Abstract
A number of lines of evidence indicate that long-lived plants have higher per-generation mutation rates than short-lived plants. This is believed to be a consequence of an increased contribution of somatic mutation, because gametes are produced from cell lineages that have undergone hundreds to hundreds of thousands of potentially imperfect mitoses prior to meiosis. Although this expectation is not itself novel, the consideration of its consequences for reproductive responses of trees has received little attention.Discussions in the literature of trees' evolutionary responses to increased selfing rates have generally assumed that a tree population's response would be similar to that observed in short-lived plants, in that an increase in population progeny selfing rate results in an increase in the proportion of inbred adults in the population. However, it is shown here that this is not likely to be true. Consequently, selfed progeny in nearly all tree populations may have no reproductive value.If somatic mutation is indeed important in trees, then its rate should be estimable. A quantitative model is developed that predicts somatic mutation rates from relative fitness of selfed progeny created via different methods of hand self-pollination. The model is based on the hypothesis that, with somatic mutation, the fitness of progeny produced by self-fertilization of a flower with its own pollen will be lower than the fitness of progeny produced by self-fertilization of a flower with pollen from a different flower on a separate primary branch. Despite extensive fieldwork involving the large tropical tree Ceiba pentandra, the flower position hypothesis could not be adequately tested.The now-classic Bateman-Mukai technique was modified to estimate per-cell division somatic mutation parameters, and was applied to pollen fitness measures along branches of the tropical tree Delonix regia. The estimated minimum per-cell division somatic mutation rate was 10-4x to 10-5x the magnitude of maximum per-meiosis mutation rates in plants. Although few data are available for synthesis, the somatic mutation rate appears to be two or more orders of magnitude lower in plants than in animals, supporting the hypothesis that plants experience strong selection to reduce the inevitable effects of somatic mutation.