Evaluation of Entomopathogenic Fungus Metarhizium ... - BioOne

15 downloads 42 Views 96KB Size Report
ABSTRACT The effect of two isolates of the entomopathogenic fungus Metarhizium anisopliae. (Metchnikoff) Sorokin (389.93 and 392.93) on root-feeding stages ...
BIOLOGICAL AND MICROBIAL CONTROL

Evaluation of Entomopathogenic Fungus Metarhizium anisopliae Against Soil-Dwelling Stages of Cabbage Maggot (Diptera: Anthomyiidae) in Glasshouse and Field Experiments and Effect of Fungicides on Fungal Activity D. CHANDLER

AND

G. DAVIDSON

Warwick HRI, University of Warwick, Wellesbourne, Warwick, CV35 9EF United Kingdom

J. Econ. Entomol. 98(6): 1856Ð1862 (2005)

ABSTRACT The effect of two isolates of the entomopathogenic fungus Metarhizium anisopliae (Metchnikoff) Sorokin (389.93 and 392.93) on root-feeding stages of cabbage root ßy, Delia radicum (L.), was studied under glasshouse and Þeld conditions. In glasshouse studies, the effect of drenching a suspension of conidia (concentration 1 ⫻ 108 ml⫺1, 40 ml per plant, applied on four occasions) onto the base of cabbage plants infested with D. radicum eggs was compared with mixing conidial suspension into compost modules (concentration 1 ⫻ 108 ml⫺1, 25 ml per plant) used to raise seedlings. Drench application reduced the mean number of larvae and pupae recovered per plant by up to 90%, but the compost module treatment had no statistically signiÞcant effect. Both application methods reduced the emergence of adult ßies from pupae by up to 92%. Most conidia applied as a drench application remained in the top 10-cm layer of compost. Applications of the fungicides iprodione and tebuconazole, which are used routinely on brassica crops, were compatible with using M. anisopliae 389.93 against D. radicum under glasshouse conditions, even though these fungicides were inhibitory to fungal growth on SDA medium. In a Þeld experiment, drench applications of M. anisopliae 389.93 to the base of caulißower plants at concentrations of 1 ⫻ 106 to 1 ⫻ 108 conidia ml⫺1 did not control D. radicum populations, although up to 30% of larval cadavers recovered supported sporulating mycelium. Drench applications often exhibited considerable lateral movement on the soil surface before penetrating the ground, which may have reduced the amount of inoculum in contact with D. radicum larvae. KEY WORDS cabbage maggot, Delia radicum, entomopathogen, Metarhizium anisopliae

CABBAGE MAGGOT, Delia radicum (L.) (Diptera: Muscidae: Anthomyiidae) is the major pest of horticultural brassicas in northern temperate regions (Finch 1989). It is part of a brassica pest complex of ⬇50 species of insects and pathogens (Finch 1993). Adult females oviposit within 5 cm of the host plant on the underside of small clumps of soil or in crevices near the soil surface (Hughes and Salter 1959). The larvae of D. radicum feed on plant roots and reduce crop yield and quality. Larvae are restricted to the plant roots and pupate close to the plant at a depth of 8 Ð12 cm (Smith 1927). Other pests in the genus include the onion maggot, Delia antiqua (Meigen); the turnip maggot, Delia floralis (Falle´ n); and the seed corn maggot, Delia platura (Meigen) (Finch 1989). Control of D. radicum is based mainly on applications of organophosphate and carbamate insecticides. However, brassica growers are under pressure to reduce inputs of chemical insecticides and alternatives are required, including microbial control agents. In previous laboratory work, we investigated the pathogenicities of a range of anamorphic entomopathogenic

fungi to larvae and adults of D. antiqua, which we used as a model for other Delia species (Davidson and Chandler 2005). Adult D. antiqua were very susceptible to infection, whereas larvae had a low susceptibility overall. Despite this, we identiÞed two isolates of Metarhizium anisopliae (Metchnikoff) Sorokin that were infective to larvae. The aim of the present work was to investigate these isolates as microbial control agents of D. radicum larvae under glasshouse and Þeld conditions and to investigate the effect on the activity of M. anisopliae of fungicides used on brassica crops. Materials and Methods Fungal Cultures. The study was done with M. anisopliae isolates 389.93 and 392.93 from the Warwick HRI collection of entomopathogenic fungal cultures. Both isolates were obtained from United Kingdom soil by using Galleria mellonella (L.) (Lepidoptera: Pyralidae) as a live insect bait (Bedding and Akhurst 1975, Zimmerman 1986) and were shown previously to be pathogenic to D. antiqua larvae (Davidson and Chan-

