Are plant populations seed-limited? A review of seed sowing experiments.

We define seed limitation to be an increase in population size following seed addition. Here, we briefly consider how theoretical models deal with seed limitation and how seed sowing experiments can be used to unravel the extent of seed limitation in natural systems. We review two types of seed addition experiments: seed augmentation studies where seeds are added to existing populations; and seed introductions where seeds are sown in unoccupied sites. Overall, approximately 50% of seed augmentation experiments show evidence of seed limitation. These studies show that seed limitation tends to occur more commonly in early successional habitats and in early successional species. Most of the studies have concentrated on simply categorising populations as seed- or microsite-limited, but we believe that seed sowing experiments could be used to reveal much more about community structure, and we discuss possible future directions. In 53% of introduction studies (where seeds were sown at sites from which the species was known to be absent) the introduced species was recorded in at least one of the experimental sites following sowing. However, of the subset of studies where both seedlings and adult plants were recorded, 64% of sites contained seedlings while only 23% contained adults. This implies that, for many species, conditions for establishment are more stringent than conditions for germination. The successful establishment of plants in unoccupied patches indicates the potential for immigration to enhance local diversity (the spatial mass effect). Few studies continued monitoring for long enough to determine whether or not self-sustaining populations were successfully established, and no study attempted to link introduction sites to a putative natural source of propagules, or considered the dynamics of the metapopulation as a whole.