THE TRADITION OF SNAKE-THREAD GLASS IN PANNONIA

: Snake-thread glass vessels were used from the late 2 nd to the mid-3 rd century AD. At least two production cent-res/distribution circles have been identified in the western part of the Roman Empire, which probably owed their existence to Syrian glassmaking artisans migrating to the west. One was located in the Rhine region, the other in Pannonia. Considering the distribution of snake-thread glass fragments known to date from Pannonia, it is striking that 52 of the currently known 112 fragments were found at Intercisa, 45 exemplars came to light at Brigetio and only a few pieces are known from other sites. It is remarkable that in both towns, glass workshops were active during the same period and there is also some evidence that snake-thread glass had been produced in the workshop of Brigetio. abstract patterns, stylised decorations, like on the beakers of the Cologne workshop. 9 These fragments, found in the centre of Tongres, date from 170–180 to 250–260 AD. 10 Some snake-thread glass vessels came to light in the northern Gaulish region, where this decoration type appeared not only on goblets, but also on bottles and plates. 11 The products of the Cologne workshop were of outstanding quality, this being the reason that it was for long believed to have been the centre where this decorative technique was invented. 12 F. Fremersdorf suggested that eastern glassmaking artisans settling in Cologne began producing them. 13 Some fragments decorated with looped serpentiform patterns are known from Britain from the late 2 nd to the early 3 rd century. 14

After the cohors I Aurelia Antoniniana milliaria Hemesenorum, a Syrian archer unit, was transferred to Intercisa, groups of eastern civilians came and settled there, and it seems likely that they had introduced the new glass style with applied trails, the so-called snake-thread glass in Pannonia. Eastern artisans were probably active in Brigetio as well. During the Severan period, there was a substantial oriental immigration to Pannonian towns, as part of the reconstruction programme after the Marcomannic Wars. 23 The newcomers contributed significantly to the economic prosperity of the Danube provinces. 24 In Pannonia, snake-thread beakers can be found primarily at Intercisa and Brigetio. Surveying the finds from other regions, particularly those parts of Pannonia that fall outside present-day Hungary, it would appear that the snake-thread beakers from Slavonia also fall into one of the four types distinguished by Barkóczi. 25 The colourless glass pieces were decorated with bird motifs and ornamental patterns, and occasionally dolphin figures. 26 Based on fragments from the cemetery of Poetovio, they can similarly be dated to the earlier 3 rd century A.D. 27 Fragments from Vindobona also fit into the categories distinguished and described by Barkóczi. 28

SNAKE-THREAD GLASS FRAGMENTS FROM BRIGETIO
Below, I shall discuss the snake-thread glass fragments found in the municipium of .
In terms of archaeological research, the civil town is the least known of the three main urban areas of Brigetio (the civil town, the legionary fortress and the canabae). It is located in a suburban area of Komárom-Szőny and its area is now mostly built up, except for the Vásártér area (the town's market place during the Early Modern Age). 29 The 25 years of research excavations at Komárom/Szőny-Vásártér, in the heart of the civil town of Brigetio, have come to an end. 30 We now have a rough picture of the general time frame and building phases of the town, 31 providing an excellent framework for the systematic assessment of the characteristic finds such as the fragments of so-called "Schlangenfaden Glas".

MAIN CHARACTERISTICS
All fragments come from cylindrical beakers with stem and foot. They have a slightly out-turned rim with fire-rounded edge, they are tall and narrow with solid globular stem and wide foot with fire-rounded edge. The foot was blown separately, the stem is narrow, short and cylindrical, occasionally with a globular knob.
The vessels are generally colourless or greenish. The decorative design consists of the following elements: self-coloured trails below the rim and on the base, looped serpentiform motifs, vertical trails occasionally folded over each other, leaves with stamped chequered and spherical motifs, and tendrils with slashed surface. The trails are either self-coloured or have opaque whitish, yellowish or bluish hues.
