Colorado Potato Beetle (Coleoptera: Chrysomelidae) Feeding, Development, and Survival to Adulthood After Continuous Exposure to Bacillus Thuringiensis Subsp

Home , Economic entomology, Instar

Colorado Potato Beetle (Coleoptera: Chrysomelidae) Feeding, Development, and Survival to Adulthood After Continuous Exposure to Bacillus thuringiensis subsp. tenebrionis-Treated Potato Foliage from the Field Author(s): Brian A. Nault, Scott D. Costa, and George G. Kennedy Source: Journal of Economic Entomology, 93(1):149-156. Published By: Entomological Society of America DOI: http://dx.doi.org/10.1603/0022-0493-93.1.149 URL: http://www.bioone.org/doi/full/10.1603/0022-0493-93.1.149

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. HORTICULTURAL ENTOMOLOGY Colorado Potato Beetle (Coleoptera: Chrysomelidae) Feeding, Development, and Survival to Adulthood After Continuous Exposure to Bacillus thuringiensis subsp. tenebrionis-Treated Potato Foliage from the Field

1 2 BRIAN A. NAULT, SCOTT D. COSTA, AND GEORGE G. KENNEDY

Department of Entomology, Box 7630, North Carolina State University, Raleigh, NC 27695

J. Econ. Entomol. 93(1): 149Ð156 (2000) ABSTRACT Colorado potato beetle, Leptinotarsa decemlineata (Say), feeding, development, and survival to adulthood were examined after continuously exposing large larvae to Bacillus thuringiensis subsp. tenebrionis-treated potato foliage from the Þeld. In laboratory assays, the overall consumption and the length of period to become prepupae were determined for larvae, which began as 3rd and 4th instars, that were offered potato leaf disks with naturally declining levels of B. thuringiensis residue. In small-cage Þeld experiments, survival to adulthood and the period to adult emergence for beetles conÞned to potato plants treated with B. thuringiensis beginning as 3rd and 4th instars also were examined. Third instars remaining on plants after a B. thuringiensis application were unlikely to feed and 4th instars consumed only Ϸ50% as much foliage as those fed untreated foliage. Many late instars subjected to B. thuringiensis-treated foliage failed to survive to adulthood; 58Ð83% of these beetles died during the larval stage. Reduced feeding and poor survival of late instars suggest that counts of large larvae after application do not provide a complete picture of the efÞcacy of the B. thuringiensis treatment. Late instar Colorado potato beetles that were exposed continually to naturally declining levels of B. thuringiensis-treated potato foliage took an average of 1.8Ð4.5 d longer to become prepupae and 4Ð8 d longer to emerge as adults compared with those provided with untreated foliage. Delayed emergence of adults that fed on B. thuringiensis-treated potatoes as late instars indicated that development was prolonged in these insects because of ingestion of a sublethal dose of B. thuringiensis.

KEY WORDS Colorado potato beetle, Bacillus thuringiensis subsp. tenebrionis, feeding, development, survival

