CHOOSiNESS AND PRODUCTIVITY IN WRENS OF
FORESTS, FRAGMENTS, AND FARMS
Kathy Winnett-Murrayembers of the family Troglodytidae (wrens)
illustrate an intriguing example of the tran-
sition in bird life history strategies along
Monteverde's steep elevational gradients. One ques-
tion is how House Wrens, which thrive in the most
disturbed habitats, maintain a reproductive rate that
is three to seven times higher than their relatives in
less disturbed habitats (Skutch 1960, Alvarez-Lopez
et al. 1984, Winnett-Murray 1986, Young 1994a). The
answer has to do with differences in the food supply
of the wrens, which I compared among House Wrens,
Plain Wrens, Rufous-and-white Wrens, and Gray-
breasted Wood-Wrens (Fig. 6.5), All four species are
opportunists, feeding on a diverse array of inverte-
brates such as worms, centipedes, slugs, and spiders.
Pastures and other early successional habitats fre-
quented by House Wrens support a larger prey bio-
mass than do the gaps, forest edges, and pristine or re-
generating forest used by the other three wren species
(Winnett-Murray 1986, 1987). The House Wren food
supply varies much less (seasonally and by location)
than it does in the habitats used by the other species.
How do House Wrens exploit their higher food avail-
ability and achieve a higher reproductive rate?
I compared the types and sizes of prey that forag-
ing birds ate with those that they fed to nestlings. All
wrens captured primarily smaller prey, and succes-
sively fewer prey in each of the larger size classes,
despite a twofold difference in average wren mass from
the smallest (House Wren; Fig. 6.10) to the largest (Ru-
fous-and-white Wren), and a similarly large difference
in their beak lengths (Winnett-Murray 1986). All spe-
cies were highly selective; they brought larger prey
items to their nest as compared to what they captured
in the field. The most striking difference was in the
higher selectivity exercised by House Wrens who were
feeding their young. The small House Wrens achieved
the largest shift in these two prey size distributions (Fig.
6.11). A similar pattern emerged in the kinds of prey
brought to nestlings (Fig. 6.12); House Wrens were much
more selective than Plain Wrens in what they brought
their nestlings, even though they caught similar prey
distributions in the field (Winnett-Murray 1986).
Nearly half of the items that House Wrens deliv-
ered to their young were large, soft-bodied insect lar-
vae, which was in far greater proportion to what they
captured in the field. House Wrens also fed their nest-
lings orthopterans (crickets, katydids, and grasshop-
pers), bees, and wasps, groups that are readily avail-
able in the open habitats where they hunt. The other
three wren species also fed insect larvae to young but
made greater use of adult beetles than did House
Wrens. The specialization of House Wrens on larvae
was not because more larvae occurred in their habi-
tats; many more larvae occur in cloud forests, where
House Wrens seldom foraged (Buskirk and Buskirk
1976). Rather, where food was predictably abundant,
House Wrens could be more choosy. The benefit of
this may be that nestlings who are fed a higher pro-
portion of large, soft-bodied prey grow faster and/or
fledge at higher relative weights (Von Bromssen and
Jansson 1980, Biermann and Sealy 1982), which
would minimize the length of the nesting cycle,
thereby increasing the opportunity for parents to
raise subsequent broods. Evidence that tropical House
Wrens are generally not food-limited is that expe-
rimental food supplementation did not enhance fledg-
ing success or weight gain (Freed 1981, Young 1994a).
Nesting in proximity to humans also seems to be a
substantial "edge" that House Wrens have over their
forest-dwelling relatives, because most (nonintro-
duced) predators of bird eggs and chicks are less com-
mon in the immediate vicinity of dwellings and pas-
tures than they are in forested habitats. Predation at
House Wren nests in or on occupied houses and barns
was substantially lower (6% of 73 offspring at 21 nests)
than predation at House Wren nests in other locations
more distant from people (21% of 57 offspring at 17
nests) (Winnett-Murray 1986).
Compared to other wrens, House Wrens are better
able to respond to habitat variability and human al-
teration of habitats by adjusting reproductive param-
eters, for example, clutch size, timing of breeding,
apportionment of parental care, patterns of adult
weight changes, and levels of aggression toward their
own and other species (Kendeigh 1941, Freed 1986a,b,
Kennedy and Power 1990, Kermott et al. 1991, Pribil
and Pieman 1991, Robinson and Rotenberry 1991,
Young 1994a). They exemplify "animal weeds" be-
cause they thrive in disturbed areas and are charac-
terized by high reproductive output and rapid dis-
persal. They also capitalize by selectively foraging on
high food levels in early successional environments.
In contrast, forest species (e.g., Gray-breasted Wood-
Wrens; Fig. 6.5) apparently have greater difficulty
finding the fewer less predictable, more cryptic prey.
Wood-wrens frequently search leaf undersides, where
understory arthropods tend to accumulate in wet208 BirdsM