Measurements of the propagation of sound in a forest have shown that signal degradation is unavoidable but to some degree predictable. Carolina Wrens (Thryothorus ludovici- anus) have a song structure suited for the estimation of distance by a comparison of the relative degradation of the components of the signal. Playback experiments using song recorded at two distances from a singing wren demonstrated that wrens can use cues other than the absolute attentuation of the sound for the estimation of the distance of the singer. The wrens responded to the near-sounding song by attack and to the far-sounding song by countersinging. The ability of the wrens to use the distance information in the song serves the same purpose as the recognition of familiar neighbors: conservation of time and energy used in territorial defense. Received 26 November 1979, accepted 22 August 1980. SEVERAL recent studies have investigated the effects of the acoustics of the en- vironment in constraining the structure of the songs used for long-distance com- munication in passerine birds. Passage through the environment degrades a song by adding amplitude fluctuations and reverberations and selectively attenuating the higher frequencies. Measurements of the propagation of sound in a forest have shown that signal degradation is unavoidable but to some degree predictable (Morton 1970, 1975; Chappuis 1971; Marten and Marler 1977; Marten et al. 1977; Richards 1978; Wiley and Richards 1978; Richards and Wiley 1980). The primary concern of most researchers has been to discover those characteristics of the structure of bird song adapted for transmitting information (primarily species and individual identity) over the greatest distance. In this study I demonstrate another adaptation of birds to song degradation: the use of the predictability of the degradation present in received signals for estimation of the distance of a conspecific signaler by the Car- olina Wren (Thryothorus ludovicianus). It seems likely that individuals of many avian species are not spaced far enough apart to place strong constraints on the structure of their signals for long-range communication. Though the territory of a breeding bird may be 100 m in diameter, the bird moves around the territory while singing, frequently changing the spacing between itself and its neighbors. To discriminate between an invader inside the territory and a neighbor or other bird outside without wasting energy in a physical interaction, an individual would do well to attend to the features in the received signal that are correlated with the distance of the signaler. The overall attenuation of the received signal, however, is probably not the best cue for the distance of the source, because changes in weather would strongly affect atmospheric absorption and scattering from microclimatic heterogeneities (Wiley and Richards 1978). Much more reliable ranging would result from comparing separate features of the received signal, either different frequency bands or different periodicities of intensity in any one band.
[1]
E. Morton,et al.
VOCAL RESPONSE TO PREDATORS IN PAIR-BONDED CAROLINA WRENS
,
1977
.
[2]
D. Borror.
Variation in Carolina Wren Songs
,
1956
.
[3]
W. Rosene.
Nesting of the Carolina wren
,
1954
.
[4]
E. Morton.
Ecological Sources of Selection on Avian Sounds
,
1975,
The American Naturalist.
[5]
A. C. Bent.
Life histories of North American nuthatches, wrens, thrashers and their allies
,
1948
.
[6]
R. H. Wiley,et al.
Recognition of Neighbors' Duets By Stripe-Backed Wrens Campylorhynchus Nuchalis
,
1977
.
[7]
R. H. Wiley,et al.
Reverberations and Amplitude Fluctuations in the Propagation of Sound in a Forest: Implications for Animal Communication
,
1980,
The American Naturalist.
[8]
W. Hays.
Statistics for the social sciences
,
1973
.