BIOPESTICIDES CONTAINING THE Coleopteran-active Ba- small larvae (Ferro and Lyon 1991). Thus, application cillus thuringiensis toxins have been examined inten- of B. thuringiensis early in the infestation episode, sively as an alternative to broad-spectrum insecticides when eggs are hatching and small larvae are the pre- for control of the Colorado potato beetle, Leptinotarsa dominant lifestage present, has been recommended decemlineata (Say), in Irish potato. Unlike conven- (Zehnder and Gelertner 1989, Zehnder et al. 1992, tional insecticides, B. thuringiensis must be ingested by Ghidiu and Zehnder 1993). the insect to be effective and it requires different Þeld Late-instar potato beetles subjected to potato foli- application strategies than conventional insecticides age treated with B. thuringiensis toxins for discrete (Bystrak et al. 1994). Research on how B. thuringiensis periods, from 18 h to 6 d, have been shown to feed less affects stage-speciÞc mortality, feeding behavior, and and die sooner than those fed untreated foliage development in potato beetles under laboratory con- (Zehnder and Gelertner 1989, Hough-Goldstein et al. ditions has provided insight on how best to use these 1991, Hoy and Hall 1993, Ferro et al. 1993, Ferro et al. toxins in the Þeld. For example, B. thuringiensis subsp. 1997). However, the effect of continuous exposure to san diego and B. thuringiensis subsp. tenebrionis toxins naturally declining levels of B. thuringiensis on foliage have been shown to be more effective in controlling from the Þeld has not been documented. For example, small (1st and 2nd instars) than large (3rd and 4th Hough-Goldstein et al. (1991) measured consumption instars) Colorado potato beetles (Zehnder and Ge- by 3rd-instar potato beetles in which the concentra- lertner 1989, Ferro et al. 1993) because large larvae tion of B. thuringiensis subsp. san diego toxin on the must consume more toxin to receive a lethal dose than potato foliage did not vary during the 6-d experiment. Ferro et al. (1993) treated potato plants in the Þeld 1 Eastern Shore Agricultural Research and Extension Center, 33446 with B. thuringiensis toxin, but foliage was excised and Research Drive, Virginia Polytechnic Institute and State University, Painter, VA 23420. maintained in the laboratory in the absence of UV light 2 Entomology Research Laboratory, P.O. Box 53400, University of before presenting it to 3rd- and 4th-instar potato bee- Vermont, Burlington, VT 05405Ð3400. tles for the duration of the 4-d experiment. UV light is

