Use of Song Repertoires among Marsh Wren Populations

Abstract

-To survey singing behaviors among populations of Marsh Wrens (Cistothorus palustris), we tape-recorded and analyzed songs of 18 individuals from 5 states. Median repertoires of song types were 33 for Illinois (n = 2 birds), 40 for North Carolina (2), 53 for New York (7), 141 for central Washington (1), and 162 for California (6) birds. North Carolina males ranked the lowest or second lowest in most measures of singing complexity, including recurrence numbers, rates, three measures of singing versatility, repertoire size, and percentage performance time. Western birds, in contrast, sang with the greatest complexity in these measures. Among the 5 populations, repertoire size correlated positively with the percentage performance time, the predictability of the next song type in a sequence, and the variability (as measured by coefficient of variation) of the syllable period in the trill of repeated song units. By comparing syllable periods within songs, we found that successive songs in frequently used transitions were more different from one another than expected by chance. Received 7 February 1986, accepted 20 July 1986. THE evolution of song repertoires persists as one of the most fascinating puzzles of vocal communication among birds. Many descriptive studies have added significantly to our understanding of these repertoires. Documenting the existence and use of repertoires among interacting birds has been an important preliminary approach (e.g. Lemon 1965, 1968; Wildenthal 1965; Lein 1978; Payne 1979; Slater 1981). Comparative surveys that relate social systems or ecology to the use of repertoires have yielded intriguing correlations among North American wrens, European Acrocephalus warblers, and European Emberiza sparrows (Kroodsma 1977, Catchpole 1980, Catchpole and McGregor 1985). Complementing these observational approaches is an increasingly elegant series of experimental studies. Removal of males and their replacement with speakers that broadcast repertoires of different sizes has revealed that repertoires may function as effective keep-out signals for territorial males (Krebs et al. 1978, Yasukawa 1981). Song playbacks to territorial or captive birds have revealed some of the complex rules that appear to be used during countersinging (e.g. Verner 1975, Todt 1981, Falls and D'Agincourt 1982, Falls et al. 1982, Wolffgramm and Todt 1982, Kramer et al. 1985, Whitney 1985). In addition, use of female displays as a bioassay for the potency of male song has provided a new technique with potential for determining the intersexual functions of these male song repertoires (Searcy and Marler 1981, 1984; King and West 1983; Catchpole et