TRANSMISSION DYNAMICS OF A NUCLEAR POLYHEDROSIS-VIRUS AND PREDICTING MORTALITY IN GYPSY-MOTH (LEPIDOPTERA, LYMANTRIIDAE) POPULATIONS

Mortality from nuclear polyhedrosis virus (NPV) among gypsy moth, Lymantria dispar (L.), neonates hatched from field-collected egg masses and larvae collected from the field was monitored at study sites in Massachusetts during 1983 through 1987. Mortality and prevalence of NPV were related to six potential predictor variables that could be obtained before egg hatch occurred in the field. These variables were calculated from egg mass density (EM Density), the number of larvae hatching from egg masses (Hatch) and the proportion of infection among hatching larvae (EM NPV). Logistic regression analyses suggest that transovum transmission is density-independent after an epizootic, whereas larva-to-larva transmission remains density-dependent. Differences may be due to the persistence of NPV on surfaces through which transmission is mediated. NPV prevalence and mortality were related to density estimates but not EM NPV or Hatch. An estimate of first instar density (L1 Density = EM Density x Hatch) provided the highest coefficient of determination, although an estimate of cadavers produced during the first wave of mortality (EM NPV x L1 Density) provided the best separation of epizootic from non-ezizootic sites. Because egg mass weight was strongly correlated with Hatch, a sampling protocol that measures egg mass density, egg mass weight, and mortality among larvae reared from field-collected egg masses may provide a reasonable predictor of subsequent NPV effect.