Seed size affects the probability of seed predation. Large seeds should be preferred over small seeds but the selective responses of seed predators may also be frequency-dependent. Seed predators may prefer the most common seed sizes (apostatic selection), the rarest ones (antiapostatic selection) or even be unresponsive to the size of seeds. Moreover, seed density may further modify the selective responses of seed predators. We expect that at a low seed density seed consumption should be concentrated on common seed sizes, and at high seed density it should be concentrated on rare seed sizes, as common ones act as a background that makes rare phenotypes more conspicuous (the effect of background). We tested this prediction in a field experiment with seeds of Cryptocarya alba (Lauraceae) at La Campana National Park, central Chile. We presented large and small seeds in two contrasting seed densities (ten and 100 seeds per m2) and at five frequencies of large seeds: 0.1, 0.25, 0.5, 0.75 and 0.9. After 28 days we measured the proportions of the two size classes in the remaining seeds. Large seeds were always preferred to small seeds, even when they were at low frequency. Contrary to the predictions, at low seed density consumption was antiapostatic whereas at high seed density selection was independent of frequency. We discuss the causes and consequences of such selective responses expressed by seed predators.
- Apostatic selection
- Frequency-independent selection
- La Campana National Park, Chile
- Mediterranean shrubland
- Seed density
- Seed predation