Combined otolith morphology and morphometry for assessing taxonomy and diversity in fossil and extant killifish...

Systematic assignment of fossil otoliths is virtually always based on studies of otolith morphology and subsequent comparisons with otoliths from collections and/or literature. Although this usually represents a practical method, comparisons and subsequent evaluation may be biased by subjective criteria used in the individual descriptions. Quantitative morphometric studies focusing on variations in the otolith morphology of extant fishes have been conducted in fisheries research, mostly based on Fourier shape analysis and related methods. However, with regard to fossil otoliths, these approaches are generally not suitable, mainly due to preservation-related problems. Here we present a new approach for quantifying otolith variation between species and populations of killifish (cyprinodontiforms) in the genera Aphanius Nardo and daggerProlebias Sauvage that can be used with both extant and fossil otoliths. Our new approach includes the definition of 10 variables from linear and angle measurements of an otolith and statistical analyses. Best results were obtained by presorting the otoliths into three groups based on sulcus shape (straight, bent, S-shaped). In this case, canonical discriminant analysis (CDA) with jackknifed cross-validation yielded an overall species classification success of 86-96%. The three groups based on sulcus shape separate according to zoogeographic patterns (i.e., Mediterranean Aphanius, Arabian Aphanius, European daggerProlebias) and probably reflect phylogenetic lineages. Application of CDA to compare otolith variation between populations resulted in an overall classification success (jackknifed) of 33-83%. High levels of variation were observed for Aphanius dispar and daggerProlebias malzi, but not for A. fasciatus and daggerP. weileri. We suggest that otolith variation between populations results predominantly from geographic separation. Combination of qualitative characters (sulcus morphology) with quantitative approaches (otolith morphometry) presents a new approach for obtaining a better understanding of the taxonomy, diversity, and zoogeography of both fossil and extant killifishes. Moreover, the method may also be suitable for assessing taxonomy and diversity in other species-rich groups like the atheriniforms and many perciforms because these groups display otolith Bauplans that are similar to those seen in killifishes.