The Wilson Journal of Ornithology 124(3):446 453. 2012 
REPERTOIRE SIZE AND SYLLABLE SHARING IN THE SONG OF 
THE CLAY-COLORED THRUSH ('TURDUS GRAYI) 
LUIS E. VARGAS-CASTRO, 1 - 2 - 3 NATALIE V. SANCHEZ. 1 AND GILBERT BARRANTES 1 
ABSTRACT.—Sons- repertoire size and extent of song sharing provide information about social interactions that occur 
in songbird species. We recorded the songs of eight male Clay-colored Thrushes (Turdus grayi ) m San Jose. Costa Rica 
during the 2008 breeding season. We classified 695 songs and 5,032 syllables using visual inspection of spectrograms and 
spectrogram correlation analysis to measure repertoire size and syllable sharing among a local group ot males. Mat 
repertoire size was 10-17 syllable types. Males shared on average 28 ± 15% (SD) syllable types from their repertoires with 
other males, but a larger proportion of syllable types remained unique to particular males. Extent ot repertoire sharing and 
distance between singing males were not related. Presence of shared and individually unique syllables in the repertoires 
indicate that imitation, and perhaps improvisation, contribute to development of the song ot Clay-colored Thrushes. 
Received 21 February 2011. Accepted 14 April 2012. 
Bird song is used in territorial defense and 
female attraction (Krebs et al. 1978. Catchpole 
1987. Catchpole and Slater 1995). Some songbird 
species accomplish both purposes with a single 
song or phrase type throughout their lives 
(Kroodsma 1996). but in many species the song 
repertoire is much larger ranging from a few 
(Krebs et al. 1978) to hundreds (Todt and Hultsch 
1996) or even thousands of song types (Kroodsma 
and Parker 1977). 
Songbird species also differ in the extent of 
song sharing between individuals of a given pop¬ 
ulation (Johnson 2006, Nicholson et al. 2007). 
Song sharing within a population is thought to be 
a result of learning, in which young individuals 
produce accurate imitations of previously heard 
adult con-specific songs (Marler 1970. Marler and 
Peters 1981. Kroodsma 1982). Usually, only pan 
of the repertoire is shared among individuals 
(Catchpole and Slater 1995). Thus, repertoire size 
and extent of song sharing arc behavioral 
characters that provide information about social 
interactions and the song learning process of a 
songbird species (Baker et al. 1986. Searcy and 
Andersson 1986. Beecher et al. 2000. Beecher and 
Burt 2004). 
Thrushes (Turdidae) in general are known for 
their long, varied, and melodious songs. The song 
ot Clay-colored Thrushes ( Tardus gmvi) consists 
of a sequence of syllables highly variable in length. 
Our objectives were to: (1) describe the temporal 
structure of songs of the Clay-colored Thrush, (2) 
' Escuela de Biologui. Universidad dc Costa Rica. 2060 
San Pedro. San Jose. Costa Rica, 
Current address: Department of Biology. University of 
Miami. Coral Gables. FL 33146, USA. 
Corresponding author; e-mail: luissum@gmail.com 
identify individual repertoire size, and (3) estimate 
the extent of syllable sharing between neighboring 
males to better understand the social interactions 
that occur among neighbors of this thrush. 
METHODS 
Study Area .—We conducted this study on the 
campus of the Universidad de Costa Rica (UCR), 
San Jose Province, Costa Rica (09 56' N, 84 02' W; 
1,200 m asl). The campus has large open areas 
with scattered trees and bushes near a stream that 
crosses the campus. There arc also two old-second 
growth forest patches of 1 and 0.5 ha. respectively. 
Female Clay-colored Thrushes nest principally on 
trees in open areas or human-made structures, while 
males defend a small territory around the nest. 
Procedures. —The Clay-colored Thrush breeds 
from March to June at this location. We recorded 
songs of eight adult males from March to May in 
2008, of which four were banded with a unique 
combination of colors and another had a unique 
mark on his chest. We recorded one banded male 
on 4 different days, three other males (2 of which 
were banded) on 3 different days, and the rest of 
the males (2 were marked) during a single day. 
Some males w ere unbanded, but w e were able to 
identify each male through time by its territory 
location and presence of individual syllable types 
in its repertoire. We obtained most songs between 
0450 and 0930 hrs, but we also used recordings 
from other times of the day. 
Song Recording .—We used a Sony M-635VK 
tape recorder and a Sennheiser ME66/K6 shotgun 
microphone to record songs. We digitized the 
recordings at a sample rate of 44.1 kHz and a 
resolution of 16 bits with Adobe Audition 1.0 
software (Adobe Systems Inc., San Jose. CA. 
446 
