Pollen grains: why so many?

My objective is the examination of selective forces that affect pollen number. Relationships among other floral traits of animal-pollinated plants, including pollen sizer stigma area and depth. and the pollen-bearing area of the pollinator may affect pollen number and also provide a model to examine how change in one trait may elicit change in other traits. The model provides a conceptual framework for appreciating intra- and inter-specific differences in these traits. An equivalent model is presented for wind-pollinated plants. For these plants the distance between putative mates map be the most important factor affecting pollen number. I briefly consider how many pollen grains must reach a stigma to assure fruit set. I use pollen-ovule ratios (P/Os) to examine how breeding system, sexual system, pollen vector, and dispersal unit influence pollen grain number. I also compare the P/Os of plants with primary and secondary pollen presentation and those that provide only pollen as a reward with those that provide nectar as part or all of the reward. There is a substantial decrease in P/O from xenogamy to facultative xenogamy to autogamy. Relative to homoecious species the P/Os of species with most other sexual systems are higher. This suggests that there is a cost associated with changes in sexual system. The P/Os of wind-pollinated plants are substantially higher than those of animal-pollinated plants, and the available data suggest there is little difference in the pollination efficiency of the various animal vectors. The P/Os of plants whose pollen is dispersed in tetrads, polyads, or pollinia are substantially lower than these of species whose pollen is dispersed as monads. There was no difference in the P/Os of plants with primary and secondary pollen presentation. The P/Os of plants that provide only pollen as a reward were higher than those that provide nectar as a reward. All of these conclusions merit additional testing as they are based on samples that are relatively small and/or systematically biased.