Meeting Abstract
One of the most fundamental species interactions is competition. However, in the fossil record competition cannot be observed or experimentally tested. Instead, paleoecological studies rely on morphological similarity to infer niche overlap. For mammals, similarity in dental morphology is used to argue for overlapping dietary niches and evidence of competition. Here, we test if competitor dentitions are more similar than those of non-competing species. To address this, we collected 535 shape descriptors and ratios of micro-CT scanned lower dentitions of 151 extant rodent species. We compared the differences in the scores of topographic variables used to infer diet, e.g. Dirichlet Normal Energy, Relief Index, and Orientation Patch Count, between 56 competitive species pairs drawn from the literature as well as between species not identified as competitors. We found that competitors are statistically closer in their morphology than non-competitors from the same dietary categories. A subset of the descriptors was then summarized using a principal component analysis. The distance was then determined between each species pair in a principal component morphospace that explained 90% of the variation and on average competitors exhibit significantly closer dental morphology than non-competitors. For 122 species, the morphospace distances were compared to Jaccard Indices (JI) calculated for the overlap in geographic ranges using NatureServe range maps for each species pairing to see if morphological similarity relates to geographic similarity. We found that morphological similarity and JI were poorly correlated suggesting geographic similarity is not well explained by morphological similarity. Our findings support the notion that competitors are more similar morphologically than non-competitors.
