Monteverde : Ecology and Conservation of a Tropical Cloud Forest

also result from (1) phylogenetic constraints on

phonological plasticity, (2) competition with other

plant species for dispersal (Wheelwright 1985a;

Wheelwright, "A Hypothesis," pp. 281-283), or (3)

seasonal constraints on seed or seedling survival.

The latter may be particularly important in Monte-

verde. Germination of most pioneers on Barro Colo-

rado Island occurs early in the wet season, and older

seedlings are better able to survive through the first

dry season than younger seedlings, especially at sites

where the density of competing seedlings is high

(Garwood 1983). The same may be true at Monte-

verde; surface soils in large gaps dry out during the

dry season, and seedlings in gaps commonly die from

water stress (K. G. Murray, unpubl. data). The evo-

lution of fruiting phenologies of pioneer plants at

Monteverde needs more empirical study, especially

on seedling demography.

WHAT HAPPENS TO SEEDS OF VERTEBRATE-DISPERSED

TREES AFTER DISPERSAL?

DanWenny

ostdispersal events, such as seed predation and
a second stage of dispersal, may play a key role
in plant recruitment. For many species, post-
dispersal seed predation is a major source of mortality
that limits population growth (Crawley 1992). Most
data are from temperate zone plants, especially on
seed preference by rodents in arid areas, although
studies on frugivory and seed dispersal have been
conducted in Monteverde (Wheelwright 1983, 1991,
Murray 1986a, b, 1988). Seed size may affect preda-
tion rates. In general, rodents prefer large seeds over
small seeds because larger seeds have more energy
than small seeds (Price and Jenkins 1986). Other fac-
tors, such as chemical content and handling time, also
influence seed choice (Kerley and Erasmus 1991).
I examined rates of cloud forest seed predation as
a function of seed characteristics by recording the fate
of naturally dispersed seeds. Most studies have used
seeds that are "experimentally dispersed" by the re-
searchers (Schupp 1988, Forget 1993). Species were

chosen to represent a variety of plant families and seed

sizes, but seeds of all species were relatively large

compared to other species in the community (Wheel-

wright et al. 1984; Table 8.5). This study was con-

ducted from May 1993 to August 1994 in a 5-ha area

(1600 m) along the Continental Divide near the inter-

section of the Valley trail (Sendero El Valle) and the

Tower trail (Sendero Las Torres).

Dispersed seeds were located by searching the

ground for freshly regurgitated, dropped, or defecated

seeds. Seeds were marked by glueing 50 cm of un-

waxed dental floss to the seed and tying flagging tape

to the floss. All sites were censused several times the

first week and once each week thereafter for 12 weeks.

If a marked seed was removed, the surrounding area

was searched until the flagging tape-dental floss as-

sembly was found. If a piece of the seed coat remained,

the seed was classified as killed by seed predators. If a

seed was partially eaten, tooth or bill marks were used

to determine the type of animal that handled the seed.

Table 8.5. Characteristics of seeds in the study of postdispersal survival.

Species

Hippocrateaceae

Salacia petenensis

Lauraceae

Ocotea endresiana

Beilschmiedia pendula

Meliaceae

Guarea glabra

Myrtaceae

Eugenia sp.

Sabiaceae

Meliosma sp.

Seed Mass

Dispersers^8 (g)

Birds? bats?

Birds

Birds

Birds, monkeys

Bats

Birds? bats?

7.35

0.75

12.86

0.76

2.26

1.75

Seed Coat

(mm) N

0.4

0.3

1.7

0.5

0.3

2.2

42

376

273

401

64

41

sDisperser classification based on Wheelwright et al. (1984), D, Wenny (pers. obs.), and C. Guindon

(pers. comm.)

286 Plant-Animal Interactions

P