Effectiveness of Heterohabditis bacteriophora strain ...

Viewer
Transcript

Biological Control 48 (2009) 232–236

Contents lists available at ScienceDirect

Biological Control journal homepage: www.elsevier.com/locate/ybcon

Effectiveness of Heterohabditis bacteriophora strain GPS11 applications targeted against different instars of the Japanese beetle Popillia japonica Kevin T. Power, Ruisheng An, Parwinder S. Grewal * Department of Entomology, The Ohio State University, Ohio Agricultural Research and Development Center, 1680 Madison Avenue, Wooster, OH 44691, USA

a r t i c l e

i n f o

Article history: Received 14 March 2008 Accepted 20 October 2008 Available online 30 October 2008 Keywords: Biological control Popillia japonica Heterohabditis bacteriophora Insect parasitic nematodes White grubs Larval instars Predictability

a b s t r a c t Field and laboratory tests were conducted from 2001 through 2007 to assess the effectiveness of entomopathogenic nematode Heterorhabditis bacteriophora strain GPS11 applications targeted against different instars of the Japanese beetle, Popillia japonica. During summer ﬂight, P. japonica adults were trapped and caged on turfgrass plots for oviposition. Larval development was monitored for the occurrence of each instar. Nematodes were applied in the ﬁeld against each developing instar at 2.5 109 infective juveniles/ha. In 2001, ﬁeld data obtained in October resulted in 75%, 53%, and 33% control with the applications targeted against the ﬁrst, second, and third instars, 69, 28, and 9 days after treatment (DAT), respectively. In 2002 ﬁeld trial, data obtained in October indicated 97%, 88%, and 0% control when the applications were targeted against the ﬁrst, second, and third instars at 66, 43, and 14 DAT, respectively. Additional plots established in 2002 to determine efﬁcacy against each instar at 14 DAT showed control of the ﬁrst, second, and third instars to be 55%, 53%, and 0%, respectively. In laboratory tests conducted in 2002, 2004, and 2007, P. japonica collected from the ﬁeld at the occurrence of each instar were exposed to H. bacteriophora at concentrations of 0, 10, 33, 100, 330, or 1000 infective juveniles/grub. Probit analysis of the mortality from three of the four sets of tests conducted showed the ﬁrst instar to be signiﬁcantly more susceptible to H. bacteriophora than the third instar at the LC50 level and all tests showed the ﬁrst instar to be signiﬁcantly more susceptible than the third instar at the LC90 level. In addition to the observed decrease in the third instar susceptibility to H. bacteriophora, soil temperatures in the mid-western United States during late September and October rapidly decline often reaching below 15 °C by the beginning of October when grubs are in the third instar stage of development. Therefore, we conclude that the applications of the nematodes made in August or September will provide higher control than those made in October, due to the more appropriate temperature for nematode activity and the presence of more susceptible larval stages. Early nematode applications may also provide an opportunity for nematodes to recycle and cause secondary infections. Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction The Japanese beetle, Popillia japonica Newman, has become an economic pest of a broad range of crops and horticultural commodities in most of the Eastern United States (Fleming, 1972) and Canada (Shetlar 1995; Vittum et al., 1999) since its introduction into New Jersey in the 1920s (Fleming, 1968). The adults are voracious foliage and fruit feeders (Fleming, 1972) and the immature grub is a major pest of turfgrass (Shetlar et al., 1995; Vittum, 1995). The grubs cause extensive feeding damage to turfgrass roots from late summer into the autumn and the following spring (Vittum et al., 1999). Rainfall or supplemental irrigation may mask the effect of root feeding by grubs (Crutchﬁeld and Potter, 1995), but the turfgrass can suffer severe uprooting from mammals and birds foraging for grubs (Vittum, 1995). Since P. japonica is a quar* Corresponding author. Fax: +1 330 263 3683. E-mail address: [email protected] (P.S. Grewal). 1049-9644/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.biocontrol.2008.10.014

