manipulated to increase the number of animal and bird species using particu-
lar habitat patches and therefore could increase the total habitat area in a land-
scape suitable for at least some of the needs of animal and bird species.
Options for biodiversity-conscious management of plant communities
might concentrate on two related, specific objectives: to increase the length of
fallow periods so that more species accumulate and (a point not touched upon
in case studies) greater numbers of individuals reach reproductive maturity
and to increase of the rate of accumulation of plant diversity so that more
diversity accumulates for a given fallow length. The former objective might be
achieved by promoting uses of fallows other than the normal ones of weed
control and the recovery of soil fertility, ones that entail longer periods of veg-
etation development, another significant challenge in the context of general
tendencies toward shortening fallows and adopting the planted fallow. A pos-
sible strategy here is fallow management for timber and nontimber forest
products (Smith et al. 2001). The latter objective could be pursued by thin-
ning to favor longer-lived or forest-dependent plant species over pioneers,
focusing specifically on reducing the degree and duration of dominance of the
vegetation by low-diversity assemblages of short- or long-lived pioneer species
(Finegan 1996). The regeneration of the species to be favored is a basic prem-
ise here and cannot by any means be guaranteed. In an ideal world, managers
would evaluate regeneration using techniques of silvicultural diagnosis, as
Finegan and Delgado (2000) have suggested in the context of forest restor-
ation through secondary succession on abandoned neotropical pastures, and
may conclude that planting is necessary for biodiversity conservation objec-
tives. In general, however, there is little or no experience in this type of
silvicultural intervention in the neotropics, although advances have been doc-
umented for temperate zones (Smith et al. 2001).
Two main areas of action suggest themselves in relation to vertebrates:
management of vegetation structure, composition, and microclimates; and
management of hunting (Bennett and Robinson 2000; Robinson and Bennett
2000; see Chapter 14, this volume).
Data from other contexts (Hartley 2002) suggest that the conservation of
more trees of the original forest than is usual, of a wider range of species,
would make an important contribution to the animal and bird diversity of
shifting cultivation communities. Different spatial configurations of con-
served trees may vary in their effectiveness in this context (Hartley 2002).
Given that shifting cultivation land is burned frequently, however, and most
tropical forest tree species are highly vulnerable to death even from ground
fires (Uhl and Kauffman 1990), trees probably would have to be conserved in
strips between fields and burning carried out with care. In terms of resistance
to fire, riparian forests also may have a special place in habitat management.
186 III. The Biodiversity of Agroforestry Systems