Seed Shadows: a Landscape View of Demographic and Genetic Effects

A seed shadow is the spatial pattern of seed distribution relative to parent trees and other conspecifics; it results from the process of seed dispersal and represents the starting template for plant regeneration. Janzen (1970) and Connell (1971) consider it the population recruitment surface. For animal-dispersed, endozoochorous species the seed shadow results primarily from movement patterns of frugivores. Presumably, frugivores can dramatically affect both the demography and genetic make-up of animal-dispersed plant species. These effects, however, have rarely been documented in an integrated way. In this chapter we focus on how frugivores influence the number and spatial pattern of propagules that reach the soil, and their simultaneous influence on gene flow via seed dispersal. We advocate an integrated view of both demographic and genetic effects to understand the role of frugivores on plant recruitment (Alvarez-Buylla et al., 1996). Given that multiple influences sequentially alter after this initial effect of frugivores (i.e. post-dispersal seed predation, germination, seedling mortality), we need to quantitatively assess the relative importance of dispersal by frugivores for plant population biology. Seed dispersal by frugivores is the link in the demographic transition between the ripe fruit crop on the trees and, after delivery, the subsequent stages of establishment of germinated seeds, seedlings, saplings and established adults, i.e. the whole recruitment cycle. Thus, seed dispersal may play a pivotal role in the demography of plant populations (Harper, 1977) by simultaneously influencing not only the numerical dynamics of recruitment from dispersal to establishment, but also the genetic make-up of the seed shadow. The difficulty of tracking the origin of frugivore-dispersed seeds has precluded a robust analysis of vertebrate seed dispersal (Levey and Sargent, 2000). Indeed, the difficulties in measuring and analysing the dispersal of seeds in natural communities has been considered an unavoidable limitation of the field (Wheelwright and Orians, 1982). Recent developments in molecular biology (Carvalho, 1998), however, have resulted in a series of molecular tools based on DNA analysis that allow analysis of gene-flow patterns via seed dispersal (Ouborg et al., 1999) and the statistical analysis of the resulting patterns of genetic structure (Schnabel et al., 1998a; Luikart and England, 1999). More specifically, for animal-dispersed species, gene flow via seeds can be estimated