0022-0493/05/1856Ð1862$04.00/0 䉷 2005 Entomological Society of America

December 2005

CHANDLER AND DAVIDSON: FUNGAL BIOCONTROL OF CABBAGE MAGGOT

dler 2005). Stock cultures of the isolates were stored in liquid nitrogen vapor (Chandler 1994). Laboratory cultures were grown on Sabouraud dextrose agar (SDA) slopes and maintained at 5⬚C for up to 6 mo. For laboratory experiments, subcultures were prepared on SDA from slope cultures and incubated at 20⬚C for 10 Ð12 d. Conidia were then harvested in sterile 0.05% Triton X-100 and Þltered through sintered glass thimbles (40 Ð100-␮m pore). Conidia were enumerated using a hemacytometer and aliquots were prepared at concentrations of 1 ⫻ 107 and 5 ⫻ 107 ml⫺1. To produce conidia for glasshouse and Þeld experiments, 10-ml aliquots of conidial suspensions were prepared from SDA subcultures and used to inoculate an autoclaved grain/perlite mixture (300 g of micronized ßaked maize, 150 g of bulgar wheat, 150 g of horticultural perlite, and 500 ml of deionized water, contained within a 5-liter mushroom spawn bag [Van Leer Packaging Systems Ltd., Poole, UK]) that was incubated at 20⬚C and a photoperiod of 16:8 (L:D) h for 28 d. Conidia were harvested by the addition of sterile 0.05% Triton X-100 to the grain/perlite mixture, which was then shaken by hand and Þltered through two layers of muslin in a Buchner funnel followed by Þltration through a milk Þlter [Lantor (UK) Ltd., Bolton, United Kingdom]. Conidia were enumerated with a hemacytometer and volumes of 0.5Ð1.0 liter were adjusted to concentrations of 1 ⫻ 106 to 1 ⫻ 108 ml⫺1. Production of conidia on the grain/perlite mixture yielded 3.0 Ð7.5 ⫻ 108 conidia g⫺1 substrate. Conidia viability was assessed by measuring the germination of conidia on SDA after incubation for 24 h at 23⬚C (Goettel and Inglis 1997) and was never ⬍89%. Insect Rearing. Rearing of D. radicum was based on the method of Finch and Coaker (1969). Adult ßies were maintained in mesh rearing cages (35 by 35 by 35 cm) at 18⬚C, a photoperiod of 16:8 (L:D) h, and 40% RH and supplied with distilled water, 10% sucrose solution, and yeast extract sandwich spread (Marmite) with brewerÕs yeast powder (30% brewerÕs yeast in soya ßour). Fly eggs were collected 12 d postemergence by placing an oviposition site, consisting of a 2-cm cube of swede (rutabaga), Brassica napus variety napobrassica on damp sand in a petri dish lid, in the cages for 24 h. Eggs were washed from the sand, ßoated on water, and collected by sieving. Larvae were reared by placing eggs on whole swedes kept in damp sand at 18⬚C and a photoperiod of 16:8 (L:D) h for 30 d. Pupae were washed from the sand and stored for up to 1 mo at 4⬚C in dampened vermiculite before they were transferred to adult ßy cages as described previously. Adult ßies began to emerge 2 to 3 d after pupae were placed in cages. Fungal Application Methods. This experiment investigated the susceptibility to M. anisopliae 389.93 and 392.93 of D. radicum larvae feeding on cabbage plants in a glasshouse experiment done from June to September 2000. The glasshouse was maintained at a minimum temperature of 18⬚C and vented at 23⬚C (temperature/relative humidity range for all experiments was 18 Ð29⬚C/28 Ð100% RH). Conidial suspensions were prepared as described previously.

1857

Two different methods of applying M. anisopliae were evaluated. In the Þrst method, conidial suspension was incorporated by hand into Levington F2 compost [The Scotts Company (UK) Ltd., Daventry, United Kingdom] at a concentration of 1 ⫻ 108 ml⫺1. The treated compost was potted into 25-ml modular Hassy trays (BHGS Ltd., Evesham, United Kingdom) that were then planted with cabbage seedlings (Brassica oleracea ÔDerby DayÕ) that had been germinated over the previous 5 d in vermiculite (20⬚C). The plants were raised in the glasshouse for 28 d, and then the modules were transferred into 11-cm-diameter pots Þlled with John Innes No. 2 compost (BHGS Ltd., Evesham, United Kingdom). Each pot contained a disk of plastic mesh (mesh size 1 mm) at the base to prevent the escape of D. radicum larvae. The plants were grown for a further 21 d before the addition of 1-d-old D. radicum eggs. Compost was scraped away from the plant stem, and the eggs were placed adjacent to the stem by using a camelÕs-hair brush, 5 mm below the soil surface and at a rate of 30 eggs per plant. The second application method consisted of weekly drenches of conidial suspension (1 ⫻ 108 ml⫺1, 40 ml per plant) that were applied to 7-wk-old plants (raised as described above but in Metarhizium-free compost) immediately after the addition of D. radicum eggs and on three further occasions (7, 14, and 21 d later). Controls were left untreated because preliminary studies indicated that 0.05% Triton X-100, incorporated into compost or applied as a drench, had no effect on the survival of D. radicum larvae. Each treatment consisted of 16 Ð24 plants, depending on the availability of D. radicum eggs. All treatments were done together as a block and the experiment was repeated three times in total. Larvae and pupae were washed from pots 28 d after the application of eggs. Dead larvae were incubated on damp Þlter paper within petri dishes at 20⬚C for 7 d and inspected for the presence of mycelium on cadavers. Pupae were incubated in damp vermiculite within Universal bottles at 20⬚C, and the numbers of adult ßies that emerged were recorded 28 d after collection. Any pupae remaining at the end of the experiment were inspected for the presence of sporulating fungal mycelium on the surface. In addition, the weight of each plant was recorded at harvest. The vertical movement of a conidial suspension drenched onto compost was investigated using the method of Storey and Gardner (1987). Movement was assessed by pouring 50 ml of a conidial suspension (5 ⫻ 107 ml⫺1) of M. anisopliae 389.93, followed by 50 ml of sterilized distilled water, onto a column (2.8 cm diameter by 30 cm in height) of water-saturated Levington F2 compost. This was contained within a polystyrene pipe that was divided in half lengthwise and held together by rubber bands, together with a muslin sheet over the bottom to hold the soil within the column but allow drainage. Efßuent was collected for 1 h, after which the column was frozen (⫺20⬚C) and cut into 5-cm lengths that were placed into 100 ml of sterile distilled water in conical ßasks and shaken for 15 min. The numbers of conidia were enumerated

