Tropical Forest Community Ecology

Treefall Gaps and Plant Species Diversity in Forests 203

will determine the percentage that require gaps to

persist in tropical forests.

In addition, most studies have failed to consider

the per capita impact of gaps on size- or age-

specific rates of reproduction of species. For exam-

ple, if gaps increase light and allow shade-tolerant

trees to become reproductive or produce more

seeds while still in the understory, then they may

promote diversity by increasing fecundity, even if

per capita species diversity is not higher than in

non-ga psites (Schnitzer 2001). This as pect of the

ga phy pothesis has been neglected and may be

particularly relevant to herbs, herbaceous vines,

shrubs, and mid-sized trees – groups that do not

typically reach the canopy, but may depend on

treefall gaps to initiate reproduction (Gentry and

Dodson 1987, Levey 1988, Denslow 1990, Dirzo

et al. 1992, Goldblum 1997, Schnitzer and Carson

2000). For example, the fecundity of forest herbs

and shrubs may be substantially higher in gaps

than in nearby intact forest (Levey 1988, Denslow

1990, Dirzoet al. 1992, Goldblum 1997). On

BCI, these groups of understory plants consti-

tute around one third of the vascular plant flora

(Figure 12.4); when combined with lianas and

pioneer tree species, they represent 65% of all

plant species on BCI and the majority of the flora

in tropical forests worldwide (Gentry and Dodson

1987). Thus, gaps may maintain the majority of

22%

12%

11%

11%

10%

Forest herbs

Shrubs

Herbaceous

vines

34%

Shade-tolerant

trees

Pioneer

trees

Lianas

Figure 12.4 Percentage of species in different
vascular plant groups on Barro Colorado Island,
Panama. Data originally from Croat (1978) and
summarized in Schnitzer and Carson (2000).

the flora in many tropical forests when both repro-

duction and diversity are examined.To adequately

determine the role of treefall gaps in maintaining

species diversity, both the growth and survival of

species as well as their reproductive output must

be considered.

Shade-tolerant trees, lianas, and

treefall gaps

Although there are still only a handful of relevant

studies, gaps do not appear to have a strong

influence on the per area or per capita diversity

of shade-tolerant tree species. Typically, shade-

tolerant trees establish prior to gap formation

and are present as advance regeneration; thus,

processes that occur prior to gap formation prob-

ably determine the composition and abundance

of species that are available to take advan-

tage of a newly formed ga p(Uhlet al. 1988,

Brown and Whitmore 1992). Additionally, many

shade-tolerant tree species are limited by low seed

productionorpoordispersal,andthustheycannot

distribute sufficient propagules into newly formed

treefall gaps to take advantage of these ephemeral,

high resource environments (see Dallinget al.

1998, Hubbellet al. 1999, Brokaw and Busing

2000, Muller-Landau Chapter 11, this volume).

Finally, shade-tolerant tree abundance and diver-

sity may be reduced in treefall gaps if gap-

phase regeneration is co-opted by lianas or palms

(Schnitzeret al. 2000).

Shade-tolerant trees may have structural

characteristics that make them particularly sus-

ceptible to competition from lianas. Shade-

tolerant tree species grow slowly and are adapted

to maximize light interception by producing many

branches, which can act as trellises for lianas,

allowing them to climb and sometimes smother

shade-tolerant trees under a blanket of foliage.

Conversely, pioneer trees and palms have char-

acteristics that may allow them to shed or avoid

lianas, such as rapid growth, smooth or peel-

ing bark, and an unbranched, monopodial trunk

(Putz 1984b). While the severity of liana com-

petition may vary with tree species identity, with

lianas affecting some tree species or guilds more

than others (e.g., Putz 1984a, Peréz-Salicru pand