LAUSR

In morphological traits, variation within species is generally considered to be lower than variation among species, although this assumption is rarely tested. This is particularly important in fields like palaeontology, where it is common to use a single individual as representative of a species due to the rarity of fossils. Here, we investigated intraspecific variation in the cochleae of harbour porpoises (Phocoena phocoena). Interspecific variation of cochlear morphology is well characterised among odontocetes (toothed whales) because of the importance of the structure in echolocation, but generally these studies use only a single cochlea to represent each species. In this study we compare variation within the cochleae of 18 specimens of P. phocoena with variations in cochlear morphology across 51 other odontocete species. Using both 3D landmark and linear measurement data, we performed Generalised Procrustes and principal component analyses to quantify shape variation. We then quantified intraspecific variation in our sample of P. phocoena by estimating disparity and the coefficient of variation for our 3D and linear data respectively. Finally, to determine whether intraspecific variation may confound the results of studies of interspecific variation, we used multivariate and univariate analyses of variance to test whether variation within the specimens of P. phocoena was significantly lower than that across odontocetes. We found low levels of intraspecific variation in the cochleae of P. phocoena, and that cochlear shape within P. phocoena was significantly less variable than across odontocetes. Although future studies should attempt to use multiple cochleae for every species, our results suggest that using just one cochlea for each species should not strongly influence the conclusions of comparative studies if our results are consistent across Cetacea.