Pathogen-Driven Outbreaks in Forest Defoliators Revisited: Building Models from Experimental Data

Models of outbreaks in forest-defoliating insects are typically built from a priori considerations and tested only with long time series of abundances. We instead present a model built from experimental data on the gypsy moth and its nuclear polyhedrosis virus, which has been extensively tested with epidemic data. These data have identified key details of the gypsy moth-virus interaction that are missing from earlier models, including seasonality in host reproduction, delays between host infection and death, and heterogeneity among hosts in their susceptibility to the virus. Allowing for these details produces models in which annual epidemics are followed by bouts of reproduction among surviving hosts and leads to quite different conclusions than earlier models. First, these models suggest that pathogen-driven outbreaks in forest defoliators occur partly because newly hatched insect larvae have higher average susceptibility than do older larvae. Second, the models show that a combination of seasonality and delays between infection and death can lead to unstable cycles in the absence of a stabilizing mechanism; these cycles, however, are stabilized by the levels of heterogeneity in susceptibility that we have observed in our experimental data. Moreover, our experimental estimates of virus transmission rates and levels of heterogeneity in susceptibility in gypsy moth populations give model dynamics that closely approximate the dynamics of real gypsy moth populations. Although we built our models from data for gypsy moth, our models are, nevertheless, quite general. Our conclusions are therefore likely to be true, not just for other defoliator-pathogen interactions, but for many host-pathogen interactions in which seasonality plays an important role. Our models thus give qualitative insight into the dynamics of host-pathogen interactions, while providing a quantitative interpretation of our gypsy moth-virus data.     

Competition between the gypsy moth, Lymantria dispar, and the northern tiger swallowtail, Papilio canadensis: interactions mediated by host plant chemistry, pathogens, and parasitoids

The gypsy moth, Lymantria dispar, and the northern tiger swallowtail, Papilio canadensis, overlap geographically as well as in their host ranges. Adult female swallowtails are incapable of distinguishing between damaged and undamaged leaves, and the opportunities for competition between these two species are numerous. We designed field and laboratory experiments to look for evidence of indirect competition between P. canadensis and L. dispar larvae. Swallowtail caterpillars were reared in the laboratory on leaves from gypsy-moth-defoliated and undefoliated trees to explore host-plant effects. We tested for pathogen-mediated interactions by rearing swallowtail larvae on both sterilized and unsterilized leaves from defoliated and undefoliated sources. In addition, we measured the effects of known gypsy moth pathogens, as well as gypsy moth body fluids, on the growth and survival of swallowtail larvae. Field experiments were designed to detect the presence of parasitoid-mediated competition, as well: we recorded parasitism of swallowtail caterpillars placed in the field either where there were no gypsy moth larvae present, or where we had artificially created dense gypsy moth populations. We found evidence that swallowtails were negatively affected by gypsy moths in several ways: defoliation by gypsy moths depressed swallowtail growth rate and survival, whether leaves were sterilized or not; sterilization significantly reduced the effect of defoliation, and gypsy moth body fluids proved lethal; and swallowtail caterpillars suffered significantly increased rates of parasitism when they were placed in the field near gypsy moth infestations.     

Effects of host heterogeneity on pathogen diversity and evolution

Phenotypic variation is common in most pathogens, yet the mechanisms that maintain this diversity are still poorly understood. We asked whether continuous host variation in susceptibility helps maintain phenotypic variation, using experiments conducted with a baculovirus that infects gypsy moth (Lymantria dispar) larvae. We found that an empirically observed tradeoff between mean transmission rate and variation in transmission, which results from host heterogeneity, promotes long‐term coexistence of two pathogen types in simulations of a population model. This tradeoff introduces an alternative strategy for the pathogen: a low‐transmission, low‐variability type can coexist with the high‐transmission type favoured by classical non‐heterogeneity models. In addition, this tradeoff can help explain the extensive phenotypic variation we observed in field‐collected pathogen isolates, in traits affecting virus fitness including transmission and environmental persistence. Similar heterogeneity tradeoffs might be a general mechanism promoting phenotypic variation in any pathogen for which hosts vary continuously in susceptibility.     

Host tree influences on the dispersal of first instar gypsy moths, Lymantria dispar (L.)

