Plant Protection Quarterly Vol.ll(1) 1996 9 There is no published information on Biological control of the black field , natural enemies of black field cricket other than pathogens. As part of a pro- Teleogyllus commodus (Walker) (: gramme evaluating pathogens for control of crickets, Reinganum et al. (1981) sur- ), using the fungal pathogen Metarhizium veyed diseases in Victoria. Live crickets anisopliae (Metsch.) Sorokin (Deuteromycotina: were sampled from 232 sites in 1979 and pathogens either apparent at the time of Hyphomycetes) sampling, or which appeared within two weeks of collection, were identified. Richard J. Milnefi, Lisa MilkerB,George G. LuttonAand Felice DeriveF Cricket paralysis virus was found most ACSIRO, Division of Entomology, GPO Box 1700, Canberra, ACT 2601, commonly (42.7% of sites) and Australia. Metarhiziurn anisopliae (Metsch.) Sorokin BAgricultureVictoria, Colac, Victoria 3250, Australia. (Deuteromycotina: Hyphomycetes), the only other pathogen found, infrequently (5.2% of sites). Neither pathogen was suf- ficiently abundant to have any influence Summary when applied as an oil-based on population density. A number of iso- The hyphomycete fungus, Metarhizium mycoinsecticide. lates of M. anisopliae were obtained dur- anisopliae, is a natural pathogen of ing the survey and many of these were black field cricket, Teleogyllus Introduction evaluated in the present study (Table 1). commodus, in Victoria. Comparison of The black field cricket, Trleogryllus Reinganum et al. (1981) concluded that 12 isolates, using Random Amplified cornrnodus (Walker) (Orthoptera: the fungus, since it was pathogenic and Polymorphic DNA (RAPDs), from this Gryllidae), is a serious pest of pastures in easily grown on inexpensive media such collected in Victoria showed that parts of Victoria and South Australia as rice, was promising for development as they were all very similar genetically causing over $8m damage in some years a mycoinsecticide, however no field trial but quite distinct from the group 3 (Anon. 1989). It is also a serious pest in results have been published. M. flavoviride isolates from grasshop- parts of the north island of New Zealand Recently, it has been shown that certain pers and locusts. Screening bioassays (Blank 1982). Populations are particularly isolates known as M.flavoviride Gams and showed that isolates from crickets were high in western Victoria where cracking Rozsypal group 3 (Prior 1995) comprise a more virulent for this pest than isolates clay soils provide essential shelter during genetically uniform and distinct group, from other sources. One of these cricket hot dry weather. T. cornmodus is and are virulent for a range of grasshop- isolates F11099 was selected for use in univoltine: over-wintering diapause eggs pers and locusts. Interestingly, natural in- field trials. An oil formulation of are laid in the soil in autumn and hatch in fections of M. flavoviride group 3 have conidia of FI1099 was tested either as a early summer (December); the nymphs only been found on grasshoppers and lo- ultra-low volume (ULV) spray or in a develop through five instars before be- custs which may suggest that they have high volume oillwater emulsion at two coming adults usually in the autumn co-evolved. Furthermore, it has been sites in Victoria. At Warrambeen, a dose (March). Populations decline with the on- found that by formulating the conidia of of 2 x loi3 conidia per hectare gave a set of cooler weather in April and few, if M. flavoviride in oil, the fungus can con- 30-40% reduction in the cricket popula- any, adults over-winter. Pasture defolia- trol locusts and grasshoppers when ap- tion after 21 days compared with 80% tion can be severe in late summer and au- plied using ultra-low volume (ULV) tech- control in the malathion plots. At tumn (February-April). The only method niques or in an oil-emulsion using high Turkeith, the higher dose of 4 x 10'j of control available to farmers is the ap- volume boom sprayers (Baker et al. 1994). conidia per hectare gave 60-70% control plication of chemical insecticides using The oil not only prevents evaporation of while the malathion again gave about malathion cereal baits (Stahle et al. 1983). water during ULV spraying but also 80% control. These results are promising Many thousands of hectares are treated aids attachment and penetration of the and show that M. anisopliae has the po- each year, often using aerial application target host by the fmgus. This enables tential to control black field cricket methods. the conidia to germinate under field

