An evaluation of three field techniques for sexing Gould's Petrels (Pterodroma leucoptera) (Procellariidae)

Abstract

Many petrels show no obvious sex-linked dimorphism in plumage or size and consequently many researchers fail to sex the living individuals they study. Several methods of sex discrimination that do not rely on plumage- or obvious size-dimorphism can be used to sex live petrels. The effectiveness of three such techniques was evaluated: body condition at the time of laying, cloacal inspection, and discriminant function analysis (DFA) of external morphometrics. Gould's Petrel (Pterodroma leucoptera leucoptera) was used as the subject species. Sexing of breeding adults on the basis of body condition at laying proved to be highly accurate (100% of birds sexed correctly) but required detailed knowledge of the breeding biology. Following training, cloacal inspection proved to be an accurate (96%) method of determining the sex of breeding adults, but not of chicks. Unlike molecular sexing, the latter two methods of sex discrimination provide immediate knowledge of the sex of individuals in the field. DFA of external morphometrics predicted the sex of adults with an accuracy of 73% and the sex of near-fledged chicks with an accuracy of 66%. However, the probability of correct assignment of sex was low in most cases and, therefore, this is the least useful of the three techniques assessed here.