Insecticide Resistance in Diamondback Moth in Florida

47 Insecticide Resistance in Diamondback Moth in Florida Gary L. Leibee and Kenneth E. Savage University of Florida, IFAS, Central Florida Research and Education Center, 2700 East Celery Avenue, Sanford, Florida 3277 1-9608, USA Abstract Topical application and leaf residue bioassays were conducted to determine the toxicity of selected insecticides to an apparently fenvalerate-resistant population of diamondback moth, Plutella xylostella (L.), from central Florida. A field study was conducted to evaluate selected insecticides for the control of diamondback moth and cabbage looper, Trichoplusia ni (Hubner), in cabbage, Brassica oleracea L. (capitata group). Laboratory strains of diamondback moth established from populations from selected treatment plots of the field study were evaluated for susceptibility to fenvalerate. The laboratory studies indicated that the diamondback moth population was resistant to fenvalerate and methomyl, but susceptible to chlorpyrifos, acephate, endosulfan, and thiodicarb. The field study indicated that chlorpyrifos, endosulfan, mevinphos, and Bacillus thuringiensis var. kurstaki were considerably more effective at suppressing diamondback moth than cypermethrin, permethrin, methomyl, and thiodicarb. Cypermethrin, permethrin, thiodicarb, and endosulfan were the most effective in controlling T. ni. Mixtures of two insecticides improved efficacy apparently due to one component controlling T. ni and not diamondback moth and the other component controlling diamondback moth and not T. ni. Examination of selected populations from the field study indicated that resistance to fenvalerate increased within one crop season with weekly use of cypermethrin and thiodicarb, but not with chlorpyrifos. Introduction The diamondback moth (DBM), Plutella xylostella (L.) (Lepidoptera: Yponomeutidae), a worldwide pest of cruciferous crops (Talekar and Griggs 1986), was not difficult to control in Florida until about the mid 1980s. Growers began complaining of failing to control DBM in crucifers in transplant (seedling) production facilities and in the field with insecticides. Since about 1985, DBM has replaced the cabbage looper, Trichoplusia ni (Hubner) (Lepidoptera: Noctuidae), as the major target of insect control activities in cruciferous crops in Florida (G. L. Leibee, personal observation). The principal insecticides of choice in Florida were fenvalerate and permethrin which until this lime (1985), provided excellent control of the cabbage caterpillar complex, which typically consisted of DBM, T. ni, imported cabbageworm, Artogeia rapae (L.) (Lepidoptera: Pieridae), and the cabbage webworm, Hellula rogatalis Hulst. (Lepidoptera: Pyralidae). Fenvalerate failed to provide adequate caterpillar control in cabbage in the spring of 1985 at the Central Florida Research and Education Center (CFREC) in Sanford, Florida (Leibee 1986a). Fenvalerate controlled DBM in the spring of 1986 at the same location, indicating an apparently susceptible population was present (Leibee 1986b). From the winter of 1986-87 to the present, pyrethroid insecticides in general provided essentially no control of DBM at the CFREC (G. L. Leibee, unpublished data). Magaro and Edelson (1990) reported that failures to control DBM in south Texas were first reported by cabbage producers in the spring of 1987. Personal communication with entomologists during the late 1980s indicated that control of DBM 427