Table 1. List of isolates and the sources of Metarhizium spp. used in this study. Isolate Species name Original host Source Place F1610 M. anisopliae termite mound CSIRO Brown Mt., New South Wales FI 985 M. flavoviride Austracris guttulosa ARSEF 324 Rockhampton, Queensland FI 1030 M. anisopliae T. cornmodus IMI 351 830 Victoria FI 1037 M. anisopliae T. cornrnodus ARSEF 448 Victoria FI 1066 M. anisopliae soil IIBC Ben 4 Benin FI 1067 M.flavoviride acridid IlBC 192-701 Benin FI 1090 M. anisopliae T. cornrnodus ARSEF 435 Warrnambool, Victoria FI 1091 M. anisopliae T. cornrnodus ARSEF 437 Victoria FI 1092 M. anisopliae T. cornrnodus ARSEF 438 Victoria FI 1093 M. anisopliae T. cornrnodus ARSEF 445 Carapook, Victoria FI 1094 M. anisopliae T. cornmodus ARSEF 440 Warrnambool, Victoria FI 1095 M. anisopliae T. cornrnodus ARSEF 441 Warrnambool, Victoria F11096 M. anisopliae T. cornrnodus ARSEF 442 Hawkesdale, Victoria FI 1097 M. anisopliae T. cornrnodus ARSEF 443 Camperdown, Victoria FI1098 M. anisopliae T. cornrnodus ARSEF 444 Byaduck, Victoria FI 1099 M. anisopliae T. cornrnodus ARSEF 445 Carapook, Victoria FI 1100 M. anisopliae T. cornrnodus ARSEF 446 Mortlake, Victoria 10 Plant Protection Quarterly Vol.ll(1) 1996 conditions of low humidity which nor- was washed twice with 500 pL of 7O0/0 experiment, the crickets were T. mally prevent or seriously delay the in- ethanol, then dried in a vacuum and cornrnodus while in the second the closely fection process (Bateman et al. 1993, resuspended in 400 pL of lOmM Tris HC1 related species T. oceanicus was used as Milner et al. 1995). Crickets are also at pH 8.0, ImM EDTA and 10 pg/mL they were more abundant in the labora- orthopteroid , and the present RNase A. The DNA was incubated at tory colonies. In previous laboratory ex- studywas undertaken to determine if the 37" for 30 minutes and then extracted periments, it had been found that these grasshopper control strategy could be ex- with 1 volume phenol : CHC1,: isoamyl species are similar in their susceptibility tended to crickets. alcohol (50:48:2), followed by CHC1,: to Metarhiziurn spp. (Milner unpublished For genetic fingerprinting, randomly isoamyl alcohol (24:l) until the interface data). amplified polymorphic DNA (RAPD) was clean. The DNA was precipitated by markers have been used to generate mo- adding 40 pL 3M sodium acetate pH 5.4 Field trials lecular markers (Welsh et al. 1991, and two volumes ethanol. The DNA was Two field trials were set up in February Williams et al. 1991) and to analvse recovered by brief centrifugation in a 1995. The first was a replicated small plot genomes and populations in a variety of microfuge, washed twice with 70% trial established at the Alcoa Landcare organisms including fungi (Fegan et al. ethanol, then dried in a vacuum. Pellets Demonstration Farm, "Warrambeen" 1993, Bidochka et al. 1994), humans and were resuspended in 5mM Tris HC1, near Colac, Victoria. Each plot was 25 x plants (Williams et al. 1991), and insects 0.5mM EDTA. 25 m and separated by an untreated (review by Haymer 1994). These DNA RAPD-PCR (polymerase chain reac- buffer of 20 m. There were four treat- polymorphisrns may arise from repetitive tion) amplifications were performed in a ments: untreated control, baited with and unique regions and indicate the de- total volume 25 pL. Each reaction con- malathion at a rate of 125 mL a.i. to 10 kg gree of relatedness between isolates. tained 2.5 pL of 10X Bresatec reaction oats ha-', sprayed with a highvolume oil/ The aims of the study were to use mo- buffer (16.6mM (NH,),SO,, 67mM Tris water emulsion of M. anisopliae at 2 x 1013 lecular fingerprinting techniques to assess HC1 at pH 8.8, 0.45% Triton X100, 200 pg conidia ha-' in 130 L ha-', and the same the genetic status of isolates from T. mL-' gelatin) 3.5mM MgCl,, 200pM each number of conidia applied using a commodus, select the most virulent isolate dATP, dTTP, dCTP, dGTP, 30pM primer, MicroUlva Plus ULV sprayer