Kovács Tibor - Stanczik Ilona (szerk.): Bronze Age tell settlements of the Great Hungarian Plain I. (Inventaria Praehistorica Hungariae 1; Budapest, 1988)

Márta SZ. MÁTHÉ: Bronze Age tells in the Berettyó valley

long, narrow ridge, but smaller elevations with mostlyBronze Age sherds on their surface have also been noted nearby. A number of Bronze Age tells (Esztár, Vésztő) accumulated over Neolithic debris. Thus the observations made at Herpály are in no way exceptional, since the Bronze Age groups pursuing a similar way of life made good use of the easily défendable "islands", the settlement mounds left by their Late Neolithic predecessors. The internal chronology of the site The first Bronze Age occupation of the Szilhalom mound is documented by the floor found at a depth of 0.67-0.78 m. Before this occupation the mound that had originally accumulated as a Neolithic tell had been a Bronze Age burial site. This became clear when our 8x5 m sounding in 1976 was extended over a further 8x3 m area in 1979. Three graves were found at a depth of 0.92-1 m under the Bronze Age layers. Two of these burials consisted of scattered ashes only, without an urn or any other grave goods. The third grave contained a one-handled jug-shaped urn with black-polished surface and a carinated cover-bowl with inturned thickened rim. The ashes were found partly in the vessels, and partly beside them (PI. 27:5-6). We distinguished three habitation levels above the layer of the graves in 1976, but only one of these could be observed in the trench opened in 1979. A comparison of the field data suggests that the area uncovered in 1979 lay on the northern slope of the mound and thus the layer corresponding to the uppermost layer of the 1976 excavation had been eroded. Level 3 of trench 1 (1976) corresponds to level 1 of trench 2 (1979). The latter was covered with a layer of burnt daub which can be correlated with level 1 of trench 1, thus trench 2 consists only of a floor and a debris layer. No daub fragments were found in level 1, but this area did not reflect the original state of the uppermost layer because of the erosion and various intrusive pits. The material of the thin Bronze Age sequence cannot be divided into chronological phases. Only the occupation period can be separated from the period represented by the graves. The earliest occupation level (level 3) consists of well beaten floor fragments (Fig. 13). As the level was disturbed by pits during the Bronze Age and later periods, it is not possible to reconstruct any kind of house. The only bit of evidence suggesting a house is a row of postholes running in a southwest to northeast direction. The postholes are 0.2 m in diameter and 0.4—0.45 m deep, and are spaced at 0.8—1.4—2 m from each other. They were probably dug for the forked posts supporting the roof construction. A mud-plastered circular hearth, 0.6 m in diameter, was found near one of the postholes. The floor remains uncovered in the western half of trench 1 had belonged to another structure. These floor remains that have survived in different states of preservation suggest a rectangular house. Two small, 0.2 m deep postholes were found at the edge of the floor. These small posts possibly served for strengthening the wall. Remains of two plastered hearths were also uncovered. The floor fragments of this level had probably belonged to two houses or other structures. The corresponding level of trench 2 yielded a layer of burnt daub fragments intermixed with broken strorage vessels. A rimmed circular hearth, partly destroyed by a pit, was also uncovered here. Floor remains corresponding to level 3 in trench 1 were found at a depth of 0.58-0.65 m after the removal of the daub fragments. The floor remains did not form a coherent surface, only a narrow strip connected it with the floor fragments of the feature noted in the western part of trench 1. Level 2 lay at a depth of 0.58—0.60 m, only about 0.1—0.15 m higher than level 3. It yielded a coherent, yellow clay floor covering almost the entire area of trench 1 (Fig 14). The size of the floor could not be determined, partly because it extended beyond the profile walls, and partly because it was cut by pits in the northeastern part of trench 1. The stratigraphical position of a collapsed, ashy hearth (or oven) füled with burnt debris is uncertain. Three parallel rows of deep postholes with a small diameter were uncovered running in a roughly north to south direction and spaced at intervals of 0.8—1—1.4 m. These postholes can be interpreted as multiple rows of forked posts supporting the roof construction of the relatively large houses. The fact that no houses with traces of internal partitioning were found argues strongly in favour of this interpretation which seems appropriate in the case of the row of postholes found in the lowermost layer of trench 1 as well. The uppermost habitation level (level 1) lay at a depth of 0.3-0.37-0.4 m (Fig. 15). It consisted of the disturbed remains of three isolated yellow clay floors under the subhumus layer. The floor extending beyond the southern profile wall and the remains in the southeastern corner may have been related, but due to their poor state of preservation this possibüity can no longer be proven. An abundance of finds was recovered from the various pits dug into the floor, including bone implements, bone and antler cheek pieces, pottery sherds and vessels, grid fragments of portable fireplaces, as well as fire-dogs. At the eastern side of trench I (Fig. 16) the narrow part of a quadrangular house floor was detected; its length was about 3 m. Apart from this cca. 0.8 m wide strip the house fell outside the excavated area. Only the presence of a few postholes indicates the wall structure.

36 Material culture The thickness of the Bronze Age deposits at Szilhalom did not exceed 0.5 m

. The pottery can generally be assigned to the Gyulavarsánd C phase as defined by Bona,

37 even though earlier types and motifs also occur. These could be assigned to phase B or even phase A on typological grounds but their number is insignificant.