Patterns in the Morphology and Distribution of Gastropods in Oneida Lake, New York, Detected Using Computer-Generated Null Hypotheses

Individuals of 23 snail species in Oneida Lake, New York, are clumped into small subsets of 141 bottom samples made by F. C. Baker (1918). The sites preferred by the 23 species seem to differ, so that the number of species present in any sample is generally much lower than expected from chance alone. The snail species that do co-occur in individual samples are more morphologically diverse than expected. The morphological similarity of snail species at a site was assessed in two ways, taxonomic relatedness (number of congeneric or confamilial pairs) and direct measurement of shell and radula (mean nearest or furthest morphological neighbor distance). Two null hypotheses were also tested, with the probability of occurrence in a sample weighted or unweighted by overall species abundance. The abundance weighted null hypothesis was rejected when shell and radula morphology were the criteria for similarity, and the unweighted null hypothesis was rejected when taxonomic relatedness was considered. No relationship was found between the morphological similarity of co-occurring species and the size of the plot, its depth, substrate, distance from shore, or plant species richness.