0022-0493/00/0149Ð0156$02.00/0 ᭧ 2000 Entomological Society of America 150 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 93, no. 1 a major factor responsible for reducing efÞcacy of B. M-Trak (B. thuringiensis subsp. tenebrionis, Mycogen, thuringiensis because it inactivates this toxin under San Diego, CA) were examined. Foliage was collected Þeld conditions (Gelertner 1990). Ferro et al. (1997) from potted potato plants, Solanum tuberosum L. (va- recently investigated the chronic effects of B. thurin- riety Kennebec), maintained outdoors. Potato plants giensis subsp. tenebrionis on large potato beetle larvae were started in the greenhouse from tuber pieces that survived a pulse exposure of this product (up to sown in small plastic pots (3.8 liters) Þlled with a 24 h), which was designed to simulate the duration of soil-less potting mixture (Metro 220). Upon emer- exposure before a rainfall event. Ferro et al. (1997) gence, the plants were fertilized with Ϸ9 g of Osmo- discovered that when large larvae were exposed to B. cote (14-14-14, N-P-K). Three-week-old plants were thuringiensis subsp. tenebrionis-treated foliage for up transferred to large plastic pots (11.8 liters) and then to 24 h, their consumption of foliage was reduced, moved outside into the full sun and allowed 2Ð3 wk of their development was delayed and their survival to additional development before initiating the experi- adult was reduced compared with beetles fed un- ments. The exterior of the large pots was painted white treated foliage. In the absence of a rainfall event soon to increase reßectance and lower soil temperatures. after an application of B. thuringiensis, large potato Plants were watered as needed and were provided beetle larvae will be exposed to residues of B. thurin- additional Osmocote fertilizer. giensis that decline over a longer period. These plants were divided into 3 groups and ar- Ferro and Gelertner (1989) reported that the abun- ranged in single rows for treatment with M-Trak using dance of small potato beetle larvae present after an either a 2.34 or 4.68 liters/ha (1 or 2 qt/acre, respec- application of B. thuringiensis subsp. san diego does not tively) rate, or were left untreated for the control. provide an adequate measure of efÞcacy because Treatments were made using a CO2-pressurized back- many 1st instars feed on the egg mass before moving pack sprayer equipped with a single row boom that to foliage and may not ingest a lethal dose of product had 3 nozzles (D3 disk per 25 core hollow cone), 1 before the counts are made. Rather, counts of large above row and 2 dropped 43 cm and directed at the larvae were shown to provide a more reliable measure row. Sprays were delivered using a volume of 234 of efÞcacy because surviving large larvae were likely liters/ha at a pressure of 276 kPa. After application, to have escaped the B. thuringiensis treatment. Bystrak 2Ð4 plants within each group were designated to 1 of et al. (1994) reported that B. thuringiensis efÞcacy is 5 replicates and used in all experiments during a single best evaluated by comparing the number of 3rd- and run. 4th-instar potato beetles remaining on plants after Larvae used in bioassays originated from eggs that application with those on untreated plants when the were collected by putting potato plants into rearing number of large larvae peak on untreated plants. An cages for 1 d. To obtain 1st instars, the eggs were abundance of large larvae on B. thuringiensis-treated excised from leaves and permitted to hatch at 27.5ЊC plants would suggest that an additional application of (Ϯ1ЊC), with a photoperiod of 14:10 (L:D) h, in a petri this product or another insecticide could be required. dish lined with dry Þlter paper (9 cm in diameter). However, this assumes that large larvae remaining on When eggs began to hatch, the Þlter paper was moist- plants after an application of B. thuringiensis cause ened slightly to encourage neonates to leave the egg signiÞcant injury. If these large larvae do not feed and mass and disperse onto the Þlter paper. Twenty of do not survive to adulthood or if their feeding is dras- these larvae were placed into a petri dish (10 by 1.5 tically reduced, counts of this lifestage after applica- cm) lined with Þlter paper (ϭexperimental unit). tion will not provide a good estimate of the level of Third and 4th instars used in the bioassay were reared crop protection provided by the B. thuringiensis ap- on potato plants growing in the greenhouse. Early 3rd plication. Additionally, if B. thuringiensis considerably and 4th instars (Ͻ24 h after molt) were collected and reduces feeding and survival of late instar potato bee- placed individually into a petri dish (ϭexperimental tles, it may be possible to delay the initial application unit) as described above. of B. thuringiensis without a signiÞcant loss in the level Fully expanded leaves were removed daily from of crop protection. Thus, the objective of our research treated plants and returned to the laboratory in plastic was to examine the amount of feeding by late instar bags. Leaf disks were cut using a #12 cork borer (1.66 potato beetles and their development and ultimate cm diameter) and the midvein was avoided. First, 3rd survival after continuous exposure to naturally declin- and 4th instars were provided with 3, 4, or 7 fresh disks, ing levels of B. thuringiensis toxin on treated foliage respectively, on a daily basis for up to 11 d, until larvae from the Þeld. became prepupae or died. As larvae grew, their food supply was increased; older foliage was discarded when fresh foliage was added. Approximately 0.5 ml of Materials and Methods distilled water was added to each of the petri dishes to All experiments were conducted and duplicated in maintain turgidity of the leaf disks. Petri dishes were Raleigh on the North Carolina State UniversityÕs cam- then put in a plastic box lined with moistened paper pus in 1995. The 1st run of the experiment occurred in towels and sealed in a plastic bag. A visual assessment June and July and the 2nd in August and September. of the leaf area consumed was taken daily using a grid Laboratory Feeding Assays. Survival, development, of 1.25-mm squares copied onto transparency Þlm. and consumption by Colorado potato beetle larvae Survival and the number of days until prepupation (1st, 3rd, and 4th instars) fed foliage treated with (i.e., C-shaped and wrinkly) also were recorded. February 2000 NAULT ET AL.: EFFECT OF B. thuringiensis ON L. decemlineata 151