Abstract. 1. In laboratory tests, first instar gypsy moths attempted dispersal more frequency when exposed to less acceptable foliage.
2. First instars from small eggs attempted dispersal less frequently than larvae from large eggs when exposed to foliage from highly acceptable or marginally acceptable hosts. Dispersal rates of larvae from medium sized eggs were intermediate.
3. These results (1–2) confirm and expand upon the findings of Capinera & Barbosa (1976).
4. In the field, data on the relative densities of larvae on different host species support the conclusion that the frequency of dispersal attempts is inversely related to host acceptability.
5. The implications of these findings for the population dynamics of the gypsy moth are discussed.     

Effects of resource quality on the population dynamics of the Indian meal moth Plodia interpunctella and its granulovirus

Although the Plodia interpunctella-granulovirus system is one of the most studied models for insect-pathogen interactions, there are relatively few precise data on the dynamics of the virus in coexisting populations of these two organisms. Previous work has suggested that resource quality, in terms of the diet supplied to P. interpunctella, has a strong effect on the population dynamics of host and pathogen. Here we investigate the impact of resource-dependent host patterns of abundance on pathogen dynamics and prevalence. In the laboratory, three populations of P. interpunctella feeding on a good quality food and infected with a granulovirus were compared with three populations also infected with a granulovirus but feeding on poor quality food. Populations feeding on good quality food produced larger adult moths, and had greater numbers of adult moths, healthy larvae, and virus-infected larvae. A higher proportion of larvae in these good quality populations were infected with virus, and adult moths exhibited cyclic fluctuations in abundance, unlike those on poor quality food. This cyclic behaviour was shown to be associated with cycles in the age structure of the larval population. Previous theoretical work suggests that these cycles may result from asymmetric competition between young and old larvae. Cyclic fluctuations in the proportion of infected larvae, that occurred on good, but not on poor quality food, were also shown to be related to cycles in the age structure of the larval population.     

Should models of disease dynamics in herbivorous insects include the effects of variability in host-plant foliage quality?

Interactions between insects and their baculovirus pathogens are often described using simple disease models. Baculoviruses, however, are transmitted when insects consume virus-contaminated foliage, and foliage variability, whether within or between host-plant species, can affect viral infectiousness. Insect-baculovirus interactions may thus be embedded in a tritrophic interaction with the insect's host plant, but disease models include only the host and the pathogen. We tested these models by measuring the transmission of a baculovirus of gypsy moths (Lymantria dispar) on red oak (Quercus rubra) and white oak (Quercus alba) in the field in six experiments over four years. In all experiments, there were only weak effects of host-tree species, and in only one did the best-fitting model include tree species effects. These weak effects of foliage variability on transmission were not due to a lack of foliage variability on viral infectiousness, because when larvae were force-fed virus-contaminated foliage, infection rates were higher on white oak. Our results suggest that feeding behavior plays an important role in baculovirus transmission and that models can usefully describe baculovirus dynamics even without including foliage variability. Our work provides a clear example of how two-species models are sometimes sufficient to describe what appear to be tritrophic interactions.     

Interpopulation quality in gypsy moths with implications for success of two pupal parasitoids: Brachymeria intermedia (Nees) and Coccygomimus turionellae (L.)*

Abstract.

• 1 Gypsy moth egg masses were collected from innocuous, release and outbreak populations and reared in the laboratory on synthetic diet under identical conditions.
• 2 Outbreak population gypsy moths hatched sooner, were smaller and less fecund than innocuous or release gypsy moths, but had a higher concentration of total carbohydrates in their haemolymph.
• 3 Pupae from each population source were submitted to parasitization by two pupal parasitoids. Emerging B.intermedia, an established parasitoid of the gypsy moth associated with outbreak populations, were largest on outbreak source gypsy moths. C. turionellae, not a usual parasitoid of the gypsy moth, were largest when emerging from innocuous or release population gypsy moths. Implications for population dynamics of the gypsy moth are discussed.