428 Leibee and Savage had become a problem in other states bordering the Gulf of Mexico and generally up the eastern seaboard of the United States. Since resistance to fenvalerate in DBM had been reported in Taiwan (Liu et al. 1982), Japan (Hama 1987), and in other areas of Southeast Asia (Cheng 1988), it was obvious that a similar problem was developing in the eastern United States. The studies reported herein were conducted to document and characterize the nature of insecticide resistance in DBM in central Florida. Laboratory studies were conducted to determine the toxicity of selected insecticides to an apparently fenvalerate-resistant population of DBM and to determine the ability of selected insecticides to increase resistance levels in the field. Field studies were conducted to evaluate selected insecticides for the control of DBM and T. ni. Materials and Methods DBM colonies: A population of DBM was allowed to build up to a moderate level in a 0.81-ha field of collards, Brassica oleracea L. (Acephala group), at the CFREC-Sanford. Prior to starting the spray program, a colony (strain A) was started from 139 pupae reared from larvae taken from the collard field from 22 April to 7 May 1987. A spray program was initiated on 13 May where fenvalerate was applied at 0.224 kg AI/ha in 467 1 of water on a weekly basis for 5 weeks. One day after the first application, it was observed that the DBM larvae were much more difficult to kill than the T. ni and A. rapae larvae that were also present in the collard field, indicating that the DBM population was already somewhat resistant to the fenvalerate. At the end of the 5-week spray program, it was apparent that fenvalerate was ineffective in controlling this population of DBM because the collard plants were completely skeletonized, and upon inspection several hours after spraying, there were no obvious effects of the spray on the larvae present. Another colony (strain B) was started from several hundred adults collected with a sweep net on 15 June, 5 days after the last spray application. Adults were used to start this second colony because the larvae that were present appeared diseased. These two colonies were maintained on rape, Brassica napus L., seedlings at 25 C and a 15-hour photoperiod and without exposure to insecticides until needed for the bioassays. Insecticides: Technical grade materials used in the topical application bioassays were fenvalerate (I. E. DuPont denemours & Co., Wilmington, Delaware), chlorpyrifos (Dow Chemical Co., Midland, Michigan), and methamidophos (Chevron Chemical Co., San Francisco, California). Formulated materials used in both the leaf residue bioassays and the field trials were chlorpyrifos (50% wettable powder [WP], Dow Chemical Co.), methomyl (215 g/l soluble liquid [L], I. E. dupont denemours & Co.), thiodicarb (383 g/l flowable liquid [F], Rhone-Poulenc), and endosulfan (359 g/l EC, FMC). Formulated materials used in the field trials only were cypermethrin (359 g/l EC, ICI Americas), permethrin (383 g/l EC, FMC), mevinphos (479 g/l EC, I. E. dupont denemours & Co.), Dipel 2X (Bacillus thuringiensis var. kurstaki, 32.0 BIU/kg wettable powder [WP], Abbott), and Javelin (B. thuringiensis var. kurstaki, 16 billion Spodoptera Units/l, Sandoz). Additional formulated materials used in the leaf residue bioassays, but not in the field, were acephate (75% soluble powder (SP), Chevron Chemical Co.) and fenvalerate (287 g/l EC, I. E. dupont denemours & Co.). Bioassays: Bioassays were conducted in March and April of 1989 when strain A was in generation 29 and 30 and strain B was in generation 27 and 28. Topical application and feeding studies were utilized to conduct the bioassays. For topical application, a 0.5 µ1 droplet of acetone containing the appropriate concentration of technical insecticide was applied with an Arnold type micro-applicator (Burkard Manufacturing Co. Ltd., England) to the middle dorsum of an anesthetized early 4th instar larva (3-4 mg). Generally, each dosage was replicated three times with 10 larvae/replicate. The number of dead and moribund larvae was recorded after 24 and 48 hours. A larva was considered moribund if it could not move one body length after prodding.

