Internet Archaeol. 10. Dye.

5.0 Land snail populations in sinkholes

Variability in the distribution of land snail taxa among habitats has implications for interpretation of change over time. The history of land snail populations in sinkholes, to which we now turn, is to a certain extent a collection of individual histories where chance local events might have great effects on the character of small land snail populations in sinkholes. Where local events resulted in extirpation of rare taxa, with no well-established populations nearby, purely local changes might permanently alter the character of the land snail population in a sinkhole through failure of extirpated taxa to recolonise. Alternatively, diversity in snail habitats through the region opens the possibility that regional environmental change might have different effects on land snail faunas from different habitats. Despite the difficulty of interpreting change in this context it is worth examining in detail stratigraphic records of sinkhole sites. The large size of most sinkhole collections and their division into many stratigraphic units yields a record of change that is unparalleled at other site types in the region.

Previous analyses were unable to establish control over the chronology of sediment deposition in sinkholes. Two recent developments make it worthwhile to re-establish an absolute chronology for land snail sequences. First is a model of the marine carbon reservoir (Stuiver et al. 1986; Stuiver and Braziunas 1993) and its local variations (Dye 1994a), which together provide the theoretical and empirical justification for calibration of ¹⁴C dates on sea bird bones. Second is development of statistical techniques for calibration that make it possible to combine archaeological information, in this case the stratigraphic positions of dated samples relative to depositional units defined by Allen (1995), with information provided by ¹⁴C dates (Buck et al. 1991; 1992; 1996). The addition of archaeological information to an analysis generally improves the results of calibration relative to an analysis of ¹⁴C dates alone, and yields a result that is archaeologically interpretable.

Another problem is that land snail sequences have been characterised almost exclusively in terms of changes in the relative proportion of native, extinct taxa in collections. The problem with this analysis is that there is no necessary relationship between a change in the relative proportion of a taxon in an excavated collection and a change in absolute abundance of that taxon in the environment. (Another problem is the unresolved issue of how many taxa listed as native, extinct are, in fact, locally extinct. This issue is critical for an analysis that relies on changes in relative proportion of native, extinct taxa.) This is referred to as the problem of 'closed arrays' (Grayson 1984, 19). The absolute abundance of a taxon whose relative proportion is declining need not be declining; it can be static or even growing. In these latter two instances, all that is required is that the absolute abundance of other taxa in collections increase more quickly, either through an increase in the population of extant taxa or through introduction of new taxa that are able to colonise and exploit unoccupied niches.

Diversity is an alternative measure useful in tracking changes over time in sub-fossil collections. Diversity has two dimensions: richness refers to the number of taxa in the collection; evenness to the distribution of individuals among taxa. Palaeontologists investigate extinction of a taxonomic class by tracking a decline over time in richness of some lower-order class. Typically, extinction of a family of organisms is illustrated by the declining number of species in the family. Evenness can be used to the same effect and is potentially more sensitive to stress in a population than is richness; changes in population structure that don't involve extirpation of a taxon will affect evenness measures but not richness. Here we investigate changes in richness and evenness of native, extinct taxa over time, comparing and contrasting patterns of stasis and change in these measures with changes in relative proportion.