antine pest, the grubs are also a major economic concern to growers engaged in the trade of nursery stock and turfgrass sod (Mannion et al., 2001). Control of P. japonica in turfgrass is targeted at the larval stage, because adult control fails to prevent grub infestations. The mainstay chemical grub control insecticides, organophosphates and carbamates, are being lost at three levels which are Federal government regulation under the Food Quality Protection Act, restrictive local government ordinances, and at the consumer level as the public is increasingly wary of conventional pesticides (Robbins et al., 2001). Therefore, there is a need for biological control agents for urban pests such as P. japonica. Entomopathogenic nematodes are naturally occurring parasites of many insect species. The steinernematids and heterorhabditids (Nematoda: Rhabditida) have emerged as excellent biological control agents (Grewal et al., 2005). Nematode production on small and large scales has led to many companies marketing nematodes as insect control agents for use in gardens and cropping systems. Of the commercially available entomopathogenic nematodes

K.T. Power et al. / Biological Control 48 (2009) 232–236

Heterorhabditis bacteriophora Poinar, a species originally found from infected Phyllophaga grubs in Utah (Poinar and Georgis, 1990), has been the most effective nematode against white grubs of this species (Klein, 1990). However, efﬁcacy tests often produced variable results (Georgis and Gaugler, 1991). Factors affecting nematode efﬁcacy included nematode species and strain, grub species, soil temperature, moisture, texture, and the natural enemy complex (Georgis and Gaugler, 1991; Kaya and Gaugler, 1993; Grant and Villani, 2003). Grewal et al. (2002) compared several species and strains of heterorhabditid and steinernematid nematodes and found that H. bacteriophora GPS11 and H. zealandica X1 strains were the most virulent against P. japonica, Cyclocephala borealis Arrow, and Anomala (= Exomala) orientalis Waterhouse. In a series of ﬁeld trials, Grewal et al. (2004) demonstrated that H. bacteriophora GPS11 and H. zealandica X1 were as effective and predictable for curative control of white grubs as the most widely used chemical insecticide, trichlorfon. The range of P. japonica control achieved with H. bacteriophora was 34–97% and with H. zealandica was 48–98% as compared with trichlorfon treatments (29–92%). Grewal et al. (2004) also noticed that applications made against the second instar often resulted in higher control than those made against third instars. Gaugler et al. (1994) showed that the third instar P. japonica grubs are capable of withstanding nematode attacks and can kill attacking nematodes. Klein (1990) suggested that early instars may be more susceptible than the third instars. However, Koppenhöfer and Fuzy (2004) conducted laboratory and greenhouse studies and found no signiﬁcant differences in P. japonica larval susceptibility to H. bacteriophora between ﬁrst and second instars, and second and third instars. Yet, the results of various ﬁeld studies show a trend toward decreased susceptibility of the third instar (Klein, 1990; Grewal et al., 2004). We hypothesized that there are differences in susceptibility of the developmental stages of P. japonica grubs and this is an important factor affecting the ﬁeld efﬁcacy of entomopathogenic nematode applications.

2. Materials and methods 2.1. Nematodes A culture of H. bacteriophora GPS11 strain has been maintained at the Entomology and Nematology laboratories at the Ohio Agricultural Research and Development Center (OARDC), Wooster, Ohio, since its isolation from an infected Cyclocephala borealis (Arrow) grub in 1998 (Grewal et al., 2002). Nematodes were cultured in late instar Galleria mellonella L. larvae and the emerging infective juveniles were harvested 10–14 days later. Infective juveniles used in all experiments were stored at 10 °C for no more than 3 weeks prior to use.