A total of seventeen glass fragments were found between 1992 and 2005, all of which were made of goodquality glass. The glass material used for their decoration was identical. Many have a leaf decoration with chequered surface as well as tendril motifs with slashed surface. One such fragment was found during the 2010 excavation season 32 and four more were brought to light in 2012. 33 The colourless fragments are decorated with colourless threads forming tendrils and leaf motifs applied to their surface. The surfaces of the tendrils were slashed and the leaves were stamped with spherical motifs. According to Barkóczi's aforementioned classification system, these fragments can be assigned to his Groups II and III. 34 The high number of fragments clearly attests to the use of snake-thread beakers in the civil town of Brigetio for a short period of time at the beginning of the 3 rd century. 23  In 2006, the Klapka György Museum of Komárom together with the Institute of Archaeological Sciences of Eötvös Loránd University undertook an excavation in the less researched area of the civil town of Brigetio, at 13 Vásártér, Komárom/Szőny. During the excavation of a 153 m² large area, the living quarters of a strip-house with various construction phases and a glass workshop associated with one of these phases were discovered. The building had five construction periods and a secondary glass workshop associated with one of these. The remains of two kilns were found in the workshop (Fig. 3.8). 35 Four pits were uncovered around the remains of two glass kilns in the back yard of a house (Fig. 3.8): one of the pits contained all the glass beads that were found during the excavation -both wasters and intact pieces -, while the other three contained molten glass drops, debris, production waste, raw glass fragments, additives, pieces of glass sticks and glass threads, some of which had tool marks on their surfaces. Fragments of vessels and glass window panes were also found. As fragments of broken glass objects were systematically collected and recycled in the Roman period, the recovered finds were not necessarily produced by the local workshop. (Further studies are needed to determine whether, in addition to the wasters and raw glass materials, a part of the material came from elsewhere or was collected locally for recycling.) It is certain that the beads, most of which were wasters except for eight intact pieces, were produced locally. The material of the raw glass pieces found in the refuse pits is identical to what most of the beads were made of. Moreover, the colour of some small glass lumps is identical to that of the beads' decoration. Thus, one may conclude that they were produced in this workshop. Since the forms, colours, and decorations of the beads found in various graves in Brigetio are similar to these, one may justifiably argue that the workshop's products were primarily intended for the population of the civil town and the canabae. At the same time, the question remains of whether the workshop, which was active for a short time only, was solely engaged in the mass production of beads, or whether it also produced glass vessels and glass window panes as well.
One hundred and twenty-one in part fragmentary and in part intact beads (Fig. 3.7), and several vessel and window pane fragments were recovered from the pits. 36 Some cylindrical beakers with flower and bird motifs of Syrian origin were also brought to light. Different colours appear on their colourless body, ranging from colourless trails to opaque yellow or white and blue ones (Figs 4.9-11). The same white and yellow can be seen on the decoration of the beads. Some other white and yellow coloured glass lumps assumed to be colouring additives were also found. The same colouring additives and the same colours dominate the decoration of both the beads and the cylindrical goblets with serpentiform trails, indicating the production of this type in the workshop (Figs 3.6-7). Thus, there is some evidence that snake-thread glass wares were produced in this workshop. Many fragments of Syrian beakers came to light from the refuse pit and the glass furnace. Over twenty small snake-thread beaker fragments come directly from the workshop, alongside the colouring agents used during their manufacture (the use of these colours can be clearly seen on both the beads and the snake-thread beakers) (Figs 3.6-7).
The glass workshop in Brigetio operated within the borders of the civil town (intra muros) for a brief period. Both glass beads and vessels were made in the two kilns, which could have been a circular melting kiln with an attached lehr.
The site of the glass workshop excavated at 13 Vásártér Square during a small rescue excavation is located immediately beside the main research excavation location of the Vásártér site. The excavation took place prior to the construction of a house. Most of the glass vessel fragments and all the glass beads were found in refuse pits beside the glass kilns. The fragments of snake-thread cylindrical beakers with stem and foot were found in the kiln (SE 58) as well as in the fill of the refuse pits beside it (SE 14, SE 15, SE 16).
The colourless body fragments are decorated with opaque (yellowish), white and turquoise blue threads. 37 The beads have similar decorations. Small white, yellow and blue chunks of colouring agents were found (Fig. 3.7), suggesting that the beakers were produced locally. Another proof for the local production of vessels is that one of the kilns was probably used as a cooling kiln. Had the glass beads contained copper as a colouring agent for red colour, they would have had to be cooled in this kiln. However, since only a few beads of this type were found, the 37 They were made of colourless glass and decorated with colourless, yellow, dark blue and turquoise blue tendril patterns. The surface of the tendrils is slashed. One fragment is decorated with a bird motif whose surface is chequered and has a dark blue-opaque colour.  kiln must have served for producing vessels instead. Thus, one may conclude that the vessels were made in Brigetio too. They are decorated with snake-threads applied to the surface in a particularly fine manner, and they constitute a group that is not fully identical with that of other beakers found in the main excavation area, given that mostly colourless glass was used for their decoration.
It seems prudent to explore the origins and distribution of snake-thread beakers. It seems to me that the Pannonian material, including the finds from the main Vásártér site in Brigetio/Szőny and the rescue excavation at 13 Vásártér Square, represents the group identified by Barag. They can be dated to the beginning of the 3 rd century AD; they are of eastern origin and their artistic quality is without doubt quite outstanding. Whether the style evolved in Syria (as it very likely did) calls for further studies. It is certainly different from the "western" Cologne style.
