242
M.F. Willson et al.
Second, covered nests might be thought to be adaptations to protect
the eggs and young from weather, nest predators, or brood parasites
(although covered nests do not necessarily prevent brood parasitism by
cowbirds; Cruz & Wiley, 1989). If the advantage of a covered nest lies in
a protective microlimate, then one might predict that covered nests would
be most frequent in the northwest, where birds often nest early, when
snow is still on the ground. Alternatively, one could argue that the rainy
climates select for protected nests and predict that covered nests would be
more frequent in northwestern and southern forests than in the midwest.
On the other hand , if the advantage of a covered nest lies in protection
from nest predators and brood parasites, one might predict them to be
most common in the midwest, where cowbirds, snakes, and avian and
mammalian predators abound, or least common in the south temperate
forests , which host relatively few species of nest robbers. None of these
patterns is seen in the array of six categories of forest, suggesting that
neither factor, taken alone , accounts for the observed pattern.
Third, some of these forests are largely evergreen, and some are largely
deciduous. One might argue that evergreenness is associated with ,longlived leaves, which are allocated higher levels of chemical and morphological protection than deciduous leaves, and that insect herbivores are
consequently less abundant on evergreen leaves. If this were true, and if
the diversity and abundance of birds is associated with the size of the
potential prey population, then leaf-gleaners should be most abundant or
diverse in deciduous stands. This is not the case, however: Although
southern forests have low levels of leaf gleaners, North American coniferous forests tend to have proportionately more abundant (though no
more diverse) populations of leaf gleaners than deciduous forests (although
the Tukey test does not detect significant differences at the predicted
points in the array). Thus, the hypothesized relationship between evergreenness and insect abundance seems not to explain much of the observed
pattern.
Fourth , conifer and broadleaf forests differ in leaf shape and in
branching patterns of branchlets and twigs, and they might offer differing
availability and diversity of noncavity nest-sites or foraging surfaces . One
might predict that nest-sites and diversity of foraging surfaces would be
greater in broadleaf forests than in conifers and that the birds using these
forests would be more diverse and abundant there. However, there tend
to be relatively more species of open-nesters and proportionately more
leaf-gleaners in conifer forests (refer to Table 11.4) .
Conclusions
Comparisons of Alaskan and Chilean Coastal Rainforests
Site diversity was somewhat higher in Chilean forest and Alaskan deciduous forest than in Alaskan conifer forests . Point-diversity was highest for
M.F. Willson et al.
Second, covered nests might be thought to be adaptations to protect
the eggs and young from weather, nest predators, or brood parasites
(although covered nests do not necessarily prevent brood parasitism by
cowbirds; Cruz & Wiley, 1989). If the advantage of a covered nest lies in
a protective microlimate, then one might predict that covered nests would
be most frequent in the northwest, where birds often nest early, when
snow is still on the ground. Alternatively, one could argue that the rainy
climates select for protected nests and predict that covered nests would be
more frequent in northwestern and southern forests than in the midwest.
On the other hand , if the advantage of a covered nest lies in protection
from nest predators and brood parasites, one might predict them to be
most common in the midwest, where cowbirds, snakes, and avian and
mammalian predators abound, or least common in the south temperate
forests , which host relatively few species of nest robbers. None of these
patterns is seen in the array of six categories of forest, suggesting that
neither factor, taken alone , accounts for the observed pattern.
Third, some of these forests are largely evergreen, and some are largely
deciduous. One might argue that evergreenness is associated with ,longlived leaves, which are allocated higher levels of chemical and morphological protection than deciduous leaves, and that insect herbivores are
consequently less abundant on evergreen leaves. If this were true, and if
the diversity and abundance of birds is associated with the size of the
potential prey population, then leaf-gleaners should be most abundant or
diverse in deciduous stands. This is not the case, however: Although
southern forests have low levels of leaf gleaners, North American coniferous forests tend to have proportionately more abundant (though no
more diverse) populations of leaf gleaners than deciduous forests (although
the Tukey test does not detect significant differences at the predicted
points in the array). Thus, the hypothesized relationship between evergreenness and insect abundance seems not to explain much of the observed
pattern.
Fourth , conifer and broadleaf forests differ in leaf shape and in
branching patterns of branchlets and twigs, and they might offer differing
availability and diversity of noncavity nest-sites or foraging surfaces . One
might predict that nest-sites and diversity of foraging surfaces would be
greater in broadleaf forests than in conifers and that the birds using these
forests would be more diverse and abundant there. However, there tend
to be relatively more species of open-nesters and proportionately more
leaf-gleaners in conifer forests (refer to Table 11.4) .
Conclusions
Comparisons of Alaskan and Chilean Coastal Rainforests
Site diversity was somewhat higher in Chilean forest and Alaskan deciduous forest than in Alaskan conifer forests . Point-diversity was highest for