233

tain actively growing turfgrass. The resultant grub populations were used in the following experiments. 2.3. Field plot tests The ﬁeld studies were located on Kentucky bluegrass lawns at the OARDC. Plots measuring 1 m by 0.9 m, separated by 0.9 m alleyways, were arranged in a complete randomized block design replicated four times. No entomopathogenic nematodes were detected in the selected plots when the soil samples collected from the area were baited with G. mellonella larvae. There were three, 20 cm in diameter, PVC cylinders in each plot in which beetles were caged (see above). Additional cylinders were maintained for monitoring of grub development and to obtain untreated larvae for laboratory bioassays (see below). When monitoring indicated that the resultant grub population had developed to the target instar, nematode applications were made. All ﬁeld applications were made at dusk. Nematodes were applied at 2.5 109 infective juveniles/ha using a modiﬁed wash bottle to the cylinder areas of each plot at an application volume of 50-ml per cylinder. The untreated control received only water. Each cylinder received 500-ml of immediate post-treatment irrigation with a sprinkling can to wash the nematodes off the grass blades and another 500-ml after all treatments were applied. Air and soil temperatures at the 2.5 cm and 5.0 cm depth were recorded at the time of each nematode application and efﬁcacy data acquisition. Efﬁcacy data were obtained by counting the number of live larvae in the soil of the cylinder areas in each plot. Soil from the cylinder area was lifted with a spade and placed in a plastic dishpan. The soil was broken apart, sifted through by hand, and the total numbers of grubs from each plot were recorded. 2.3.1. 2001 Field trial Separate sets of plots were treated when monitoring indicated the occurrence of each instar on August 2 (ﬁrst instar), September 12 (second instar), and October 1 (third instar). Physical conditions of the study site were: Turf 100% Kentucky bluegrass, level, silt loam soil, pH 7.7, cation exchange capacity 14, and no thatch. Environmental conditions during the ﬁrst instar applications at 1900 h on August 2, were: cloudy, air temperature 25 °C, soil temperature 25 °C at 2.5 cm and 24 °C at 5.0 cm depth. Environmental conditions during the second instar applications at 1830 h on September 12, were: cloudy, air temperature 20 °C, soil temperature 17 °C at 2.5 cm and 16 °C at 5.0 cm depth. Environmental conditions during the third instar applications at 1800 h on October 1, were: cloudy, air temperature 11 °C, soil temperature 14 °C at 2.5 cm and 13 °C at 5.0 cm depth. Efﬁcacy data were obtained on October 10, at 69, 28, and 9 days after treatment (DAT), respectively. Environmental conditions during acquisition of data were: sunny, air temperature 16 °C, soil temperature 13 °C at 2.5 cm and 5.0 cm depth.

2.2. Japanese beetle populations At the ﬁrst sign of P. japonica ﬂight, the beetles were collected daily using Trécé Catch Can traps with feeding lures containing PEG (phenol ethyl propionate, eugenol, and geraniol) following the method described by Klein et al. (2000). The sex ratio was determined from sub-samples (2 30 beetles) of each day’s catch. When monitoring indicated that the ﬂight comprised of at least 50% females, the beetles were used for caging. PVC cylinders, 20 cm in diameter by 15 cm tall, inserted half way into established turfgrass plots were used as oviposition cages. Forty beetles measured volumetrically and consisting of P50% females were placed in each cylinder and covered with nylon 20 mesh screen to induce egg laying within the cylinder area. Caged beetles were fed fresh apple slices twice a week. The screens were removed 14 days after caging. The entire area received supplemental irrigation to main-

2.3.2. 2002 Field trial Physical conditions of the study site were: Turf 100% Kentucky bluegrass, level, silt loam soil, pH 7.7, cation exchange capacity 14, and no thatch. Nematode applications were made to plots when monitoring indicated the occurrence of each instar on August 13, September 5, and October 4. Environmental conditions during the ﬁrst instar applications at 1930 h on August 13, were: cloudy, air temperature 25 °C, soil temperature 22 °C at 2.5 cm and 23 °C at 5.0 cm depth. Environmental conditions during the second instar applications at 1830 h on September 5, were: cloudy, air temperature 20 °C, soil temperature 19 °C at 2.5 cm and 5.0 cm depth. Environmental conditions during the third instar applications at 1730 h on October 4, were: cloudy, air temperature 21 °C, soil temperature 20 °C at 2.5 cm and 21 °C at 5.0 cm depth. Half of the plots were evaluated at 14 DAT and the second half

