Song Similarity in Populations of Fox Sparrows: A Rejection of Naugler's and Smith's Conclusions

Abstract

Naugler and Smith (1991) used data on song in Fox Sparrows (Passerella iliaca) to test hypotheses about the evolution of the structure of song in insular avian populations. Their conclusions are incorrect for a host of reasons. The authors appear to have misused or ignored information within the references they cite and they appear to have relied on second-party assessments (Miller 1982) of data within these references. It will become clear that the review process for this manuscript failed in its assessment of the manuscript's accuracy and its contribution to ornithology. Naugler and Smith presented two hypotheses concerning possible effects of insularity on song complexity: (1) The song complexity hypothesis predicts song in insular populations may become more complex than mainland populations due to weaker selective pressures resulting from reduced interspecific constraints (competitive release ?). (2) The founder effect hypothesis predicts song complexity may be less in insular than mainland populations if a develops during the founding of the population. A bottleneck would result in an incomplete sampling of mainland song complexity, thus, exposing future generations to a depauperate species specific auditory field from which to model song. Following the procedures for measuring song complexity presented by Miller (1982), Naugler and Smith generated statistics for four measures of song complexity from their study population of migratory Fox Sparrows on Bon Portage Island, Nova Scotia. These were: (1) size of song repertoire (number of structurally distinct song units sung by an individual), (2) number of syllables/song, (3) number of syllable types/song, and (4) syllable diversity [number of different syllables (syllable types) in a song/total number of syllables in a song]. They then tested the two hypotheses by comparing their statistics with what they assumed to be similar statistics they or Miller (1982) generated from two other non-isolated Fox Sparrow populations reported on by Martin (1977, 1979) and Blacquiere (1979). In brief, these comparisons led them to conclude that song complexity was less in the Bon Portage population and that this condition resulted from a bottleneck ef-