370 
THE WILSON JOURNAL OF ORNITHOLOGY . Vol. 123. No. 2. June 2011 
TABLE 2. The proportion male and proportion adult of 
wintering Bicknell’s Thrush captured at sites on Hispaniola 
characterized by differing levels of understory density. 
Understory density 
>10.000 stems/hn <8.000 stems/ha 
Total birds captured 53 74 
Proportion male 0.76 0.55 
Proportion adult 0.77 0.69 
of this species. A review of adult sex ratios in 
birds found that species listed as Globally 
Threatened by the IUCN had male-skewed sex 
ratios in comparison to non-threalened species 
(Donald 2007). This review offered a series of 
five non-independent explanations, each of which 
highlighted conservation implications. Among 
these was the possibility lhal intersexual compe¬ 
tition during the non-breeding season forces 
smaller-bodied females to occupy inferior-quality 
winter habitat (Benkman 1997, Marra and Holmes 
2001). Under these circumstances, winter habitat 
quality could limit female survivorship with 
implications for overall population dynamics 
(Marra and Holmes 2001). 
Our data suggest that Bicknell's Thrushes have 
some sexual habitat segregation on Hispaniola. 
We currently cannot evaluate the extent to which 
female thrushes may be limited in wintering areas, 
but our results suggest that future investigation 
should pursue this possibility. Our data also 
indicate that certain areas are preferentially 
occupied by males and that differential habitat 
occupancy may be proximatcly related to charac¬ 
teristics of the forest understory. Males were 
numerically dominant in high-efevation forests 
with a thick understory of vines and small trees 
occurring at densities >10,000 stems/ha. The 
understory vegetation at other sites, where sex 
ratios did not differ significantly from parity, was 
relatively more open occurring at mean densities 
<8.000 stems per ha. Bicknell’s Thrush compete 
during the winter for exclusive access to territo¬ 
ries (Townsend et al. 2010 ), and it is possible that 
understory density is a key component of habitat 
quality. A radiotelemetry study of wintering 
Bicknell’s Thrush indicated that individuals in 
the Sierra de Bahoruco (a male-biased site) were 
detected most frequently in the densest microhab¬ 
itats (Townsend et al. 2010). Thrushes wintering 
at this site primarily consumed arthropods, 
whereas thrushes at an open understory site in 
the Cordillera Septentrional (an equal sex-ratio 
site) mainly consumed fruit (Townsend et al 
2010). It is possible that territory holders in male 
dominated habitats derive distinct benefits from 
dense understory thickets, including protection 
from predators and a steady, abundant source of 
arthropods, which may serve as a higher qualm 
winter food source than fruit (Long and Stauffer 
2003, Diggs 2008). 
The proportion male did not vary significant!) 
with elevation across Hispaniola; however, the two 
most male-biased sites were in high-elevation cloud 
forest. It is possible that elevation in isolated areas 
of Hispaniola provides a buffer against human 
agricultural disturbance allowing for persistence of 
intact, dense-understory cloud forest (Latta ct al. 
2003). The densest patches of understory within this 
forest type are preferred by Bicknell’s Thrush 
(Townsend et al. 2010) and are frequently the result 
of storm-related blowdowns. This suggests the need 
to better understand the interactions among severe 
weather dynamics, intact forest that is isolated from 
human agricultural activity, and quality Bicknell’s 
Thrush habitat. 
The geographic variation in sex ratios of 
Bicknell’s Thrush across Hispaniola highlights 
the importance of sampling a broad array of 
habitats and locations when assessing a species 
winter social structure. We suggest that drawing 
conclusions about the distribution of males and 
females from studies at only one or two sites of 
similar habitat might fail to reveal broader 
population-level patterns. The sex ratio among 
American Redstarts (Setophaga ruticUla) in 
Jamaica, for example, varied from 50:50 at only 
two of six study sites along a gradient from 
natural to agricultural habitats (Johnson et al. 
2006). Only two of seven geographic areas in our 
study were strongly male-biased with the remain¬ 
der having no significant variation from an equal 
sex ratio. Sex ratios and habitat quality are likely 
to vary along a wide gradient of available habitats 
with many permutations of intermediate quality 
(Latta and Faaborg 2002, Johnson et al. 2006). 
Intermediate sites are likely to influence fitness 
and demographics for a large portion of a given 
species' population. 
The proportion of adults in the population. m 
contrast to sex ratios, had no association with 
forest structure and little variation among sites. 
Evidence from our study suggests Bicknell s 
Thrush on Hispaniola do not segregate by age 
class. The evidence also suggests relatively low 