Insecticide Resistance in Florida 429 Larvae were collected from the colonies, sorted to size, placed in the glass petri dishes lined with filter paper and covered. Larvae were provided with pieces of rape seedlings until anesthesia and dosing. All anesthesia, dosing, and maintenance of the larvae were conducted with the larvae in the petri dish. Prior to anesthesia, the larvae were dislodged from the rape seedlings with a camel-hair brush and the plant material removed to facilitate positioning of the larvae. The larvae were anesthetized with carbon dioxide for 8-10 min in a clear polystyrene plastic box with lid. The anesthetized larvae were removed from the box, dosed, and provided with three rape seedlings per petri dish. Leaf residue bioassays were conducted by placing early 4th instar larvae (3-4 mg) in glass petri dishes containing pieces of rape seedlings coated with dried residues of formulated insecticides and recording the number of dead or moribund larvae after 24 and 48 hours. The rape seedlings were dipped as whole plants for 5 sec in serial dilutions of the formulated insecticide and allowed to dry. Enough plants were dipped with each dose to provide extra for replenishing the food supply, if necessary, during the bioassay. Each petri dish contained three seedlings (generally two cotyledon leaves and one true leaf completely expanded) excised well into the dipped part of the stem. The petri dish bottoms were lined with filter paper. The range of doses for each insecticide included concentrations representing the low and high recommended field rates in 467 1 water/ha and concentrations above and below these. Field study: Golden Acre Yellows Resistant cabbage was transplanted from seedbeds on 20 and 21 April 1989 into Myakka fine sand at the CFRE-Sanford, Florida. Plots consisted of four 15.2 m rows with a 0.76 m row-spacing and about a 0.28 m plant spacing. Four rows were left unplanted between each plot to provide a separation of 3.8 m. Plots were arranged in four blocks and the blocks were separated by 7.6 m alleyways. All the treatments, except for endosulfan, were assigned to the plots in a randomized complete block design with four replicates. The endosulfan treatments were an afterthought and were assigned to plots at the same end of each block originally not to be used because of high soil moisture resulting from poor drainage. Standard cultural practices for the area were used for fertilization and weed control. Sprays were applied with a tractor-mounted, compressed-air sprayer. Three hollow-cone nozzles (D2-25) were used per row; one overhead and one drop on each side. The delivery rate of spray was 467 l/ha with a boom pressure of about 3163 g/cm² (45 psi) and a speed of 3.2 km/hour. Application dates were 11, 16, 23 May and 1, 8, 16 June 1989. Insecticides and rates (expressed as kilograms of active ingredient [AI] per hectare unless stated otherwise) used in treatments consisted of chlorpyrifos at 1.12, cypermethrin at 0.067, Javelin at 1. 17 1, methomyl at 1.01, permethrin at 0.22, thiodicarb at 0.84 and 1.12, mevinphos at 1.12, and endosulfan at 1.12. Treatments consisting of a mixture of two insecticides and rates listed above were cypermethrin with chlorpyrifos, cypermethrin with Javelin, chlorpyrifos with Javelin, and thiodicarb with mevinphos. An additional treatment of chlorpyrifos mixed with Dipel 2X at 0.56 kg/ha was included. An untreated check was included. The wetting agent X-77 (Chevron Chemical Co.) was used in all treatments at the rate of 0.62 ml/l of spray. Plants were sampled on 22 and 31 May, and 7 and 14 June 1989 to determine larval numbers. Four randomly selected plants per plot (two from the middle of each center row) were