K.T. Power et al. / Biological Control 48 (2009) 232–236

were evaluated on October 18 as the grubs began to overwinter which was 66, 43, and 14 DAT, respectively. Environmental conditions during acquisition of 14 DAT data were: August 27, air temperature 22 °C, soil temperature 23 °C at 2.5 cm and 22 °C at 5.0 cm deep; September 12, air temperature 21 °C, soil temperature 23 °C at 2.5 cm and 5.0 cm depth; October 18, air temperature 8 °C, soil temperature 8 °C at 2.5 cm and 11 °C at 5.0 cm depth. 2.4. Laboratory studies Laboratory studies were conducted in separate years at the occurrence of each instar. P. japonica larvae were harvested from oviposition cylinders installed and infested with adults as described in Section 2.2. Tests run in 2002 and 2004 were conducted using 24 well plates. In response to concerns that the well plates excessively conﬁned the larvae and may have impacted results, a method holding the larvae in 30-ml specimen cups was also used in 2007. Soil used in the laboratory studies was sandy loam soil (74% sand, 22% silt, and 4% clay, 4.4% organic matter, pH 6.7, cation exchange capacity 11.8), constructed by blending a natural silt loam soil, sand and peat using a 1:1:1 ratio by volume. One day prior to initiation of each test, 1.5 g of seed consisting of perennial ryegrass and Dutch white clover in a 4:1 ratio by weight was added to the soil moistened to 25% saturation and held at 25 °C in a polyethylene bag in order to pre-germinate the seed to provide food for the P. japonica larvae. Mortality data were recorded 5 DAT. In the well-plate tests, 20 larvae per nematode treatment and 40 per untreated control were individually placed in 24 well plates containing 1.5 g of the pre-moistened soil at 25% saturation per well. The volume of soil was sufﬁcient for grubs to burrow and those which did not burrow into the soil were replaced. Nematode concentrations of 0 (control), 10, 33, 100, 330, and 1000 infective juveniles were applied to the soil surface using a micro-pipette with sufﬁcient water to moisten the soil up to 50% saturation/well. The plates were then covered and held at 21 °C until data were taken. In the specimen cup tests, 20 larvae per nematode treatment and 40 per untreated control were individually placed in 30-ml cups containing 20 g of pre-moistened soil at 25% saturation. Grubs which did not burrow into the soil were replaced. Nematode concentrations of 0 (control), 10, 33, 100, 330, and 1000 infective juveniles were applied to the soil surface of each vial using a micro-pipette with water sufﬁcient to moisten the soil to up to 50% saturation/cup. Each cup was capped and held at 21 °C until data were taken. 2.5. Statistical analysis The ﬁeld plot data were log 10 (X + 1) transformed and subjected to the analysis of variance. Plot means were separated by LSD test at p = 0.05 using 2000 Statistica Software 6.0, StatSoft Inc. (1999). Probit analysis was run on the dose mortality data acquired from the laboratory tests using Minitab 15.1.0.0. (Minitab Inc., 2006).

3. Results 3.1. 2001 Field trial The H. bacteriophora treatment targeted against the ﬁrst instar grubs caused 75% reduction in population by the end of the season which was signiﬁcantly different from the control (F = 4.9; df = 3, 12; p = 0.019) (Fig. 1). The reduction achieved from treatments made against the second (53%) and third (33%) instar was not signiﬁcant. However, all treatments showed a trend toward decreased efﬁcacy with each subsequent instar.

30

c

25

Mean grubs/plot

234

bc

20

b 15

10

a

5

0 August 2 (first)

September 12 October 1 (second) (third)

Untreated

Date of treatment (targeted instar) Fig. 1. 2001 Field trial results on the effectiveness of Heterorhabditis bacteriophora against Popillia japonica grubs evaluated at the end of the season. Mean grubs per plot following H. bacteriophora treatments applied on dates that target each of the P. japonica development instars.

