THE EVOLUTIONARY DEVELOPMENT OF CROWN GALLOANSERAE SKULLS: TRACKING CHANGES IN GALLIFORM AND ANSERIFORM CRANIAL DEVELOPMENT

One of the most remarkable features of the animal kingdom is its diversity of forms in it, which has allowed animals to thrive in a wide range of niches throughout their evolutionary history. Understanding the origins of disparity remains one of the major goals of evolutionary biology. Of the variety of approaches that can be used in understanding the origins of adult cranial disparity, the evolutionary developmental approach is a relatively unexplored one. The evolutionary developmental approach prioritises the role of development as the process that produces organismal form, which includes the dichotomy of constraint and plasticity. To unpack how developmental constraint and plasticity interact to produce adult cranial disparity, I have focused on the development of Galloanserae skulls. Galloanserae is a clade of birds that includes Galliformes (such as chickens and quails), Anseriformes (such as ducks and geese), and their close relatives. Galloanserae are distinctive in their very close relatedness and very distinct skull shapes. Most studies of galloanseran cranial disparity have focused on early craniofacial morphogenetic differences and overlook subsequent divergence in skull formation and growth. Such oversight hinders an understanding of the relative role of embryonic and post-hatching divergence in producing distinct adult skull morphologies along with the constraints influencing this divergence. To fill that gap, I compared skull formation and growth in representative species of Galloanserae from initial skull ossification to the adult stage. I began by constructing robust morphology-based staging criteria for nearhatching embryos of two species of anseriforms, mallard duck (Anas platyrhynchos) and swan goose (Anser cygnoides), because skull ossification extends into these stages and therefore these criteria are necessary for robust comparisons. Subsequently, I qualitatively compared the ossification of each element of the skull of chicken (Gallus gallus), quail (Coturnix japonica), mallard (Anas platyrhynchos), and swan goose (Anser cygnoides). For the comparison to be meaningful, I constructed a skull-based staging system to discretise the skull formation process, based on shared aspects of galloanseran ossification sequence. The skull staging system yielded five stages, which I describe in detail in Chapter 2. Moreover, I examine some of the predictions of von Baerian divergence, noting the relative timing of the formation of clade- and species-specific features. These comparisons revealed that the relative order of osteological feature formation seems to be, at least in part, independent of how phylogenetically widespread the feature is; indeed, most shared osteological features among galloanserans seem to form at post-hatching stages. Alternatively, it seems that the order of element ossification, and the direction of ossification within each element are relatively constrained, and that these may dictate the relative timing of feature formation. These findings led me to propose a new model of evolutionary developmental divergence, which I dub the braiding hourglass model. Subsequently, I compared the post-hatching development of 30 representative species of Galloanserae, with some stem galloanseran fossils, by comparing the geometry of hatchling and adult crania belonging to these species. The comparison revealed the ontogeny of most crown Anseriformes diverges from crown Galliformes, with an intermediate ontogeny in the enigmatic anseriform clade Anhimidae (the sister group to all other anseriforms, represented by Chauna). Moreover, I found that anseriform cranial shape and ontogeny are more variable relative to galliforms, indicating that anseriform ontogeny is more plastic. Furthermore, I found that adult anseriform crania are peramorphic relative to galliforms. Subsequently, I examined the role of heterochrony in the production of cranial disparity between Galliformes and Anseriformes. I found that allometric trajectories of both clades overlap in rate and direction, but anseriform ontogeny is longer and has an earlier onset relative to galliforms. Finally, with the incorporation of the stem galloanseran fossil Asteriornis and the stem anseriform fossil Presbyornis, I found that the cranial morphology of extant galliforms is generally more plesiomorphic than that of anseriforms. This implies that the distinctive aspects of anseriform cranial ontogeny may have arisen from a galliform-like ancestral ontogeny.