sampled by carefully cutting the stem below the portion of the plant containing the bud or head and the next four youngest leaves (wrapper leaves in the head stage). The stem was cut below all the foliage if only four leaves were present. The four plants were placed in a plastic bag and transported back to the processing area. Each sample of four plants was placed into a Berlese funnel and subjected to heat for 24 hours. When heads were present, the infested head leaves were pulled away from the head and the uninfested portion of the head was cut out to reduce the amount of plant material that went into the funnel. Larvae were collected into 70% ethyl alcohol. All species except DBM were categorized according to size (small, medium, and large). DBM was categorized according to instar by head capsule size. Ten mature plants were rated for damage on 16 June 1989 using a scale of 1-6 similar to that of Greene et al. (1969). Percent marketability for normal market conditions (no head damage) was based on the proportion of

430 Leibee and Savage plants having a damage rating of 3.0 or less. Percent marketability for exceptional market conditions (slight amount of head damage) was based on the proportion of plants having a damage rating of 4.0 or less. Damage ratings were taken from only one plot of the endosulfan treatments due to severe stunting in the other three replicates resulting from excessive soil moisture. Insecticide resistance development study: Colonies of DBM were established from the untreated check, chlorpyrifos, cypermethrin, and thiodicarb plots of the field trial. Larvae and pupae were collected from each replicate of a treatment and combined to provide at least 100 individuals to start each colony. Collections were made from the untreated check plots on 22 June, the cypermethrin plots on 23 June, and the chlorpyrifos and thiodicarb plots on 26 June 1989. The colonies were maintained on rape seedlings into the F2 generation until there were enough individuals to determine the toxicity of fenvalerate to each colony using the leaf residue procedure described above. Statistical analysis: Concentration-mortality response data were subjected to probit analysis (Finney 197 1) where applicable. All concentration-mortality response data were corrected for control mortality (Abbot 1925) unless indicated otherwise. Data from the field trial were subjected to analysis of variance and means were separated by Duncan s (1955) Multiple Range Test (P = 0.05 level). Damage ratings were not transformed. Percentage data were transformed using arcsine. Larval count data were transformed using log (X + 1.5) for DBM and square root (X + 1.5) for T. ni. Actual means are presented in tables. Results and Discussion Bioassays: There was a significant difference in toxicity of fenvalerate to DBM strains A and B using topical application based on nonoverlap of the 95 % fiducial limits (Table 1). The LD50 values of 0.96 and 36.75 µg/larva for strains A and B, respectively, indicated that strain B was about 38 times more resistant to fenvalerate than strain A and that both strains were comparable to the super-resistant strains described by Hama (1987). A very stable kind of resistance is indicated in that strains A and B have retained this high level of fenvalerate resistance after 29 and 27 generations, respectively, without selection. Highly stable fenvalerate resistance in DBM has been indicated in Japan (Hama 1988) and Taiwan (Chen and Sun 1986). There were no significant differences in the toxicity of chlorpyrifos and methamidophos to strains A and B based on overlap of the 95% fiducial limits (Table 1). But, the LD50 values indicated that chlorpyrifos was about 2.7 times