Functional significance of highly variable colouration in the shore skink (Oligosoma smithi) : a thesis presented in partial fulfilment of the requirements for the degree of Doctor in Philosophy in Ecology at Massey University, Albany, New Zealand
Variation in animal colouration is attributed to several biological functions, a key one being camouflage. Background-matching is a camouflage strategy where prey conceal themselves from predators by resembling their immediate backgrounds. Achieving optimal background-matching can be challenging, particularly in 1) visual backgrounds that form a mosaic caused by spatial variations in habitat characteristics, and 2) varying predator abundance or behaviour. Additionally, crypsis can be affected by alternative and potentially antagonistic functions, such as intraspecific signalling and thermoregulation. This thesis aimed to investigate the selective processes that affect prey colouration for background-matching in a heterogeneous environment. Specifically, I focused on the influence of habitat gradients, predator behaviour, and the potential conflict between camouflage requirements and thermoregulation or intraspecific signalling. Firstly, I conducted a detailed survey on the colour and colour patterns of a wild shore skink population (Oligosoma smithi) within a continuous heterogeneous habitat at Tāwharanui Regional Park. This population’s body colouration showed a significant association with a vegetation gradient, consistent with selection for background-matching. However, field experiments also revealed that predation risk for the more common colour pattern variants was double that of the rarer variant’s regardless of background type, consistent with predictions for apostatic selection (negative frequency-dependent selection). Secondly, I demonstrated that population colouration can respond to a change of habitat. One year after a translocation of shore skinks to an island habitat with a disjunct two-patch background, the population’s colours matched the simple substrate type (bare rocky stones) more than the complex substrate (high vegetation cover on sand). Skinks were darker, less intense in colour, and had lower colour pattern diversity compared to the founder and source populations at Tāwharanui. This study highlighted the potential significance of considering camouflage requirements of a species in human-induced translocations. Thirdly, in an analysis of seasonal effects to camouflage, I found no evidence that background-matching in the Tāwharanui population was compromised by differences in body colours between breeding and non-breeding seasons. This is likely because colours associated with intraspecific signalling (i.e. that exhibited age-dependence and sexual dichromatism) were located in the ventral body regions of skinks that would typically be hidden from predators. Finally, across 17 populations, shore skink colouration showed patterns of spatial variation consistent with thermal melanism (thermoregulation) and island syndrome. Despite the strong correlation of maximum monthly temperature on colours and latitude on colour patterns, I suggest that the significantly darker island populations were caused by a combination of local adaptation (i.e. crypsis) and non-selective forces (e.g., genetic drift). Overall, my thesis provides new insight on how different selection processes maintain dramatic colouration within a species, and marks the first quantitative research on colouration in New Zealand reptiles.