A multi-peak performance landscape for scale biting in an adaptive radiation of pupfishes

The physical interactions between organisms and their environment ultimately shape diversification rates, but the contributions of biomechanics to evolutionary divergence are frequently overlooked. Here, we estimated a performance landscape for biting in an adaptive radiation of Cyprinodon pupfishes, including scale-biting and molluscivore specialists, and compared performance peaks with previous estimates of the fitness landscape in this system. We used high-speed video to film feeding strikes on gelatin cubes by scale eater, molluscivore, generalist and hybrid pupfishes and measured bite dimensions. We then measured five kinematic variables from 227 strikes using the SLEAP machine-learning model. We found a complex performance landscape with two distinct peaks best predicted gel-biting performance, corresponding to a significant non-linear interaction between peak gape and peak jaw protrusion. Only scale eaters and their hybrids were able to perform strikes within the highest performance peak, characterized by larger peak gapes and greater jaw protrusion. A performance valley separated this peak from a lower performance peak accessible to all species, characterized by smaller peak gapes and less jaw protrusion. However, most individuals exhibited substantial variation in strike kinematics and species could not be reliably distinguished by their strikes, indicating many-to-many mapping of morphology to performance. The two performance peaks observed in the lab were partially consistent with estimates of a two-peak fitness landscape measured in the wild, with the exception of the new performance peak for scale eaters. We thus reveal a new bimodal non-linear biomechanical model that connects morphology to performance to fitness in a sympatric radiation of trophic niche specialists.