Tempo of evolution in a Neogene bryozoan: are trends in single morphologic characters misleading?

Abstract

The extent to which perceived patterns of evolution are affected by the use of single characters versus overall morphology or measured versus counted and coded characters is explored empirically, employing multiple-character data from closely spaced sequential populations of the Neogene bryozoan Metrarabdotos. In all nine species examined, the pattern of evolution in overall morphology revealed by discriminant analysis is so highly punctuated that changes within species do not vary significantly from zero. Among the 46 single characters in the nine species, a few statistically significant temporal trends do occur, as apparent departures from the static pattern in overall morphology. However, these exceptions are so poorly related to the morphology that distinguishes inferred ancestor and descendant species from each other, and are so few in number, that they can be interpreted as chance variation from a general condition of no change within species. There is no difference in behavior between measured characters and counted or coded ones, in part because the distinction between the two types of characters can be artificial in bryozoans and other modular organisms. The results suggest that interpretation of single-character changes, in isolation rather than as statistical samples of the change in overall morphology, should be made with caution.