Measurement of Host Curvature by The Parasitoid Wasp Trichogramma Minutum, and Its Effect on Host Examination and Progeny Allocation

ABSTRACT The parasitoid wasp Trichogramma minutum Riley (Hymenoptera: Chalcidoidea: Trichogrammatidae) adjusts its progeny allocation and host examining behaviour to the exposed volume of its host – an insect egg. The female wasp examines a potential host by walking over its surface while drumming its antennae against it, spending a longer time examining larger hosts. If wasps of different sizes are presented with glass beads of the same diameter, examining time increases significantly the larger the bead is relative to the wasp. Since the wasps’ velocities do not differ with wasp size, the total distance travelled during host examination also increases with relative host diameter. When large and small wasps are placed on large and small hosts such that the relative curvature is constant, examining time and total path length are the same. Smaller wasps take more steps to travel the same distance, therefore the duration of the examination walk, rather than step number or relative distance, is adjusted to host curvature. This adjustment of examination time may ensure that the surface of hosts of different sizes are examined adequately, despite differences in total area. Progeny allocation, usually dependent only on exposed host surface area, can be affected by the wasps’ measure of curvature. When hosts are mounted on the tips of thin wires rather than on the substrate, no finite cues dependent on surface area are available to the wasps. Such hosts receive only 25 % more progeny, either in the light or in total darkness, than hosts which are resting on the substrate, demonstrating that an upper limit to clutch size is set by cues dependent on curvature. The structure of the compound eyes suggests that the visual acuity of Trichogramma is insufficient to account for their ability to discriminate between small differences in host curvature. By placing large and small wasps on point-mounted hosts of fixed diameter, we show that the clutch size limit is set according to the relative size of the host. These observations are discussed in terms of a mechanosensory mechanism of curvature measurement.