Recently, several issues have been raised concerning the use of comparative or interspecific data with standard statistical analyses. Trait values measured in phylogenetically related groups of organisms are not statistically independent of each other, and must be corrected for degree of relatedness. In addition, there is still much debate concerning which type of evolutionary relationships (and thus, evolutionary statistics) are of greatest interest to comparative biologists. Although several solutions to these problems have been proposed, little is known about how the proposed solutions relate to one another. All of the proposed methods require that the phylogeny of the organisms and an estimate of the expected amount of change for each trait be available. As it is often difficult to obtain accurate estimates of expected amounts of character change, several techniques have been proposed to infer this information using independent estimates of time or by using the comparative data themselves. Again, little is known about the relative abilities of the various techniques. During this study the P.I. will use a combination of mathematical and computer simulation techniques to compare the available methods for both analyzing comparative data and estimating expected amounts of change in terms of both their biological and statistical properties.
