Sexing Wrens Troglodytes troglodytes indigenus using morphological measurements and discriminant analysis

Abstract

Within the breeding population of a deciduous woodland, male Wrens averaged slightly larger than females in wing length, foot length, head plus bill length and body mass. First-year and older birds were similar for three of these morphometric variables, but the first-year group were shorter-winged than the older birds (by 3%). Despite this, wing length still proved to be a useful discriminator of sex when the data were pooled for all ages (74% of males exceeding the maximum female value of 49.8 mm). A discriminant function utilizing wing length and head plus bill length correctly sexed 96% of individuals in the sample from which it was derived (n = 49 males, 36 females). This function was converted into a predictive equation for the probability of an unknown case being male, and a plot of wing length against head plus bill length along with probability contours which enabled the sexing of unknown cases without the need for calculations. Measurements of the morphological variables investigated were found to be highly repeatable for a single worker, with the exception of body mass which differed substantially between the sexes but which was only moderately repeatable. The methods provided should prove valuable for sexing British mainland Troglodytes troglodytes indigenus, and probably also Western European T. t. troglodytes, but should not be applied to the larger, offshore island races. © 1998 British Trust for Ornithology.