more toxic to strain A as it was to strain B. All the leaf residue concentration-response data were subjected to probit analysis, but only the data for fenvalerate, acephate, and thiodicarb fit the probit model well enough to present the LD50 and slope values (Table 2). There were no significant differences in the toxicity of fenvalerate, acephate, and thiodicarb to strains A and B using the leaf residue bioassay based on overlap of the 95% fiducial limits. But, the LD50 values indicated that fenvalerate was about twice as toxic to strain A than to strain B. Even though the leaf residue concentration-mortality response data for methomyl, chlorpyrifos, and endosulfan did not fit the probit model, the data for concentrations representing 0.5, 1.0, and 2.0 times the maximum recommended field rate were considered valuable when presented with similar data for fenvalerate, acephate, and thiodicarb (Table 3). Results indicated very little difference between strains A and B in response to the insecticides. Data suggest that thiodicarb, acephate, chlorpyrifos, and endosulfan at the maximum recommended field rate would be effective in controlling this resistant strain of DBM where fenvalerate and methomyl would not. The great differences in the relative toxicity of fenvalerate to strains A and B indicated by the two different bioassay methods might be due to differences in the routes of entry of the toxicant inherent in the two bioassay methods. Topical application relies on contact toxicity only where leaf residue relies on ingestion of the toxicant and probably some toxicity from contact

Insecticide Resistance in Florida 43 1 Table I. Toxicity of fenvalerate (48 hours), chlorpyrifos (24 hours), and methamidophos (24 hours) to two strains (A and B) of DBM using topical application. Insecticide and strain n Slope ± SE LD50 (ppm) 95% FL (generation) Fenvalerate A (F29) I77 0.44 + 0.09 I9 16.98 283.20-9152.03 B (F27) I80 2.84 + 0.69 73490.94 20515.52-15841 1.64 Chlorpyrifos A (F30) I80 2.01 0.35 5234.39 2665.07-8615.22 B (F28) I80 1.24 + 0.24 I4 149.43 8244.15-26645.65 Methamidophos A (F30) I80 1.02 + 0.17 6748.73 2842.90-14459.00 B (F28) I80 0.91 + 0.21 680 I.80 1663.70-22307.00ª fiducial limits. - Table 2. Toxicity of fenvalerate (48 hours), acephate (24 hours), and thiodicarb (48 hours) to two strains (A and B) of DBM using a leaf residue bioassay. Insecticide and strain n Slope ± SE LD50 (g AI/I) 95% FL (generation) Fenvalerate A (F29) I80 0.02 + 0.01 2.43 I.4 1-4.93 B (F27) I80 0.02 + 0.01 4.76 3.37- I 0.36 Acephate A (F31) 210 1.45 + 0.32 0.28 0.06-0.54 B (F29) 210 1.42 + 0.26 0.50 0.19-0.84 Thiodicarb A (F3I) I77 1.15 + 0.29 0.63 B (F29) I80 1.63 + 0.31 0.74 0.10-I. I8 0.26-1.23 Table 3. Concentration-mortality (48 hours) response of two strains of DBM exposed to leaf residues of fenvalerate, methomyl, thiodicarb, acephate, chlorpyrifos, and endosulfan representing 0.5, 1.0 and 2.0 times the recommended field rates. Insecticide Mortality (%) (SE) Strain 0.5X I.0xª 2.0x Fenvalerate A 1 1 (8) 26 (12) 54 (9) B 27 (8) 36 (22) 66 (I 8) Methomyl A 65 (7) 72 (8) 77 (12) B 52 (10) 53 (12) 93 (7) Thiodicarb A 41 (9) 90 (10) 76 (9) B 71 (7) 93 (3) 82 (10) Acephate A 92 (4) 96 (4) 100 (0) B 100 (0) 87 (8) 100 (0) Chlorpyrifos A 72 (16) 82 (12) 78 (7) B 65 (0) 88 (7) 84 (10) Endosulfan A 86 (4) 100 (0) 89 (6) B 82 (4) 93 (4) 96 (4) ª1.OX for fenvalerate. methomyl, thiodicarb, acephate. chlorpyrifos, and endosulfan is 0.48, 2.16, I.92, 2.40, 2.40 and 2.40 g AI/l, respectively.