3.2. 2002 Field trial The H. bacteriophora treatments targeted at the ﬁrst and second instars caused signiﬁcant reduction (F = 10.16; df = 1, 6; p = 0.19) for ﬁrst instar and F = 3.94; df = 1, 6; p = 0.036 for second instar) in grub survival 14 DAT, resulting in 55% and 53% control, respectively (Fig. 2A). The applications made against the third instar showed no control 14 DAT. By the time of the ﬁnal reading on October 15, the applications made against the ﬁrst and second instar grubs caused signiﬁcant reduction (F = 2.75; df = 3, 12; p = 0.08) resulting in 97% (66 DAT) and 88% (43 DAT) control, respectively (Fig. 2B). However, nematode treatments made against the third instars were ineffective. 3.3. Laboratory studies Results of the probit analysis of the grub mortality data from the 2002 well-plate tests indicated a trend toward decreased susceptibility with each successive instar (Table 1). The LC50 ﬁducial limits overlapped for the ﬁrst and second instars, but not for the third instar, showing that the third instar is signiﬁcantly less susceptible than the ﬁrst and second instars. The LC90 was 314 nematodes for the ﬁrst, 800 for the second and 1373 for the third instar, showing a trend towards decreasing susceptibility with each successive instar. Here, the ﬁrst instars were signiﬁcantly more susceptible than the second and third instars based on the nonoverlap of the ﬁducial limits. Results of the 2004 well-plate tests showed the same trend. However, either the population was more susceptible overall or the nematodes were more virulent than those of the 2002 test. The LC50 was 10 nematodes for the ﬁrst, 17 for the second, and 194 for the third instar. At both the LC50 and LC90 levels, there was no difference between the ﬁrst and second instars, but both were signiﬁcantly more susceptible than the third instar. Results of the 2007 tests also suggest that the population was overall more susceptible than those of the 2002 test. The well-plate tests showed the ﬁrst instar to be more susceptible than the second and third instars, except the ﬁducial limits overlapped for all instars at the LC50 level. At LC90 level there was a signiﬁcant difference between the ﬁrst instar and the second and the ﬁrst and the third instars, but not between the second and third instars. Results of the specimen cup tests run concurrently in 2007 with the well-plate test also showed the trend of decreasing susceptibility

K.T. Power et al. / Biological Control 48 (2009) 232–236

the ﬁducial limits overlapped for the second and third instars at both the LC50 and LC90 levels.

A 100 b

4. Discussion

Mean grubs/plot

80

0%

60 88% 97%

a

40

b

20

a

a

a

0 August 13 (first)

September 5 (second)

October14 (third)

Date of treatment (targeted instar)

B

14

b b

12 10

Mean grubs/plot

235

8 6 4 a

2 a

0 August 13 (first)

September 5 (second)

October 4 (third)

Untreated

Date of treatment (targeted instar) Fig. 2. 2002 Field trial results on the effectiveness of Heterorhabditis bacteriophora against Popillia japonica grubs. Mean grubs per plot following H. bacteriophora treatments applied on dates that targeted the three P. japonica instars. (A) Evaluations 14 days after treatment (DAT). (B) Evaluations at the end of the season. Black bar = H. bacteriophora treated, Grey bar = Untreated control.

Table 1 Estimated LC50 and LC90 values for Popillia japonica larval mortality following laboratory exposure to different concentrations of Heterorhabditis bacteriophora, GPS11 strain. Target stage (control mortality)

LC50 (ﬁducial limits)

LC90 (ﬁducial limits)

2002; 24-well-plate First instar (15%) Second instar (10%) Third instar (0%)

106 (65–162) 147 (25–277) 785 (623–1053)

314 (235–488) 800 (575–1359) 1373 (1093–1909)

2004; 24-well-plate First instar (15%) Second instar (5%) Third instar (5%)

10 (5–14) 17 (13–23) 194 (138–309)

27 (21–40) 35 (29–50) 486 (353–838)

2007; 24-well-plate First instar (15%) Second instar (10%) Third instar (10%)

14 (8–23) 49 (3–91) 49 (15–87)

41 (29–70) 225 (183–436) 201 (101–327)

2007; specimen cups First instar (25%) Second instar (10%) Third instar (10%)

9 ( 15–26) 127 (40–229) 221 (152–365)

84 (58–156) 595 (426–1012) 760 (561–1203)

with each subsequent instar. Again, the ﬁrst instar was signiﬁcantly more susceptible than the second and third instars, but

