Format: ORAL

Peter Chesson1,2

1 Department of Ecology and Evolutionary Biology, the University of Arizona, USA 2 Department of Life Sciences and Center for Global Change Biology, National Chung Hsing University, Taiwan

Arid regions support diverse floras of annual plants. Such regions are characterized by highly variable rainfall. The flexibility of the annual plant life cycle is suited to such variable conditions because annual plants can take advantage of short periods of rainfall to germinate, grow and set seed. A striking feature of such annual plant communities, however, is high levels of temporal variation in species composition. This variation can be traced to species-specific germination requirements, especially sensitivity to temperature when rain falls, which differs between species. More than 70 years ago it was suggested that temporal changes in composition have an important role in maintaining the species diversity of these systems. These annual plants are generally assumed to compete intensely for soil moisture and nutrients, but temporal variation in composition means that the various species are competing for these resources at different times. Many observations and experiments have confirmed these patterns, yet definitive tests that temporal variation maintains diversity is lacking. Such tests use covariance measures to test whether temporal patterns in germination and growth do, over time, intensify average intraspecific competition relative to interspecific competition, thus promoting species coexistence. However, such systems are commonly subject to high levels of seed predation from rodents, ants, and birds, which can limit competition, undermining temporal variation as a coexistence mechanism. Species-specific patterns of seed predation, however, provide an alternative diversity maintenance mechanism, yet studies of seed predation patterns are too limited to support this mechanism. Quantitative coexistence theory suggests observations, experiments and statistics that would provide strong tests of these hypotheses, and quantify their relative and absolute contributions to species coexistence. The evidence from the existing literature on these hypotheses will be reviewed, including evidence from biological invasions. Suitable studies for strong tests of the various hypotheses will be proposed.