A Model for Fruiting Display: Seed Dispersal by Birds for Mulberry Trees

The effectiveness of fruiting displays of mulberry trees (Morus) for seed dispersal by birds was studied in eastern Kansas. Alternative models were developed in which the number of fruits eaten by birds and the number of seeds dispersed from a tree were hypothesized to be functions of the number of fruits on a tree, reward energy in an individual fruit, and total reward energy on the tree. The null hypothesis was that birds eat fruit randomly on trees regardless of these numerical or energetic characteristics. The models predict an increase in seeds dispersed with an increase in fruiting display, with each of the three models predicting a different behavioral response to increases in fruiting display. When empirical data were analyzed on a weekly basis, the null hypothesis was rejected and total reward energy on the tree did a much better job of explaining the number of fruits eaten by birds and the number of seeds dispersed than did number of fruits on a tree or reward energy per fruit. The latter two variables are components of the first, more significant, variable. In successive weeks smaller displays resulted in the same number of seeds being dispersed. Decreasing availability of insect food probably is at least partially responsible for this increase in fruiting display efficiency. Mulberry trees are monoecious with highly variable sex ratios, and the results predict selection for reproduction at an early age in Morus and equal fitness for all sex ratios.