many plants into abandoned agricultural lands could
be limited, and succession of these areas might differ
from that of treefall gaps. In landscapes that are highly
fragmented, some plant species may decline in abun-
dance, perhaps to the point that they no longer sup-
port fruit-eating animal populations (Howe 1984).
Some plant species are more heavily used by birds
than others. Particularly important are fruits of the
avocado family (the Lauraceae), which are rich in lip-
ids and amino acids and which provide critical com-
ponents of the diet of two of the most threatened spe-
cies in the area, the Resplendent Quetzal and the
Three-wattled Bellbird (Wheelwright 1983, Wheel-
wright et al. 1984). Although the Lauraceae are often
found in regenerating pastures, it is not known which
factors influence the successful recruitment of these
species.
Monteverde pastures vary in size and shape, and
the number and identity of tree species that farmers
leave in the pasture to serve as shade for cows. My
research focused on whether these differences influ-
ence the pace or pattern of forest regeneration. I had
observed that shade trees were used as perches by
many bird species, presumably because they provided
fruits and protection from predators. I also investi-
gated the potential for these shade trees to attract fruit-
eating birds, and thereby increase the frequency with
which birds disperse seeds into the pasture near the
shade tree. The shade trees I observed were mostly
fruiting trees (usually a species of fig [Ficus sp.] or
Lauraceae), but a few were wind-dispersed species
(e.g., Heliocarpus appendiculatus) that do not offer
fruit that birds would find suitable. I asked if the spe-
cies of shade tree influenced the species of plants that
were dispersed into the pasture, and the overall rate
of pasture growth.
In spring of 1988, I documented regeneration of
lauraceous trees and other species in abandoned pas-
tures. I asked whether (1) lauraceous trees colonize
abandoned pastures and (2) shade trees influence the
number and species identity of recruiting plants in
pastures. I counted the number of individuals in the
Lauraceae, and the total numbers of all species of
plants in 5 x 5 m sampling quadrats placed randomly
at the pasture edge, in the open, or under shade trees
(a total of 12 quadrats/site). I examined three pastures
in each of three successional age groups, defined by
the length of time since the last cows were removed
from the pasture: 2-5 years, 11-15 years, and >25
years. For comparison, I censused the numbers of
lauraceous plants and other juvenile shrubs and trees
under tree canopies and in more open areas in the
primary forest of the MCFP and in forest fragments
near each of the pastures where I worked.
Lauraceous trees recruit to abandoned pastures,
although they are rare in early successional pastures
(Fig. 12.8). The total number of stems of all species
reached peak density in the mid-successional pas-
tures, but juvenile lauraceous trees reached densities
typical of primary forest only in later successionalFigure 12.8. Density of stems as a
function of time since pasture
abandonment, and compared with
densities in primary forest, for all
species combined and for Lauraceae,
the early-successional species
Hampea and the mid-successional
species Conostegia. In all cases, there
is a significant relationship between
time since abandonment and the
number of stems (seedlings +
saplings) of each species group.
(Kruskal-Wallis ANOVA, N = 144, p <
.001 in all cases). Error bars in this
and all other figures represent ±1
standard error.443 Conservation Biology