     

A field release of genetically engineered gypsy moth (Lymantria dispar L.) nuclear polyhedrosis virus (LdNPV)

The gypsy moth (Lymantria dispar L.) nuclear polyhedrosis virus was genetically engineered for nonpersistence by removal of the gene coding for polyhedrin production and stabilized using a coocclusion process. A beta-galactosidase marker gene was inserted into the genetically engineered virus (LdGEV) so that infected larvae could be tested for its presence using a colorimetric assay. In 1993, LdGEV-infected gypsy moths were released in a forested plot in Massachusetts to test for spread and persistence. A similar forested plot 2 km away served as a control. For 3 years (1993-1995), gypsy moths were established in the two plots in Massachusetts to serve as test and control populations. Each week, larvae were collected from both plots. These field-collected larvae were reared individually, checked for mortality, and then tested for the presence of beta-galactosidase. Other gypsy moth larvae were confined on LdGEV-contaminated foliage for 1 week and then treated as the field-collected larvae. The LdGEV was sought in bark, litter, and soil samples collected from each plot. To verify the presence of the LdGEV, polymerase chain reaction, slot blot DNA hybridization, and restriction enzyme analysis were also used on larval samples. Field-collected larvae infected with the engineered virus were recovered in the release plot in 1993, but not in subsequent years; no field-collected larvae from the control plot contained the engineered virus. Larvae confined on LdGEV-contaminated foliage were killed by the virus. No LdGEV was recovered from bark, litter, or soil samples from either of the plots.     

Prey‐processing by avian predators enhances virus transmission in the gypsy moth

The Lymantria dispar nucleopolyhedrovirus (LdNPV) is one of the most important regulators of gypsy moth populations, but some aspects of its transmission remain poorly understood, particularly its high rate of spatial spread and ability to persist in low‐density populations. We tested the role of predatory birds in the transmission of this virus using experimental gypsy moth populations in an aviary. Predatory birds captured virus‐infected caterpillars and facilitated viral dispersal via two processes: 1) by ingesting infected caterpillars and passing viral occlusion bodies (OBs) through their guts, and 2) by scattering OBs during predator‐specific processing behaviors, a mechanism documented here for the first time. The relative importance of both pathways differed by predator species. After eating virus‐infected gypsy moth larvae, red‐eyed vireos and black‐capped chickadees passed more gypsy moth nucleopolyhedrovirus in feces than did gray catbirds. During prey‐processing, the repetitive beating of caterpillars by red‐eyed vireos, a behavior that was rarely utilized by chickadees and catbirds, resulted in the scattering of infectious virus. Due to the combination of efficient gut passage and virus spread from prey beating, higher rates of transmission occurred in experimental gypsy moth populations exposed to red‐eyed vireos than those exposed to catbirds or chickadees. Our results show that effective virus transmission was achieved when virus was vectored by predatory birds through a combination of both behavioral and physiological traits.     

Forest edges enhance mate‐finding in the invasive European gypsy moth,Lymantria dispar

Habitat type, fragmentation, and edge effects can play important roles in the mate‐finding abilities of many species. These effects can be particularly pronounced in low‐density populations, which are often found at the margins of species' ranges or at the leading edge of an invasion. The European gypsy moth, Lymantria dispar (L.) (Lepidoptera: Erebidae), is a non‐native insect defoliator in the USA and Canada, where flightless females attract male moths through pheromone production and local extirpation of low‐density populations can be due to mate‐finding failure. To assess the effects of habitat edges on the ability of gypsy moths to find mates, we conducted a release experiment with male gypsy moths using female‐baited trap arrays in fields, at forest edges, and in the forest interior. Reduced mate‐finding was expected in fields and near forest edges based on geographic variation in invasion rates, male flight behavior, and pheromone plume dynamics. However, we found that mate‐finding was highest at forest edges, reduced in fields, and lowest within the forest interior. Within an array, traps closest to the forest edge also had the highest mate‐finding, suggesting that habitat characteristics can influence male flight direction in addition to pheromone cues. These results suggest that a moderate level of forest fragmentation enhances mate‐finding ability in the gypsy moth. Understanding the relationship between habitat heterogeneity and mate‐finding success in invasive species can inform predictions of future spread and assist with management plans that target mating disruption.     

