AVIS-IBIS

Birds of Indian Subcontinent

Morphological sexing of passerines: not valid over larger geographical scales

Publication Type:Journal Article
Year of Publication:2010
Authors:Ellrich, H, Salewski, V, Fiedler, W
Journal:Journal of Ornithology
Volume:151
Issue:2
Date Published:2010
ISBN Number:2193-7192
Keywords:Acrocephalidae, Acrocephalus, Acrocephalus scirpaceus, Denmark, Emberiza, Emberiza godlewskii, Emberiza schoeniclus, Emberizidae, Erithacus, Erithacus rubecula, Germany, Muscicapidae, Notiocichla scirpacea, Phylloscopidae, Phylloscopus, Phylloscopus trochilus, Schoeniclus, Schoeniclus schoeniclus, Sollicitus scirpaceus, Sylvia, Sylvia borin, Sylviidae, Trochilus, United Kingdom
Abstract:Sex determination of birds is important for many ecological studies but is often difficult in species with monomorphic plumage. Morphology often provides a possibility for sex determination, but the characters need to be verified. We tested whether five passerine species can be sexed according to standard morphological measurements applying a forward logistic regression with sex determined by molecular analysis as the dependent variable. Furthermore, we tested whether the results can be used on a larger geographic scale by applying morphological sexing methods gained by similar studies from other regions to our data set. Of the five species of this study only Garden Warblers Sylvia borin could not be sexed morphologically. In the Robin Erithacus rubecula , 87.2% of all individuals were sexed correctly. For Reed Warblers Acrocephalus scirpaceus , Willow Warblers Phylloscopus trochilus and Reed Buntings Emberiza schoeniclus , the respective values were 77.6, 89.4 and 86.4%. When the logistic regression functions from similar studies on Robins and Reed Buntings in Denmark and Scotland were applied to the birds from south-western Germany, they performed less well compared to the original dataset of these studies and compared to the logistic regression function of our own study. The same was the case for Willow Warblers when a wing length criterion used in Great Britain was applied to the birds of our study. These discrepancies may have several explanations: (1) the models are optimised for the dataset from which they were extracted, (2) inter-ringer variation in measurements, (3) the use of different age cohorts, (4) different morphology due to different habitat availability around the study site, or, most likely, (5) different morphology due to different migratory behaviour. We recommend that morphological sex differentiation methods similar to this study (1) be only used population specific, (2) only with one age cohort and (3) to adjust the extracted equations from time to time.
URL:http://dx.doi.org/10.1007/s10336-009-0478-z
Scratchpads developed and conceived by (alphabetical): Ed Baker, Katherine Bouton Alice Heaton Dimitris Koureas, Laurence Livermore, Dave Roberts, Simon Rycroft, Ben Scott, Vince Smith