1858

JOURNAL OF ECONOMIC ENTOMOLOGY

using a hemacytometer. The experiment used three replicate soil columns and was repeated three times. Effects of Fungicides. Six fungicides used routinely on brassica crops were evaluated for their effect on the colony extension of M. anisopliae 389.93 and 392.93 (Table 3). The fungicides were chlorothalonil (a protectant chlorophenyl, Rover; Sipcam UK Ltd., Cambridge, United Kingdom), difenoconazole (a protectant and curative diphenyl-ether triazole, Plover; Novartis Crop Protection UK Ltd., Cambridge, United Kingdom), iprodione (a protectant dicarboximide, Rovral WP; Aventis CropScience UK Ltd., Halifax, United Kingdom), tebuconazole (a systemic conazole, Folicur; Bayer plc, Cambridge, United Kingdom), tolclofos-methyl (a protectant organophosphate, Basilex; The Scotts Company (UK) Ltd., Daventry, United Kingdom), and triadimefon (a systemic conazole, Bayleton; Bayer plc, Cambridge, United Kingdom). Fungicides were incorporated into SDA at 0.1, 1, and 10⫻ the manufacturersÕ recommended application rate immediately before pouring. Conidia suspensions were prepared at a concentration of 1 ⫻ 107 ml⫺1, and 100 ␮l of suspension was spread evenly over SDA (15 ml) in petri dishes (90 mm in diameter, triple vented) that were then incubated in the dark at 20⬚C for 48 h. Plugs (6 mm) were cut from these cultures with a ßame-sterilized cork borer and placed upside down in the center of petri dishes of SDA plus fungicide (15 ml of SDA per dish), one plug per dish, and two dishes for each isolate. The dishes were sealed in polyethylene bags and incubated at 20⬚C for 28 d. Colony diameters were measured with a ruler using two cardinal diameters, every 3 to 4 d for the duration of the experiment. Mean colony radius was plotted against time, and colony extension rate was calculated during the linear phase (Fargues et al. 1992). The experiment was repeated three times. Iprodione and tebuconazole were examined in February 2002 for their effect on the efÞcacy of M. anisopliae 389.93 against D. radicum feeding on cabbage plants growing in pots under glasshouse conditions as described previously. Cabbage seedlings, B. oleracea Derby Day were inoculated with 30 D. radicum eggs and treated immediately with a foliar spray of iprodione or tebuconazole or a soil drench application of iprodione. The fungicides were applied at the manufacturersÕ recommended rate (1 g/liter for tebuconazole and 2.5 ml/liter for iprodione). The foliar applications were done to runoff by using a hand held sprayer (Cherwell, Southam, United Kingdom) and the soil drench application was done at a volume of 40 ml per pot. Conidial suspensions (1 ⫻ 108 ml⫺1, 40 ml) were applied to the plants immediately after the fungicide treatments and on three further occasions at 7, 14 ,and 21d thereafter. A positive control comprising plants treated only with M. anisopliae 389.93 also was included, together with an untreated control. Fungicide-only treatments were not included. The experiment was done according to a randomized block design, with six blocks. Each block included all Þve treatments with six plants per treatment. Four weeks after the application of eggs, the plants were harvested

