The pupal cocoons of two southern African wild silkmoth species, Gonometa postica and G. rufobrunnea (Lepidoptera: Lasiocampidae), are composed of high quality silk and have potential as a commercially viable resource. However, limited ecological research has been done on these species, and their population dynamics is especially poorly known. A steady and predictable supply of cocoons is paramount to the economic sustainability of a wild silk industry. There is thus an urgent need for documenting and understanding the population dynamics of southern Africa’s Gonometa species. Here, the temporal and spatial variation of pupal (and thus cocoon) abundance, as well as associated natural enemies, are described for both Gonometa species for the first time. The larval parasitoid species emerging from parasitised pupae were quantitatively associated with species-specific emergence holes, making field-identification of these species possible. Eleven sites in total were sampled, over four generations, across the region where both species have historically reached high population densities. Apparent spatial synchrony in pupal abundance found between sites suggests that climate is responsible for observed population size fluctuations. As predicted from their life history traits, temporal variability was lower than expected for classically eruptive species. Gonometa species thus have an intermediate position on the population dynamics gradient. In turn, the responses of natural enemies were not predictable from Gonometa species defensive traits, but appear to be mediated by between-species cocoon strength differences. Using data on the number of G. postica pupae per tree and associated parasitism at several sites, the importance of the degree of spatial explicitness in the quantification of aggregation and the detection of density dependence was illustrated. The spatially explicit method gave different results and more information regarding the spatial pattern of pupal abundance and parasitism than non- and semi-explicit methods. Similarly, the detection of density dependence in parasitism rates was affected by the use of spatially explicit data, with the spatial explicit approach giving different and more biologically informative results than traditional, non-spatially explicit methods. This has marked implications for previous insect-host - parasitoid studies aimed at detecting density dependence. The variability in cocoon size, a surrogate for larval performance, adult fecundity and silk yield, revealed that gender, followed by species, contributed most to observed size differences, with no clear differences between generations or localities. Finally, the between-host plant and withinhost plant distribution ofG. postica and G. rufobrunnea pupae was quantified, chiefly investigating the deterministic nature of the choice of pupation site. The distribution of both species at these scales was found to be markedly non-random, with pupae generally preferring specific tree characteristics and micro-sites. These results now provide the basis for recommending an appropriate utilisation strategy for southern Africa’s wild silk moths. Based on the spatial and temporal variability in pupal abundance observed, a constant and predictable cocoon supply for natural harvesting is unlikely. Long-term, broad-scale documentation of Gonometa species population cycles may make it possible to predict cocoon availability in the future. Until such research is done, it is recommended that the current practise of only collecting cocoons from which moths have emerged be continued. Simultaneously, artificial rearing and seeding as alternative utilisation strategies should be experimentally explored based on the information gathered and patterns identified here.
Thesis (DPhil (Entomology))--University of Pretoria, 2007.