THE EVENNESS AND RICHNESS COMPONENTS OF TAXONOMIC DIVERSITY
I propose a new evenness metric, Ess, based on the sum of squared deviations from the mean proportion. The general expression for this metric is given below, where s is the number of taxa and p is the proportion of the i th taxon. This new metric is simple to calculate and simulations show that it significantly outperforms other commonly used evenness metrics in its ability to estimate the population parameter from small samples. From a single sample, it is possible to bracket the population value with approximately 95% confidence (equation shown is maximum estimate). Simulations also demonstrate that sampled richness is more strongly influenced by the evenness of the population than by the richness of the population, particularly when sample size is relatively small. To apply this new metric to paleontological data, I examined the evenness of North American Cambrian-Ordovician benthic marine assemblages. Preliminary data suggest that mean assemblage evenness increased substantially from the Cambrian to the Ordovician. This result has implications for the accuracy of taxonomic diversity estimates during the Ordovician radiation and for ecological changes taking place in benthic communities. While richness has motivated many paleobiological studies, evenness is quantitatively attractive and has promise in furthering our understanding of Phanerozoic biodiversity and ecology.