Developmental origins of mosaic evolution in the avian skull

Speakers: Professor and Research Leader A. Goswami and Postdoc R.N. Felice, Life Science at The Natural History Museum, UK
Host: Professor Guojie Zhang, Section for Ecology and Evolution, BIO-UCPH

Phenotypic integration and modularity are pervasive characteristics of organisms. Interactions among morphological traits, termed phenotypic integration, can be readily identified through quantitative analysis of geometric morphometric data from living and extinct organisms. These analyses demonstrate that anatomical structures are typically formed of smaller regions of semiautonomous, highly-integrated traits, termed modules. These interactions have been hypothesized to reflect genetic, developmental, and functional relationships and to be a fundamental influence on morphological evolution on small to large time scales. Simulations using covariance matrices derived from landmark data for diverse vertebrate taxa confirm that trait integration can influence the trajectory and magnitude of response to selection. At a macroevolutionary scale, high phenotypic integration produces both more and less disparate organisms, and most often the latter, than would be expected under unconstrained evolution, thereby increasing morphological range, but also homoplasy and convergence. However, this effect may not translate simply to evolutionary rates.

Here, we discuss the macroevolutionary consequences of phenotypic integration for cranial evolution in vertebrates, focusing on birds. We use extremely high-dimensional geometric morphometrics to quantify modularity and evolutionary rates across regions of the avian skull with a broad comparative sample of over 350 extant species and over 750 surface sliding semi-landmarks. We demonstrate that bird skull is highly modular and that rate of evolution is negatively correlated with integration, suggesting that high correlations constrain phenotypic evolution. Individual skull regions exhibit mosaic evolution, with heterogenous rates across the avian tree of life. We also show how these patterns are linked to the embryonic development of the skull. Finally, we leverage our detailed 3-D data to reconstruct a hypothesis of the ancestral crown bird, giving new insight into the deep history of the clade.