A separate experiment was conducted for each in- treatments. The accumulation of degree-days within a star. Each experiment had 3 treatments (0, 2.34, and replication (cage) was weighted in the analysis by the 4.68 liter/ha rate of B. thuringiensis) that were ar- number of adults that ultimately had emerged. Treat- ranged in a randomized complete block design repli- ment means for overall percentage survival to adult cated 5 times. Additionally, there were 5 subsamples emergence were compared using LSMEANS (P Ͻ (i.e., 5 petri dishes) per replication. The means of data 0.05). Each run of the experiment was analyzed sep- for subsamples within a replication were used in the arately because initial analysis with run as a Þxed analyses. Cumulative foliage consumption and days to component indicated a signiÞcant effect of run or its prepupation were weighted in the analysis by consid- interaction with treatment. ering the number of larvae surviving to any given day or to prepupation, respectively. The percentage sur- Results and Discussion vival and days to become prepupae, and the total leaf area consumed were analyzed using the general linear First-instar Colorado potato beetles exposed con- model of analysis of variance (ANOVA), unless anal- tinually to B. thuringiensis-treated potato foliage fed ysis was inappropriate because of little or no survival very little and most died within 48 h (Fig. 1 A and B). in the B. thuringiensis treatments (SAS Institute 1990). During the Þrst 24 h, consumption by 1st instars on Because the period in which each run was conducted foliage treated with B. thuringiensis at the 2.34 and 4.68 occurred under a distinct set of environmental con- liters/ha rates was 85 and 87% less than for those given ditions to which the treated plants were exposed, run untreated foliage, respectively (F ϭ 263.61; df ϭ 2, 16; was considered Þxed and mean square error for F-tests P ϭ 0.0001) (Fig. 1A). At 48 h, consumption by larvae was used in the analyses. Treatment means on certain on foliage treated with B. thuringiensis at the 2.34 and days were compared using the LSMEANS statement at 4.68 liters/ha rates was 99Ð99.5% less, respectively, P Ͻ 0.05 in SAS. than in the untreated controls (F ϭ 1657.93; df ϭ 2, 16; Small Field-Cage Experiments. The survival of 3rd P ϭ 0.0001) (Fig. 1A). Survival of 1st instars after 24 and 4th instars to adult emergence after exposure to and 48 h on untreated foliage was signiÞcantly greater caged potato foliage treated with B. thuringiensis at 0, than survival of those conÞned on B. thuringiensis- 2.34, or 4.68 liters/ha was examined. A separate ex- treated foliage (F ϭ 106.54; df ϭ 2, 16; P ϭ 0.0001 and periment was conducted for each instar. The plants F ϭ 570.26; df ϭ 2, 16; P ϭ 0.0001, respectively). were grown in the same manner as those used for the Although 30Ð50% of 1st instars that were exposed to B. laboratory assay. However, a bag made of Þberglass thuringiensis-treated foliage were still alive after 24 h, window screen was inserted into each pot when the very few continued living after 48 h (Fig. 1B) and none plants were transferred to the 11.4-liter plastic pots. were alive after 72 h. The bag extended 76 cm above the pot rim, was held Cumulative foliage consumption by larvae exposed erect by 2 bamboo sticks inside the cage, and was to B. thuringiensis-treated potato foliage beginning as sealed with clothespins. The cages were rolled down 3rd instars was signiÞcantly less than consumption by below the level of the plant before spraying the plants those given untreated foliage (92 and 89% less by day with B. thuringiensis as described for the laboratory 5 for the 2.34 and 4.68 liters/ha rates, respectively) assay. (F ϭ 192.65; df ϭ 2, 13; P ϭ 0.0001) (Fig. 2A). Con- Test insects were produced as described for the sumption comparisons between treatments beyond laboratory assay and 20 larvae were released into a day 5 were omitted because larval survival in the B. cage after the B. thuringiensis spray had dried. Larvae thuringiensis treatments was very low. On day 5, sur- were allowed to complete development and pupate vival of larvae that were subjected to the 2.34 and 4.68 without being disturbed. Just before the expected liters/ha rate of B. thuringiensis-treated foliage begin- emergence of adults, all foliage within cages was re- ning as 3rd instars did not differ signiÞcantly (42 and moved. The cages were examined daily for adult emer- 17%, respectively), but their survival was signiÞcantly gence and any adults found were counted and then lower than survival of those fed untreated foliage removed until emergence ended. The experiment was (90%) (F ϭ 43.59; df ϭ 2, 16; P ϭ 0.0001) (Fig. 2B). terminated when no further adult emergence was de- Overall, signiÞcantly fewer larvae fed B. thuringien- tected over a 10-d period. Weather data for the Ra- sis-treated potato foliage beginning as 3rd instars sur- leigh, NC, area was obtained from the National Cli- vived to become prepupae (8 and 6% for the 2.34 and matic Data Center to estimate degree-days (DD) to 4.68 liters/ha rates, respectively); whereas, 82% of adult emergence by the method of Allen (1976). The those fed untreated foliage survived to become pre- temperatures used for the lower and upper develop- pupae (F ϭ 108.23; df ϭ 2, 16; P ϭ 0.0001) (Fig. 4B). ment thresholds were 11.0 and 35.0ЊC, respectively Larvae conÞned to B. thuringiensis-treated foliage be- (Wegorek 1959). Experiments had 3 treatments (0, ginning as 3rd instars and surviving to become pre- 2.34, and 4.68 liters/ha rate of M-Trak) that were pupae tended to take longer to reach prepupation than arranged in a randomized complete block design rep- those fed untreated foliage, although this difference licated 5 times. Percentage survival to adult emer- was not signiÞcant (F ϭ 34.54; df ϭ 2, 1; P ϭ 0.1195) gence and degree days to adult emergence were an- (Fig. 4A). The lack of statistical signiÞcance is likely alyzed in SAS using the general linear model of caused by the very low survival of larvae that became ANOVA, unless analysis was inappropriate because of prepupae from those that were continually exposed to little or no survival of beetles in the B. thuringiensis B. thuringiensis-treated foliage. This decreased the to- 152 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 93, no. 1