Vol. 98, no. 6

and weighed. Larvae and pupae were washed from the pots and counted and incubated as described above. Field Experiment. The susceptibility of D. radicum to M. anisopliae 389.93 was evaluated in a Þeld experiment done in 2002 at the Warwick HRI research farm. Caulißower seedlings (B. oleraceaÔCassius F1⬘) were raised in the glasshouse in Levington F2 compost in 25-ml modular Hassy trays and transplanted into the Þeld (sandy loam of the Wick series, 2% organic matter content) at 6 wk old, on 12 June 2002. Five treatments were used: fungal conidia drenched onto plants at concentrations of 1 ⫻ 106, 1 ⫻ 107, and 1 ⫻ 108 ml⫺1, and two untreated controls. Treatments were applied to plots of 25 plants grown in 5 by 5 squares with a plant spacing of 50 cm. The plots were arranged according to a Latin Square design in which the treatments were replicated Þve times, i.e., Þve plots by Þve plots. Plots were separated by two rows of guard plants. Immediately after transplanting, plants were covered with horticultural ßeece until 11 July 2002, 2 d before the date predicted for 50% of second generation D. radicum adult females to oviposit according to the Warwick HRI D. radicum forecast (Finch et al. 1996). Immediately after the ßeece was removed, conidial suspensions (40 ml) were drenched onto the base of the plants. Further drenches were applied on three occasions at 7, 14, and 21d thereafter. Seven days after the last treatment, the nine central plants from each plot were cut at soil level and soil cores (600 ml) were taken around the plant stem, to a depth of 15 cm, by using an auger. Samples were placed into clear polyethylene bags (31 by 21 cm), sealed, and stored at 4⬚C in the dark for a maximum of 2 wk before processing. Larvae and pupae were then washed from the samples. Sporulating larval cadavers and the emergence of adult ßies from pupae were recorded as described previously. Data Analysis. Data were analyzed using the Genstat statistical package (Genstat 2002). The effect of the fungal treatments on D. radicum populations in the glasshouse and Þeld experiments was expressed in two ways: 1) the mean number of live larvae and pupae recovered per plant (this Þgure was used to indicate the level of control in the root zone), and 2) the proportion of adults that emerged from the pupae recovered per plant (this Þgure was used to indicate mortality in the pupal stage). These data were analyzed by analysis of variance (ANOVA) incorporating a log10 (⫹0.375) transformation for counts of the number of live larvae and pupae, and an angular transformation of the proportion of emerged adults, to stabilize the residual variance (Snedecor and Cochran 1989). The numbers of D. radicum cadavers that supported fungal mycelium were not analyzed because it has been shown that death of D. radicum by entomopathogenic fungi is not always followed by fungal colonization of the cadaver (Vanninen et al. 1999a). The fresh weights of plants from the glasshouse and Þeld experiments were analyzed by ANOVA, incorporating a log10 transformation. The effect of fungicides on the colony extension rates of fungal isolates was analyzed by ANOVA.

December 2005

CHANDLER AND DAVIDSON: FUNGAL BIOCONTROL OF CABBAGE MAGGOT

1859

Table 1. Effect of M. anisopliae 389.93 and 392.93 on the mean numbers (back transformed) of D. radicum recovered from pots of cabbage plants grown under glasshouse conditions Treatment Control agent

Application method

No. larvae ⫹ % adult emergence % larval cadavers % pupal cadavers Plant wt (g) pupae/plant from pupae with sporulating with sporulating (transformed (transformed data) (transformed data) mycelium mycelium data)

Control M. anisopliae 389.93 Compost mixture Drench M. anisopliae 392.93 Compost mixture Drench LSD (P ⬍ 0.05, ControlÐtreatment df ⫽ 10) TreatmentÐtreatment

6.6 (0.82) 8.7 (0.94) 0.7 (⫺0.15) 7.1 (0.85) 3.3 (0.52) (0.216) (0.250)

10.5 (18.88) 4.6 (12.35) 2.6 (9.35) 1.0 (5.79) 0.8 (5.28) (4.931)

0 50.5 71.1 68.2 61.0

0 2.3 2.5 3.9 5.2

38.0 (1.58) 33.1 (1.52) 55.0 (1.74) 38.0 (1.58) 40.7 (1.61) (0.129)

(5.693)

(0.150)

Transformed data shown in parentheses.

Results and Discussion Fungal Application Methods. Approximately 20% of D. radicum eggs placed on control plants were recovered as larvae and pupae (Table 1), which is in keeping with the survival rates from eggs to pupae recorded by Finch and Ackley (1977) for D. radicum on B. oleracea under glasshouse conditions. Drenches of M. anisopliae 389.93 and 392.93 reduced the mean number of D. radicum larvae and pupae recovered per plant compared with the control (t10 ⫽ 10.30, P ⬍ 0.01 and t10 ⫽ 3.20, P ⬍ 0.01, respectively), with M. anisopliae 389.93 causing a 90% reduction, whereas M. anisopliae 392.93 caused a 50% reduction (Table 1). The drench application of M. anisopliae 389.93 also increased mean plant weight (t8 ⫽ 2.90, P ⬍ 0.05). On the basis of these results, M. anisopliae 389.93 as a drench treatment was selected for further studies against D. radicum. In contrast, incorporating conidia into compost modules at planting had no effect on numbers of larvae and pupae recovered per plant for either isolate. Further work is required to determine whether the viability of conidia incorporated into compost declined over the course of the experiment, which could have accounted for the lack of control. However, all the fungal treatments reduced the emergence of adult ßies from the recovered pupae (t9 ⫽ 2.99, P ⬍ 0.05 and t9 ⫽ 4.37, P ⬍ 0.01 and t9 ⫽ 6.00, P ⬍ 0.01 and t9 ⫽ 6.24, P ⬍ 0.01, respectively), with M. anisopliae 392.93 causing ⬎90% reduction in emergence compared with the control. The proportion of larval cadavers recovered that supported sporulating mycelium ranged from 50.5 to 71.1% for the fungal treatments, whereas the proportion of pupae that supported sporulating mycelium ranged from only 2.3 to 5.2%. It is possible that the high degree of sclerotization of the puparium acted as a barrier to fungal outgrowth and prevented sporulation on the surface of pupal cadavers. The fate of the adult D. radicum that emerged from pupae was not followed in this study. However, Poprawski et al. (1985) noted that adult female D. antiqua that emerged from pupae treated with anamorphic entomopathogenic fungi, including M. anisopliae, were unable to oviposit, even though mating was observed. Similarly, Meadow et al. (2000) showed that treating adult D. radicum with B. bassiana prevented oviposition, in addition to causing rapid death.

