among species (Snow 1965, Levin and Anderson 1970,
Stiles 1977).
To understand seasonal rhythms in the forests of
Monteverde, I have studied reproduction in 23 small-
fruited species of trees whose seeds are dispersed by
birds (Wheelwright 1985a, 1986) since 1979. I have
focused on a single guild (a group of species that in-
teract strongly because they share ecological require-
ments; Root 1967). This guild, composed of trees
in the family Lauraceae, all provide similar floral
and fruit resources and depend upon a similar set
of pollinators and seed dispersers. The goals of
the study are to quantify the proportion of a given
plant population that flowers or fruits in a particu-
lar year, the seasonal synchrony of reproduction
between in-dividuals within populations and be-
tween species, the degree to which flower and fruit
production varies from year to year, and community-
wide phenological differences between flowering
and fruiting.
Almost all of the Lauraceae of Monteverde produce
panicles (branched inflorescences) of small, yellow-
ish green flowers that yield small amounts of nectar.
Their major insect visitors (and presumed pollinators)
are generalist bees, wasps, and flies. Lauraceae fruits
also vary little among species in terms of their basic
structure—all have a large single seed surrounded by
a dense oily pulp. Most fruits turn black when ripe,
and in many species, the pedicel (stem) holding the
fruit becomes conspicuously swollen and red. Al-
though fruits of different species are structurally simi-
lar, the variation in size among species is large; fruits
range in diameter from 0.6 to 2.3 cm (Mazer and
Wheelwright 1993).
During the first year of my study, I monitored re-
production in 286 trees along a 7-km transect at bi-
weekly intervals. During the second year, I observed
trees at monthly intervals. Thereafter, censuses have
been conducted once a year at the same time each
year (early dry season). Each species is surprisingly
predictable and synchronous in its flowering
and fruiting phenology. Different species reproduce
at characteristic times of the year, with little varia-
tion among years despite appreciable annual varia-
tion in rainfall. Most individuals of a particular spe-
cies flower or fruit within several weeks of other
members of the population. The proportion of each
population that reproduces each year and the mag-
nitude of flower or fruit production, however, vary
widely from year to year (20-100% of the individu-
als of a particular species reproduce). The average
number of flowers or fruits produced per tree within
species also differs markedly among years. There is
also high variation across species in flowering and
fruiting patterns, with some species reproducing
annually, biennially, or erratically (Wheelwright
1986).
In terms of community patterns, flowering within
the Lauraceae is more evenly distributed throughout
the year than fruiting. In every month of the year, at
least two species are in flower, with no more than
eight species flowering in a given month (Fig. 8.20A).
Fruiting is relatively aggregated. In December, it is
difficult to find ripe lauraceous fruits, whereas just
before the beginning of the rainy season as many as
13 species are in fruit (Fig. 8.20B). Studies of herba-
ceous and woody plant species from the temperate
zone and tropics suggest that staggered or minimally
overlapping phenologies may be more likely for flow-
ering than for fruiting (Frankie et al. 1974, Hilty 1980,
Wheelwright 1985a).
Natural selection may be greater for temporal di-
vergence among plant species in flowering than it
is for fruiting. Plant species that flower simulta-
neously potentially compete for a limited number
of pollinators, just as fruiting plants may compete
for seed dispersers. However, different plant species
that flower at the same time potentially have another
disadvantage: they may suffer reproductive losses to
stigma-clogging by foreign pollen, waste of pollen
that is transferred to the wrong species, and hybrid-
ization, penalties that have no analog in fruit-
ing. Selection against divergence in phenologies
may be less for flowering than fruiting. After polli-
nation has occurred, the developing embryo is still
nourished and protected by the maternal plant;
development associated with flowering can theo-
retically proceed at any time of the year. In contrast,
after fruits have ripened, the dispersed seed and
seedling are at the mercy of the environment once
they are independent of the parent plant. Unless
the seed or seedling can survive desiccation, nutri-
ent scarcity, shade, and predation, the timing of fruit-
ing may be restricted to certain favorable times of
year.
I controlled for phylogeny by comparing related
plant species, but phylogeny itself may play an im-
portant role in determining phenology. It may be that
lauraceous species that flower during November in
Monteverde do so not because of interspecific com-
petition for pollinators but simply because they origi-
nated in the southern hemisphere and have retained
an evolutionary legacy of flowering in the spring. The
systematics of the Lauraceae is even more uncertain
than that of most tropical plant families (Burger and
van der Werff 1990). We need to understand evolu-
tionary relationships and the current and historical
geographic distributions of tropical tree species in
order to explain the timing of flowering and fruiting
in tropical forests.282 Plant-Animal Interactions