been quantitatively studied, but inspection of published sonograms suggests
that individuals may vary from one another in frequency and temporal
characteristics of all song components, even when the basic structure of
notes is quite similar, a possible example of by-product distinctiveness.
Songs sung by different males within a song group show striking similarity
in the frequency contour from the song’s middle portion, and this feature
differs from other song groups on a lek (Wiley 1971). Visual comparison of
sonograms indicates that song at the lek studied by Snow (1968) differs
from that at the lek studied by Wiley (1971). Both Snow and Wiley suggest
that vocal learning leads to the song sharing they observed. To date, no
experimental tests of the learning hypothesis have been conducted on little
hermits.
Dialect variation has also been found in the song of Anna’s hummingbird, C. anna (Mirsky 1976; Baptista and Schuchmann 1990).The differences
between island and mainland populations include the structure of syllables
and temporal phrasing of the song (Mirsky 1976). Mainland males show
no response to playback of island songs, whereas they respond strongly to
playback of mainland songs (Mirsky 1976), indicating that dialects are
recognized as distinct by the birds. Visual comparison of sonograms from
southern (Mirsky 1976) and northern California (Baptista and Schuchmann
1990) shows differences that are much less marked. The island population
is only about 100 birds and was probably founded quite recently; song varies
little among individuals. Mirsky (1976) speculates that the oddity of island
song results from the initial founding event coupled with a subsequent bottleneck. Repeated bottlenecks may have reduced genetic and cultural variation, resulting in strong cultural drift. This hypothesis would be supported
if a song resembling that found in the island song were part of the mainland repertoire.
Gaunt et al. (1994) obtained indirect evidence of vocal learning by comparing the songs of neighbors to more distant individuals in two species, the
sparkling violet-ear, Colibri coruscans, and green violet-ear, C. thalassinus.
They find that neighbors’ songs are more highly correlated with each other
than with those of nonneighbors. The degree of acoustic difference correlates with geographic distance between neighborhoods. Oddly, the withinindividual correlation for C. coruscans is quite low (0.28), and the authors
do not comment on this in their paper. They find occasional strong note
similarities between distant birds in C. thalassinus but argue that these similarities are not as strong as those between neighbors. The lack of genetic
data makes interpretation of these patterns inconclusive.
More direct evidence that song is acquired through learning comes from
a single study of song development in isolated Anna hummingbirds (Fig.
4.10; Baptista and Schuchmann 1990). One male reared without exposure
to conspecific song developed a song that differed significantly from that of
wild adults in frequency range and syllable duration, although it retained
several species-specific characteristics. Three males reared with acoustic
188
J.W. Boughman and C.F. Moss
that individuals may vary from one another in frequency and temporal
characteristics of all song components, even when the basic structure of
notes is quite similar, a possible example of by-product distinctiveness.
Songs sung by different males within a song group show striking similarity
in the frequency contour from the song’s middle portion, and this feature
differs from other song groups on a lek (Wiley 1971). Visual comparison of
sonograms indicates that song at the lek studied by Snow (1968) differs
from that at the lek studied by Wiley (1971). Both Snow and Wiley suggest
that vocal learning leads to the song sharing they observed. To date, no
experimental tests of the learning hypothesis have been conducted on little
hermits.
Dialect variation has also been found in the song of Anna’s hummingbird, C. anna (Mirsky 1976; Baptista and Schuchmann 1990).The differences
between island and mainland populations include the structure of syllables
and temporal phrasing of the song (Mirsky 1976). Mainland males show
no response to playback of island songs, whereas they respond strongly to
playback of mainland songs (Mirsky 1976), indicating that dialects are
recognized as distinct by the birds. Visual comparison of sonograms from
southern (Mirsky 1976) and northern California (Baptista and Schuchmann
1990) shows differences that are much less marked. The island population
is only about 100 birds and was probably founded quite recently; song varies
little among individuals. Mirsky (1976) speculates that the oddity of island
song results from the initial founding event coupled with a subsequent bottleneck. Repeated bottlenecks may have reduced genetic and cultural variation, resulting in strong cultural drift. This hypothesis would be supported
if a song resembling that found in the island song were part of the mainland repertoire.
Gaunt et al. (1994) obtained indirect evidence of vocal learning by comparing the songs of neighbors to more distant individuals in two species, the
sparkling violet-ear, Colibri coruscans, and green violet-ear, C. thalassinus.
They find that neighbors’ songs are more highly correlated with each other
than with those of nonneighbors. The degree of acoustic difference correlates with geographic distance between neighborhoods. Oddly, the withinindividual correlation for C. coruscans is quite low (0.28), and the authors
do not comment on this in their paper. They find occasional strong note
similarities between distant birds in C. thalassinus but argue that these similarities are not as strong as those between neighbors. The lack of genetic
data makes interpretation of these patterns inconclusive.
More direct evidence that song is acquired through learning comes from
a single study of song development in isolated Anna hummingbirds (Fig.
4.10; Baptista and Schuchmann 1990). One male reared without exposure
to conspecific song developed a song that differed significantly from that of
wild adults in frequency range and syllable duration, although it retained
several species-specific characteristics. Three males reared with acoustic
188
J.W. Boughman and C.F. Moss