Most conidia drenched onto the surface of compost remained in the top 10-cm layer, and particularly in the top 5-cm layer, although a small proportion percolated through all the layers examined (Table 2). It is possible therefore that conidia drenched onto compost formed a concentrated layer near the surface, increasing the exposure of D. radicum neonates to fungal inoculum as they moved to the root zone. Concentrating fungal inoculum at the soil surface was suggested by Vanninen et al. (1999b) as a possible strategy for increasing fungal infection of D. radicum larvae in soil. In contrast to the results of this study, Vanninen et al. (1999a) observed no control of D. radicum or D. floralis larvae on brassica plants under glasshouse conditions by using a single drench application of Beauveria bassiana (Balsamo) Vuillemin, Paecilomyces fumosoroseus (Wize) Brown & Smith, or M. anisopliae, although M. anisopliae caused 40 Ð50% larval mortality in a laboratory bioassay. This may indicate that repeated drench applications are necessary for effective control. Effects of Fungicides. All the fungicides were inhibitory to fungal colony extension, although there were differences in effect depending upon the type of fungicide and its concentration (Table 3). The greatest inhibition was caused by tebuconazole, and this fungicide was selected for examination against M. anisopliae 389.93 under glasshouse conditions. Iprodione had less of an effect, but it was still chosen for the glasshouse experiment because it is used as both a foliar spray and a soil drench treatment, enabling the impact of these different application methods to be compared. Table 2. compost

Movement of M. anisopliae 389.93 conidia through

Depth (cm)

% Conidia recovered from compost (SEM)

0Ð5 5Ð10 10Ð15 15Ð20 20Ð25 25Ð30 Efßuent

62.3 (11.67) 21.0 (7.22) 6.5 (2.23) 2.3 (0.74) 1.1 (0.29) 1.5 (0.56) 5.4 (2.28)

1860

JOURNAL OF ECONOMIC ENTOMOLOGY

Table 3. Effect of six fungicides on the colony extension rates of entomopathogenic fungi (millimeters dayⴚ1) Treatment Chlorothalonil Difenoconazole Iprodione Tebuconazole Tolclofus-methyl Triadimefon

Rate

M. anisopliae 389.93

M. anisopliae 392.93

2.00 1.25 1.11 1.61 0.02 0 1.68 1.42 0.85 0 0 0 1.80 1.34 1.16 1.70 1.09 0.98 2.13

1.86 1.28 1.15 1.16 0 0 1.92 1.39 0.54 0 0 0 2.74 1.09 0.72 1.82 0.99 0.86 3.42

0.1 1 10 0.1 1 10 0.1 1 10 0.1 1 10 0.1 1 10 0.1 1 10

Control

Fungicides were applied at 0.1, 1, and 10⫻ the manufacturersÕ recommended rate. Least signiÞcant difference (P ⬍ 0.05, df ⫽ 74) ⫽ 0.370.

In the glasshouse, M. anisopliae 389.93 on its own reduced the number of D. radicum larvae and pupae recovered per plant (t20 ⫽ 2.49, P ⬍ 0.05) (Table 4) and also reduced the proportion of adults that emerged from pupae (t20 ⫽ 5.62, P ⬍ 0.01). In addition, application of M. anisopliae 389.93 on its own improved mean plant weight (t20 ⫽ 3.10, P ⬍ 0.01). However, the size of these effects was less than in the previous experiment. For example, there was a 20% reduction in the mean number of D. radicum larvae and pupae recovered per plant, compared with a 90% reduction in the previous experiment. The emergence of adult ßies from pupae recovered in the control also seemed to be higher. The reason for the difference in fungal efÞcacy between the two experiments is not known. However, high levels of emergence of adults in the control may indicate that the D. radicum population was in a better physiological condition in this year and hence better able to resist fungal infection. Overall, the fungicides were compatible with the use of M. anisopliae 389.93 in the glasshouse, although