at 2 L ha-'. for this host and to undertake initial field 10-15 ng DNA and 2.4 U Taq polymerase The plots were arranged in a randomized tests of the efficacy of the selected isolate (Bresatec). The reactions were covered by block with four replicate plots per treat- in oil-based sprays. a drop of light mineral oil. Primers were ment. The treatments were applied late in those used by Fegan et al. (1993) from the evening on February 15 as the very Materials and methods Random Primer kits H and F (Operon hot, windy, day-time conditions were un- Technologies, Alameda, California). Re- suitable for spraying. Isolates actions were placed in a thermocycler The other trial was set up at "Turkeith", A list of isolates used in this study are (Corbett Research, Australia) and DNA a property located about 20 km south of given in Table 1. The isolates from T. was amplified using the following tem- Warrambeen and with a history of seri- commodus were all from the ARSEF col- perature cycles: one cycle at 94" for five ous cricket infestations. There were two lection (USDA Insect Pathogenic Fungi minutes, 40°C for t-wo minutes, 72" for unreplicated plots each 50 x 50 m sepa- Culture Collection at Ithaca, New York) three minutes, then 39 cycles at 94°C for rated by a 20 m untreated buffer. One plot and originally derived from the survey one minute, 40°C for one minute 30 sec- was treated with the same high volume for pathogens of black field cricket onds, and 72" for two minutes. spray as at Warrambeen but at double the (Reinganum et al. 1981). Other isolates Products were separated in 1.3% TBE dose - 4 x 1013conidia ha-', while the other were either from the CSIRO Insect Patho- agarose gels (Sambrook et al. 1989). Gels plot was treated with malathion as previ- gen Culture Collection or from the Inter- were run at 7.5 V cm-' for two hours, ously. The residual paddock was used as national Institute of Biological Control, stained with ethidium bromide and pho- an untreated control. The plots were Ascot, United Kingdom. tographed on a UV transilluminater (UVP treated on the day after those at Inc.). "Warrambeen" during the afternoon Genetic analysis when the weather was cool, cloudy and To extract the DNA, the methods of Bioassay threatening rain. Curran et al. (1994) were modified Five isolates (Table 2) were grown on The viability of the M. anisopliae conidia slightly. Briefly, 50-100 mg of dried plates of Sabourauds Dextrose Agar sup- was checked~byplating on Sabouraud's mycelium was ground to a fine powder plemented with 1% yeast extract. After dextrose agar with yeast extract (SDAYE) using a mortar and pestle. The ground about three weeks at 25"C, the conidia plates andfound to-be >90 %. Laboratory mycelium was transferred to a 1.5 mL were harvested into Propar 12 oil tests also confirmed the infectivity of the microcentrifuge tube containing 750 pL (Ampol) and sonicated to produce a ho- spray as used in the field trials for crick- proteinase K buffer (0.1M Tris HC1 pH mogenous suspension. The concentration ets. The M. anisopliae was formulated in 8.5, 0.05 EDTA, 0.2M NaCl, 1% SDS and of conidia was determined using a Propar 12. Laboratory studies have 2 mg mL-2proteinase K) and incubated at Petroff-Hausser particle counter and ad- shown this mineral oil to be non-toxic for 65°C for 30 minutes. The DNA was ex- justed to 3 x lo7 conidia mL-'. Two deci- M. anisopliae conidia. For the high volume tracted by hand using an equal volume of mal dilutions were made in Propar 12. spray, Wetter TX (Monsanto) was added water saturated phenol. The phases were Adult crickets were inoculated by placing as an emulsifying agent and water added separated by spinning for two minutes in a 5pL drop of Propar 12 using a to give a 20:80 oil : water emulsion. a microfuge, and the aqueous phase was micropipette on the ventral thorax and The population density of crickets was transferred to a clean tube. The extraction were then incubated individually in 20 determined using folded hessian bags procedure was repeated using 1 volume mL plastic containers at 28" . Mortality (Murphy and Stahle 1985).The bags, five phenol : chloroform : isoamyl alcohol was recorded every day for 12 days and per plot at Warrambeen and 12 per plot at (50:48:2),then chloroform : iso amyl alco- the dead insects placed in moist chambers Turkeith, were placed over cracks and re- hol (24:l) until the interface was clean. to promote sporulation. Twenty insects moved periodically to count the number DNA was precipitated using 1 volume of were used for each of the three doses, 20 of crickets. At Warrambeen the popula- isopropanol. The DNA was recovered by were treated as blank oil controls and 20 tion was assessed at 2,3,15,21,33, and 49 centrifuging for two minutes. The pellet were left as untreated controls. In the first days post treatment and at Turkeith 2,7, Planlt Protection Quarterly Vol.ll(1) 1996 11 14, 20, 36 and 48 days. The population analysis. Lanes 2-16 in Figure 1 contain control mortality. However, the fact that density is expressed as number of crickets PCR products from the cricket isolates. only three insects sporulated when dosed per bag. In addition, samples of live crick- The bands were identical in mobility and with FI985 supports previous laboratory ets were collected at 2 and 7 days post- intensity, indicating the reproducibility of data suggesting that neither species of treatment at Warrambeen and 6 days this technique. Lanes 17-19 correspond to Teleogyllus is susceptible to FI985. In the post-treatment at Turkeith. These crickets isolates of Metarhizium isolated from lo- experiment with T. commodus only 15% of were incubated in the laboratory at 28°C custs. They differed from the cricket iso- the untreated controls died by da) 12 and for 21 days and the mortality recorded. lates for all primers tested. Some primers FI1099 again proved to be the most viru- All dead crickets were placed ir, moist such as H02 (Figure 1) identified minor lent isolate. However many of the crickets chambers to promote sporulation thus variation within the cricket or the locust even though killed by the M. anisopliae confirming that the insects were killed by isolates, however, all the primers clearly treatment (based on a comparison with M. anisopliae. distinguished these two groups. Interest- control mortality) failed to sporulate. The population data were analysed by ingly where multiple isolates of FI1067 was also pathogenic but conidia ANOVA, following transformation Metarhizium from a single host have been did not develop on any of the cadavers (dn+l), using the Minitab 10 software analysed using RAPDs, (Fegan et al. 1993, while FI1066 was the least pathogenic. package. At Turkeith, each bag was re- Driver and Milner unpublished data), the With FI1099, both species of cricket died garded as an independent sampling unit variability has been much greater than of M. anisopliae 5-7 days after inoculation for the analysis. that exhibited by the cricket isolates in at the highest dose and after 6-11 days at this study. the lowest dose at the incubation tem- Results perature of 28°C. Based on these results, Bioassay and the propensity of FI1099 to sporulate Genetic a~zalysis Only two isolates, FI610 and FI1099, well on artificial media, this isolate was The DNA extracted from isolates from caused both high mortality and produced selected for field trials. black field crickets and locusts were sub- large numbers of sporulating cadavers jected to RAPD analysis using primers (Table 2) with FI1099 producing a higher Field trials H01, H02, F06, F07, F08 and F10. Fif- proportion of sporulating cadavers at all At Warrambeen, the number of crickets teen isolates from crickets and three iso- doses. The mortality data for T. oceanicus remained constant until day 15 when the lates from locusts were included in the are difficult to interpret due to the high population in the malathion plots de- clined to 0.5 per bag compared with 3.8 OP-HOI per bag on average in the control plots idiffereice signifirant P <0.05) (~abie3), both M. anisopliae treatments were also lower than the controls by day 15, though not significantly different (P >0.05).After day 15, there was a gradual reinvasion of the insecticide plots and by day 49 the number was nearly at the level of the con- trol plots. The numbers in M. anisopliae treated plots continued to decline to day 21 indicating continued M. anisopliae in- duced mortality. The ULV application was more effective than the high volume application in reducing