Fig. 2. (A) Mean Ϯ SEM cumulative leaf area consumed per L. decemlineata larva and (B) percentage survival of these larvae when fed potato foliage that was treated with either a Ϯ 2.34 or 4.68 liters/ha rate of B. thuringiensis subsp. tenebrionis Fig. 1. (A) Mean SEM leaf area consumed per 1st- (M-Trak) or was untreated. On day 0, all larvae were early instar L. decemlineata and (B) percentage survival of these 3rd instars. Means followed by the same letter within the larvae when fed potato foliage that was treated with either a same day after application are not signiÞcantly different (P Ͼ 2.34 or 4.68 liters/ha rate of B. thuringiensis subsp. tenebrionis 0.05; LSMEANS). Means represent data from 2 experiments. (M-Trak) or was untreated. Consumption and survival were recorded after 24 and 48 h. Means followed by the same letter within the same period after application are not signiÞcantly Ͼ survival than those fed untreated foliage (Fig.3Aand different (P 0.05; LSMEANS). Means represent data from B, 4 A and B). After 3 d, all 4th instars fed untreated 2 experiments. foliage had become prepupae and had consumed sig- niÞcantly more leaf area than those that had been tal number of observations and resulted in a single continually exposed to B. thuringiensis-treated potato degree of freedom in the error term. foliage (91 and 92% greater for the 2.34 and 4.68 li- Fourth instars exposed to B. thuringiensis-treated ters/ha rates, respectively) (F ϭ 351.49; df ϭ 2, 16; P ϭ foliage fed less, developed more slowly, and had lower 0.0001) (Fig. 3A). At this time, survival among larvae February 2000 NAULT ET AL.: EFFECT OF B. thuringiensis ON L. decemlineata 153

