In the area of the Roman bath, part of which houses today the archaeological museum, we excavated an area to the north of the museum, ca. 90 m2 in surface area. We conducted this excavation so that we recover evidence on the dating of the Roman bath, possible later phases and its architecture. The area to which we concentrated was the only suitable for excavation, as it was free of modern interventions, and clearly defined architecturally. Part of the Roman bath, and more specifically of its southwestern side, had been excavated by A. Orlandos in 1935 in order to be restored and serve as an archaeological museum. In the context of these excavations a substantial fill ca. 1.60 m thick had been removed, in which a number of sculptures were found that Orlandos set on display in the new museum. The presence of hypocausts around the three halls that had been dug, as betrayed by the characteristic vaults and the results of the geophysical prospection, and the vertical rectangular recesses visible on the inner side of the walls which meant to receive the clay pipes (tubuli) which carried the hot air from the hypocast to the halls, suggest that these were the hot and warm rooms of the bath complex (caldaria and tepidaria). On the basis of the intensive surface survey that we conducted in 2000s and the reconstruction of the city-grid we estimate that the bath complex covered a total area of ca. 3.400 m2.
|
|
Trench 20, measuring 9 x 10 m, was designed so that it includes the entire space defined by Wall 6002 (to the north), Wall 6004 (to the east), Walls 6005 and 6028 (to the south) and Wall 6003 (to the west). From these walls, the north wall, the northern part of the west wall and the northern end of the east wall were visible prior to excavation, standing up to a height of 2 m above ground level. The line of the south wall was known thanks to the geophysical prospection. We divided the trench in two sub-trenches, the western one (Trench 20.1) and the eastern one (Trench 20.2), initially with the same dimensions (9 x 5 m), but after the discovery of Wall 6012 to a depth of just 10 cm, the boundary between the two sub-trenches was redefined so that it falls along the line of Wall 6012. As a result, Trench 20.1 is wider by ca. 1.5 m than Trench 20.2.
In Trench 20.1, after removing topsoil (Contexts 6000 and 6009), Wall 6019 was revealed, of E-W orientation, which separated the trench in two sections, the northern and the southern one. The wall has a doorway at its middle, 1 m wide, with stone jambs and a built threshold (Context 6039). Wall 6019 bonds with Wall 6012 to the east, both being 0.45 m wide. They feature the same masonry, consisting of bricks, rubble and reused stones set in mortar. These two walls are built against the outer walls of the trench, Contexts 6002 and 6003, which are clearly earlier in date. The north wall of the trench (Context 6002) is also 0.45 m wide and has a masonry of layers of bricks, rubble and mortar and reused ashlars, occasionally of very large size, placed in the lower part of the wall. The west wall of the trench (Context 6003) is wider, 0.60 m, and has a neater masonry in opus mixtum, with alternating courses of bricks (opus testaceum) and small stones bonded with mortar (irregular opus vittatum). Excavation in Trench 20.1 up to a depth of 1.7 m brought to light an industrial installation of Late Roman times as well as remains of a building earlier than the Roman Bath.
|
|
At the northern room of trench, under Context 6018 in which a coin was also found, we encountered a deposit 0.30 m of average thickness, filled with lime inclusions (Context 6021), particularly at the NW and SE corners of the room. From Context 6021 also comes a coin which is under conservation. By removing these contexts, the various components of the industrial installations started to appear: a stone well head placed upside down (Context 6010), a Doric column drum cut on its upper part (Context 6008), a circular marble base, probably from a table, placed upside down (Context 6110), and two constructions at the NW and SE corner (Context 6023 and 6022) made of parallel rows of tiles placed as pantiles. At the western half of the room a floor is preserved made of bricks (Context 6051), each measuring 0.31 x 0.31 m, i.e. pedales. The inverted wellhead (Context 6010), of conical shape, has a diameter of 0.94 m on its upper part (which was its original base) and 0.80 m on its lower part (which was its rim), and a height of 0.75 m. On its outer surface it has two vertical orthogonal cuttings, diametrically opposite to each other, that would have originally served the placement of a wooden frame supporting the lifting mechanism. The construction at the SE corner (Context 6022), measuring 1.8 x 1.4 m and 0.20 m in height from the floor of the room, consists of two rows of three Laconian pantiles and corresponding cover tiles which, though, are fragmentarily preserved (Fig. 5). The construction likely extended to the north with one more row of three tiles, the imprint of which we were able to discern. The construction at the NW corner of the room (Context 6023), measuring 1.25 x 1.15 m, consists of two rows of cover tiles placed with their concave side upwards. This construction is not level but slopes to the east so that its eastern end meets the floor of the room. The column drum with the Doric fluting (Context 6008), sits on the floor, just 0.80 m to the south of the tile construction described above. It is 0.53 m in diameter and 0.50 m in height. The space between Context 6023 and the column drum was covered by a solid deposit of lime (Context 6024) with striations on its surface running parallel to the rows of tiles of Context 6023. Deposits rich in ash and burnt olive pits (Contexts 6021, 6022, 6049, 6064, 6065) covered most of the room. In addition, two of these deposits (Contexts 6064, 6065), clearly defined, offered evidence of intense firing, which suggest that they are fire pits. Based on pottery these deposits date to the 4th-5th century CE, a period to which we should assign the industrial installation.