numbers though neither were statistical significantly lower than controls or any sampling occasion. However, the live cricket samples gave 20-30 % infection supporting the conclu- sion that the Metarhizium treatment had produced the population reductions ob- served in the field. As with the insecticide treated plots, there was evidence of reinvasion after day 21 and by day 49 the population densities were similar to those in the controls. The dose of Metarhizium applied at Turkeith was double that -ked at Warrambeen and the fungus gave a much higher level of control (Table 4). The in- secticide treatment produced the most substantial population reduction (to 0.75 crickets pe; bag by day 14). There was some reinvasion, however, even by day 48 the population in the insecticide treated plot was still significantly lower than in the control (Table 4). This slower Figure 1. RAPD banding patterns from isolates of Metarhizium from black rate of reinvasion than at Warrambeen field crickets (lanes 2-16) and acridid grasshoppers (lanes 17-19) as was probably due to the larger plot size. revealed using H01 and H02 oligonucleotide probes. Note lanes 1 and 20 The apparent increase in population den- are h Hind 111 molecular markers. sity in the control and treatment plots 12 Plant Protection Quarterly Vol.ll(1) 1996 between day 14 and day 20 in the spraying. Observations in the field at the require a combination of wet weather and Metarhizium was probably an artefact as end of April, after the experiment had low density cricket populations in which the increase in numbers of crickets in the been completed, suggested that cannibalism is unlikely. control plots suggest that the weather on reinvasion had occurred in both the fun- March 7 was particularly favourable for gus and insecticide plots after day 48. Discussion sampling crickets under the bags. Never- At neither site was there any evidence The M. anisopliae fungus is being devel- theless, the data suggest that the fungus of transmission of the disease between oped as a mycoinsecticide for a range of gave a population reduction of around crickets. Observations in the laboratory target insect pests in many countries. Re- 70% while the insecticide gave about 80% suggest that infected crickets may be less cently two Metarhizium-based products, reduction, comparable to that reported by mobile and aggressive than uninfected BioPath and BioBlast, have been regis- Murphy and Stahle (1985) in a baiting trial crickets and may be attacked and eaten tered for use against cockroaches and ter- with malathion. The live cricket sample before the fungus has developed to the mites respectively in the USA. In Aus- from Turkeith gave 43% mortality with infectious stage. Conidia, the infective tralia BioGreen, a product for control of 36% sporulation. This was taken seven stage, are not produced until several days the red-headed pasture cockchafer, is be- days after spraying and seems to have un- after death. In addition, field conditions ing considered for registration by the Na- derestimated the effect of the fungus, sug- may be too dry to promote sporulation. tional Registration Authority. Character- gesting that some crickets picked up a le- Field transmission, as implied by the find- istics of Metarhizium which make this fun- thal dose more than seven days after ings of Reinganum et al. (1981), may gus attractive for development as a mycoinsecticide are that it is safe to pro- Table 2. Per cent mortality of five isolates of Metarhizium spp. against duce and use, is readily mass produced spp. (n=20) assessed 12 days after treatment and subsequent on simple substrates such as rice, pro- per cent sporulation in parenthesis. duces conidia which are robust and thus amenable to formulation and storage, and Isolate Target 150 000 15 000 1500 is genetically diverse with specific strains conidia/cricket conidia/cricket conidia/cricket often having a restricted host range. In FI 1099 T. commodusa 95 (50) 75 (25) 75 (25) order to develop a mycoinsecticide based on M. anisopliae for a particular target it is FI 1066 T. commodus 50 (5) 10 (0) 20 (0) necessary to select a virulent strain which FI 1067 T. commodus 50 (0) 40 (0) 35 (0) FI 1099 T. oceanicusb 100 (70) 95 (60) 95 (55) sporulates profusely