Baculoviruses are fatal, infectious diseases that drive population cycles in many forest insects. In these insects, baculoviruses are species-specific and are transmitted when larvae accidentally consume foliage that contains virus-infected cadavers of conspecifics. Laboratory studies have shown that gypsy moth larvae can detect and avoid cadavers infected with the baculovirus, LdNPV, but the effects of this and other behaviors on infection risk in the field are poorly known.

To understand how avoidance affects the risk of infection, I first conducted a field experiment in which individual larvae were challenged with leaves infected with cadavers. I recorded the distance eaten to a cadaver and the infection outcome for all individuals. In the lab, as a preliminary test of whether there is a genetic component to feeding behavior, I compared the avoidance behavior of larvae from full-sibling groups. Within each group, half were fed clean disks and half were fed virus-infected disks. Larvae fed for 10 hours and feeding progress was documented at six time intervals. A mixed effects model was used to determine if the leaf area eaten varied among larval families.

Results/Conclusions

The field data show that infection risk dramatically decreases as the distance eaten to an LdNPV-infected cadaver increases (p = 0.027, df = 68). The spatial scale of infection is thus small enough that it is at least possible that larval feeding behavior, in terms of cadaver avoidance, can affect the risk of infection. In the lab experiment, the best statistical model, as chosen by AIC analysis includes effects of full-sibling family, suggesting that the behavior may be heritable. These results indicate that feeding behavior has the potential to play an important role in gypsy moth-virus dynamics. Current work involves the creation of an individual based model to better understand how variability in larval feeding behavior shapes viral epidemics.