B. thuringiensis-treated foliage (day 8) was 40Ð56% less than overall consumption by those fed untreated foliage (day 3) (Fig. 3A). Similarly, overall survival of 4th instars fed B. thuringiensis-treated foliage (day 8) was 46Ð52% less than overall survival of those fed untreated foliage (day 3) (Fig. 4B). Survival to prepupation was signiÞcantly lower for 4th instars fed B. thuringiensis-treated than untreated foliage (62 and 79% lower for the 2.34 and 4.68 liters/ha rates, respectively) (F ϭ 28.03; df ϭ 2, 16; P ϭ 0.0001) (Fig. 4B). Further, 4th instars that were fed B. thuringiensis-treated foliage took signiÞcantly longer to become prepupae (nearly 5 d) than those fed untreated foliage (F ϭ 182.12; df ϭ 2, 10; P ϭ 0.0001) (Fig. 4A). Fewer late instars that were conÞned to B. thuringiensis-treated potato plants survived to the adult stage than those conÞned to untreated plants (run 1, 3rd instar, F ϭ 208.79; df ϭ 2, 8; P ϭ 0.0001; run 1, 4th instar, F ϭ 222.99; df ϭ 2, 8; P ϭ 0.0001; run 2, 3rd instar, F ϭ 59.55; df ϭ 2, 8; P ϭ 0.0001; run 2, 4th instar, F ϭ 56.28; df ϭ 2, 8; P ϭ 0.0001, for Fig. 5 AÐD, respectively). No 3rd instars subjected to B. thuringiensis-treated plants survived to adulthood in the 1st run of the experiment (Fig. 5A). Larvae that survived to adulthood on B. thuringiensis-treated plants began emerging later than those that survived to adulthood on untreated plants. In the 2nd run of the experiment, initiation of adult emergence was delayed by 100 DD (Х 4 d) among beetles conÞned to B. thuringiensis-treated potato plants beginning in the 3rd instar (Fig. 5C). Initial adult emergence in cages that contained B. thuringiensis-treated potatoes, which were infested initially with early 4th instars, was delayed by 90 (Fig. 5B) and 100Ð150 cumulative degree-days (ϭ6Ð8 d) (Fig. 5D) in the 1st and 2nd run of the experiment, respectively. Late instar Colorado potato beetles fed less when exposed continually to naturally declining levels of B. thuringiensis-treated potato foliage than when they were provided with untreated foliage from the Þeld. Other researchers have reported a reduction in feeding by late instar potato beetles when subjected to B. thuringiensis-treated potato foliage, but in their studies larvae were fed a constant dose of B. thuringiensis for Fig. 3. (A) Mean Ϯ SEM cumulative leaf area consumed various periods (from 0.5 to 6 d). For example, Ze- per 4th-instar L. decemlineata and (B) percentage survival of hnder and Gelertner (1989) have shown that 3rd- 4th instars when fed potato foliage that was treated with instar potato beetles subjected to B. thuringiensis either a 2.34 or 4.68 liters/ha rate of B. thuringiensis subsp. subsp. san diego-treated potato foliage for 24 h fed tenebrionis (M-Trak) or was untreated. On day 0, all larvae 38Ð74% less than those fed untreated foliage after 3 d. were early 4th instars. Means followed by the same letter Ferro et al. (1997) reported that total consumption by within the same day after application are not signiÞcantly different and NS denotes no signiÞcant difference (P Ͼ 0.05; 4th instars that were exposed to B. thuringiensis subsp. LSMEANS). Means represent data from 2 experiments. tenebrionis-treated potato foliage for 24 h was 54% less than consumption by those fed untreated foliage; however, total consumption by 3rd instars exposed to fed untreated and B. thuringiensis-treated foliage did treated and untreated foliage in the same manner did not differ signiÞcantly (F ϭ 0.99; df ϭ 2, 16; P ϭ 0.3915) not differ signiÞcantly. Hough-Goldstein et al. (1991) (Fig. 3B). After8dofexposure to B. thuringiensis- observed that 3rd-instar potato beetles fed Ϸ75Ð95% treated foliage, survival and consumption were not less over 6 d when given constant exposure to foliage affected by rate of B. thuringiensis (F ϭ 0.31; df ϭ 1, treated at the labeled-rate for B. thuringiensis subsp. 8; P ϭ 0.5958 and F ϭ 0.00; df ϭ 1, 4; P ϭ 0.9583, san diego than when fed untreated foliage. Our results respectively). Overall consumption by 4th instars fed using naturally declining levels of B. thuringiensis in- 154 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 93, no. 1

