Abstract:
Chondrichthyans occupy a basal place in vertebrate evolution and offer a relatively unexplored opportunity to study the evolution of vertebrate brains. This thesis examines the brain morphology of elasmobranchs (sharks, skates, and rays) and holocephalans in relation to phylogeny and ecology (Section I), with particular emphasis on the evolution of the cerebellum from a cerebellar-like precursor and the implications this has for its function (Section II). Section I There is significant variation in both brain size and complexity across the chondrichthyan phylogeny. Relative brain size (expressed as encephalization quotients and residuals) and the relative development of six major brain areas (the telencephalon, diencephalon, mesencephalon, cerebellum, cerebellar-like structures, and medulla) were assessed. These techniques were applied to 46 chondrichthyan species from 25 families that represent a range of different lifestyles and occupy a number of habitats. A visual grading index (1-5) was created to quantify the structural complexity of the cerebellum based on length, number, and depth of folds. Although the variation in brain size, morphology, and complexity is due in part to phylogeny, as basal groups have smaller brains, less structural hypertrophy, and lower foliation indices, there was also substantial variation within and across clades that did not track with phylogenetic relationships. Ecological correlations with the relative development of different brain areas as well as the complexity of the cerebellar corpus were supported by cluster analysis and are suggestive of a range of 'cerebrotypes'. These correlations suggest that relative brain development reflects the dimensionality of the environment and/or agile prey capture in addition to phylogeny. Section II Cerebellar-like structures are well understood functionally. If the cerebellum evolved from a cerebellar-like precursor, essential elements of cerebellar-like functionality should have carried through to the cerebellum. These predictions were tested in a pectoral fin lift reflex in sharks and in a human motor learning task. The pectoral fin lift reflex failed to show the predicted learning, whereas the human learning task did reveal the predicted response. These results are discussed and placed within the context of the debate around cerebellar function.