|
|
|
|
In the southern room of Trench 20.1, excavation of the upper layers (Contexts 6014, 6027) contained relatively little pottery, of the late 4th and first half of the 5th century CE, but in Context 6027 we found many slugs and lead fragments together with charcoal, burnt olive pits and 4 coins, three Sikyonian issues of which one dates from the 4th and two from the 3rd century BCE, and a coin of the emperor Honorius (early 5th cent. CE. Distinct layers of ash were also encountered at lower levels (Contexts 6033, 6039, 6048, 6050, 6075, 6078) and must relate to the small, double kiln that we discovered at the NE corner of the room (Context 6035), or also to smaller fire pits as the ones discovered in the northern room. From the two adjacent kilns, the eastern one, 0.80 m (Length) x 0.30 m (Width) is better preserved. It is tripartite on the N-S axis, with the entrance to the south and the combustion chamber to the north. Besides the abundant ash in the layer within and around the double kiln (Context 6034) chips of bluish granite and a number of glass wasters were found as well as a fragment of a glass panel. These finds suggest the production of glass in this space during late-Roman times. From the ash deposits (Contexts 6033, 6048, 6050) we also recovered 4 coins which are under conservation. The continuation of the excavation here at lower levels (Context 6070) exposed a distinct deposit rich in lime, measuring ca. 1.5 x 1 m, directly south of the threshold 6037 and to the west of the double kiln. Directly west of this deposit a large fragment of mortar, ca. 0.45 x 0.50 m, is preserved – perhaps part of a mortared floor, and further to the west, at the NW corner of the room, a pit 0.8 m in diameter, with traces of ceramic walls (Context 6111).
On the southern side of the room we exposed the stone threshold (Context 6087) of Wall 6005, 1.1 m in original width and on the eastern continuation of Wall 6005 a cutting 1.28 m long, likely meant to accommodate a second entrance. Wall 6005, 0.60 m wide and of a max. height of 1.27 m, is built with bricks (opus testaceum) and belongs to the original phase of the Bath complex as it bonds with the western (Context 6003) and the eastern (Context 6004) wall of Trench 20. Based on ceramics, Context 6070 dates to the 3rd century CE. Two coins were found in the same context which are still under conservation. By removing Context 6070 two intersecting walls were revealed to the north of Wall 6005, oriented E-W and N-S. They are 0.84 m wide, and are made of two rows of stone blocks. The top surface of these walls is 0.35 m lower than the threshold level of Wall 6005, and one of the two continues further to the south, under the southern wall of Trench 20. Manifestly both walls belong to a building earlier than the Roman bath. By continuing digging between the south wall of the room (Context 6005) and the earlier wall running to its north (Context 6094), we exposed the foundation level of Wall 6005 as well as part of a Hellenistic pithos placed upside down (Context 6123) with its southern side going under the foundation of the Roman wall. Unfortunately, the little diagnostic pottery found in these layers, of the first half of the 2nd century CE (Context 6121) and of the second half of the 3rd century CE (Context 6100) is not enough to establish a secure terminus post quem for the construction of the Roman wall.
