A CHEILOSTOME BRYOZOAN STORY OF FITNESS AND PHENOTYPES ACROSS TWO MILLION YEARS

Natural selection can be quantified and modelled as statistical associations between fitness and phenotypes. Although the fossil record offers unique insights into long-term phenotypic evolution, successfully operationalizing the study of selection on data from the fossil record is challenging. Estimating fitness components using fossil populations allows linking observed patterns of evolution and selective forces in the past, but these are not available for most groups of organisms. Cheilostome bryozoans, a group of calcified colonial marine invertebrates with an exceptional fossil record, offer a unique opportunity to overcome this barrier: either the measure of the density of skeletal brooding structures/ovicells within a colony of genetically identical feeding modules/autozooids) or outcomes of overgrowth competition can be used as proxies for fitness components (i.e. fecundity and partial mortality, respectively).

Using multiple fossil populations of closely related (four species of the genus Microporella) and non-related species (Antarctothoa tongima) of cheilostome bryozoans spanning more than 2 million years of the Pleistocene of the Wanganui Basin in New Zealand, we investigate multivariate signals of trait-fitness associations for three phenotypic traits (autozooid size and shape and ovicell size). Thus, we ask how trait-fitness associations vary across time, among species/lineages, when either fecundity or overgrowth competition are used as a proxy for fitness, and whether they can predict phenotypic evolutionary outcomes. We also test how paleotemperature and its variability contribute to the substantial variation in each of the three quantitative traits that we observe for all species across time intervals.