Induced plant defenses breached? Phytochemical induction protects an herbivore from disease

Although wound-induced responses in plants are widespread, neither the ecological nor the evolutionary significance of phytochemical induction is clear. Several studies have shown, for example, that induced responses can act against both plant pathogens and herbivores simultaneously. We present the first evidence that phytochemical induction can inhibit a pathogen of the herbivore responsible for the defoliation. In 1990, we generated leaf damage by enclosing gypsy moth larvae on branches of red oak trees. We then inoculated a second cohort of larvae with a nuclear polyhedrosis virus (LdNPV) on foliage from the damaged branches. Larvae were less susceptible to virus consumed on foliage from branches with increasing levels of defoliation, and with higher concentrations of gallotannin. Defoliation itself was not related to any of our chemistry measures. Field sampling supported the results of our experiments: death from virus among feral larvae collected from unmanipulated trees was also negatively correlated with defoliation. In 1991, defoliation and gallotannin were again found to inhibit the virus. In addition, gallotannin concentrations were found to be positively correlated with defoliation the previous year. Compared with previous results that demonstrated a delecterious effect of induction on gypsy moth pupal weight and fecundity, the inhibition of the virus should confer an advantage to the gypsy moth. Since leaf damage levels increase as gypsy moth density increases, and since leaf damage inhibits the gypsy moth virus, there is the potential for positive feedback in the system. If phytochemical induction in red oak can inhibit an animal pathogen such as LdNPV, it suggests to us that induction in red oak is a generalized response to tissue damage rather than an adaptive defense against herbivores.     

Effects of phenotypic plasticity on pathogen transmission in the field in a Lepidoptera-NPV system

In models of insect–pathogen interactions, the transmission parameter (ν) is the term that describes the efficiency with which pathogens are transmitted between hosts. There are two components to the transmission parameter, namely the rate at which the host encounters pathogens (contact rate) and the rate at which contact between host and pathogen results in infection (host susceptibility). Here it is shown that in larvae of Spodoptera exempta (Lepidoptera: Noctuidae), in which rearing density triggers the expression of one of two alternative phenotypes, the high-density morph is associated with an increase in larval activity. This response is likely to result in an increase in the contact rate between hosts and pathogens. Rearing density is also known to affect susceptibility of S. exempta to pathogens, with the high-density morph showing increased resistance to a baculovirus. In order to determine whether density-dependent differences observed in the laboratory might affect transmission in the wild, a field trial was carried out to estimate the transmission parameter for S. exempta and its nuclear polyhedrosis virus (NPV). The transmission parameter was found to be significantly higher among larvae reared in isolation than among those reared in crowds. Models of insect–pathogen interactions, in which the transmission parameter is assumed to be constant, will therefore not fully describe the S. exempta-NPV system. The finding that crowding can influence transmission in this way has major implications for both the long-term population dynamics and the invasion dynamics of insect–pathogen systems. Received: 14 June 1999 / Accepted: 22 March 2000     

Horizontal and vertical transmission of wild-type and recombinant Helicoverpa armigera single-nucleocapsid nucleopolyhedrovirus

Transmission plays a central role in the ecology of baculoviruses and the population dynamics of their hosts. Here, we report on the horizontal and vertical transmission dynamics of wild-type Helicoverpa armigera single-nucleocapsid nucleopolyhedrovirus (HaSNPV-WT) and a genetically modified variant (HaSNPV-AaIT) with enhanced speed of action through the expression of an insect-selective scorpion toxin (AaIT). In caged field plots, horizontal transmission of both HaSNPV variants was greatest when inoculated 3rd instar larvae were used as infectors, transmission was intermediate with 2nd instar infectors and lowest with 1st instar infectors. Transmission was greater at a higher density of infectors (1 per plant) than at a lower density (1 per 4 plants); however, the transmission coefficient (number of new infections per initial infector) was lower at the higher density of infectors than at the lower density. HaSNPV-AaIT exhibited a significantly lower rate of transmission than HaSNPV-WT in the field cages. This was also the case in open field experiments. In the laboratory, the vertical transmission of HaSNPV-AaIT from infected females to offspring of 16.7+/-2.1% was significantly lower than that of HaSNPV-WT (30.9+/-2.9%). Likewise, in the field, vertical transmission of HaSNPV-AaIT (8.4+/-1.1%) was significantly lower than that of HaSNPV-WT (12.6+/-2.0%). The results indicate that the recombinant virus will be transmitted at lower rates in H. armigera populations than the wild-type virus. This may potentially affect negatively its long-term efficacy as compared to wild-type virus, but contributing positively to its biosafety.     

