CAN WE SEE THE FOREST FOR THE TREES? FAUNAL STABILITY AND SPATIO-TEMPORAL SCALE IN THE DEVONIAN HAMILTON GROUP, NEW YORK STATE

Brett and Baird introduced the concept of coordinated stasis about a decade ago after a number of years of field observation of New York Silurian-Devonian sequences. Earlier studies have cast some doubt on the validity of coordinated stasis in the Hamilton Group, but issues of temporal and spatial scale could not be directly addressed with the data available at that time. In this study, we test for taxonomic and ecologic stability within the Hamilton Group at the formation, member, and horizon levels over a broad geographic area. We use a dataset consisting of the distributions of 247 species from 92 samples dispersed through 13 horizons, 10 members and 4 formations within the Hamilton Group. Replicate samples from a single horizon collected over a wide geographic range allow us to address the effects of spatial patchiness while exploring faunal composition at higher scales. Multi-Dimensional Scaling and Analysis of Similarities are used to test for taxonomic stability of biotic assemblages within and between formations, members, and horizons.

As expected from the results of previous studies, variability within horizons and geographic variation within members allow us to reject the null hypothesis that samples are randomly distributed, suggesting that the assemblages within horizons and members are often distinct geographically. However, at the formation level we cannot reject the null hypothesis: despite variation within, formations are statistically indistinguishable based on taxonomic composition. In addition, we tested for ecological stability by coding taxa for their substrate and feeding characteristics and repeating the analysis at various scales. The results are similar to those of the taxonomic tests in that the formation analysis suggests stability, but the finer-scale tests do not. As the original definition of coordinated stasis describes a regional pattern at the scale of formations, we find no basis to reject it. Perhaps the question should not be ‘does coordinated stasis occur?' but rather, ‘at what scale is the pattern apparent?' At high temporal and spatial resolutions, fossil assemblages appear to be variable. It is only at the larger scale of regional formations that stability becomes statistically perceptible, reinforcing the field impression that these faunas are stable for millions of years.