Provisioning strategies, intraspecific parasitism and guild structure in solitary wasps (Hymenoptera: Sphecidae, Pompilidae)
Group-living, within the Hymenoptera, is restricted to taxa in which females return to their nests repeatedly with food for their offspring (multiple provisioning), increasing the opportunities for helping (e.g. by guarding the nest). Some solitary wasps provide only one large prey item per cell (single provisioning). Ammophila sabulosa (Sphecidae) females provision half of their cells singly and half multiply. The costs and benefits of these two alternatives are compared in terms of : a. Total weight of prey (caterpillars) provided. b. Time taken to capture and transport prey. c. The probability of intraspecific theft of prey from cells. d. The incidence of parasitism by miltogrammine flies. There is no evidence that either strategy is more successful than the other. An advantage of flexible provisioning is suggested by a study of Anoplius viaticus (Pompilidae), females of which provision all cells singly with spiders. Despite provisioning prey of a smaller size range than does Ammophila, total prey weight in Anoplius cells is more variable and dependent upon seasonal changes in prey availability. Ammophila females steal caterpillars from each others' nests, and both they and Anoplius females brood-parasitize cells of conspecifics. Intraspecific parasitism as an alternative nesting strategy is discussed and compared with work on other solitary wasps. Patterns of resource utilization are documented for the guild of spider-hunting pompilid wasps at one site. Data were collected on : a. Microhabitat utilization. b. Seasonal occurrence. c. Female size distributions. d. Prey species and sizes. Observed patterns of utilization and overlap are compared with 'neutral' models of guild structure in which various aspects of the species x resources utilized matrix are randomized. Some of the results are consistent with the idea that interspecific competition has partly determined pompilid guild structure, but problems of methodology and interpretation are discussed.