Archaeomalacology

Molluscs in former environments of human behaviour

D. Bar-Yosef

• Publisher: Oxbow Books
• Year: 2005
• ISBN:
  9781782979050
• Topic: Social Sciences
• Subtopic: Archaeology
• Index: Social Sciences

Molluscs are the most common invertebrate remains found at archaeological sites, but archaeomalacology (the study of molluscs in archaeological contexts) is a relatively new archaeological discipline and the field of zooarchaeology is seen by many as one mainly focused on the remains of vertebrates. The papers in this volume hope to redress this balance, bringing molluscan studies into mainstream zooarchaeological and archaeological debate, and resulting in a monograph with a truly international flavour.

Europe

9th ICAZ Conference, Durham 2002

Archaeomalacology: Molluscs in former environments of human behaviour

(ed. Daniella E. Bar-Yosef Mayer) pp. 56–62

6. Marine mussel shells – wear is the evidence

Jan Light

Marine mussel shells (Mytilus spp.) are common components of shell assemblages from archaeological sites worldwide and this reflects, amongst other things, their palatability, their wide zone of intertidal distribution with the potential for exploitation during a major interval within the tidal cycle, and their global distribution. At a Romano-British site at Fistral Bay in North Cornwall large numbers of mussel shells were retrieved from midden deposits, with limpets (Patella spp.) and smaller numbers of the dog whelk (Nucella lapillus). Whilst the limpet and dog whelk shells were in an excellent state of preservation, the mussels were flaky and fragmented, and many bore abraded and facetted surfaces around the umbones and on the convex shell exteriors. These wear patterns on the shells were initially interpreted as evidence that the shells had been employed for some utilitarian purpose; for example, for buffing or burnishing processes. Examination of the local mussel colonies in the modern-day environment has revealed the agent of abrasion to be in situ environmental processes. An understanding of the ecology, habitat and provenance of marine mollusc shells is necessary to distinguish between natural processes and the hand of humans when analysing modified shells from archaeological sites.

Introduction

Archaeological sites worldwide yield marine mollusc shells in varying conditions – they may be complete or fragmented, with good preservation of sculpture and ornament, or bearing evidence of abrasion in the open environment. It is also possible that they may be naturally or artificially perforated, or show traces of working by the human hand. In any analytical work carried out on such deposits, a fundamental question to be addressed in their interpretation is whether they were natural deposits unassociated with human intervention, bait and/or food remains, or materials brought to a site for utilitarian, artefactual or decorative purposes.

In Europe and southwest Asia archaeological sites provide widespread evidence of shellfish exploitation. Molluscs were used as a source of animal protein as early as Upper Pleistocene times (Brothwell and Brothwell 1998) and chief amongst the mollusc groups exploited were winkles, whelks and limpets (gastropods) and oysters, cockles, various clams and mussels (bivalves).

This practice continued well into the Holocene. It is also common in Africa and Australia (see Meehan 1982).

On rocky shores, no bivalve is more common than the mussel, and few bivalves are so widely distributed globally. They are euryhaline and have a tolerance for exposed rocks as well as more sheltered shores. They can extend over and consolidate sand and gravel beaches. Soft substrate populations are maintained by larval settlement and metamorphosis in suitable depressions on the substrate or between adult mussels.

At a site overlooking Fistral Bay, North Cornwall (Figure 1) excavations carried out in 1998 by the Cornwall Archaeological Unit (CAU) revealed settlement remains dating primarily to the Romano-British period (2nd – 4th C AD) (Reynolds 1999). A striking feature of the site was the numerous middens and an extensive cultivation soil layer. A total of 21 middens was recorded during the excavation and of these, 14 were excavated and sampled to varying levels of completeness (Reynolds 1999, Appendix 7.1). The shell deposits were dominated by three marine mollusc genera: Patella spp. (limpets), Mytilus spp. (mussels) and Nucella lapillus L., 1758 (dog whelks). Other marine mollusc species were scant and sporadic. The calcareous sandy nature of the overlying soils resulted in an excellent state of preservation of the large quantity of animal remains which were present: bone, molluscs and crab shell (Light 2001). The site is described in detail elsewhere (Light 2003).

Fig. 1. Map indicating the site of the Atlantic Road excavation together with Bronze and Iron Age sites in the surrounding area (after Nowakowski, 1991).

Mussels comprised a significant portion of the middens at the Fistral Bay site. Examination and analysis of the mussels revealed modified morphological features that have the appearance of being made by humans. However, comparison with mussels from Fistral Bay enabled a reevaluation of the midden material to be made in the light of present day environmental processes.

Methodology and data

A general outline of excavation methods and shell analysis techniques is described in Light (2003). The mussel shells (Mytilus spp.) examined for the analysis consisted of valves and umbonal fragments which were selected during processing at CAU in Truro, Cornwall, where the residue had been retained in storage. To obtain an estimate of numbers of mussel individuals present, shells from two midden contexts (Table 1, middens 14 and 18) were weighed in bulk. Twenty samples of 100 valves/umbonal fragments from each midden were weighed to derive mean weight per shell. Five samples of 100 fragments were picked randomly from 4 bags of mussel shell from middens 14 and 18 (40 subsamples), from which mean valve/umbonal fragment weights ranging from 0.6 g to 0.9 g were derived. For each bag, an overall mean fragment weight was established within this range by visual assessment. Mean fragment weight was then used to estimate the total number of valves/ umbones per bag. These data were extrapolated to derive the estimates of total number of valves (and MNI) in the deposits, as shown in Table 1. Although two species of mussel, Mytilus edulis L., 1758 and M. galloprovincialis Lamarck, 1818, were recognised in the Fistral Bay archaeological assemblage, no attempt was made to distinguish between the two species for analytical purposes.

As part of the analysis and interpretation of the marine shells from the Romano-British site at Atlantic Road, the provenance needed to be considered. It is highly probable that the marine shells in the shell midden assemblages were collected locally. With this in mind, a site visit was made to the shore in order to examine the present-day populations of the mussels, together with the limpets and dog whelks, in order to compare them with shells of those species in the archaeological assemblages from Atlantic Road. A particular requirement was to examine the mussel populations to confirm the identification of two mussel species in the samples: Mytilus edulis and M. galloprovincialis. In recent times, the latter species was only first recognised in England by Hepper (1957) from the north coast of Cornwall and Devon and called ‘the Padstow Mussel.’ Subsequent work (Lewis and Seed 1969; Seed 1971) confirmed it as distinct from M. edulis.

Table 1. Marine molluscs from Atlantic Road contexts/samples identified as midden deposits, with extrapolated total contents given as shell numbers.

Totals represent actual and extrapolaed (italicised) shell abundances. These are derived from actual weights and/or numbers recorded during the analysis and multiplied, where appropriate, by a factor based on individual estimates on entire (but not excavated) context contents See the Archive Report, Appendix 7.1 (Reynolds 1999) for weights/volumes and estim...