Pollinator limitation, cost of reproduction, and fitness in plants: A demographic approach

Many orchid species exhibit very low levels of fruit set. Because experimental hand pollination has resulted in a significant increase in fruit set relative to natural levels in many species, low fruit set in orchids has been generally explained in terms of pollinator limitation. However, recent studies have shown that experimentally increased fruit production may result in significant reductions in subsequent growth and reproduction. These results stressed the importance of quantifying the trade-off between reproduction and growth, and the effect of the trade-off on overall plant fitness. In this study, a 2-yr experimental assessment of the effects of pollination intensity and resource levels on future growth and flowering, and a 3-yr study of the population dynamics of the orchid Tolumnia variegata were combined in a matrix model for this species. Complete hand-pollination increased mean fruit set 80-100 times over the natural value, and resulted in a significant decrease in growth and flower production. Based on these experimental results, simulations were performed on the model, representing variation in schedules of pollination, recruitment, and resources, and the effects of the variation on overall fitness (defined as the population growth rate) were analyzed. The simulations predicted that a small increase in seedling production associated with the dramatic increase in fruit set would be sufficient to overcome the cost of fruiting, and would result in strong selection for increased fruit production. This conclusion does not explain the widespread occurrence of low fruit set in orchids. An alternative explanation is proposed. To test it, studies on the transition from seed to seedlings in nature are required.