A stochastic model of parasitism of the mexican bean beetle, Epilachna varivestis Mulsant (Coleoptera: Coccinellidae), in soybeans by Pediobius foveolatus (Crawford) (Hymenoptera: Eulophidae)

The Mexican bean beetle (MBB), Epilachna varwestis MULSANT, has long been an important pest of snap beans and lima beans (Phaseolus spp.), and in recent years its importance as a pest of soybeans (Glycine max (L.) MERRILL) has increased, especially in the Southeastern U.S.A. and along the eastern coast as far north as Maryland and Delaware. This pest is essentially free of naturally-occurring parasitoids and diseases. All larval stages and adults cause defoliation of host plants. Although several species of arthropod predators feed on the eggs and larvae of MBB (WADDILL and SHEPARD 1974; WADDILL and SHEPARD 1975), predation is often ineffective in preventing economic loss by the pest. A computer simulation model of populations of MBB in soybeans (MEXSIM) has been developed and has predictive value in terms of identifying peak infestations (WADDILL et al. 1976). Certain biological information, such as MBB growth, development, overwintering, etc. in the soybean ecosystem, has been provided (BERNHARDT and SHEPARD 1978a, b) in order to refine the model. Releases of the imported parasitoid, Pediobius foveolatus (CRAWFORD) near the coastal areas of Maryland, have shown promise for suppression of MBB populations (STEVENS et al. 1975a), but no quantitative information is available relative to optimum numbers of parasitoids needed for reduction of MBB population density to subeconomic levels. This information is important because P. foveolatus does not overwinter in the U. S. A. Thus laboratory colonies of both MBB and P. foveolatus must be maintained during winter months and numbers of the parasitoid increased for eventual "field releases in the spring. Therefore, it is desirable to optimize the numbers of parasitoids