Song Pattern Variation in the Sage Sparrow (Amphispiza belli): Dialects or Epiphenomena?

Abstract

Abstract I investigated variation in the songs of Sage Sparrows (Amphispiza belli) breeding in the northwestern Great Basin over a 3-yr period. Individuals sang a single song type, which did not change during the day, through the breeding season, or between years. Although different individuals and populations differed in song parameters such as duration, delivery rate, and notes and note types per song, these variations exhibited no overall patterns with respect to geography or habitat. Similarity matrices based upon the sequential arrangement of note types in songs were used to draw comparisons within and between populations. Within-population vocal similarity of individuals varied considerably, from populations in which most neighboring individuals sang similar or identical songs to situations in which most individuals sang quite different songs; in one population, song variation was arrayed in "neighborhoods" of 3-5 individuals sharing similar songs, which differed markedly from those of adjacent clusters of individuals. Song variation between populations was significantly greater than that within populations. Some populations located close to one another shared many song elements and patterns, but other nearby populations were totally different and showed greater similarity with populations located some distance away. The degree of song similarity between populations was neither a simple function of the physical distance between the populations, nor was it related to the distribution of topographic barriers in this region. Populations with greater between-individual dissimilarity in song patterns were also more variable from one year to the next and occurred in habitats of lower vegetation stature and greater patchiness. Generally, however, the patterns of vocal variation within and between populations were unrelated to features of habitat structure, to the densities of Sage Sparrows or of other coexisting species, or to avian community attributes. I suggest that Sage Sparrows may have a simple, single-song repertoire because the song serves simple functions, and selection favoring a more elaborate song may be absent. The variety of patterns of within- and between-population vocal similarity may be associated with differences in population turnover rates, song ontogeny, dominance relationships, and dispersal, all of which require documentation to establish the foundations of vocal variation in this system (and others). The variety of population patterns, however, argues against a simplistic categorization of this species as dialectal or nondialectal. Much of the variation that is expressed may be a consequence of chance events and represents "epiphenomena" that are random with respect to natural selection.