The results of well-plate and the 30-ml cup tests indicate that the ﬁrst instar P. japonica is signiﬁcantly more susceptible to H. bacteriophora than the third instar with one exception whereas the trend was not signiﬁcant between the three instars at the LC50 level. At the LC50 level, the second instar P. japonica was signiﬁcantly more susceptible than the third in two of the three wellplate tests. At the LC90 level, the second instar P. japonica was signiﬁcantly more susceptible than the third in one of the three well-plate tests. However, in the 30-ml cup test a signiﬁcant decrease in susceptibility with each subsequent instar was found. All ﬁeld trials resulted in greater mortality of the ﬁrst instar than that of the third instar, and a consistent trend that the second instar data lay between the ﬁrst and third instars. Since our tests directly compared the susceptibility of the ﬁrst, second and third instars, the data show instar susceptibility differences not evaluated by Koppenhöfer and Fuzy (2004). Their P. japonica experiments were only paired comparisons between ﬁrst and second instars and between second and third instars. Corroborating the relative susceptibility of the second and third instar scarabs in our data, Lee et al. (2002) and Koppenhöfer and Fuzy (2004) found that the second instar A. orientalis was more susceptible to heterorhabditids than the third instar. In addition, Lee et al. (2002) found that all steinernematid isolates tested caused 50–80% mortality of A. orientalis second instars and only 15–30% of the third instars. Conversely, Smits et al. (1994) found that the susceptibility of Phyllopertha horticola L. to heterorhabditids and steinernematids generally increased with larval development, speculating that larger larvae may attract more nematodes and the larger body openings may be more accessible to nematodes. Similarly, Fujiie et al. (1993) reported that the third instar Anomala cuprea Hope were more susceptible to Steinernema kushidai Mamiya than earlier instars. Perhaps, the body openings of early instar P. japonica are accessible to H. bacteriophora since it is smaller than some of the steinernematids such as Steinernema glaseri Steiner. However, the evasive and aggressive behavior of the third instar P. japonica may be more effective at resisting nematode infection than the early instars. The results of the 2001 ﬁeld trial showing signiﬁcantly greater control of the ﬁrst instar over the third instar may have been confounded by the difference in the days after treatment in which the data were acquired. The acquisition of the data from the treatments targeted against the ﬁrst instar occurred at 69 DAT, second instar at 28 DAT, and the third instar at 9 DAT. It was possible in the treatments with extended DAT, that the nematodes may have recycled (reproduced) in the infected grubs and the emerged nematodes may have infected and killed additional grubs (secondary infections). Therefore, the protocol of the 2002 trial was designed to limit the inﬂuence of DAT time factor and possible recycling of nematodes by including a second set of plots for acquiring data 14 days after each instar targeted application. Since it is also possible that soil temperature (see below) could impact the results of the ﬁeld trials (Grewal et al., 1994; Georgis and Gaugler, 1991), dose mortality tests were conducted concurrently in the laboratory. Taking data at the end of the season again in 2002 following the 2001 protocol, allowed comparison of the results of the two studies. Prolonged development of the P. japonica grubs within the cylinders in 2002 and the onset of cold autumn weather prevented observations of the third instar targeted treatment beyond 14 DAT. The ﬁrst and second instar targeted applications produced only 53% and 50% reduction at 14 DAT, but increased to 97% and 88%, respectively, at the end of the season. The 53% control

