Ray Mariano Chatterji

Convergent evolution can undermine parsimony as a criterion for choosing among phylogenetic hypotheses, resulting in a challenge to find consensus among conflicting trees. This challenge is exemplified in the studies of sea turtle (Pan-Chelonioidea) evolution. Despite the wide interest in this iconic clade, rigorous analyses of character acquisition are currently limited by poorly resolved, largely parsimony-based phylogenies. In particular, there is uncertainty regarding the placement of the extinct family Protostegidae, which has been recovered in several different parts of the tree and even outside Pan-Chelonioidea as a separate radiation of marine turtles. The relationships among the stem of Cheloniidae are also far from a consensus, despite it having a relatively good fossil record. Here we investigate the evolution of sea turtles using a Bayesian tip-dating approach, a technique not previously applied to this group. We also conduct a biogeographic analysis using DEC and DECJ models. We find Protostegidae to be sister to the rest of Pan-Chelonioidea, implying convergent evolution of the sea turtle flipper and other marine adaptations. We find that Chelonioidea probably originated in the North Atlantic, and that the now global distribution of most sea turtle species suggests that the pelagic life phase of extant species is unique to crown Cheloniidae.

Convergence provides strong evidence for adaptation, reflecting similar responses to recurring selection pressures. Waterfowl (order Anseriformes) have repeatedly evolved morphotypes putatively correlated with foraging ecology (e.g., dabbler, grazer, diver), making them an ideal model for testing hypotheses of adaptive evolution. Here, we estimated the most robust time-calibrated phylogenetic tree for the waterfowl to date and quantified the shape of the skull, femur, tibiotarsus, and tarsometatarsus of 118 extant waterfowl species using geometric morphometrics. Multivariate generalized evolutionary models provide strong support for convergent phenotypic evolution across lineages that share dietary characteristics. We recovered a significant relationship between the evolution of diet and foraging behavior and the evolution of skull shape. We found a stronger correlation between the evolution of foraging behavior and the evolution of hindlimb shape, but diet evolution was also significantly related to hindlimb shape. Derived terrestrial and diving lineages consistently showed greater rates of phenotypic evolution than surface swimming and wading lineages. We also find evidence for integration between the hindlimb elements and the skull. These results suggest that the morphological diversity of waterfowl is largely explained by adaptive evolution in response to different foraging ecologies.

Establishing the relationships between organismal phenotypes and their environment is a key component to understanding evolutionary history. Comparative evaluations of extant and extinct species can reveal how lineages have adapted to changing environmental conditions over time. However, the understanding of paleoecologies is predicated on a robust understanding of how modern species have been shaped by adaptation. Waterfowl may present an ideal group to study adaptive evolution as much of their morphology is apparently shaped by their dietary ecology. Here we use a large geometric morphometric dataset of waterfowl combined with random forest, a supervised machine learning algorithm, and linear discriminate analysis, to predict the dietary ecologies of nine extinct waterfowl species. We find that both model types reliably predict ecology for extinct species with well-established diets. Interestingly, we also found that the Hawaiian moa-nalo and the New Zealand Cnemiornis calcitrans likely occupied ecological niches no longer present in modern waterfowl as they were not morphologically or ecologically convergent with modern geese as previously asserted. Our study demonstrates that waterfowl are an excellent model group for the study of adaptive evolution, and underscores the utility of predictive modelling for paleontological studies.

Convergence provides strong evidence for adaptive evolution as it reflects shared adaptive responses to the same selection pressures. The waterfowl (order Anseriformes) are an ideal group in which to study convergent evolution as they have repeatedly evolved morphotypes putatively correlated with diet (i.e., dabbler, grazer, diver). Here, we construct the most robust evolutionary hypothesis to date for waterfowl and reveal widespread morphological convergence across the order. We quantified the shape of the skull and hindlimb elements (femur, tibiotarsus, and tarsometatarsus) of 118 species of extant waterfowl using geometric morphometrics. Multivariate generalized evolutionary models provide strong support for a relationship between dietary ecology and skull shape, and evidence for convergent evolution across lineages that share dietary niches. Foraging behavior better explained the evolution of hindlimb shape, but diet still contributed significantly. We also find preliminary evidence for integration across all three measured hindlimb elements with each other and with the skull. We demonstrate that dietary ecology drives morphological convergence within waterfowl, that this convergent evolution involves multiple integrated skeletal structures, and that morphological changes are associated with shifts in the rate of phenotypic evolution.

Despite only comprising seven species, extant sea turtles (Cheloniidae and Dermochelyidae) display great ecological diversity, with most species inhabiting a unique dietary niche as adults. This adult diversity is remarkable given that all species share the same dietary niche as juveniles. These ontogenetic shifts in diet, as well as a dramatic increase in body size, make sea turtles an excellent group to examine how morphological diversity arises by allometric processes and life habit specialisation. Using three-dimensional geometric morphometrics, we characterise ontogenetic allometry in the skulls of all seven species and evaluate variation in the context of phylogenetic history and diet. Among the sample, the olive ridley (Lepidochelys olivacea) has a seemingly average sea turtle skull shape and generalised diet, whereas the green (Chelonia mydas) and hawksbill (Eretmochelys imbricata) show different extremes of snout shape associated with their modes of food gathering (grazing vs. grasping, respectively). Our ontogenetic findings corroborate previous suggestions that the skull of the leatherback (Dermochelys coriacea) is paedomorphic, having similar skull proportions to hatchlings of other sea turtle species and retaining a hatchling-like diet of relatively soft bodied organisms. The flatback sea turtle (Natator depressus) shows a similar but less extreme pattern. By contrast, the loggerhead sea turtle (Caretta caretta) shows a peramorphic signal associated with increased jaw muscle volumes that allow predation on hard shelled prey. The Kemp’s ridley (Lepidochelys kempii) has a peramorphic skull shape compared to its sister species the olive ridley, and a diet that includes harder prey items such as crabs. We suggest that diet may be a significant factor in driving skull shape differences among species. Although the small number of species limits statistical power, differences among skull shape, size, and diet are consistent with the hypothesis that shifts in allometric trajectory facilitated diversification in skull shape as observed in an increasing number of vertebrate groups.

Chelonioidea (sea turtles) are a group where available morphological evidence for crown-group relationships are incongruent with those established using molecular data. However, morphological surveys of crown-group taxa tend to focus on a recurring subset of the extant species. The Australian flatback sea turtle, Natator depressus, is often excluded from comparisons and it is the most poorly known of the seven extant species of Chelonioidea. Previous descriptions of its skull morphology are limited and conflict. Here we describe three skulls of adult N. depressus and re-examine the phylogenetic relationships according to morphological character data. Using X-ray micro Computed Tomography we describe internal structures of the braincase and identify new phylogenetically informative characters not previously reported. Phylogenetic analysis using a Bayesian approach strongly supports a sister-group relationship between Chelonia mydas and N. depressus, a topology that was not supported by previous analyses of morphological data but one that matches the topology supported by analysis of molecular data. Our results highlight the general need to sample the morphological anatomy of crown-group taxa more thoroughly before concluding that morphological and molecular evidence are incongruous.