pastures. I report the occurrence of two other species
for contrast, Hampea, a bird-dispersed species, and
Conostegia spp., which are bat- and bird-dispersed
species. Hampea reached peak density in early suc-
cessional pastures, whereas Conostegia was most
common in mid-successional pastures. Rapid growth
of these two fleshy-fruited species may play a critical
role in pasture regeneration. Conostegia provided an
almost continuous canopy cover after 10 years, pro-
viding shelter for many species that do not eat its fruit,
but which perch in its branches.
Lauraceous plants were restricted to forest edge or
beneath shade trees until later successional stages. In
contrast, the total number of stems (seedlings, sap-
lings, and adults of all species) were evenly distrib-
uted throughout the pasture after 10 years (Fig. 12.9).
Because Conostegia grows up quickly in abandoned
pastures, its distribution may influence the subse-
quent distribution of other plant species. Birds are
attracted to Conostegia for cover and for fruit and,
while perched, may regurgitate and spread lauraceous
seeds.
The species of tree that was left as a shade tree influ-
enced the recruitment of Lauraceae to pastures (Fig.
12.10). The largest number of lauraceous seedlings were
found beneath lauraceous trees. Most lauraceous seed-
lings and saplings were the same species as the shade
tree, but other species of Lauraceae were also found
beneath these trees. Only fig trees had close to as many
lauraceous seedlings growing beneath their canopies. To
ensure the recruitment of lauraceous trees into aban-
doned pastures, one or more lauraceous shade trees
should be left. Both figs and Lauraceae had many more
young plants beneath their canopies than did wind-dis-
persed species, suggesting that for overall recruitment,
a tree with edible fruit attracts more dispersers.
My study was a "snapshot" census of pastures at
different ages. The ideal study would document how
each pasture changes over time. Understanding which
individuals grow to maturity will provide guidance
for adopting management practices that accelerate
pasture regeneration. My data suggest that planting of
particular species or setting up artificial perches may
increase the influx of seeds into more open areas.
In some temperate areas, the position of trees, snags,
and other perching areas is correlated with enhanced
deposition of seeds (McDonnell and Stiles 1983,
McClanahan and Wolfe 1993, Robinson and Handel
1993). As primary habitats dwindle, how we influ-
ence regeneration of fallow lands to enhance con-
servation and management objectives become in-
creasingly important. Analysis of pasture regeneration
gives insight into variables that influence recruitment
into fallow pastures. Accelerating the dispersal of tar-
geted plant species by birds or other animal dispers-
ers may be a key component of directed regeneration
projects.Acknowledgments This work was supported by an
OTS 88-1 postcourse research grant. I am grateful
to Bob Lawton, Peter Feinsinger, Doug Levey, Bob
Podolsky, and Nat Wheelwright for inspirational dis-
cussions and to William Haber and Willow Zuchowski
for help with plant identification.444 Conservation Biology