368 
THE WILSON JOURNAL OF ORNITHOLOGY . Vol. 123. No. 2. June 2011 
toward endangered and extinct status (Donald 
2007). 
The goal of our study was to document the sex 
ratio of wintering populations of BicknelTs 
Thrush for comparison to that of breeding 
populations. This information could lay the 
framework for investigating where females are 
limited, if the winter population is equally male- 
biased. We classified male : female ratios and age 
ratios of mist-netted thrushes at seven sites across 
Hispaniola. We also quantified local vegetation 
characteristics at each site to evaluate the 
potential influence of forest structure on sex and 
age ratios. Our primary objectives were to; (I) 
document sex and age ratios of Bicknell’s Thrush 
across an array of sites that spanned most major 
geographic areas in which the species is known to 
occur in Hispaniola, (2) describe any deviation in 
sex and age ratios among these sites and between 
winter and breeding areas, and (3) characterize 
and compare the relationship between forest 
structure and sex and age ratios at each site. 
METHODS 
Field and Laboratory Procedures .—We cap¬ 
tured Bicknell s thrushes at seven rain forest and 
cloud forest sites between January and March 
2004 on Hispaniola. (1) Cordillera Central, the 
largest and highest elevation mountain range in 
the Dominican Republic with a well-managed 
system of preserves (Perdomo and Arias 2008). 
(2) Sierra de Bahoruco. a mountain range in the 
south-central part of the island that is largely 
protected as a national park (Townsend et al. 
2010). (3) Massif de la Hotte. a mountain range on 
the southwestern end of Haiti's Tiburon Peninsula 
that is tenuously protected as the Pic Macaya 
Biosphere Reserve (Rimmer et al. 2005). (4) 
Sierra Martin Garcia, a peninsular mountain range 
on the southcentral coast of the Dominican 
Republic that is partially protected as a national 
park. (5) Sierra de Neiba. a mountain range along 
the Haiti-Dominican Republic border that has 
been developed extensively for agriculture. The 
western high peaks area is formally protected as a 
national park in the Dominican Republic but 
illegal settlers and loggers are common (Rimmer 
et al. 2004, Townsend and Rimmer 2006). (6) 
Cordillera Septentrional, a mid-elevation moun¬ 
tain range in northcentral Dominican Republic 
with protected areas managed by cooperative 
agreements between government and non-govern¬ 
mental agencies (Townsend et al. 2010). (7) Los 
Haitises, a swath of low-elevation, coastal rain¬ 
forest in eastern Dominican Republic protected as 
a national park (Perdomo and Arias 2008). 
We captured thrushes in 6- and 12-m mesh inist 
nets primarily by playback of conspecilie viyal- 
izations. We examined rectrices and greater 
covert tips to classify age of birds (Collier and 
Wallace 1089). Birds classified as “adults" were 
> I year of age. whereas birds classified as "first- 
winter’ ’ were bom during the previous breeding 
season. We collected blood from all birds for 
molecular analysis by puncturing the brachial vein 
with sterile 27-gauge needles. Blood was harvest¬ 
ed in capillary tubes and stored in Queen's lysis 
buffer (Seutin et al. 1991). DNA was extracted in 
the laboratory, using Perfect gDNA Blood Mini 
kits (Eppendorf) following the manufacturer's 
protocol. We amplified homologous sections of 
sex chromosome-based chromo-heliease-DNA- 
bitiding (CHD) genes by polymerase chain 
reaction (PCR) and viewed the PCR product on 
an agarose gel to distinguish single CHD-Z male 
bands from double CHD-Z and CHD-W female 
bands (Griffiths ct al. 1998). 
Vegetation structure was characterized for each 
geographic area where we captured BicknelTs 
Thrush using a modified version of the James and 
Shugart (1970) sampling method for forested 
habitat. We established five 5-m radius circular 
plots within each geographic area at random 
distances and bearings from points of thrush 
capture. Each circular plot was divided into 
quadrants along cardinal directions. We counted 
all trees, shrubs, and herbaceous plants within 
each quadrant crossing a plane at chest height 
(~ 1.5 m) in the following categories: shrubs 
<2 cm diameter at breast height (DBH). small 
trees 2-10 cm DBH, medium trees 11-20 cm 
DBH. and large trees >20 cm DBH. Mean 
measurements at each site were scaled to estimate 
vegetative cover per hectare in each category and 
total tree cover basal area. 
Statistical Analyses .—We used Chi-square (/.*) 
goodness of fit (GOF) tests to examine variation 
in the proportion male and proportion adult. We 
analyzed the relationship between elevation and 
proportion niale/proportion adult using linear 
regression. We examined the relationship between 
proportion male, proportion adult, and forest 
vegetation characteristics using a generalized 
linear model (GLM) with mean proportion male 
or adult in each geographic area as the response 
variable and the four vegetation categories as the 