dicate that large larvae remaining on plants after application are unlikely to feed (e.g., 3rd instars) or will feed but consume only Ϸ50% as much foliage as large larvae (e.g., 4th instars) on untreated plants. Because large larvae remaining on B. thuringiensis-treated potatoes may never feed again or will consume much less foliage than normal, counts of large larvae several days after application do not provide a complete picture of the effectiveness of the B. thuringiensis treatment. Monitoring changes in defoliation after application would be a more meaningful assessment of B. thuringiensis efÞcacy. Many late instar potato beetles exposed to B. thu- ringiensisÐtreated foliage failed to survive to adulthood. Ferro et al. (1997) reported that survival of Colorado potato beetles to adulthood ranged from 0 to 37% after they were brießy exposed (from 0.5 to 24 h) to B. thuringiensis subsp. tenebrionis-treated potato leaves as late instars. In their study, most of the mortality occurred during prepupation or during the pupal stage (46Ð100%), whereas in our experiments most of the mortality occurred during the larval stage (58Ð 83%). Because B. thuringiensis at the labeled rate was very effective against 3rd instars and somewhat effective against 4th instars, the timing of the initial B. thuringiensis spray could be delayed until early 3rd instars Þrst appear, rather than during the period up to peak egg mass hatch. This initial application would occur later in the infestation episode than the initial application that is currently recommended (Ghidiu and Zehnder 1993). However, the current timing rec- ommendation was based on results from research where older, less effective formulations of B. thuringiensis (e.g., M-One) were used. Unquestionably, formulations of B. thuringiensis have been improved through the years (Ferro et al. 1993). Delaying the initial application of B. thuringiensis could result in a reduction in the number of applications needed to effectively manage Colorado potato beetle infesta- tions, and consequently in reduced costs and lower selection pressure on the population for resistance to B. thuringiensis. One constraint of this approach is that a period of rainy weather could make it impossible to apply B. thuringiensis in time to prevent severe defoliation. Delayed emergence of adults that fed on B. thuringiensis-treated foliage as late instars indicates that development was prolonged in these insects because of ingestion of a sublethal dose of B. thuringiensis. In North Carolina, delayed emergence of 1st-generation Fig. 4. (A) Mean Ϯ SEM number of days required for adults in early-planted potato Þelds could result in a either 3rd or 4th instar L. decemlineata to become prepupae majority of the population emerging after the crop is and (B) the mean Ϯ SEM percentage survival to become harvested. Past studies have shown that 1st-generation prepupae for these larvae after they were conÞned on potato adults denied access to potato foliage are not likely to foliage that was treated with either a 2.34 or 4.68 liters/ha rate survive through the winter (Nault et al. 1996). Thus, of B. thuringiensis subsp. tenebrionis (M-Trak) or was un- using a B. thuringiensis foliar spray in conjunction with treated. Experiments were initiated when larvae were either early 3rd or 4th instars. Means followed by the same letter early planting and early harvesting could signiÞcantly within the same instar are not signiÞcantly different (P Ͼ reduce the size of the overwintering potato beetle 0.05; LSMEANS). Means represent data from 2 experiments. population. February 2000 NAULT ET AL.: EFFECT OF B. thuringiensis ON L. decemlineata 155

Fig. 5. Mean Ϯ SEM percentage of cumulative L. decemlineata adult emergence through time after larvae were conÞned on potato foliage that was treated with either a 2.34 or 4.68 liters/ha rate of B. thuringiensis subsp. tenebrionis (M-Trak) or was untreated. (A) Data from run 1 initiated with early 3rd instars and (B) early 4th instars. (C) Data from run 2 initiated with early 3rd instars and (D) early 4th instars. Cumulative degree-days were determined using methods described by Allen (1976). Means followed by the same letter are not signiÞcantly different (P Ͼ 0.05; LSMEANS). 156 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 93, no. 1

