Three leading entomopathogens isolated from Madagascar's migratory locust, Locusta migratoria capito Sauss, were evaluated in field and laboratory tests. In a field trial in Madagascar in 1994, two isolates of Metarhizium flavoviride Gams and Rozsypal (SP3 and SP9) and an isolate of Beauveria bassiana (Balsamo) Vuillemin (SP16) were tested against L. migratoria capito. Locusts from field plots treated with SP9 experienced 100% mortality in 8 days, a higher death rate than that found in locusts treated with M. flavoviride SP3 or B. bassiana SP16. However, locusts treated with M. flavoviride SP3 or B. bassiana SP16 had significantly higher mortality than did the untreated controls. In separate field and laboratory trials in Cape Verde in 1994, SP9 was also tested against the Senegalese grasshopper, Oedaleus senegalensis Krauss. Oedaleus senegalensis treated in small-scale field plots with SP9 experienced 100% mortality in 8 days, a significantly higher death rate than that of the untreated controls. An extensive laboratory bioassay with SP9 revealed a dose–response for rate of mortality to O. senegalensis. Results from these trials in Madagascar and Cape Verde suggest that one or more of the Malagasy strains evaluated have good potential for biocontrol of locusts and grasshoppers.

The biology, ecology, disease etiology, and biological control potential of different members of the Entomophaga grylli species complex are discussed. This complex is represented by several pathotypes that include members that produce both conidia and resting spores within a single season, and members that produce only resting spores. This complex is known as a major pathogen of acridids from most areas of the world where populations of these insects are found, including Africa, Asia, Australia, Europe, North America, and South America. Pathogens from this species complex commonly cause disease epizootics in their host populations and are known to reduce significantly outbreaks of grasshoppers, particularly following periods of rain or high humidity. Specific factors that either limit or enhance disease processes and host mortality are discussed in relation to both epizootiology and biological control programmes. Recent biological control efforts are discussed and the potential of using members of the E. grylli species complex in both augmentation and introduction programmes is considered.

Entomophaga grylli (Fresenius) Batko (North American pathotype 1) is a fungal pathogen of the clearwinged grasshopper, Camnula pellucida (Scudder). We present results from a field experiment conducted in Arizona in 1984, designed to investigate factors associated with seasonal patterns of cadaver persistence and sporulation by E. grylli. Rangeland plots at two sites were monitored daily for 8 weeks for the appearance of new cadavers of diseased grasshoppers during a natural epizootic. Cadavers were individually marked and revisited on subsequent days, when it was noted whether or not conidial sporulation was underway. Environmental variables were recorded by electronic data loggers. Daily probabilities of cadaver disappearance and fungal sporulation were analysed in relation to site, date, and various measures of cadaver status, sporulation history, and environmental variables by logistic regression analysis. The average daily rate of cadaver disappearance was 0.22, yielding an expected time to 50% disappearance of 2.8 days. The environmental factor most significantly associated with cadaver disappearance was rainfall, and the most important host factor was age of the cadaver. The probability that conidia would be discharged from a cadaver over the next 24 h was most dependent on whether or not conidial sporulation was underway already. This probably reflects a state of readiness for sporulation on the part of the fungus. Although the probability of sporulation declined with increasing age of a cadaver, high rates of sporulation were predicted under conditions of prolonged leaf wetness and high humidity at night, regardless of age of the cadaver. These results, together with the observation that in some cadavers sequences of sporulation were interspersed with periods of no sporulation, suggest that E. grylli may undergo cycles of dehydration and rehydration, in which conidial production is interrupted and then resumes in response to changing environmental conditions.

Protozoa known to infect Acrididae include certain Amoebida (Phylum Rhizopoda), Eugregarinida, Neogregarinida (Phylum Apicomplexa) and at least one species from the Phylum Ciliophora, but are mainly Microsporida (Phylum Microspora). Among the Microsporida, Nosema locustae Canning has been the most common subject of large-scale field testing as a potential microbial control of grasshoppers and locusts. Although there have been demonstrations of significant impacts of Protozoa such as Nosema spp. on survival, development, reproduction and feeding of grasshoppers and locusts, these candidates have not met the requirements for fast-acting control of Acrididae. However, recently discovered and as yet undiscovered Microspora and other Protozoa may offer reliable non-chemical control of grasshoppers and locusts. Research on candidates such as Nosema locustae has not been wasted effort; at a minimum, research invested toward development of methods of formulation, application, assessment and ecological fate of these potential microbial control agents has provided useful information needed to prepare for the eventuality of discovery of more fast-acting protozoa, and has contributed to development of field methods required for testing of other microbial agents. With improvements in understanding of insect behavioral and physiological responses, satisfactory short-term efficacy (perhaps resulting from combined formulations with behavioral modifiers or stressors) may allow N. locustae, N. cuneatum Henry, N. acridophagus Henry or Johenrea locustae Lange et al. to also serve a role in limiting acridid abundance and activity, either as an agent of biological control in sites requiring special care, or within a rangeland IPM context.

Initial promising results with the microsporidium Nosema locustae Canning as a means of controlling grasshoppers have been followed by some cases of limited field performance. However, at sublethal doses this entomopathogen can reduce grasshopper feeding, reproduction and development, suggesting that the damage caused to range, forages and crops might be reduced if the prevalence of infection in the field could be increased over previously achieved levels. In a field experiment, we treated plots totalling 780 ha with bran bait containing N. locustae at either 2.5 or 5.0 × 109 spores per ha in each of two consecutive years, and grasshopper populations and the prevalence and degree of infection were monitored every two weeks during the summers, for three years. Little or no background infection occurred in the untreated plots. Infections of grasshoppers with N. locustae in the treated plots during the first year of application, typically increased from less than 5% by the 4th week to 10–15% by the 8th week after application. The higher application rates resulted in only slightly higher percentages infected. Most of the infections were classified by microscopic examination as trace to moderate, although heavy infections were observed in the second year. Up to 35% of some Melanoplus species collected from the treated plots were found to contain N. locustae spores, but intensive monitoring indicated population reductions that were inconsistent among sites and generally less than 50%. Other species, notably Aeropedellus clavatus, Ageneotettix deorum, Bruneria brunnea and Camnula pellucida, failed to acquire significant infection by N. locustae. The presence of these apparently less susceptible species did not account for the poor overall performance of the treatments. Two annual applications of N. locustae did not greatly increase the rate or severity of infections in grasshopper populations, although at one of the three sites, trace infections were found in the treated plots six years later.Although Nosema may eventually have value as a component of an integrated grasshopper pest management system, the low virulence cannot be readily overcome by repeated applications.