![](/wp-content/uploads/2021/03/abstractheadergraphic2014.gif)
Meeting Abstract
110.4 Tuesday, Jan. 7 08:45 Lower morphological but greater functional diversity in aquatic emydid turtles, relative to terrestrial species STAYTON, CT; Bucknell University tstayton@bucknell.edu
Morphological variation is frequently used as a proxy for functional variation. However, a variety of mechanisms can decouple these two forms of variation. Here I investigate whether greater morphological variation in the shells of terrestrial emydid turtles, relative to aquatic emydids, is also associated with greater functional variation. Morphology was quantified with geometric morphometrics, and function was quantified by finite element analysis (FEA) of shell models generated using the morphological data. Twelve load cases on the carapace and plastron were modeled and average and maximum stresses among all elements were extracted and used as functional data. Variation in morphology and function was quantified as per-taxon disparity for both aquatic and terrestrial species. Aquatic emydids show much higher levels of functional variation across all functional axes. This implies that aquatic taxa occupy a region of phenotypic space characterized by a relatively “steep” functional landscape. Previously, low morphological variation in aquatic emydids was attributed to a trade-off between shell strength (optimized by a tall, rounded shell) and hydrodynamic efficiency (optimized by a flat shell); it was assumed that only a limited number of shell shapes could produce adequate performance in both functions. The large amount of functional variation observed in aquatic taxa in the present study does not contradict this assumption, but it does indicate that trade-offs are not incompatible with functional diversity. Aquatic turtles appear to be diversifying along a morphological limited but functionally rich gradient, with strong shells (as strong as those of terrestrial taxa) on one end and streamlined shells on the other, while maintaining acceptable levels of performance in both functions.