236

K.T. Power et al. / Biological Control 48 (2009) 232–236

achieved by the second instar targeted treatment in the 2001 trial was at 28 DAT, while the 88% control achieved in the 2002 trial was at 43 DAT. Again, we observed increased grub mortality due to the nematodes as the DAT increased, thus suggesting the possibility of secondary infections. The overall decrease in the survival of P. japonica overtime is commonly observed and may be due to predation or competition. The 2001 and 2002 late season ﬁeld data showed a trend of decreasing susceptibility with each successive instar. Although in 2002, the 14 DAT data showed signiﬁcantly greater control of the ﬁrst and second instars than for the third instar, decreasing soil temperatures may have inﬂuenced the effectiveness of the nematodes against the third instar grubs. The soil temperatures in 2002 did not drop as readily on the subsequent application dates as in 2001. Soil temperatures in 2001 at the 2.5 cm depth occurring at the ﬁrst, second and third instar applications were 25 °C, 17 °C, and 14 °C, respectively, and reaching 11 °C on the day of the ﬁnal observations. Infectivity of H. bacteriophora declines below 15 °C with no infection below 12 °C (Grewal et al., 1994). Soil temperatures at the ﬁrst, second, and third instar applications in 2002 remained higher at 22 °C, 19 °C, and 20 °C, respectively. However, the weather rapidly turned quite cold following the third instar application resulting in a soil temperature of 8 °C on the day of the ﬁnal observations. Nonetheless, both years of laboratory data support the trend in the ﬁeld data showing the ﬁrst and second instars to be signiﬁcantly more susceptible at the LC50 and the third instar to be signiﬁcantly less susceptible than the ﬁrst instar at the LC90 levels. The degree of control achieved when ﬁeld applications were directed at the ﬁrst and second instars indicates the viability of H. bacteriophora strain GPS11 as both a preventive and curative biological control agent. It is nearly impossible to survey for the ﬁrst instar grubs and make a determination whether to treat. The ﬁrst instars are too small to be easily seen and the cavities that grubs eventually form in the soil, which facilitate grub detection by allowing the soil to break apart during a survey, are barely constructed at this stage (K. Power, personal observation). Therefore, applications targeted against the ﬁrst instar are essentially preventive. Quite differently, the second instar is large enough to detect, so that a survey could reveal if grub density is high enough to warrant treatment. Controlling the population density prior to the development of the third instar not only limits root pruning, but may also limit attractiveness of the site to foraging birds and mammals as well. In summary, the results of both ﬁeld trials show that the applications made against the ﬁrst instar are most efﬁcacious, the applications made against the second instar were effective and the third instar were not effective. The laboratory data mitigated the confounding variables in the ﬁeld and corroborated the trend in susceptibility of the instars. We conclude that the ﬁrst and second instars of P. japonica are more susceptible to H. bacteriophora than the third instar. Thus, for the greatest efﬁcacy after considering the difﬁculty in detecting the ﬁrst instars, we recommend H. bacteriophora applications be made against grubs in turfgrass at the second instar of P. japonica rather than the third instar. Applications of nematodes targeted against the second instars will produce higher grub control not only due to greater susceptibility of the younger instars, but also due to the conducive soil temperatures in August and early September. There is also potential for recycling by the nematodes to cause secondary infections. By late September and

