The Life Cycle and Spatial Distribution of the Aquatic Parasitoid Agriotypus armatus (Hymenoptera: Agriotypidae) and its Caddis Host Silo pallipes (Trichoptera: Goeridae)

Abstract

(1) The life cycle of Silo pallipes (Fabricius) took one year in a small, stony stream. Adults were present chiefly in June and July; a few being taken in May, August, September. Oviposition occurred on large stones just below the water surface; eggs hatched within 19-35 days. The five larval instars built eleven stony-cases. Growth was rapid in the early instars; most larvae were in instar IV or V by November. Pupation commenced in spring; emerging adults were taken from May to August. (2) Changes in population density followed a similar pattern in each year-class. Density increased during egg hatching, reached a maximum in July, decreased exponentially at a fairly constant rate from July to May, and then decreased rapidly during pupation and emergence from June to August. Only 5% of the initial population in the egg stage reached the pupal stage. 67-68% of the pupal population emerged as adults, 22-23% died and 9-10% were parasitized by Agriotypus armatus Curtis. (3) The life cycle of A. armatus also took one year. Adults were present in May and June, and laid their eggs in the pupal cases of their host. The five larval instars fed externally on their living host and grew rapidly to the pupal stage by August. There was only one parasitoid per host. All parasitoids were in the adult stage by October, overwintered as resting adults within the pupal cases and emerged in spring. Only 4-7% of the initial population in the egg stage reached the adult stage in spring. (4) Three methods were used to analyse spatial distribution. The simple index of dispersion, tested by X2, was too insensitive to detect clumping in most samples. Although the negative binomial was fitted successfully to most samples, the statistic 1/ could not be used as an index of clumping because it either followed a curvilinear relationship with increasing population density (Silo larvae), or was highly variable (Silo pupae, Agriotypus). The third, and most successful, method was Taylor's power law which showed that the spatial distribution of both species was density dependent. Relative clumping remained fairly constant between sites, year-classes and larval instars of Silo, but increased markedly between the larval and pupal stages. Although the aquatic stages of Agriotypus were also clumped, the degree of relative clumping was less than that of their pupal host.