Bivalve conjoin analyses: assessing site integrity

Conjoin (also refitting or cross-mending) analyses of stone artefact, ceramic and faunal assemblageshave long been employed to assess the integrity of various archaeological deposits (see, forexample, collected papers in Cziesla et al. 1990; Hofman and Enloe 1992). Only two systematicstudies have been conducted in Australia, both concerning rockshelters in the Central QueenslandHighlands (Richardson 1992, 1996; Stern 1980; see also Leavesley and Allen 1998). No comparablestudies are available for open coastal midden sites despite explicit and implicit reference to this sitetype as stratigraphically problematic (e.g. Gillespie and Polach 1979; Lourandos 1996, 1997; Roberts1991; Stone 1989, 1992, 1995). In this chapter I present results of experiments which (a) establishmethods for effective conjoin analysis of Anadara trapezia (mud ark or Sydney cockle) valvesrecovered from coastal middens; (b) apply the methods to a case study of the A. trapezia assemblageexcavated from the mid-Holocene Seven Mile Creek Mound as part of this study; and (c) assess thereplicability of these methods through a blind test. The chapter also briefly outlines the range ofconjoin analyses undertaken for archaeological purposes and discusses some limitations of thesestudies. This is the first known application of bivalve conjoin analysis to an archaeological depositin Australia and the first known application anywhere of these methods to A. trapezia. The overallaim of the experiments was to evaluate the potential of using bivalve conjoin analyses to assess theintegrity of excavated shell midden deposits on the southern Curtis Coast.

Chapter 5 of 'Coastal Themes: An Archaeology of the Southern Curtis Coast, Queensland'