Acknowledgments Gelertner, W. D. 1990. Bacillus thuringiensis, bioengineer- ing and the future of bioinsecticides. Brighton crop pro- Assistance with data collection by B. Bumgarner, M. tection conference: pests and diseases. Br. Crop Protec- Campbell, C. Cheshire, V. Covington, and M. Nault was tion Council, Surrey. 2: 617Ð624. appreciated greatly. This research was supported by the Ghidiu, G. M., and G. W. Zehnder. 1993. Timing of the USDA-ARS and the North Carolina Agricultural Research initial spray application of Bacillus thuringiensis for con- Service under SpeciÞc Cooperative Agreement Nos. 58-1275- trol of the Colorado potato beetle (Coleoptera: Chry- 4-033 and 58-5352-5-665, entitled “Effect of Harvest Date and somelidae) in potatoes. Biol. Control 3: 348Ð352. B.t.t. Treatments on Colorado Potato Beetle Overwintering Hough-Goldstein, J., A. M. Tisler, G. W. Zehnder, and K. A. Survival and Susceptibility to Fungal Pathogens” and “Effect Uyeda. 1991. Colorado potato beetle (Coleoptera: of B.t.t. Treatments and Harvest Date on Defoliation of Po- Chrysomelidae) consumption of foliage treated with Ba- tato and Overwintering Survival of Colorado Potato Beetle,” cillus thuringiensis var. san diego and various feeding stim- respectively. ulants. J. Econ. Entomol. 84: 87Ð93. Hoy, C. W., and F. R. Hall. 1993. Feeding behavior of Plute- References Cited lla xylostella and Leptinotarsa decemlineata on leaves treated with Bacillus thuringiensis and esfenvalerate. Pes- Allen, J. C. 1976. A modiÞed sine wave method for calcu- tic. Sci. 38: 335Ð340. lating degree days. Environ. Entomol. 5: 388Ð396. Nault, B. A., N. M. French II, and G. G. Kennedy. 1996. Bystrak, P. G., S. Sanborn, and G. Zehnder. 1994. Methods Inßuence of European corn borer (Lepidoptera: Pyrali- of optimizing Þeld performance of Bacillus thuringiensis dae) damage to potato and foliage availability on over- endotoxins against Colorado potato beetle, pp. 386Ð402. winter survival of Þrst-generation Colorado potato beetle In G. W. Zehnder, M. L. Powelson, and K. V. Raman adults (Coleoptera: Chrysomelidae) in North Carolina. J. [eds.], Advances in potato pest biology and management. Econ. Entomol. 89:124Ð130. APS, St. Paul, MN. SAS Institute. 1990. SAS/STAT userÕs guide, version 6.12. Ferro, D. N., and W. D. Gelertner. 1989. Toxicity of a new SAS Institute, Cary, NC. strain of Bacillus thuringiensis to Colorado potato beetle Wegorek, W. G. 1959. Stonka ziemniaczana (Leptinotarsa (Coleoptera: Chrysomelidae). J. Econ. Entomol. 82: 750Ð decemlineata Say). Pr. Nauk. Inst. Ochr. Rosl. 1:1Ð178. 755. Zehnder, G. W., and W. D. Gelertner. 1989. Activity of the Ferro, D. N., and S. M. Lyon. 1991. Colorado potato beetle M-ONE formulation of a new strain of Bacillus thurin- (Coleoptera: Chrysomelidae) larval mortality: operative giensis against the Colorado potato beetle (Coleoptera: effects of Bacillus thuringiensis subsp. san diego. J. Econ. Chrysomelidae): relationship between susceptibility and Entomol. 84: 806Ð809. insect life stage. J. Econ. Entomol. 82: 756Ð761. Ferro, D. N., Q.-C. Yuan, A. Slocombe, and A. F. Tuttle. 1993. Zehnder, G. W., G. M. Ghidiu, and J. Speese III. 1992. Use Residual activity of insecticides under Þeld conditions for of the occurrence of peak Colorado potato beetle (Co- controlling the Colorado potato beetle (Coleoptera: leoptera: Chrysomelidae) egg hatch for timing of Bacillus Chrysomelidae). J. Econ. Entomol. 86: 511Ð516. thuringiensis spray applications in potatoes. J. Econ. En- Ferro, D. N., A. C. Slocombe, and C. T. Mercier. 1997. Col- tomol. 85: 281Ð288. orado potato beetle (Coleoptera: Chrysomelidae): residual mortality and artiÞcial weathering of formulated Ba- cillus thuringiensis subsp. tenebrionis. J. Econ. Entomol. Received for publication 3 February 1999; accepted 8 Sep- 90: 574Ð582. tember 1999.