It is obvious that the western side of Trench 20 was reused in late Roman times as an industrial space that involved firing at high temperature, lime, a basin, constructions with rows of tiles placed as pantiles, a stone well head placed upside down, a column drum and a circular marble base. On the basis of the glass wasters an activity that we can reconstruct with a fair degree of certainty is glass-making, whereas the lime deposits suggest the presence of a lime kiln in the surrounding area.
|
In Trench 20.2 (Fig. 1), removal of top layers (Contexts 6001, 6006) disclosed the trace of the eastern (Context 6004), southern (Context 6028) and western (Context 6012) walls of the trench as well as two distinct deposits, on the northern (Context 6016) and southern (Context 6017) sides of the trench. These deposits were clearly richer in finds (ceramics, tile fragments, bones, metal objects etc.) than the upper ones, dating to the Late Roman period. In addition, from Context 6017 comes a coin of the emperor Valentinian II (375-392 CE). By removing these two deposits we came upon a layer of large tiles of various sizes that covered almost the entire surface of the trench (Context 6030). Digging through this layer of the late 4th-late 5th century CE, we revealed cross-wall 6040 as well as the remains of a fire-pit, ca. 0.76 x 0.60 m, roughly at the center-west of the trench (Context 6046), with concentration of ash, charcoal, burnt olive pits, glass fragments and pottery. Three coins were found in this context, an issue of the emperor Julian (355-361 CE), an issue of the emperor Theodosius II (401-450 CE), and a Sikyonian issue of the first half of the 2nd century BCE. Wall 6040, 0.37 m wide and 0.35 m of preserved height, is built with irregular stones, tiles and mortar. It clearly postdates the eastern wall of the trench (Context 6004) but may continue westwards under Wall 6012 which appears to cut through it.
In lower layers (Contexts 6040, 6042, 6043, 6044, 6045), and more specifically under Contexts 6044 and 6045, we located part of a floor consisting of packed soil, which covered the southern half of the trench, and corresponds with the level of the threshold preserved on the west side of Wall 6040. In addition, two coins come from these contexts that await their conservation. We continued digging north of Wall 6040 by removing a very large fill of browning red soft earth, 1.3 m high, full of pottery, tile fragments and pieces of mortar (Contexts 6060, 6062, 6063, 6079, 6103, 6105). From Context 6079 come 9 whole vessels and 11 coins, the majority of which could not be identified because of their high corrosion. Among these coins a 5th century BCE silver coin of Sikyon stands out. The removal of the fill revealed the northern and eastern boundary of a cistern, with its floor and side-walls covered with thick, hydraulic mortar. Three pillars are preserved in situ, two parallel ones on the E-W axis and one further to the south. To the west and to the south, the cistern extends under the later Walls 6012 and 6040. We discovered the southern border of the cistern by digging to the south of Wall 6040, below the level of one more Late-Roman floor (Context 6073). In the layers that we removed from here (Contexts 6067, 6071, 6072, 6074, 6076, 6077) and date from the 3rd century CE, we found fallen stones of large size and two coins. The southern wall of the cistern (Context 6081), 0.78 m wide, is built with large stones of irregular size without mortar (Fig. 28). The wall continues toward the west at least as far as Wall 6012, and must belong to an earlier building, probably of the Hellenistic period. The continuation of the excavation in the space between Walls 6041 and 6081 brought to light the southern part of the cistern. In the thick deposit 6082, which is the continuation of the cistern’s fill with abundant pottery and other discarded material, we found two more pillars in situ and 10 coins. The northern edge of the two pillars, which are parallel on the E-W axis, goes under Wall 6040.
|
|
On the N-S axis the cistern is 4.45 m long and its disclosed width (on the E-W axis) is 2.07 m. This is clearly less than its original width, since the cistern extends westwards under Wall 6012 and under the later levels of Trench 20.1. At its NE corner the cistern has a built staircase (Context 6061), 0.75 m wide and 1 m high, with 4 steps, each 0.25 m high and 0.20-0.25 m deep. The staircase, the side-walls and the floor of the cistern are covered with thick hydraulic mortar (opus signinum). In order to avoid leakage problems, the corners of the cistern are all beveled. Five stone pillars of square shape, all but one 0.44 m of a side and of maximum preserved height of 1 m, supported the cover of the cistern which must have been wooden and flat. Two are parallel on the north side (Contexts 6115 and 6116), two more are parallel on the south side (Contexts 6112 and 6113) and one stands at the center (Context 6114). Based on the height of the staircase, the maximum preserved height of the pillars and the height of the mortar on the side-walls we estimate that the height of the cistern must not have exceeded 1.3 m. How the cistern was supplied with water is not clear. One possibility is that it collected rainwater from the roof the building, and the other that it was supplied by an aqueduct, but no outlet has been observed on the exposed walls of the cistern. The second basic question is whether the cistern was part of the Roman bath complex or not. Its north wall is part of the north wall of Trench 20 (Context 6002) but its eastern wall (Context 6062) adjoins the eastern wall of the trench (Context 6004). The reason for which they have not used the eastern wall of the trench as the eastern boundary of the cistern is not clear yet. One possible explanation could be that the cistern is earlier than the Roman bath, all the more since the purpose of an underground cistern in a Roman bath is not easily understood. Usually the cisterns of Roman baths are higher than the bathing installations so that they can channel the water under enough pressure. However, in our case the structural relation of the walls of the cistern to the walls of the Roman bath favor the probability the cistern to have been installed after the Roman bath.
