A major factor in the evolutionary success of the elasmobranchs (sharks, skates and rays) has been the development of a sophisticated battery of sense organs and large brains. The non-visual senses have received a great deal of attention from biologists, often at the detriment of vision. Thus this thesis aimed to use a neuroethological approach, to investigate both the 'importance' of vision in relation to other senses in elasmobranchs, using sensory brain morphology (Section I), and adaptations for spatial and colour vision (Section II).
The comparative study of brain morphology in 29 species of elasmobranch revealed a range of sensory strategies. Vision appears more important in sharks than in rays, and enlarged optic tecta were more prevalent in active reef-associated and pelagic sharks, as opposed to benthic species and scavengers. In rays, olfaction was more important, although some species relied more heavily on vision and the octavolateralis senses. Generally, well-developed octavolateralis brain areas were associated with enlarged optic tecta. Analysis of brain mass to body mass relationships revealed the carcharhinid and sphymid sharks, and the dasyatid rays, possess the largest brains. The telencephalon and cerebellum were well-developed in those species with larger brains, but there were no correlations between the relative development of the sensory brain areas and brain size, or the relative development of the telencephalon or the cerebellum. A similar approach revealed substantial changes in the relative size of the sensory brain areas in juvenile and adult elasmobranchs. In particular, the relative size of the optic tectum declined from juveniles to adults, whereas the size of the olfactory bulbs increased, suggesting that vision is relatively more important than olfaction to juveniles, and vice versa in adults. This apparent shift in sensory orientation with development may be correlated with shifts in habitat and diet. Finally, a comparison of gross brain morphology in large pelagic sharks and teleosts showed that open-ocean sharks appear to have evolved different sensory strategies than teleosts. In sharks, the olfactory bulbs and octavolateralis areas are particularly well-developed, whereas in teleosts the optic tectum is the largest sensory brain area, suggesting that pelagic teleosts rely more heavily on vision than sharks inhabiting the same environment. In the sharks two integration centres, the telencephalon and the corpus cerebellum were much larger and better developed.
Visual optics in 12 elasmobranchs were investigated using cryo-sectioned eyes. In sharks, the eyes appeared symmetrical about the centrally placed optical axis, whereas the rays exhibited a vertical ocular asymmetry, and a dorsally placed optical axis. A range of lens shapes was exhibited, with the equatorial lens diameters ranging from 1% less than to 21% greater than the axial diameter. Matthiessen's ratio (the distance from the centre of the lens to the retina and the lens radius) ranged from 2.34:1 to 3,33:1 with average ratios for sharks and rays of 2.80:1 and 2.72:1, respectively. Despite these ratios being higher than that frequently quoted for teleosts (2.55:1), the differences in spatial resolving power, calculated using ganglion cell spacing and Matthiessen's ratio (to estimate focal length), were negligible (<1 cycle deg-1). No obvious correlations were found among variations in lens shape and size, Matthiessen's ratio and behavioural ecology, although the influence of the contractile iris and consequent variation in pupil size and shape (present in elasmobranchs but not teleosts) was not considered, and is an important direction for fixture research.
The analysis of Niss1-stained retinal wholemounts revealed interspecific differences in the position of specialisations for acute vision in eight species. A dorsal horizontal streak containing multiple areae appears to be a common adaptation in benthic, relatively immobile or inactive elasmobranchs. In more active species, the streak (and the areae within it) becomes more central, and in some species, less elongate. Alopias superciliosus is unusual, having a ventral streak containing central areae, which is an adaptation for sampling the upper visual field. Estimates of spatial resolving power ranged from 2.00 cycles deg-1 to 10.30 cycles deg-1 These differences reflect variation in habitat and feeding strategies, with slow moving and/or nocturnal or benthic species (including benthically orientated yet active species) having a lower spatial resolving power ( 2 - 5 cycles d e g -1),than active, arrhythmic open water species with higher resolving powers (6 - 10.30 cycles deg-1). Retrograde labelling from the optic nerve using biotinylated dextran amine (BDA) was not successful in determining the overall proportion of amacrine cells to ganglion cells in Chiloscyllium punctatum. However, comparison between the total number of cells in a 20mm2 area in the central retina (an unspecialised region) in a BDA-labelled retina and a cresyl violet stained retina revealed that 76% of the cells in this area were non-ganglion cells, which is in accordance with previous work.
Using microspectrophotometry, the ray Rhinobatos typus was found to possess two cone visual pigments, with wavelength of maximum absorbance (λmax) values at 476nm and 564nm, and one rod pigment, with a λmax at 504nm. The shark Hemiscyllium ocellatum, possessed a rod and a cone visual pigment with identical λmax values (500nm). This is the first microspectrophotometric evidence of two spectrally distinct cone visual pigments in an elasmobranch, and suggests that some elasmobranchs have the potential for colour vision. Work has started to identify and sequence the opsin proteins from a range of elasmobranchs. Preliminary results based on partial sequences from two species of ray, R. typus and Dasyatis kuhlii, indicate that both Rh1 (Rod photoreceptor) and LWS/MWS Long/medium wavelength sensitive opsins are present, while an Rhl opsin has been identified in the shark Carcharhinus amblyrhynchos.