It is particularly interesting that in Pannonia, snake-thread beakers have principally been found in settlement contexts (examples are known from Brigetio and Intercisa), and it would appear that they were not deposited in burials. There is evidence for glass workshops operating in both towns. The Pannonian finds are mostly cylindrical beakers with stem and foot, only three globular bottles are known (from Intercisa, Savaria and Bakonyszentlászló) (Fig. 1.3). Considering the distribution of snake-thread glass fragments known to date from Pannonia, it is striking that 52 of the currently known 112 fragments were found at Intercisa (Figs1.2-3, Figs 4-5), 45 pieces came to light at Brigetio (Figs 4.9-12, Figs 5.13-27, Figs 6.28-45, Figs 7.46-47, Figs 8.48-60) and only a few pieces are known from other sites (three each from Aquincum and Gorsium, two each from Carnuntum and Aquincum, and one whit each from Vindobona, Matrica and Majs alongside an unprovenanced piece) (Fig. 1.1). A new find must here be mentioned from Savaria, where 110 graves of the northern cemetery were excavated by P. Kiss in 2006. 38 The investigated cemetery section was used from the later 1 st century to the end of 2 nd century AD. Among the grave goods of one of the cremation burials roofed with tiles was a glass bottle with applied base ring and snakethread decoration. The colourless globular flask is decorated with self-coloured slender spiral tendrils ending in leaves and the figure of a bird. The surface of the leaves and the bird is decorated with chequered impressions. The other grave goods of Grave 38 were a square bottle, two other globular flasks, a square, indented flask, a pear-shaped jar and two conical unguent bottles. 39 This seems to be particularly interesting and calls for further studies because no snake-thread glass vessels have yet been found in Pannonian burials and its date is too early.
At Intercisa, five glass kilns as well as 220 kilograms of waste (raw glass, molten glass and semi-finished products) were uncovered by Zs. Visy. This workshop was active until the 260s AD. 40 After the Roman cohors I Aurelia Antoniniana milliaria Hemesenorum was transferred to Intercisa, a high number of civil settlers probably arrived in several successive waves from the recruitment area. On the testimony of the grave inscriptions and the personal names, a large number of people of eastern origin settled in the canabae of Intercisa, to whom we may attribute the introduction of glassmaking in Pannonia. As we have seen, this was the very period when the first snake-thread beakers appeared in Brigetio and when the production of similar beakers in the Brigetio workshop started. Thus, one may tentatively conclude that Intercisa and Brigetio were the major production centres of snakethread beakers in Pannonia, and that the activity of these workshops was connected to glassmaking artisans migrating here from the east. The Pannonian finds form a coherent assemblage 41 representing a short time period of 25 to 30 years, either after the Marcomannic Wars or, more likely, between 230 and 260, during the reign of Alexander Severus, which saw the arrival and settlement of a new wave of eastern immigrants. The glass trails are often decorated with spherical and beehive-like patterns, stamped on their surface, a practice that is attested in Syria, but does not seem to have reached Cologne. In sum, these decorative patterns and techniques were most probably brought to Pannonia by the eastern glassmaking artisans active in the Brigetio and Intercisa workshops at the beginning of the 3 rd century. Based on the currently known fragments, we may assert that the discontinuous patterns of upwardpointing leaves with veins is known only at Intercisa, where the widest range of both monochrome and polychrome patterns were also employed: bird-shaped and ornamental motifs, spiral trails with smooth surface, tendrils and  41 Taking into account only the Pannoninan decoration types and their possible combinations; cf. BarKóCzi 1981. The style of the beakers found in Pannonia clearly differs from that of the finds from Cologne. The Cologne workshop is probably earlier and can be associated with glassmaking artisans, who had settled there during the time of Marcus Aurelius. While the Pannonian finds are mostly beakers with stem and foot, applied decorations were used on various types of vessels in Cologne. The shape of the leaves and the structure of the motifs are different on the Pannonian finds, whereas on the Cologne finds, the surfaces of the leaves were not impressed with chequered or beehive patterns, and the slashed decoration as well as large-leaved tendrils -typical for the Pannonian ones -are also missing. leaves with chequered or hemispherical design, and tendrils with slashed surface. A smaller variety of patterns was observed at Brigetio, where the following patterns are not attested: discontinuous leaves the chequered design of leaves and the monochrome bird pattern. Ornamental motifs with smooth surfaces in both monochrome and polychrome, tendrils with slashed surfaces and leaves with hemispherical patterns are attested in very high numbers. The tooled decoration of the surface is mainly characteristic of colourless glass trails, although not exclusively. One interesting question is whether monochrome and polychrome wares were produced simultaneously or in different periods. In Brigetio, polychrome decoration is common in the glass workshop, while monochrome variants dominate the finds from the Vásártér site. It should be noted that both polychrome and colourless trails can be observed on certain fragments, indicating that these are unlikely to date from separate periods. horizontal opaque turquoise trail below the rim (Fig. 5.18) 37. Brigetio/Vásártér 13: Inv. no. 2006.V13.218.4, fragment of a cylindrical beaker, colourless; decoration: opaque yellow curved trail (Fig. 5.24) 38. Brigetio/Vásártér 13: Inv. no. 2006.V13.218.5, fragment of a cylindrical beaker, colourless; decoration: self-coloured trails with slashed surface and L-shaped trail (Fig. 5.14) 39. Brigetio/Vásártér 13: Inv. no. 2006.V13.135.58, fragment of a cylindrical beaker, colourless; decoration: part of an opaque dark blue bird figure with chequered design (Fig. 4.9) 40. Brigetio/Vásártér 13: Inv. no. 2006.V13.185.1, fragment of a cylindrical beaker with stem and foot, colourless; decoration: opaque yellow leaf motif with curved end and slashed surface (Fig. 4.11, Fig. 6