reports toucans (Ramphastidae) and cotingas (Cotingidae) from fallows and
crop fields and observes that their use of these habitats is related to generalist
feeding habits. Birds of both families are principally frugivores, and many
species take a wide range of fruit types that may include those of pioneer
plants (Stiles and Skutch 1989). Species richness in families that include con-
sumers of small berries, such as Pipridae (manakins) and Thraupidae (tan-
agers), may similarly be maintained or increased in fallows (Johns 1991;
Andrade and Rubio-Torgler 1994). Such increases may be at least partially
linked to increased availability of the small berries of species of plant families
such as Rubiaceae, Melastomataceae, and Piperaceae, which can be both more
abundant (Laska 1997; Guariguata and Dupuy 1997) and more fecund
(Levey et al. 1994) in disturbed habitats than in the shaded forest understory.
However, it is important to emphasize that patterns of habitat use by frugivo-
rous birds in human-influenced landscapes do not always correlate well with
patterns of fruit availability and therefore must be related to other factors
(Restrepo et al. 1999).
Microclimatic variation between tropical forest environments with different
degrees of disturbance, and its effects on populations and communities of
plants at least, is well documented and has been the subject of several reviews
over the years (e.g., Clark 1990). Shifting cultivation clearings are large in the
context of tropical forest canopy gaps, and it is obvious that their microclimates
will differ markedly from those of the forest understory and forest treefall gaps.
The regeneration of vegetation can quickly buffer microclimatic change in
large clearings, however (Fetcher et al. 1985), and it is unfortunate that there
appear to be no published comparative studies of microclimate in different-
aged fallows and primary forest. Do microclimatic differences between primary
forest and fallow affect the distributions of some forest vertebrates over shifting
cultivation landscapes in the same way as they affect plant regeneration pat-
terns? Karr and Freemark (1983) demonstrated that habitat use patterns of
many neotropical rainforest birds are partly related to spatial and temporal
microclimatic gradients, suggesting that physiological stress, rather than a
microclimatically mediated pattern in food availability, was the main causal fac-
tor. This suggestion apparently has been interpreted as a tested hypothesis by
subsequent authors (Johns 1991; Andrade and Rubio-Torgler 1994).
Hunting, rather than habitat structure and quality, is likely to be the sig-
nificant factor in determining some vertebrate community characteristics in
many shifting cultivation landscapes (see Chapter 14, this volume). The
apparently greater effect of shifting cultivation on primates indicated for
neotropical sites, in comparison with those from Africa and Asia, may have
been a consequence of historical or contemporary hunting rather than habitat
factors (Cowlishaw and Dunbar 2000). The high biomass of small and
medium-sized mammals in African shifting cultivation landscapes is sustained
even under significant hunting pressure, but if large mammals are absent from
180 III. The Biodiversity of Agroforestry Systems