6.0 The Environmental Evidence

by J. Hamshaw-Thomas and D. Jaques, Environmental Archaeology Unit, University of York

With contributions by J. Carrott, K. Dobney, A. Hall, M. Issitt, C. Johnstone and F. Large

Introduction | Vertebrate remains | Methods | Preservation | Species representation and quantification | The economic base | Carcass representation and butchery | Husbandry and economy | Biometry | Pathology | Animal burials | The bird bones | Discussion

6.1 Introduction

A total of eighty-five ‘environmental' samples (66 GBAs,15 BSs, one SRs and three spot samples, using the classification of Dobney et al. 1992), five boxes of hand-collected molluscs, fourteen boxes of hand-collected animal bone and a single box of human bone were assessed for their bioarchaeological potential by the Environmental Archaeology Unit in 1994 (Carrott et al. 1994).

The results of this assessment showed that plant and invertebrate remains were present in approximately one-third of the 66 samples assessed but, with the exception of one context (1277), in numbers too small to be of interpretative value.

The flot from Sample 49, context 1277, produced at least twenty plant taxa. Weeds predominated, but there were also quite frequent rush (Juncus) seeds and some probable indicators of grassland, though the assemblage was too small to say whether this might have been turf growing near the site or material from hay and/or stable manure.

The insect assemblage from context 1277 was also too small to be of definite interpretative value but, may have indicated a stable manure community in the early stages of formation. There was no dominant beetle group but rather a mixed community of mostly ground-dwelling species.

Following the assessment, no further work was recommended on material from the sediment samples, as it was felt that little additional information would be retrieved. The human remains, two cremations and an inhumation, were recorded during the assessment phase. No further work was considered necessary on this material. Further work was recommended on the molluscs (Carrott et al. 1994, 7). However, this aspect of the analysis was not possible within the financial constraints of the project.

From the assessment, two small groups of vertebrate remains, the first dating to the later 3rd century and the second of early 4th century date, were identified as assemblages worthy of further investigation. The first group related to intense activity alongside the nearby Roman road, whilst the second group was recovered from a series of levelling and dumps which extended from a roadside building and sealed the later 3rd century activity. The remainder of this report is concerned with the analysis of the vertebrate remains from these two groups.

6.2 Vertebrate remains

The following account is based on material excavated from 52 contexts (39 from Trench 1 and 13 from Trench 2). Thirty of these contexts date from the 3rd century, with the remaining 22 dating from the early 4th century. These contexts were selected on the basis of their tight date and the presence of vertebrate material as discussed in the assessment report (Carrot et al. 1994). The 3rd century deposits are from Periods 3 and 4 (context groups 1.6-1.11 and 2.6-2.11), whilst those from the early 4th century are in Period 5 (context groups 1.12 and 2.12).

The total sample consisted of 399 recorded mammalian fragments and eight avian fragments. Most (92%) of the material is from Trench 1. One hundred and sixty-eight fragments (42%) are from Phase 1, and 231 fragments (58%) are from Phase 2. In view of the small sample size and the homogeneous nature of the assemblage, the data from both phases have been combined.

The material analysed for this report was all hand-collected. This will inevitably bias the results in favour of more robust and denser fragments, thereby increasing the representation of larger mammals. Sieving can recover smaller skeletal elements and bones from young individuals, as well as small mammal, fish and bird remains. Small amounts of bone were recovered from some of the samples but, on assessment, were found to contribute little additional information to the material recovered by hand.

Only three samples (11, 12 and 46) from two deposits (1094 and 1255) produced material worthy of comment; these are discussed below. The rest of the sieved material has not been included in this analysis.

6.2.1 Methods

The recording methodology followed the protocol described in detail by Dobney et al. (in preparation). Minor deviations from this protocol were as follows: no measurements were taken on any teeth and the only loose teeth recorded were M3 and Dp4; no horncore measurements were taken; and no distinction was made between sheep and goat bones.

Detailed recording of all identifiable fragments of major domesticates or large wild mammals followed the diagnostic zones scheme outlined by Dobney and Rielly (1988).

Within the assemblage of mammal bones, only the following skeletal elements were recorded: horncore, mandible, M3 and Dp4, scapula, distal humerus, distal radius, distal metacarpal, pelvis, distal femur, distal tibia, distal metacarpal, astragalus, calcaneum and first and third phalanges.

Recording of mandibular tooth wear for caprovids followed Payne (1973), whilst that for cattle and pigs followed Grant (1982). Cattle mandibles were assigned to the general age categories outlined by O'Connor (1988). Age assessments based on the fusion of long-bone epiphyses followed Silver (1969).

Measurements (unless otherwise specified) followed von den Driesch (1976). Additional measurements, not detailed by von den Driesch, followed those outlined by Davis (1992) and Dobney et al. (in preparation).

6.2.2 Preservation

The preservation of the material did not show any significant variations between contexts, periods or excavation areas. The assemblage could be described as well preserved, although there was variability in both the angularity (nature of the broken surfaces) and the colour of the fragments. Black staining was noted on some bone surfaces. A small number of fragments appeared highly abraded compared to the rest, and this may indicate the presence of residual material in some contexts.

In general, there was limited direct modification of the bone surfaces; burning was rare and only 13% of the bones showed evidence of butchery. A higher proportion of the fragments (20%) showed signs of dog gnawing. No evidence of cat or rodent gnawing was observed. A single caprovid astragalus appeared to have passed through the digestive tract of, presumably, a dog. As noted in the initial assessment (Carrott et al. 1994), these are all characteristics of material which has been discarded but not immediately buried.


© Internet Archaeology URL:
Last updated: Tue Nov 28 2000