POLLINATING SEED EATERS - WHY IS ACTIVE POLLINATION SO RARE

Insects whose larvae depend on developing seed are very common, but ac tive pollination, meaning that the insect possesses specific structure s and behaviors for the purpose of assuring pollination, is only known to have evolved twice in such insects, namely in yucca moths and in f ig wasps. This rarity could be due to high cost of pollination, phylog enetic constraints, alternative life history shifts to reduce or avoid risk of seed nondevelopment, or ecological factors such as co-pollina tors that can satiate pollen requirements and mask variation in pollin ator effectiveness among ovipositing seed eaters. Ecological costs of being a pollinator were measured for a yucca moth species and were fou nd to be low: active time allocated to pollen pickup and deposit was o n average 4.1%, an average of 0.42% of female body mass was allocated to specific structures for pollen manipulation, and the average pollen load weighed <4% of moth body mass. These estimates suggest that ecol ogical costs need not be a major obstacle to evolution of active polli nation. In contrast, recent combined ecological-phylogenetic analyses for the yucca moth family suggest that the evolution of active pollina tion and transition to mutualism depended largely on preadaptations, a nd that few traits were truly novel. If general, active pollination wo uld be predicted to be likely to evolve only in lineages with life his tories that facilitate mutualism. Alternative outcomes to evolution of active pollination include delayed oviposition, detection of floral p ollination status, egg placement that allows the larva to select a fru it, and modified egg dispersion strategies to balance the cost of poll ination. The historical significance of these factors can be assessed only when mechanisms are documented in many lineages and analyzed in a phylogenetic framework. Active fungal inoculation among arthropods is ecologically analogous to active pollination, and offers a complement for comparative analyses. Specific structures for spore transport hav e evolved many times in at least three orders of insects, and several times in mites. The large number of independent lineages of active pol len and fungus dispersers jointly provide a platform for testing hypot heses about, e.g., the role of preadaptations in evolution of mutualis m, reversals of mutualism, and the role of mutualism in diversificatio n.