Introduction

All theories concerning metapopulation, source–sink, and metacommunity dynamics require dispersal, seed-bank dynamics, and seedling establishment to structure populations and communities (Hubbell, 2001; Leibold et al., 2004). Our understanding of dispersal (Howe & Smallwood, 1982; Nathan & Muller-Landau, 2000; Levine & Murrell, 2003), seed germination ecology (Baskin & Baskin, 1998; Fenner & Thompson, 2005), and seed-bank dynamics (Leck et al., 1989a), as well as the structure and dynamics of adult plants and communities, emphasize the need to bring the seedling life history stage to the fore. This is particularly important for the concept of recruitment limitation (Hurtt & Pacala, 1995).

Seedlings are clearly a vulnerable stage, shifting in short time periods from complete dependence on maternal reserves to physiological independence. Variation in seed size, carbon allocation patterns, and seedling structure and physiology has considerable influence on the potential for individual seedlings to survive to establishment. Dispersed across a variable habitat, mortality results from discordance in those characters and the environment, limiting potential establishment of seedlings. Their high mortality results from drought, herbivory, and disease (Moles & Westoby, 2006b; Fenner & Thompson, 2005; Kitajima, 2007), although many other factors also determine seedling success.

Reducing vulnerability may be accomplished by maternal investment in the seed that shepherds the seedling toward independence, or by facilitation by a nurse plant or other microhabitat (rock crack). Typically, the nurse plant is not the parent, raising the question of implications for community associations and long-term dynamics.