432 Leibee and Savage with the leaf residue. The difference in toxicity to fenvalerate between strains A and B with topical application could indicate slower cuticular penetration in strain B. Noppun et al. (1989) showed reduced cuticular penetration of fenvalerate in resistant strains of DBM. Ingestion of fenvalerate with the leaf material might remove differences in cuticular penetration as a factor in toxicity. Also, the similarity between strains A and B indicated by the leaf residue bioassay might be due to loss of resistance in strain B over time due to lack of selection with fenvalerate to maintain the original levels of resistance. Field study: DBM and T. ni were the predominant species present throughout the experiment. Numbers of DBM were different (P < 0.05) among treatments on the last three sample dates (Table 4). There were few differences (P <0.05) in the numbers of T. ni among treatments on the first sample date and no differences (P > 0.05) on the last three sample dates. Numbers of T. ni appeared to be very low on each sample date, but a considerable amount of damage in several treatments was characteristic of that produced by T. ni. Since T. ni larvae can consume 18.4 times more cabbage than DBM larvae (East et al. 1989), the numbers of T. ni present over the entire sampling period were considered high enough in several treatments to cause significant damage. When the numbers of T. ni were combined over all sample dates, there were more and greater differences (P < 0.05) among treatments (Table 4) which corresponded well with amounts of damage present (Table 5). In the treatments consisting of a single insecticide, chlorpyrifos, mevinphos, endosulfan, and Javelin were considerably more effective at reducing numbers of DBM than cypermethrin, permethrin, methomyl, and thiodicarb, suggesting a higher level of resistance to pyrethroids and carbamates than to organophosphates, endosulfan, and B. thuringiensis (Table 4). Except for thiodicarb, the effects of chlorpyrifos, endosulfan, and methomyl on the numbers of DBM in the field study were in agreement with the results of the leaf residue study (Table 3). Cypermethrin, permethrin, thiodicarb, and endosulfan were most effective at reducing numbers of T. ni. Endosulfan was the only insecticide that was effective against both species and produced the most marketable cabbage in the experiment among the treatments consisting of a single insecticide (Table 5). The remaining single insecticide treatments provided essentially no marketable cabbage under normal market conditions. The insecticide mixtures resulted in a significant (P <0.05) decrease in damage when compared to the damage in the untreated check and that of components of the mixtures, except for cypermethrin + Javelin. Based on the frequency of damage ratings of 3.0 or less, a significant (P <0.05) increase in marketability over the untreated check was indicated in the treatments consisting of the following mixtures listed in order of decreasing marketability: cypermethrin + chlorpyrifos, thiodicarb + mevinphos, and chlorpyrifos + Dipel 2X. Cypermethrin + chlorpyrifos was significantly (P <0.05) better than all other mixtures. Based on the frequency of damage ratings of 4.0 or less, which represents a relatively high level of insect control, all mixtures resulted in a significant increase in marketability over the untreated check with the increases becoming quite high in all but the cypermethrin + Javelin. In general, based on larval numbers, the increase in efficacy from the mixtures appears to be due to one component controlling DBM and not T. ni and the other component controlling T. ni, but not DBM. The cypermethrin-chlorpyrifos mixture resulted in the greatest reduction in damage probably because the cypermethrin alone provided the best T. ni control and the chlorpyrifos alone provided the best DBM control. Insecticide resistance development study: The toxicity of fenvalerate to DBM strains established from the untreated check, chlorpyrifos, cypermethrin, and thiodicarb plots indicated an increase in resistance to fenvalerate had developed in the cypermethrin and thiodicarb plots relative to the untreated check and chlorpyrifos plots (Table 6). This increase in resistance to fenvalerate occurred over the 5-week spray period during which no more than three generations of DBM could have occurred based on the shortest development time of 14.55 days from egg

