Morphological Variation in the Shearwater Genus Puffinus.

Abstract

Through the concurrent collection of morphological (2,694 specimens), ecological, geographic and climatic data, this study investigates the relationships between ecology and morphology in the shearwater genus Puffinus. Multivariate methods are used to describe patterns of trait variation in terms of sexual, geographic, specific and subspecific variation. There is no geographic variation in the magnitude of sexual dimorphism, thus indicating that any variation in selective pressures being exerted on male and female Puffinus do not differ significantly over the species' range. Species for which a significant difference is found between the sexes, exhibited low levels of sexual size dimorphism (5-7%), expressed only in bill depth dimensions; male bills are deeper. Bill dimensions exhibit greater phenotypic variation than the wings or feet. This pattern of phenotypic variation is the same between the sexes, indicating that the selective forces responsible for this pattern appear to be similar in both sexes. Only species with widespread distributions exhibit significant geographic variation: all P. assimilis, P. lherminieri and P. pacificus morphometrics exhibit such variation in both sexes, but only wing length in male P. griseus and female P. nativitatis. The roles of climatic (sea surface and air temperatures) and geographical (latitude, longitude and inter-population distance) factors as potential agents of natural selection on the body dimensions of P. assimilis, P. lherminieri and P. pacificus are investigated. Patterns of morphological variation represented by phenotypic distances are unrelated to the geographic proximity of colonies. Puffinus lherminieri and P. pacificus exhibit the trend in body size predicted by Bergmann's rule, however there is no evidence to support Allen's rule. Puffinus species support the premise of multidimensional niche segregation. Both morphological and ecological divergence occurs between sympatric species as a possible isolating mechanisms to reduce competition for resources and thus enable coexistence. The subspecies of P. assimilis and P. lherminieri differ significantly in all morphometrics but, following Amadon's 75% rule, the measurements do not provide enough discriminatory power to correctly classify 75% of individuals to the respective subspecies. Consequently morphometrics alone should not be used to assign subspecific status of individuals in P. assimilis and P. lherminieri. Body size and foraging strategy influence breeding parameters: larger species have comparatively shorter laying periods but longer mean incubation shifts and nestling periods compared with smaller species; inshore feeders have shorter mean incubation periods and nestling periods compared with offshore feeders. The Puffinus genus does not conform to the hypothesis that inshore feeding species begin breeding at an earlier age. While migratory populations of P. assimilis have relatively longer wings than those in non-migratory populations, this trend is not found in P. pacificus. There is no relationship between migratory habit and sexual size dimorphism amongst Puffinus species. Due to the interrelated nature of ecological parameters such as food, climate, distribution, life history and breeding biology, it is unlikely that a single factor is responsible for the observed morphological variation. However food, through its influence on competition, dispersal, growth, fecundity and survival, may play an important role in the relationships found in this study between the morphology and ecology of Puffinus individuals.