Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)




Keith L. Dixon


Structure of songs and organization of singing in 133 Fox Sparrows Passerella iliaca were studied during 2 breeding seasons in 3 geographically separate populations in northern Utah and southern Idaho. The structure of songs was analyzed with the aid of an audiospectrograph. The organization of the singing of songs in 56 birds was analyzed by applying Markov chain analyses to the sequences of songs uttered.

Songs were composed of sy lla bl e- types , of which 49 were recognized. Syllabl e- types could occur sing ly or be seria lly repeated within songs, but they were never fractured so that onl y a portion of one wo ul d be in evidence. Songs were categorized into 5 major types (A, B, C, D, and E) on the basis of the uniformity amo ng individuals in the sequences of sy llable-types which were used to form so ngs . Most song-types were easily characterized by a particular sequence of syllable- types used in forming the terminal portions of the songs, but song-type D was most easily characterized by a sequence of syllable-types near the beginning of the song. Although the sequences of syllable-types forming songs were sufficiently distinct so that the songs could be assigned to a particular major song-type, there were consistent variations among individuals in the sequences of syllable-types composing their songs. Such variants were termed song-versions. The variation in the syllable-types composing songs tended to be restricted to the first halves of the songs. About one half of all the individuals recorded sang more than one version of some particular song-type, usually B or C.

Individual birds used a mean number of 8.2 syllable-types in the formation of each song. The mean number of syllable-types used in forming song-types A, B, C, D, and E in 1973 and 1974 were 7.5 -7.7, 8.5-7.6, 7.8-7.8, 9.4-9.4, and 9.0-7.5, respective ly. Few variations were evident among individuals in the mean number of syllable-types or song-types that constituted their repertoires. Differences in the mean number of syllable-types composing similar song-types were al so of little magnitude. Significant differences were evident in the number of syllable-types possessed by individuals having repertories of 2, 3, and 4 songs. Those birds which possessed the largest repertoires of songs exhibited the greatest number of syllable-types. Six color-banded individuals did not change the size or structure of their syllable-type or song-type repertoires during the year or between years.

Singing was organized into discrete bouts in which each song of an individual tended to be presented with equal frequency of occurrence. The ordering of songs within singing bouts occurred in particular sequences, with each song being sung once. After a bird had sung all of its songs once, it would begin the sequence over again. The order in which a bird presented its songs did not change with the passage of time, it was not related to the song-types the bird possessed, and it did not appear to be affected by the sequence of songs being sung by neighboring Fox Sparrows. Markov cha in analysis of the ordering of songs described the sequencing as a first-order Markov chain in all but three birds. A higher order Markov chain was most appropriate for those three birds which were not described by a first-order Markov chain.

Intra- and interpopulation variations in most of the parameters of song which were considered demonstrated little variation within any of the 3 populations between years or among populations in either of the 2 years . Cluster analyses of the geographic distribution of syllabletypes and song-types reiterated that the incidence of syllable-types and song-types tended to be uniform within and among the populations. The most distinctive group of individuals, based upon the presence or absence of syllable-types and song-types, was the northernmost population. The southernmost population of birds tended to demonstrate the most variability in their possession of syllable-types and song-types.

Comparison of the structure of song in Fox Sparrows with other species of the Emberizidae showed that Fox Sparrows' song structuring was not directly analogous to that of any other emberizid, although the structures of Fox Sparrow songs and syllable-types were not sufficiently different that they could not be recognized as belonging to a member of the Emberizidae. The structure of Fox Sparrow song is most similar to that of their nearest relatives, Melospiza, especially M. melodia, whereas song structuring in Fox Sparrows is less similar to that in the species of the genera Zonotrichia and Junco.

The variations which were present in the structure of individuals' songs and the geographic distributions of syllable-types and song-types were considered to reflect geographic variation rather than dialects. It is proposed that Fox Sparrows learn their songs early in life, as does Zonotrichia leucophrys, and that song may encode messages which allow others to recognize the singer's sex, location, marital status, motivation, and species and individual identity. It is suspected that Fox Sparrow songs do not have great capability of conveying the population affiliation of the singer. It is proposed that the various songs of individual Fox Sparrows are of equal valence with respect to intraspecific interactions, and that this suspicion associated with other factors concerning the organization of singi ng in Fox Sparrows indicates the order in which a bird presents its songs is learned early in life and it is retained unaltered.

Included in

Biology Commons