The material that we recovered from the thick fill of the cistern is one of the richest deposits of our excavations at Sikyon. We recorded in total 14,443 sherds, weighing 414.15 kg, that are distributed as follows:
Fine pottery: 1536 — 17,19kg
Coarse pottery: 9557 — 353,04kg
Cooking pottery: 3350 — 43,92kg
The ceramic material is not all of the same period but represents four chronological phases. The overwhelming majority dates from the Late Hellenistic period and is almost identical to the material found in the filling of the well of the West Stoa. The difference is that the cistern had material from three more periods: a) of the middle of the 1st century CE with characteristic examples the local and imported thin-walled cups, the Italian sigillata and certain coarse vessels, b) of the first half of the 2nd century CE, which includes Eastern sigillata B types (e.g. Type 80 and Type 58), and c) pottery of the 3rd century CE with characteristic examples of pedestal kraters. The material from these three phases was by no means separated stratigraphically, in fact a lot of the late Hellenistic material came from the upper layers of the fill. This compels us to assume that the latest artifacts, i.e. those of the 3rd century CE, offer a terminus ante quem for the abandonment and subsequent filling of the cistern with material coming from various contexts of the surrounding area, hence its chronological differentiation.
|
To the south of Wall 6081, we continued the excavation beyond the south wall of the trench (Context 6028) until we reached the foundation level of Walls 6028 and 6004 in order to find evidence on the date of these walls that we consider to belong to the first phase of the bathing complex. The eastern wall (Context 6004) bonds both with the northern (Context 6002) and the southern (Context 6028) wall of the trench. It has a width of 0.58 m, a preserved height of 1.20 m and is built in opus mixtum, namely with opus testaceum and opus vittatum, just like the western wall of Trench 20. The higher level, which is built in opus vittatum, is preserved only at the northern section of the wall. The south wall of Trench 20.2 (Context 6028) is of similar width and construction but preserves only its lower part, in opus testaceum, to a height of 1.16 m. Excavation down to the foundation level of these two walls, which consists of a layer of hard mortar (Contexts 6080, 6084, 6089), yielded few diagnostic sherds, mostly of the first half of the 2nd century CE as well as one coin. In conclusion, the excavation to the south and the southeast sides of Trench 20 down to the level of the foundation of Walls 6005 and 6004 produced no definite evidence regarding the date of these walls that belong to the first phase of the Roman bath. However, the little dated material that we do have from these contexts points to a date in the first half of the 2nd century CE.
The geophysical prospection in the area of the Bath used a Ground Penetrating Radar GSSI SIR-3000 with an antenna of 400MHz. The goal of the prospection was to map subsurface features of the bathing complex. A number of measurements were taken on a grid basis along parallel transects of 0.50 m apart. This year’s measurements supplement the ones taken in 2023, when four grids had been surveyed as well as many independent transects. This year three new grids were investigated (nos. 5-7), and measurements along independent transects were taken, mostly in the field to the east of the fenced site of the museum where trees prohibited the implementation of a grid. In all grids, linear features, oriented N-S and E-W, have been identified, and must belong to walls of the Roman bath. Some of these walls align with standing or visible walls of the complex. A major question that we were hoping to answer with the geophysical prospection is the eastern limit of the Bath complex, as the western, northern and southern limits are more or less known. Grid no. 7 and the independent transects measured further to the east were proven helpful in this regard, as they have established the presence of parts of the Bath in this area, although the eastern limit of the bath remains wanting.