Insecticide Resistance in Florida 433 Table 4. Effects of selected insecticidal treatments on larval numbers for DBM and T. ni on cabbage. Insecticide and rate (kg AI or I formulation / ha) No. 2nd-4th instar DBM larvae/4 plantsa 31 May 7 June 14 June Total no. T. ni larvae/4 plantsª for all sample dates Untreated check 7.3 ab 5.0 cde I I.0 bcd 6.3 ab Chlorpyrifos I. I2 0.8 fgh 2.3 e-h 3.3 def 6.0 ab Cypermethrin 0.067 7.8 a 10.0 a 31.8 a 0.8 de Javelin 1.17 I 3.5 bcd 3.7 c-f 5.5 cde 4.5 a-d Cypermethrin 0.067 + Chlorpyrifos I. I2 0.3 gh 1.7 e-h 2.8 ef 1.8 b-e Cypermethrin 0.067 + Javelin 1.17 I 1.3 e-h 6.7 bcd 14.3 abc 1.8 cde Chlorpyrifos 1.12 + Dipel 2X 0.56 Chlorpyrifos I. 12 + I.8 d-g I.7 e-h 3.5 def 5.5 abc Javelin 1. 17 I 0.8 fgh 1.0 fgh 2.0 ef 4.8 a-d Methomyl 1.01 6.5 ab 2.3 d-g 17.5 abc 3.8 a-e Permethrin 0.22 6.8 ab 9.3 abc 24.8 ab 0.5 e Thiodicarb I. I2 2.8 def 13.0 ab 21.8 ab 0.5 e Thiodicarb 0.84 6.0 abc 7.7 abc 33.0 a 2.0 b-e Thiodicarb 0.84 + Mevinphos I. I2 1.0 fgh 1.3 fgh 8.5 b-e 1.5 cde Mevinphos I. I2 3.0 cde 1.0 fgh 6.3 def 6.3 a Endosulfan I. I2 0.8 fgh 1.0 fgh 3.5 def 2.8 a-e ªMeans in the same column followed by the same letter are not significantly different (P = 0.05; Duncan's 1955 Multiple Range Test). Table 5. Effects of selected insecticidal treatments for the control of DBM and T. ni on damage ratings (DR) and two levels of marketability in cabbage. Insecticide and rate (kg AI or I formulation / ha) Untreated check Chlorpyrifos 1. I2 Cypermethrin 0.067 Javelin I. I7 I Cypermethrin 0.067 + Chlorpyrifos I. I2 Cypermethrin 0.067 + Javelin I. 17 I Chlorpyrifos I. I2 + Dipel 2X 0.56 Chlorpyrifos 1.12 + Javelin I. 17 I Methomyl 1.01 Permethrin 0.22 Thiodicarb I. I2 Thiodicarb 0.84 Thiodicarb 0.84 + Mevinphos I. I2 Mevinphos 1. I2 Endosulfan I. I2 Damage ratinga (DR I 3) % marketablea 6.0 a 0.0 f 0.0 d 5.2 cd 2.5 ef 10.0 cd 6.0 a 0.0 f 0.0 d 5.1 d 0.0 f 20.0 bc 3.5 ef 40.0 d 95.0 a 4.9 d 0.0 f 40.0 b 4.0 e 17.5 e 82.5 a 3.9 e 12.5 ef 95.0 a 5.9 ab 0.0 f 0.0 d 6.0 a 0.0 f 0.0 d 5.7 abc 0.0 f 0.0 d 5.7 abc 0.0 f 0.0 d 3.7 ef 22.5 e 87.5 a 5.4 bcd 0.0 f 5.0 cd 3.4-50.0-100.0 - ªMeans in a column followed by the same letter are not significantly different (P = 0.05; Duncan's [ I955] multiple range test). -' indicates unreplicated. See text.

434 Leibee and Savage to adult reported for DBM by Sarnthoy et al. (1989). This is in agreement with Chen and Sun (1986) who found that resistance to fenvalerate in DBM could be increased to an extremely high level within four generations in the laboratory with intensive selection with fenvalerate. These results suggest that cypermethrin and thiodicarb were more effective than chlorpyrifos at selecting for resistance to fenvalerate. Table 6. Mortality (24 hours) of four strains of DBM exposed to fenvalerate leaf residues. The strains were established from mature cabbage field plots that were untreated (UTC) and treated with chlorpyrifos, cypermethrin, and thiodicarb. Mortality (%) (SE)b Rate of UTC Chlorpyrifos Cypermethrin Thiodicarb fenvalerateª strain strain strain strain 0.0x 10.0 (3.3) 2.7 (I.7) 7.6 (I.6) 6.7 (6.7) 0.5X 17.1 (5.4) 13.3 (3.8) 1.7 (1.7) 0.0 (0.0) I.ox 33.5 (5.6) 20.8 (7. I) 12.1 (2.0) 8.1 (1.8) 2.0x 36.4 (2.0) 21.8 (2.2) 12.1 (2.0) 9.5 (2.4) 4.0X 36.4 (8.4) 32.9 (3.8) 16.9 (4.3) 20.7 (4.0) 8.0X 42.2 (1.1) 34.5 (I.2) 25.7 (3. I) 30.9 (7.9) ª1,OX = 0.48 g AI/l. Not corrected for control mortality Conclusions: A DBM population in central Florida was found to be highly resistant to fenvalerate. Laboratory studies indicated that the population was also resistant to methomyl, but susceptible to chlorpyrifos, acephate, endosulfan, and thiodicarb. The field study indicated that chlorpyrifos, endosulfan, mevinphos, and B. thuringiensis were considerably more effective at suppressing DBM than cypermethrin, permethrin, methomyl, and thiodicarb. Cypermethrin, permethrin, thiodicarb, and endosulfan were the most effective in controlling T. ni. Mixtures of two insecticides were shown to improve efficacy apparently due to one component controlling T. ni and not DBM and the other component controlling DBM and not T. ni. Laboratory examination of populations from the field study indicated that resistance to fenvalerate in DBM increased within one crop season with weekly use of cypermethrin and thiodicarb, but not with chlorpyrifos. References Abbot, W. S. 1925. A method of computing the effectiveness of an insecticide. J. Econ. Entomol., 18, 265-267. Chen, J. S. and Sun, C. N. 1986. Resistance of diamondback moth (Lepidoptera: Plutellidae) to a combination of fenvalerate and piperonyl butoxide. J. Econ. Entomol., 79, 22-30. Cheng, E. Y. 1988. Problems of control of insecticide-resistant Plutella xylostella. Pestic. Sci., 23, 177-188. Duncan, D. B. 1955. Multiple range and multiple F tests. Biometrics, 11, 1-42. East, D. A., Edelson, J. V., and Cartwright, B. 1989. Relative cabbage consumption by the cabbage looper (Lepidoptera: Noctuidae), beet armyworm (Lepidoptera: Noctuidae), and diamondback moth (Lepidoptera: Plutellidae). J. Econ. Entomol., 82, 1367-1369. Finney, D. J. 1971. Probit analysis, 3rd ed. Cambridge University Press, Cambridge, England, 333 p. Greene, G. L., Genung, W. G., Workman, R. B., and Kelsheimer, E. G. 1969. Cabbagelooper control in Florida-a cooperative program. J. Econ. Entomol., 62, 798-800. Hama, Y. 1987. Development of pyrethroid resistance in the diamondback moth, Plutella xylostella Linné (Lepidoptera: Yponomeutidae). Appl. Entomol. Zool., 22, 166-175. - 1988. Stability of fenvalerate resistance in the diamondback moth, Plutella xylostella Linné (Lepidoptera: Yponomeutidae). Jpn. J. Appl. Entomol. Zool., 32, 210-214.

Insecticide Resistance in Florida 435 Leibee, G. L. 1986a. Caterpillar control on cabbage, spring 1985. CFREC-Sanford Research Report SAN 86-10. University of Florida, IFAS. 2 p. - 1986b. Caterpillar control on cabbage, spring 1986. CFREC-Sanford Research Report SAN 87-10. University of Florida, IFAS. 3 p. Liu, M. Y., Tzeng, Y. J., and Sun, C. N. 1981. Diamondback moth resistance to several synthetic pyrethroids. J. Econ. Entomol., 74, 393-396. Magaro, J. J., and Edelson, J. V. 1990. Diamondback moth (Lepidoptera: Plutellidae) in south Texas: a technique for resistance monitoring in the field. J. Econ. Entomol., 83, 1201-1206. Noppun, V., Saito, T., and Miyata, T. 1989. Cuticular penetration of S-fenvalerate in fenvalerateresistant and susceptible strains of the diamondback moth, Plutella xylostella (L.). Pestic. Biochem. Physiol., 33, 83-87. Sarnthoy, O., Keinmeesuke, P., Sinchaisri, N., and Nakasuji, F. 1989. Development and reproductive rate of diamondback moth Plutella xylostella from Thailand. Appl. Entomol. Zool., 24, 202-208. Talekar, N. S., and Griggs, T. D. (ed.) 1986. Diamondback moth management: Proceedings of the first international workshop. Asian Vegetable Research and Development Center, Shanhua, Taiwan, 471 p.