Specialized carnivory is relatively uncommon across mammals, and bats constitute one of the few groups in which this diet has evolved multiple times. While size and morphological adaptations for carnivory have been identified in other taxa, it is unclear what phenotypic traits characterize the relatively recent evolution of carnivory in bats. To address this gap, we apply geometric morphometric and phylogenetic comparative analyses to elucidate which characters are associated with ecological divergence of carnivorous bats from insectivorous ancestors, and if there is morphological convergence among independent origins of carnivory within bats, and with other carnivorous mammals. We find that carnivorous bats are larger and converged to occupy a subset of the insectivorous morphospace, characterized by skull shapes that enhance bite force at relatively wide gapes. Piscivorous bats are morphologically distinct, with cranial shapes that enable high bite force at narrow gapes, which is necessary for processing fish prey. All animal-eating species exhibit positive allometry in rostrum elongation with respect to skull size, which could allow larger bats to take relatively larger prey. The skull shapes of carnivorous bats share similarities with generalized carnivorans, but tend to be more suited for increased bite force production at the expense of gape, when compared with specialized carnivorans.

Tooth classes are an innovation that has contributed to the evolutionary success of mammals. However, our understanding of the mechanisms by which tooth classes diversified remain limited. We use the evolutionary radiation of noctilionoid bats to show how the tooth developmental program evolved during the adaptation to new diet types. Combining morphological, developmental and mathematical modeling approaches, we demonstrate that tooth classes develop through independent developmental cascades that deviate from classical models. We show that the diversification of tooth number and size is driven by jaw growth rate modulation, explaining the rapid gain/loss of teeth in this clade. Finally, we mathematically model the successive appearance of tooth buds, supporting the hypothesis that growth acts as a key driver of the evolution of tooth number and size. Our work reveal how growth, by tinkering with reaction/diffusion processes, drives the diversification of tooth classes and other repeated structure during adaptive radiations.