Determining the major Bt refuge crops for cotton bollworm in North China

Evaluation of the effectiveness of refuge strategies involved in cotton bollworm Bt resistance management would be aided by technologies that allow monitoring and quantification of key factors that affect the process under field conditions. We hypothesized that characterization of stable carbon and nitrogen isotopes in adult bollworm, Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) moths may aid in determining the larval host that they developed upon. We found moths reared from larvae fed on peanut, soybean or cotton, respectively, could be differentiated using isotopic analyses that also corresponded to their respective larval host origins. These techniques were then used to classify feral second‐generation bollworm moths caught in Bt cotton (Gossypium hirsutum) fields into different populations based on their isotopic signatures. In 2006–2007 feral moths captured in Bt cotton fields predominantly correlated with the peanut (Arachis hypogea) having served as their larval host, indicating this is the most important refuge crop for Bt‐susceptible bollworm individuals (providing 58%?64% individuals) during independent moth peaks for the second generation in North China. The remaining feral moths correlated with soybean (Glycine max) (0?10%); other C₃ plant (20%?22%) and non‐C₃ plant (12%?14%) host types also provided some Bt‐sensitive moths. Field observations showed that peanut constitutes the primary refuge crop contributing to sustaining Bt‐susceptible moths dispersing into cotton in North China. These results suggest that peanut may be a more effective refuge to sustain Bt‐susceptible bollworm individuals and reduce the risk of development of a Bt‐resistant biotype.     

What causes outbreaks of the gypsy moth in North America?

The gypsy moth has been present in North America for more than 100 years, and in many of the areas where it has become established outbreaks occur with varying degrees of periodicity. There also exists extensive spatial synchrony in the onset of outbreaks over large geographic regions. Density-dependent mortality clearly limits high-density populations, but there is little evidence for strong regulation of low-density populations. Predation by small mammals appears to be the major source of mortality affecting low-density populations, but because these are generalist predators and gypsy moths are a less preferred food item, mammals do not appear to regulate populations in a density-dependent fashion. Instead, predation levels appear to be primarily determined by small mammal abundance, which is in turn closely linked to the production of acorns that are a major source of food for overwintering predator populations. Mast production by host oak trees is typically variable among years, but considerable spatial synchrony in masting exists over large geographic areas. Thus, it appears that the temporal and spatial patterns of mast production may be responsible for the episodic and spatially synchronous behavior of gypsy moth outbreaks in North America. This multitrophic relationship among mast, predators, and gypsy moths represents a very different explanation of forest insect outbreak dynamics than the more widely applied theories based upon predator–prey cycles or feedbacks with host foliage quality. Received: September 8, 1999 / Accepted: September 20, 2000     

Testing a key assumption of host‐pathogen theory: density and disease transmission

Conventional disease theory suggests that extinction with density‐dependent transmission is unlikely as the threshold host density (KT) is greater than zero. Extinction may result if transmission is frequency dependent or the pathogen has an environmental reservoir. Given the importance of understanding how pathogens affect species richness and diversity there are few empirical tests of these conclusions. We used an Ambystoma tigrinum–Ambystoma tigrinum virus (ATV) model system in the laboratory to examine disease transmission dynamics. Susceptible A. tigrinum larvae were exposed to three different densities and proportions of infected larvae for 24 h. We then housed susceptible hosts individually for 28 days and monitored them for infection. The density of infected hosts to which susceptible hosts were exposed was the best predictor of infection (p=0.037). There was no effect of host clutch on the probability of becoming infected (p=0.67). Larvae in the highest density treatments died sooner than larvae in lower density treatments (p<0.001). Asymptomatic but infected hosts shed sufficient virus into the water in a 24‐h period to infect susceptible hosts without any direct contact between individuals. ATV transmission was best described by a power function, leading to the prediction that extinction of A. tigrinum as a result of this pathogen is unlikely. Indeed, field observations show that larval salamander populations that experience ATV‐driven epidemics may decrease, but not to extinction, and then recover. Disease is proposed as a possible explanation for the global decline of amphibians. Ranaviruses infect many amphibian populations, but based on our results may not be a general cause of declines to extinction. In contrast, frequency dependent transmission, environmental reservoirs and alternative hosts may be the most likely explanation for the enigmatic decline, at times to extinction, of some amphibian populations as a result of emerging infectious diseases, like the chytrid fungus Batrachochytrium dendrobatidis.     

