The reliability of morphometric discriminant functions in determining the sex of Chilean flamingos Phoenicopterus chilensis

Abstract

Monomorphic birds cannot be sexed visually and discriminant functions on the basis of external morphological variations are frequently used. Our objective was to evaluate the reliability of sex classification functions created from structural measurements of Chilean flamingos <i>Phoenicopterus chilensis</i> museum skins for the gender assignment of live birds. Five meas-urements were used to develop four discriminant functions: culmen, bill height and width, tarsus length and middle toe claw. The functions were tested on a sample of live flamingos from a zoo. The best classification for museum flamingos was given by a function using tarsus length, bill width and middle toe claw (97%). However, this function did not give the best classification for the zoo-based flamingos (81%) which had the best sex assignment by a function including measurements of tarsus, culmen and bill height and width (85%). This shows that a function giving good results in the sample from which it originated may not be as good when applied to another group of animals. Our study emphasizes the need for assessing the accuracy of a function by testing it with other methods to ensure its suitability when being applied.