Vol. 98, no. 6

there were some mixed effects. The number of D. radicum larvae and pupae recovered per plant was reduced when M. anisopliae 389.93 was used with foliar sprays of iprodione or tebuconazole (t20 ⫽ 2.74, P ⬍ 0.05 and t20 ⫽ 5.97, P ⬍ 0.01, respectively) (Table 4). However, a statistically signiÞcant reduction did not occur when M. anisopliae 389.93 was used with a soil drench application of iprodione, although this may have been confounded by lower fungal efÞcacy compared with the previous experiment. The proportion of D. radicum adults that emerged from pupae was reduced when M. anisopliae 389.93 was used with each of the three fungicide treatments (foliar spray of iprodione: t20 ⫽ 5.77, P ⬍ 0.01; foliar spray of tebuconazole: t20 ⫽ 7.46, P ⬍ 0.01; soil drench of iprodione: t20 ⫽ 4.02, P ⬍ 0.01). The sizes of the reductions were similar to that obtained with the fungus alone. Mean plant weight was increased when M. anisopliae 389.93 was used with the soil drench and foliar spray of iprodione (t20 ⫽ 3.58, P ⬍ 0.01 and t20 ⫽ 2.45, P ⬍ 0.05, respectively), but it was not increased when it was used with the foliar spray of tebuconazole. Lack of inhibition of M. anisopliae 389.93 by fungicides under glasshouse conditions may have been due to spatial separation of the fungicides and conidia. Only a small number of other studies have been done in this area, but they suggest that laboratory tests do not predict reliably the effects of fungicides on the plant scale performance of entomopathogenic fungi, although fungicides that are harmless in vitro are likely to remain so in the Þeld (Loria et al. 1983, Majchrowicz and Poprawski 1993, Jaros-Su et al. 1999). Similar Þndings have been found with mycoparasitic fungi used for the biocontrol of plant disease (Budge and Whipps 2001). The results of the current study corroborate these Þndings and stress the importance of glasshouse or Þeld scale evaluations in determining the compatibility of entomopathogenic fungi with agrochemicals. Field Experiment. In contrast to the glasshouse experiments, application of M. anisopliae 389.93 in the Þeld caused no reduction in the total number of D. radicum larvae and pupae or to the proportion of adults that emerged per plant (Table 5). However, up to 30% of the larval cadavers recovered from plants treated with the fungus supported sporulating myce-

Table 4. Effect of iprodione or tebuconazole fungicides on the mean numbers (backtransformed) of D. radicum recovered from pots of cabbage plants treated with M. anisopliae 389.93 and grown under glasshouse conditions Treatment Control agent

Fungicide

Control M. anisopliae 389.93 M. anisopliae 389.93 Iprodione (soil drench) M. anisopliae 389.93 Iprodione (foliar spray) M. anisopliae 389.93 Tebuconazole (foliar spray) Least signiÞcant difference (P ⬍ 0.05, df ⫽ 20)

No. larvae ⫹ % adult emergence % larval cadavers % pupal cadavers Plant wt, g pupae/plant from pupae with sporulating with sporulating (transformed (transformed data) (transformed data) mycelium mycelium data) 17.4 (1.24) 14.4 (1.16) 15.1 (1.18) 14.1 (1.15) 11.2 (1.05) (0.065)

52.6 (46.50) 22.9 (28.60) 22.2 (28.10) 14.9 (22.70) 30.8 (33.70) (6.650)

0 28.9 30.6 22.2 19.4

0 10.7 8.9 19.1 8.3

6.9 (0.84) 11.0 (1.04) 11.8 (1.07) 10.0 (1.00) 5.5 (0.74) (0.135)

December 2005

CHANDLER AND DAVIDSON: FUNGAL BIOCONTROL OF CABBAGE MAGGOT

1861

Table 5. Effect of drenches of M. anisopliae 389.93 on mean numbers (backtransformed) of D. radicum recovered from cauliflower plants in a field experiment Treatment Control agent Control M. anisopliae 389.93 Least signiÞcant difference (P ⬍ 0.05, df ⫽ 12)

Concn

No. larvae ⫹ pupae per plant

% adult emergence/plant

% larval cadavers with sporulating mycelium

1 ⫻ 106 ml⫺1 1 ⫻ 107 ml⫺1 1 ⫻ 108 ml⫺1 ControlÐtreatment TreatmentÐtreatment

7.1 (0.85) 6.8 (0.83) 7.6 (0.88) 6.2 (0.79) (0.178) (0.206)

23.0 (28.68) 23.9 (29.28) 20.5 (26.93) 16.9 (24.24) (5.216) (6.023)

0 6 26.4 30.3

Transformed data are given in parentheses.