Parasitoid-mediated transmission of an iridescent virus

We examined the interaction between an invertebrate iridescent virus (IIV) isolated from Spodoptera frugiperda (J.E. Smith) and the solitary ichneumonid endoparasitoid Eiphosoma vitticolle Cresson. In choice tests, parasitoids examined and stung significantly more virus infected than healthy larvae, apparently due to a lack of defense reaction in virus infected hosts. Parasitoid-mediated virus transmission was observed in 100% of the female parasitoids that stung a virus infected host in the laboratory. Each female parasitoid transmitted the virus to an average (+/-SE) of 3.7+/-0.3 larvae immediately after stinging an infected larva. Caged field experiments supported this result; virus transmission to healthy larvae only occurred in cages containing infected hosts (as inoculum) and parasitoids (as vectors). The virus was highly detrimental to parasitoid development because of premature host death and lethal infection of the developing endoparasitoid. Female parasitoids that emerged from virus infected hosts did not transmit the virus to healthy hosts. We suggest that the polyphagous habits of many noctuid parasitoids combined with the catholic host range of most IIVs may represent a mechanism for the transmission of IIVs between different host species in the field.     

Horizontal and vertical transmission of a Nosema sp. (Microsporidia) from Lymantria dispar (L.) (Lepidoptera: Lymantriidae)

The gypsy moth, Lymantria dispar L. (Lepidoptera, Lymantriidae), a serious defoliator of deciduous trees, is an economically important pest when population densities are high. Outbreaking populations are, however, subject to some moderating influences in the form of entomopathogens, including several species of microsporidia. In this study, we conducted laboratory experiments to investigate the transmission of an unusual Nosema sp. isolated from L. dispar in Schweinfurt, Germany; this isolate infects only the silk glands and, to a lesser extent, Malpighian tubules of the larval host. The latent period ended between 8 and 15 days after oral inoculation and spores were continuously released in the feces of infected larvae until pupation. Exclusion of feces from the rearing cages resulted in a 58% decrease in horizontal transmission. The silk of only 2 of 25 infected larvae contained microsporidian spores. When larvae were exposed to silk that was artificially contaminated with Nosema sp., 5% became infected. No evidence was found for venereal or transovum (including transovarial) transmission of this parasite.     

Effects of phytochemical variation in quaking aspen Populus tremuloides clones on gypsy moth Lymantria dispar performance in the field and laboratory

1. This study investigated how phytochemical variation among clones of quaking aspen Populus tremuloides, growing in a common habitat, affects the growth and fecundity of a model herbivore. 2. Gypsy moth Lymantria dispar larvae were reared from egg hatch to pupation on 10 aspen clones in the field or on excised foliage in the laboratory. Foliage was collected from each clone, and concentrations of phenolic glycosides, condensed tannins, nitrogen, and water were determined. 3. Herbivore fitness parameters and aspen phytochemical concentrations varied significantly among clones. In both the field and laboratory, larvae reared on clones containing high concentrations of phenolic glycosides exhibited prolonged developmental times and reduced pupal weights and fecundity. Herbivore performance parameters were also related positively to foliar nitrogen concentrations in the laboratory. Food consumption, but neither growth nor reproductive parameters, were related positively to condensed tannin concentrations. 4. In this study, foliar concentrations of phenolic glycosides were implicated as a significant determinant of food quality for gypsy moths, consistent with results of previous laboratory experiments. Additionally, this study documents a case in which host plant variation at a local level influences the performance and possibly the distribution and abundance of an important herbivore.     

Biomodal patterns of mortality from nuclear polyhedrosis virus in gypsy moth (Lymantria dispar) populations

A bimodal temporal pattern of mortality caused by the nuclear polyhedrosis virus (NPV) was observed in nine gypsy moth (Lymantria dispar) populations of varying densities. In all cases, peak mortality from NPV occurred during the second wave (late larval instars) and the highest mortality occurred in high density populations. Patterns of NPV mortality were established several weeks before being expressed. There was no discernible correlation between weekly mortality rates and temperature, rainfall, or total solar radiation. The bimodality was also apparent in NPV contamination on foliage which was measured by bioassay. A similar pattern was observed in the laboratory among larvae reared in groups from field-collected egg masses and from eggs artificially contaminated with NPV from a laboratory population. As in field populations, the period of low mortality from NPV between the two waves occurred when most larvae were late third and fourth instars. Larvae reared individually did not exhibit the second wave of mortality.