Abstract
Convergent evolution is widely viewed as strong evidence for the influence of natural selection on the origin of phenotypic design. However, the emerging evo‐devo synthesis has highlighted other processes that may bias and direct phenotypic evolution in the presence of environmental and genetic variation. Developmental biases on the production of phenotypic variation may channel the evolution of convergent forms by limiting the range of phenotypes produced during ontogeny. Here, we study the evolution and convergence of brachycephalic and dolichocephalic skull shapes among 133 species of Neotropical electric fishes (Gymnotiformes: Teleostei) and identify potential developmental biases on phenotypic evolution. We plot the ontogenetic trajectories of neurocranial phenotypes in 17 species and document developmental modularity between the face and braincase regions of the skull. We recover a significant relationship between developmental covariation and relative skull length and a significant relationship between developmental covariation and ontogenetic disparity. We demonstrate that modularity and integration bias the production of phenotypes along the brachycephalic and dolichocephalic skull axis and contribute to multiple, independent evolutionary transformations to highly brachycephalic and dolichocephalic skull morphologies.
Highlights
Convergent evolution is the independent phylogenetic origin of a similar form or function in different taxa and is often viewed as strong evidence for the influence of natural selection on molding organismal phenotypes (Futuyma, 1998; Gallant et al, 2014)
We study the interface between developmental modularity and integration and the consequences of each on patterns of neurocranial shape diversity and variation using two-d imensional geometric morphometrics
We find that the early-burst model does not provide the best fit to our data and instead find that the OU model provides the best fit to our data, further suggesting that our use of principal components (PC) do not appreciably bias our interpretation of convergent evolution in relative skull length (Table S3)
Summary
Convergent evolution is the independent phylogenetic origin of a similar form or function in different taxa and is often viewed as strong evidence for the influence of natural selection on molding organismal phenotypes (Futuyma, 1998; Gallant et al, 2014). It is likely that the disruption of these signaling molecules from the forebrain is a plastic response to a mutation of one or more genes within the large network of genetic interactions that govern skull development This pleiotropic network is expected to have a large mutational target size, such that a mutation in any of several candidate genes would result in a similar truncated response (Boell, 2013; Houle, 1998). This plastic response would bias the phenotypic variation toward the production of brachycephalic skulls. We hypothesize that developmental biases may have contributed to widespread homoplasy in relative skull length among extant gymnotiform species (Sadleir & Makovicky, 2008; Sanger et al, 2012; Wroe & Milne, 2007)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
