Monteverde : Ecology and Conservation of a Tropical Cloud Forest

The pollination system of B. involucrata shares
several features with those of other bee-pollinated
Begonia species. A marked discrimination against
the rewardless female flowers and a positive rela-
tionship between petal size of artificial flowers
and attractiveness to pollinators have been docu-
mented in B. oaxacana, B. tonduzii, and B. urophylla
(Schemske et al. 1996, J. Agren and D. Schemske,
unpubl. data). Experimental data and field observa-

tions of these and other bee-pollinated species indi-

cate that several factors, including the size and ori-

entation of flowers and scent production, may be

important for pollinator visitation in Begonia. The

deceit pollination system of many Begonia and the

variety of mating systems found within the genus

(Agren and Schemske 1993) make begonias useful for

understanding plant reproduction and pollination

biology.

Figure 8.18. (above left) Average visitation rate to male- and female-phase inflorescences of Begonia involucrata by

the bees Trigona grandipennis in 1990 (N - 220 visits observed) and Meiipona fasciata panamica in 1992 (N = 56).

Data from Agren and Schemske (1991) and Schemske and Agren (1995). Figure 8.19. (above right) Pollinator

approaches to artificial female flowers in a population of Begonia involucrata at Monteverde. The mean flower size

is equal to the mean of male flowers, and the small and large flowers are, respectively, 2 S.D. smaller and 2 S.D.

larger than the mean male (based on width of the petallike sepal). The sepal areas for the three flower sizes were

86 mm^2 ,139 mm^2 , and 204 mm^2 , respectively. The solid bars indicate the number of approaches followed by a

rejection; the hatched bars indicate the number of approaches that were followed by a visit. (From Schemske and

Agren 1995)

A HYPOTHESIS ABOUT THE TIMING OF FLOWERING AND

FRUITING IN COMPETING TROPICAL TREES

Nathaniel T, Wheelwright

easonal changes in the tropics tend to be more

subtle than those of the temperate zone, where

annual variation in climate is more extreme.

Nonetheless, tropical plants replace senescent leaves,

burst into flower, and ripen their fruits at distinct

times of the year. The timing of reproduction in tropi-

cal plants affects the life histories of animals that feed

on nectar, pollen, fruits, and seeds, just as the season-

ally changing behaviors and abundance of animal pol-

linators, seed dispersers, and seed predators influence

plant reproductive success. A major challenge in com-

munity ecology is to explain the timing (phenology)

of major events in the life histories of tropical plants

and to understand the relative role of abiotic (physi-

cal) factors versus biological factors. For example,

abiotic factors such as day length, rainfall, or tempera-

ture might constrain plants of different species to re-

produce relatively synchronously during a single fa-

vorable time of year. On the other hand, biological

factors such as competition among plants for a lim-

ited number of pollinators or seed dispersers should

select for nonoverlapping or staggered phenologies

281 Plant-Animal Interactions

S