MORPHOLOGICAL DISPARITY LACKS CLEAR BIOGEOGRAPHIC STRUCTURE: A CASE STUDY USING MODERN COCCOLITHOPHORES (PRYMNESIOPHYCEAE)

To date, there has been limited support for the hypothesis that morphological disparity varies over geographic space in different groups of organisms (Roy et al. 2001; McClain 2005;Neige 2003;Jønsson et al. 2015). Although we might expect to see biogeographic patterns of disparity like those of taxonomic richness, should they have similar underlying forces, it’s possible that environmental variation is too fine grained to see beyond the level of populations and species. I study the morphological disparity of extant coccolithophores– an extreme example of complex mineralized morphology in phytoplankton– to test its relationship to species richness and characterize the nature of disparity across different biomes and latitudes in the modern ocean. I analyze 71 coccolith characters and 41 coccosphere characters measured in >125 species, spanning 13 clades and all extant orders, within a biogeographic context. I find a weak positive relationship between disparity and richness across clades and biomes using both coccolith and coccosphere characters. There is no clear pattern when comparing by latitude. The morphological breadth of families portrayed by principal coordinate axes (first 2 axes explain ~30% of variance in both “coccolithospace” and “coccospherospace”) shows some overlap in “coccolithospace” and considerable overlap in “coccoshperospace”; however canonical variates analysis on these PCO axes separates all clades in both spaces. Latitudinal assemblages also overlap in both spaces, but CVA is able to separate them to some extent. Biomes however have a great deal of overlap in both spaces and can’t be distinguished from one another using CVA. It’s unlikely that this is a result of the near-cosmopolitan extent of species, as excluding shared taxa in biome to biome pair-wise discriminant analysis yields similar results as including shared taxa. It’s possible that there are minute variations in disparity over space, perhaps at the level of populations, but that signal may be lost when working with coarser resolution. Lastly, it’s important to bear in mind that the modern is a single time slice and while– to a first approximation– nearly every morphology is found in every region at present, things may have been different in the past.