lium. There also were indications of a positive relationship between the concentration of the conidial suspension applied to plants and the proportion of larval cadavers that supported sporulating mycelium as well as a negative relationship with the proportion of adult ßies that emerged from pupae. Vanninen et al. (1999b) were similarly unable to control D. radicum and D. floralis larvae in the Þeld with Þve different species of entomopathogenic fungi, including M. anisopliae. They suggested this was because D. radicum and D. floralis larvae have evolved a high degree of resistance to infection, because they inhabit an environment of decomposing plant material that contains large numbers of microorganisms. Low susceptibility to fungal infection has been observed for larvae of tsetse ßies and fruit ßies (Kaaya and Munyinyi 1995, De La Rosa et al. 2002) and may be a feature of the Diptera. However, other factors are likely to have inßuenced fungal efÞcacy in the Þeld in the current study, because we were able to control D. radicum under glasshouse conditions. These could include soil type, temperature, moisture, and biotic factors, which are all known to affect the distribution, persistence, or infectivity of entomopathogenic fungi (Storey and Gardner 1988, Inglis et al. 2001). Vanninen et al. (1999b) speculated that D. radicum larvae may use isothiocyanates produced by brassica plants as a defense against fungal infection. These compounds have been shown to inhibit the growth of entomopathogenic fungi in laboratory studies (Inyang et al. 1999, Klingen et al. 2002), but brassicaceous plants had no effect on the survival of conidia in soil in a growth chamber experiment (Klingen et al. 2002). Fungistatic effects caused by other soil microorganisms also could be important because entomopathogenic fungi are poor competitors in soil (Lockwood and Filonow 1981, Studdert and Kaya 1990, Pereira et al. 1993), although it might be possible to overcome this using a novel formulation (Groden and Lockwood 1991). It was noticed that the drenches applied in the Þeld often exhibited considerable lateral movement across the ground before penetrating the soil, which is likely to have reduced the amount of inoculum acquired by D. radicum larvae, despite that previous glasshouse experiments indicated that M. anisopliae had to be applied as a drench to be effective. Hence, improvements in application method may result in increased control in the Þeld.

In conclusion, this study has shown that M. anisopliae can control populations of D. radicum larvae feeding on the roots of brassica plants under glasshouse conditions and is unlikely to be affected adversely by chemical fungicides. Control of D. radicum larvae in Þeld conditions is more problematic and further research is needed if a reliable and cost-effective fungal control agent is to be developed. In particular, research should concentrate on identifying isolates with greater virulence, developing an effective application method, and investigating in more detail the distribution, persistence, and infectivity of inoculum in soil. Acknowledgments We thank Kath Phelps (Warwick HRI) for specialist advice on experimental design and Rosemary Collier (Warwick HRI) for critically evaluating the manuscript. This work was funded by the United Kingdom Department for Environment, Food, and Rural Affairs.

References Cited Bedding, R. A., and R. J. Akhurst. 1975. A simple technique for the detection of insect parasitic rhabditid nematodes in soil. Nematologica 21: 109 Ð116. Budge, S. P., and J. M. Whipps. 2001. Potential for integrated control of Sclerotinia sclerotiorum in glasshouse lettuce using Coniothyrium minitans and reduced fungicide application. Phytopathology 91: 221Ð227. Chandler, D. 1994. Cryopreservation of fungal spores using porous beads. Mycol. Res. 98: 525Ð526. Davidson, G., and D. Chandler. 2005. Laboratory Evaluation of Entomopathogenic Fungi Against Larvae and Adults of Onion Maggot (Diptera: Anthomyiidae). 98: 1848 Ð1855. De La Rosa, W., F. L. Lopez, and P. Liedo. 2002. Beauveria bassiana as a pathogen of the Mexican fruit ßy (Diptera: Tephritidae) under laboratory conditions. J. Econ. Entomol. 95: 36 Ð 43. Fargues, J., N. K. Maniania, J. C. Delmas, and N. Smits. 1992. Inßuence de la temperature sur la croissance in vitro dÕhyphomycetes entomopathogenes. Agronomie 12: 557Ð 564. Finch, S. 1989. Ecological considerations in the management of Delia pest species in vegetable crops. Annu. Rev. Entomol. 34: 117Ð137. Finch, S. 1993. Integrated pest management of the cabbage root ßy and the carrot ßy. Crop Prot. 12: 423Ð 430.

