Roman shipbuilding tradition of a probably southeastern European origin, heavily influenced by the technology of sewn ships from the area around the Adriatic Sea. The technology of sewing the entire planking was used along the coast of Central Dalmatia and the northern Adriatic during the late Republican Period and the first centuries of the Empire (Beltrame 2000; Marlier 2002). The early transfer and the subsequent adaptation of this technology from seagoing ships to cargo vessels for inland transport is attested by the approximately 30-metre long barge from Lipe at the Ljubljana Moor, dated to the early 1st century AD. The construction of a prototype of the later Gallo-Roman barges reveals a remarkable detail that is, with the exception of the Greek vessels, practically absent in the Roman sewn vessels with rounded hulls (Beltrame 2000: 93). The detail in question is the cylindrical wooden treenails that were used to hold some of the planks together (Gaspari 1998: 534, Taf. 72: 45, 100) by which the sewing process was facilitated and the longitudinal strength of the vessel increased; the latter being a very important characteristic in river transport. The barge from Sisak is much later in date than the Ljubljana barge and is the only recorded find of a high capacity vessel of the Roman Imperial Period in the mid- and lower Danube regions. Both the similarity of the Sisak barge with the contemporaneous Gallo-Roman barges and the rationalization of the building process by replacing the timeconsuming sewing of the planks with the use of iron clamps, are clearly visible. The need for increased rigidity that probably originated from the increased transport capacities dictated by the Roman military and the economy, is visible also in the massive construction and in the less precise execution of the constructional elements. The vessel from Sisak was undoubtedly intended for the transport of heavy or bulk cargo, perhaps brick or iron from the ore deposits in western Bosnia, which is attested to this day by large quantities of slag. Down the Una, Sana and Japra rivers and further down the Sava and Kupa Rivers, iron was transported as ingots or shapeless pieces to Siscia where the general administration of ferrariarum Delmaticarum et Pannonicarum was stationed (Koš¹eviº 1995: 23 – 24; Loliº 2003: 134). The use of navigable waterways for the shipping of the goods is attested by the find of pieces of bronze statues in the riverbed of the Kupa near Karlovac, which are doubtlessly the remains of a cargo intended for the metallurgic workshops in Siscia (Šariº 1983). References Arnold, B., 1992, Batellerie gallo-romaine sur le lac de Neuchâtel. Éditions du Ruau, Archéologie Neuchâteloise 13. Saint-Blaise. Beltrame, C., 2000, Sutiles naves of Roman Age. New Evidence and Technological Comparisons with Pre-Roman Sewn Boats. In V. J. Litwin (ed), Down the River to the Sea. Proceedings of the Eight International Symposium on Boat and Ship Archaeology Gdansk 1997, 91–96. Polish Maritime Museum. Gdansk. Bockius, R., 1998, Zur rekonstruktion des römisches Plattbodenschiffes aus Woerden. Jahrbuch Römisch-Germanischen Zentralmuseums, Mainz 43.2: 511–533. Bockius, R., 2000, Antike Schwergutfrachter – Zeugnisse römischer Schiffbaus und Gütertransports. In Steinbruch and Bergwerk (eds). Denkmªler Römischer Technikgeschichte zwischen Eifel und Rhein. Vulkanpark-Forschungen 2, 110–134. Mainz. Bockius, R., 2002, Die römerzeitlichen Shiffsfunde von Oberstimm in Bayern. Monographien Römisch-Germanisches Zentralmuseums 30. Mainz. Bockius, R., 2003, A Roman river barge (?) found in the Danube near Prahovo, Serbia. In C. Beltrame (ed), Boats, Ships and Shipyards. Proceedings of the Ninth International Symposium on Boat and Ship Archaeology Venice 2000, 169– 176. Oxbow. Oxford. Brunšmid, J., 1909, Rimski vojni¹ki diplom iz Siska. Vjesnik hrvatskog arheološkog društva n. s. XI: 23–39. Zagreb. Burkowsky, Z., 2000, Pregled zaštitnih arheoloških istraživanja 1990–2000. Gradski muzej Sisak. Sisak. CIL, Corpus Inscriptionum Latinarum. De Weerd, M. D., 1988, Shepen voor Zwammerdam. Academisch Proefschrift. Universiteit van Amsterdam. Amsterdam. Durman, A., 1992, O geostrateškom položaju Siska. Opuscula Archaeologica 16: 117–131. Gaspari, A., 1998, Das Frachtschiff aus Lipe im Moor von Laibach (Ljubljana). Jahrbuch Römisch-Germanischen Zentralmuseums Mainz 45.2: 527–550. Koš¹eviº, R., 1995, Finds and metalwork production. In R. Koš¹eviº and R. Makjaniº (eds), Siscia, Pannonia Superior, 16–42, BAR Int. S621. Oxford. 46 Roman river barge from Sisak (Siscia), Croatia Loliº, T., 2003, Colonia Flavia Siscia. In M. Šašel Kos, P. Scherrer (eds), The Autonomous Towns of Noricum and Pannonia. Situla 41: 131–152. Narodni Muzej Slovenije. Ljubljana. Marlier, S., 2002, La question de la survivance des bateaux cousus de l´Adriatique. In L. Rivet and M. Scillano (eds), Vivre, produire et échanger: Reflets Méditerranéens. Mélanges offerts à Bernard Lou, 21–32. M. Mergoil. Montagnac. Steffy, J. R., 1994, Wooden Ship Building and the Interpretation of Shipwrecks. Texas A & M University Press. College Station, Texas. Šariº, I., 1983, Rimski bron¹ani material izva»en iz rijeke Kupe kod Karlovca. Obavijesti Hrvatskog arheološkog društva 15.2: 17. Zagreb. 289 Šariº, I., 1986, Zaštitno arheološko istraživanje lokaliteta “Kovnica” u Sisku 1985. godine. Obavijesti Hrvatskog arheološkog društva 18.1: 28–29. Zagreb. Šašel, J., 1974, RE suppl., XIV, s. v. Siscia, 702–741. Šašel Kos, M., 2005, Appian and Illyricum. Situla 43. Narodni Muzej Slovenije. Ljubljana. Vrbanoviº, S., 1981, Prilog prouºavanju topografije Siscije. Arheološka istraživanja u Zagrebu i njegovoj okolici. Izdanja Hrvatskog arheološkog društva 6: 187–200. Zagreb. Vukoviº, 1994, Siscija. Vizija rimskoga grada u Panoniji. Sisak. Wiewegh, Z., 2001, Rimska keramika iz Siska s lokaliteta “Kovnica”. Opuscula Archaeologica 25: 89–92. Zagreb.