October when the less susceptible third instars are present, the soil temperatures are also too cool for infection to occur. References Crutchﬁeld, B.A., Potter, D.A., 1995. Tolerance of cool-season turfgrasses to feeding by Japanese beetle and southern masked chafer (Coleoptera: Scarabaeidae) grubs. Journal of Economic Entomology 88, 1049–1056. Fleming, W.E., 1968. Biological control of the Japanese beetle. US department of agriculture. Technical Bulletin 1383, 1–78. Fleming, W.E., 1972. Biology of the Japanese beetle. US department of agriculture. Technical Bulletin 1449, 1–129. Fujiie, A., Yokoyama, T., Fujikata, M., Sawada, M., Hasegawa, M., 1993. Pathogenicity of an entomopathogenic nematode Steinernema kushidai Mamiya (Nematoda: Steinernematidae) on Anomala cuprea (Coleoptera: Scarabaeidae). Japanese Journal of Applied Entomology and Zoology 37, 53–60. Gaugler, R., Wang, Y., Campbell, J.F., 1994. Aggressive and evasive behaviors in Popillia japonica (Coleoptera: Scarabaeidae) larvae: defense against entomopathogenic nematode attack. Journal of Invertebrate Pathology 64, 193–199. Georgis, R., Gaugler, R., 1991. Predictability in biological control using entomopathogenic nematodes. Journal of Economic Entomology 84, 713– 720. Grant, J.A., Villani, M.G., 2003. Soil moisture effects on entomopathogenic nematodes. Environmental Entomology 32, 80–87. Grewal, P.S., Selvan, S., Gaugler, R., 1998. Thermal adaptation of entomopathogenic nematodes: Niche breadth for infection, establishment, and reproduction. Journal of Thermal Biology 19, 245–253. Grewal, P.S., Grewal, S.K., Malik, V.S., Klein, M.G., 2002. Differences in the susceptibility of introduced and native white grub species to entomopathogenic nematodes from various geographic localities. Biological Control 24, 230–237. Grewal, P.S., Power, K.T., Grewal, S.K., Suggars, A., Haupricht, S., 2004. Enhanced predictability in biological control of white grubs with new strains of entomopathogenic nematodes. Biological Control 30, 73–82. Grewal, P.S., Ehlers, R.U., Shapiro-Ilan, D.I. (Eds.), 2005. Nematodes as Biocontrol Agents. CABI Publishing, CAB International, Wallingford, Oxfordshire, UK. Kaya, H.K., Gaugler, R., 1993. Entomopathogenic nematodes. Annual Review of Entomology 38, 181–206. Klein, M.G., 1990. Efﬁcacy against soil-inhabiting insect pests. In: Gaugler, R., Kaya, H.K. (Eds.), Entomopathogenic Nematodes in Biological Control. CRC Press, Boca Raton, FL, pp. 195–214. Klein, M.G., Grewal, P.S., Jackson, T.A., 2000. Lawn, turf and grassland pests. In: Lacey, L.A., Kaya, H.K. (Eds.), Field Manual of Techniques in Invertebrate Pathology. Kluwer Academic Publishers, Netherlands, pp. 681–706. Koppenhöfer, A.M., Fuzy, E.M., 2004. Effect of white grub developmental stage on susceptibility to entomopathogenic nematodes. Journal of Economic Entomology 97, 1842–1849. Lee, D.W., Choo, H.Y., Kaya, H.K., Lee, S.M., Smitley, D.R., Shin, H.K., Park, C.G., 2002. Laboratory and ﬁeld evaluation of Korean entomopathogenic nematodes isolates against the oriental beetle Exomala orientalis (Coleoptera: Scarabaeidae). Journal of Economic Entomology 95, 918–926. Mannion, C.M., McLane, W., Klein, M.G., Moyseenko, J., Oliver, J.B., Cowan, D., 2001. Management of early instar Japanese beetle (Coleoptera: Scarabaeidae) in ﬁeldgrown nursery crops. Journal of Economic Entomology 94, 1151–1161. Minitab. Release 15.1.0.0. Minitab Inc. 2006. Poinar, G.O., Georgis, R., 1990. Characterization and ﬁeld application of Heterohabditis bacteriophora strain HP88 (Heterohabditidae: Rhabditida). Revue de Nematologie 13, 387–393. Robbins, P., Polderman, A., Birkenholtz, T., 2001. Lawns and toxins: an ecology of the city. The International Journal of Urban Policy and Planning 18, 369–380. Shetlar, D.J., 1995. Turfgrass insect and mite management. In: Watschke, T., Dernoeden, P.H., Shetlar, D.J. (Eds.), Managing Turfgrass Pests. Lewis Publishers, Boca Raton, FL, pp. 171–343. Smits, P.H., Wiegers, G.L., Vlug, H.J., 1994. Selection of insect parasitic nematodes for biological control of the garden chafer, Phyllopertha hirticola. Entomologia Experimentalis et Applicata 70, 77–82. Statsoft. 1999. CSS: Statistica. Statsoft, Tulsa, OK. 1999. Vittum, P.J., 1995. Japanese Beetle. In: Brandenburg, R.L., Villani, M.G. (Eds.), Handbook of Turfgrass Insect Pests. Entomological Society of America, Lanham, MD, pp. 66–69. Vittum, P.J., Villani, M.G., Tashiro, H., 1999. Turfgrass Insects of the United States and Canada. Cornell University Press, Ithaca, NY.