in vitro, and to de- velop an application strategy that pro- FI 985 T. oceanicus 100 (5) 75 (5) 95 (5) FI 610 T. oceanicus 100 (50) 95 (35) 100 (40) vides acceptable and consistent pest con- trol. a control mortality of untreated control 15OlO. Given that M. anisopliae is especially di- control mortality of crickets treated with oil was 75% and untreated loo%, however, verse genetically, it was surprising that no control crickets sporulated. the genetic analysis of 12 isolates from disparate sites in Victoria showed that they were all practically identical. This Table 3. Changes in population density of Teleopyllus commodus suggests that the strain may be quite spe- (expressed as average number of crickets per ba&at Warrambeen. cific for black field crickets. One of these Date Days post- High volume Low volume Malathion Control isolates, FI1099, was selected for use in treatment Metarhizium Metarhizium bait the field trials as it was virulent for Teleogryllus in the screening experiments 9 February and sporulated well on the rice used for 14 February mass production of the fungus. Labora- 16 February tory assays have shown that this isolate 17 February will not infect Locusta migratoria L. 1 March (Orthoptera: Acrididae) nor Tenebrio 7 March molitor L. (Coleoptera: Tenebrionidae) at 23 March doses which give a high level of kill on 4 April crickets (Milner and Prior, unpublished data), however further work is needed to * indicates significantly lower population than controls at 5% level. determine the level of host specificity ex- ns indicates not significant. hibited by this isolate. The present study has demonstrated Table 4. Changes in population density of Teleogryllus commodus that isolate FI1099 can be mass produced, (expressed as average number per bag) at Turkeith. formulated in oil and applied to control a field population of black field crickets. Date Days post- High volume Malathion Control The doses used in these trials are some treatment Metarhizium bait tenfold higher than those found effective 17 February 2 2.33 ns 1.83 ns 3.16 against wingless grasshopper, Phaula- 22 February 7 2.33* 3.41 ns 5.00 cridium vittatum (Sjost.) (Baker et a/. 1994) 1 March 14 1.42+ 0.75' 3.00 and are likely to be uneconomic. Further 7 March 20 3.33* 2.33* 8.67 research is needed to achieve effective 23 March 36 3.67 ns 3.58 ns 6.75 control at a much lower dose, i.e. by us- 4 April 48 3.42+ 1.83* 9.08 ing a more virulent isolate and develop- ing more efficient application technology. * indicates significantly lower than controls at 5% level. (Note: malathion and Given the low level of genetic diversity Metarhizium plots never significantly different from each other). of isolates naturally occurring on T. ns indicates not significant. comrnodus in Victoria, it seems unlikely ~t Protection Quarterly Vol.ll(1) 1996 13 that a more virulent isolate will be found References Teleogyl!us commodus, in the Western locally. This leaves three options for de- Anon. (1989). The economic impact of district of Victoria. Journal of Inverte- veloping a better isolate: passage of a pasture weeds, pests, and diseases on brate Pathology 38, 153-60. cricket isolate such as FI1099 through the Australian wool industry, 205 pp. Sambrook, J., Fritsch, E.F. and Maniatis, black field cricket and selecting for a more (Sloane Cook and King Pty. Ltd.). T. (1989). Molecular cloning: a labora- virulent isolate; obtain a more virulent Baker, G.L., Milner, R.J., Lutton, G.G. and tory manual. (Cold Spring Harbour isolate from another host, another part of Watson, D.M. (1994). Preliminary field Press, New York, USA). Australia or another country; or geneti- trial on the control of Phaulacridium Stahle, P., Murphy, G. and Conley, D. cally engineer an isolate. At the present vittatum (Sjostedt) (Orthoptera: (1983). Control of field crickets. Agnote time only the first option can be readily Acrididae) populations with Meta- 2046/83, Victorian Department of Ag- achieved and will be pursued with the rhizium flavoviride Gams and Rozsypal riculture, pp. 2. aim of field testing a passaged isolate in (Deuteromycotina: Hyphomycetes). Welsh, J., Petersen, C. and McClelland, M. the 1995/6 season. Screening of exotic iso- Journal of the Australian Entomologicd (1991). DNA polymorphisrns amplified lates has not given results to Society 33, 190-2. by arbitary primers are useful as ge- date but will continue as the opportunity Bateman, R.P., Carey, M., Moore, D. and netic markers. Nucleic Acids Research 19, arises. Improving the virulence of M. Prior, C. (1993). The enhanced infectiv- 7213-8. anisopliae by genetic engineering is not yet ity of Metarhizium flavoviride in oil for- Williams, J.G.K., Kubelik, A.R., Llivak, possible and might have problems From mulations to desert locusts at low K.J. , Rafalski, J.A. and Tingey, S.V. an environmental point of view if it had a humidities. Annals of Applied Biology (1991). DNA polymorphisms amplified wider host and/or was thought to 122, 145-52. by arbitary primers are useful as ge- threaten non-pest species of cricket. Bidochka, M.J., McDonald, M.A., St. netic markers. Nucleic Acid Research 18, Crickets are nocturnal insects which Leger, R.J. and Roberts, D.W. (1994). 6531-5. hide during the day in deep cracks. This Differentiation of species and strains of makes application difficult both with entomopathogenic fungi by random chemical &~dbiological insecticides since amplification of polymorphic DNA. very Few crickets are directly exposed to a Current Genetics 25, 107-13. spray at any one time. The use of baits has Blank, R.H. (1982). Control of Teleogyllus proved effective with chemicals because commodus using insecticidal baits and the pesticide persists on the oats (or other other agents. Proceedings of the Aus- cer;al) and as. the crickets progressively tralasian Conference on the Ecology of feed on these baits the population is Grassland Invertebrates, Adelaide 1981, killed. Baits take two weeks or longer to 355-63. be effective, which may reflect the fact Curran, J., Driver, F., Ballard, J.W.O. and that only a small proportion of crickets Milner, R.J. (1994). Phylogeny of come onto the surfaceto feed each night. Metarhizium: analysis of ribosomal Baits were not used in the application of DNA sequence data. Mycologicd Re- M. anisopliae, baits for two reasons: firstly, se~rch98, 547-52. crickets were not infected in the labora- Fegan, M., Manners, J.M., Maclean, D.J., tory when exposed to oats treated with Irwin, J.A.G, Samuels, K.D.Z., Holdom, fungal spores; and secondly, the fungus, D.G. and Li, D.P. (1993). Random am- unlike the chemical, does not act as a plified polymorphic DNA markers re- stomach poison. Nevertheless, recent veal a high degree of genetic diversity laboratory studies have given promising in the entomopathogenic fungus results using rice baited with Metarhizium Metarhizium anisopliae var. anisopliae. spores and this strategy will be field Journal of General Microbiology 139, tested. A combination of an isolate with 2075-81. increased virulence and an improved ap- Haymer, D.S. (1994). Arbitary (RAPD) plication strategy it is hoped will provide primer sequences used in insect studies. a higher level of control at a lower dose. Insect Molecular Biology 3, 191-4. Milner, R.J., Baker, G.L., Hooper, G.H.S. Acknowledgments and Prior, C. (In press). Development of Ian and Trish Taylor of Warrambeen and a mycoinsecticide for the Australian Lachlan Gordon of Turkeith kindly pro- plague locust. Conference on Future vided the two field sites. Financial sup- Methods in Locust Control, Bamako, port was provided by Alcoa through Mali, March 1995. Kathy Junor and by Graham Woods Murphy G. and Stahle, P. (1985). Exarni- Agencies through David Looker. Valu- nation of baiting techniques against the able comments on the manuscript were black field cricket Teleogyllus provided by Drs. Gordon Berg, Thom commodus. Proceedings of the 4th Aus- Boyce and Roger Farrow. The construc- tralasian Conference on Grassland In- tive advice and help of Dr. Chris Prior is vertebrate Ecology, Canterbury, New gratefully acknowledged. Research on the Zealand, May 1985, 293-5. genetic aspects of pathogenicity in Prior, C. (In press). Safety of Metarhizium Metarhizium is being funded by Rural In- isolates. Conference on Future Methods in dustries Research and Development Cor- Locust Control, Bamako, Mali, March poration. 1995. Reinganum, C., Gagen, S.J., Sexton, S.B. and Vellacott, H.P. (1981). A survey for pathogens OF the black field cricket,