Avian wingtip shape reconsidered: wingtip shape indices and morphological adaptations to migration

In this paper we review the existing methods of quantifying avian wingtip shape, and compare their efficiency at detecting morphological adaptations to migration. We use multivariate methods to derive two novel measures of avian wingtip shape, pointedness C-2 and convexity C-3, based on measurements of primary feather lengths. Size-constrained components analysis, a modified form of principal components analysis, is used to ensure that the measures are independent of isometric size, and have a consistent interpretation in terms of the geometric shape of the wingtip. Our measures of pointedness and convexity can be calculated easily for both live birds and museum skins, and can be applied to any ecomorphological or functional analysis of avian wingtip shape. This approach circumvents many of the interpretational problems associated with previous wingtip shape indices that are often based on less accurate wing measurements. To test the suggested interpretations of previously published wingtip shape indices, we use a comparative interspecific analysis to determine the interrelations of our new shape measures with these published indices, and with aerodynamic parameters which have known functional significance. Published indices do not always measure the quantities that they are claimed to do, and are beset with awkward terminological inconsistencies. We assess the efficacy and utility of these wingtip shape measures with regard to predicted and well-known morphological adaptations in the wings for migration. Once phylogenetic bias and extraneous ecological factors are controlled, the majority of published wingtip shape indices are unable to detect morphological differences between migratory and nonmigratory species. Our measures of pointedness and convexity confirm that migrants have wingtips that are relatively more pointed and more convex; they also have wings of relatively larger aspect ratio. The biomechanical implications of these adaptations for different flight behaviours are discussed. An appendix discusses some of the statistical problems involved in the analysis of size and shape, and introduces the size-constrained components analysis (SCCA) method, which is applicable to any study of morphological variation.