1862

JOURNAL OF ECONOMIC ENTOMOLOGY

Finch, S., and T. H. Coaker. 1969. A method for the continuous rearing of the cabbage root ßy Erioischia brassicae (Bouche) and some observations on the biology. Bull. Entomol. Res. 58: 619 Ð 627. Finch, S., and C. M. Ackley. 1977. Cultivated and wild host plants supporting populations of the cabbage root ßy. Ann. Appl. Biol. 85: 13Ð22. Finch, S., R. H. Collier, and K. Phelps. 1996. A review of work done to forecast pest insect attacks in UK horticultural crops. Crop Prot. 15: 353Ð357. Genstat. 2002. The guide to Genstat: part 2: statistics. VSN International Ltd., Oxford, United Kingdom. Goettel, M. S., and G. D. Inglis. 1997. Fungi: hyphomycetes, pp. 213Ð249. In L. Lacey [ed.], Manual of techniques in insect pathology. Academic, San Diego, CA. Groden, E., and J. L. Lockwood. 1991. Effects of soil fungistasis on Beauveria bassiana and its relationship to disease incidence in the Colorado potato beetle, Leptinotarsa decemlineata, in Michigan and Rhode Island soils. J. Invertebr. Pathol. 57: 7Ð16. Hughes, R. D., and D. D. Salter. 1959. Natural mortality of Erioischia brassicae (Bouche´ ) (Dipt., Anthomyiidae) during the immature stages of the Þrst generation. J. Anim. Ecol. 28: 231Ð241. Inglis, G. D., M. S. Goettel, T. M. Butt and H. Strasser. 2001. Use of hyphomycetous fungi for managing insect pests, pp. 23Ð 69. In T. M. Butt, C. W. Jackson, and N. Magan [eds.], Fungi as biocontrol agents; progress, problems and potential. CABI Publishing, Wallingford, United Kingdom. Inyang, E. N., T. M. Butt, K. J. Doughty, A. D. Todd, and S. Archer. 1999. The effects of isothiocyanates on the growth of the entomopathogenic fungus Metarhizium anisopliae and its infection of the mustard beetle. Mycol. Res. 103: 974 Ð980. Jaros-Su J., E. Groden, and J. Zhang. 1999. Effects of selected fungicides and the timing of fungicide application on Beauveria bassiana-induced mortality of the Colorado potato beetle (Coleoptera: Chrysomelidae). Biol. Control 15: 259 Ð269. Kaaya, G. P., and D. M. Munyinyi. 1995. Biocontrol potential of the entomogenous fungi Beauveria bassiana and Metarhizium anisopliae for tsetse ßies (Glossina sp.) at developmental sites. J. Invertebr. Pathol. 66: 237Ð241. Klingen, I., A. Hajek, R. Meadow, and J.A.A. Renwick. 2002. Effect of brassicaceous plants on the survival and infectivity of insect pathogenic fungi. Biocontrol 47: 411Ð 425. Lockwood, J. L., and A. B. Filonow. 1981. Response of fungi to nutrient limiting conditions and to inhibitory substances in natural habitats, pp 1Ð 61. In M. Alexander [ed.], Advances in microbial ecology, vol. 5. Plenum, New York.

Vol. 98, no. 6

Loria, R., S. Galaini, and D. W. Roberts. 1983. Survival of inoculum of the entomopathogenic fungus Beauveria bassiana as inßuenced by fungicides. Environ. Entomol. 12: 1724 Ð1726. Majchrowicz, I., and T. J. Poprawski. 1993. Effects in vitro of nine fungicides on growth of entomopathogenic fungi. Biocontrol Sci. Technol. 3: 321Ð336. Meadow, R., J. D. Vandenberg, and A. M. Shelton. 2000. Exchange of inoculum of Beauveria bassiana (Bals.) Vuill. (Hyphomycetes) between adult ßies of the cabbage maggot, Delia radicum L. (Diptera: Anthomyiidae). Biocontrol Sci. Technol. 10: 479 Ð 485. Pereira, R. M., J. L. Stimac, and S. B. Alves. 1993. Soil antagonism affecting the dose response of workers of the red imported Þre ant, Solenopsis invicta, to Beauveria bassiana conidia. J. Invertebr. Pathol. 61: 156 Ð161. Poprawski, T. J., P. H. Robert, I. Majchrowicz, and G. Boivin. 1985. Susceptibility of Delia antiqua (Diptera: Anthomyiidae) to eleven isolates of entomopathogenic hyphomycetes. Environ. Entomol. 14: 557Ð561. Smith, K. M. 1927. A study of Hylemyia (Chortophila) brassicae (Bouche´ ), the cabbage root ßy and its parasites, with notes on some other Dipterous pests of cruciferous plants. Ann. Appl. Biol. 14: 312Ð329. Snedecor, G. W., and W. G. Cochran. 1989. Statistical methods. Iowa State University Press, Ames. Storey, G. K., and W. A. Gardner. 1987. Vertical movement of commercially formulated Beauveria bassiana conidia through four Georgia soil types. Environ. Entomol. 16: 178 Ð181. Storey, G. K., and W. A. Gardner. 1988. Movement of an aqueous spray of Beauveria bassiana into the proÞle of four Georgia soils. Environ. Entomol. 17: 135Ð139. Studdert, J. P., and H. K. Kaya. 1990. Water potential, temperatures and clay coating of Beauveria bassiana conidia: effect on Spodoptera exigua pupal mortality in two soil types. J. Invertebr. Pathol. 56: 327Ð336. Vanninen, I., H. Hokkanen, and J. Tyni-Juslin. 1999a. Attempts to control cabbage root ßies Delia radicum L. and Delia floralis (Fall.) (Dipt., Anthomyiidae) with entomopathogenic fungi: laboratory and greenhouse tests. J. Appl. Entomol. 123: 107Ð113. Vanninen, I. H. Hokkanen, and J. Tyni-Juslin. 1999b. Screening of Þeld performance of entomopathogenic fungi and nematodes against cabbage root ßies (Delia radicum L. and D. floralis (Fall.); Diptera, Anthomyiidae). Acta Agric. Scand. B-S P. 49: 167Ð183. Zimmerman, G. 1986. The ÔGalleria bait methodÕ for detection of entomopathogenic fungi in soil. Z. Angew. Entomol. 102: 213Ð215. Received 23 February 2005; accepted 19 June 2005.