Extinction in multispecies and spatially explicit models of habitat destruction.

proceeds, some species may be driven to extinction only Loss of habitat due to human activities is a major threat to reestablish in the community following more extensive to biological diversity (Ehrlich and Ehrlich 1981; Wilson habitat destruction that drives more competitive species 1988). Recent models of competition among sessile orto extinction. Thus, the models of Tilman et al. (1994, ganisms predict that as habitat is destroyed, the species 1997) and others allow a much richer collection of results that are first to become extinct will be those that are the than initially suspected. superior competitors (Nee and May 1992; Tilman et al. Second, Tilman et al.’s (1994, 1997) analytical model 1994, 1997). In these models, competition occurs among does not account for the spatial location of sites on the individuals for suitable sites that can be occupied by at landscape; dispersing propagules are equally likely to land most one individual. Species are ordered in a strict comin any site in the universe of habitat. Here, I investigate petitive hierarchy. If a propagule lands in a site occupied the consequences of spatial structure by analyzing the by an individual of a species lower in the hierarchy, it simple case in which dispersing propagules are more immediately replaces the inferior competitor. In order to likely to land in sites closer to their source. In this case, allow coexistence of many species, these models assume dispersal of propagules depends on both the number of a trade-off between competitive and colonization abilipropagules produced and the distance that they travel ties, with those species higher on the competitive hierarfrom their source. In contrast, spatially implicit models chy assumed to be less able to make propagules to colocharacterize the colonization ability of species only by the nize sites that are empty or occupied by an inferior number of propagules produced, because dispersal discompetitor. This trade-off allows coexistence, because the tance is effectively infinite. For the spatially explicit case, inferior competitors can survive as fugitive species. I assume that habitat destruction reduces the total size of The result that habitat destruction can lead to the excontiguous habitat but does not fragment the area of tinction of the best competitor is a consequence of the suitable habitat. I show that when there is a trade-off beassumed trade-off between competitive and colonization tween the number of propagules produced and competiability. Habitat destruction occurs through the loss of tive abilities, the consequence of habitat destruction is habitable sites, thereby decreasing the effective colonizasimilar to that exhibited by the spatially implicit model: tion rate of all species. This loss is felt most heavily by although the best competitor may become extinct first the poorest colonizers (the best competitors). With habiwith the reduction in the size of the area of suitable habitat destruction, the reduction in the effective colonization tat, this is not necessarily the case. Furthermore, even