OUP user menu

Effect of Bt Corn for Corn Rootworm Control on Nontarget Soil Microarthropods and Nematodes

(CC)
Mohammad A. Al-Deeb, Gerald E. Wilde, John M. Blair, Tim C. Todd
DOI: http://dx.doi.org/10.1603/0046-225X-32.4.859 859-865 First published online: 1 August 2003

Abstract

The western corn rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae), is a major pest of corn in Kansas. Planting Bt corn hybrids resistant to this pest is being tested as a method to control the larval stage of corn rootworms. These hybrids express Cry3Bb1 toxin and are expected to only directly impact chrysomelids and possibly related taxa. Soil samples were examined to evaluate the effect of Bt corn for corn rootworm control on soil microarthropods and nematodes in Kansas in 2000 and 2001. Soil samples from soil close to Bt corn and to its isoline were taken on three occasions (early, mid, and late season) from eight locations in 2000 and three locations in 2001. Soil mites and Collembola were extracted using a modified Tullgren high-gradient extractor. Nematodes were extracted using a centrifugal-flotation procedure. In general, numbers of soil mites (Prostigmata, Mesostigmata, and Oribatei), Collembola, and nematodes were similar in soil planted with Bt corn and soil planted with its isoline.

  • Bt corn
  • nontarget organisms
  • Cry3Bb1

Introduction

THE LARVAE OF SEVERAL Diabrotica species (Coleoptera: Chrysomelidae) that feed on the roots of corn (Zea mays L.) are commonly called corn rootworms. The western corn rootworm, Diabrotica virgifera virgifera LeConte, is one of the major pests of corn in Kansas (Higgins et al. 2000). It was first recorded in Kansas in 1868, and until 1955 had a very limited distribution in the United States (Chiang 1973). Since then, it has spread northward and eastward to inhabit most of the Corn Belt states (Pedigo 2002). Damage from larval feeding reduces yield, and plants weakened by the feeding of corn rootworms are easily blown down (lodged) by strong winds and are difficult to harvest (Spike and Tollefson 1991). Adults eat the foliage and silks of corn, and additional damage is done to corn and cucumbers when they carry and transmit organisms causing bacterial wilt to these plants (Davidson and Lyon 1987). Crop losses and control costs attributed to these pests reach $1 billion annually in the United States (Metcalf 1986).

It is important to control corn rootworms without damaging existing, beneficial soil fauna. Soil microarthropods play an important role in soil, as they affect many aspects of the distribution, abundance, and activity of soil fungi and bacteria. They selectively graze on fungi, disperse fungal inoculum, and stimulate microbial growth (Lussenhop 1992, Anderson 1988). They also increase nitrogen mineralization, nutrient leaching, and shoot N contents in some plants (Bardgett and Chan 1998), and play an integral role in the decomposition of litter and nutrient cycling. As such, they are important mediators of food-web stability (Moore et al. 1988).

The majority of microarthropods occur in the top 5–10 cm of soils. Mites and Collembola generally constitute 90–95% of microarthropod samples, although the relative abundance of different groups varies in different biomes (Coleman et al. 1999). Mites, spring-tails, and earthworms are the most important soil fauna transforming the above-ground litter entering the soil (Brussaard 1998). Collembolans are ubiquitous members of the soil fauna (Coleman and Crossley 1996) that feed on decaying plant material, fungi, and bacteria (Borror et al. 1989). Soil mites (Acari) are the most abundant microarthropods in many types of soils (Coleman and Crossley 1996) and have a high degree of habitat density (Krantz 1970). In soil, they are comprised of saprophagous, microphytophagous, and predatory species that form the numerically dominant component of the arthropod mesofauna (Evans 1992). Four suborders of mites occur frequently in soils: the Oribatei, Prostigmata, Mesostigmata, and Astigmata (Coleman and Crossley 1996).

Soil nematodes, like soil microarthropods, are useful indicators of environmental conditions and general ecosystem health (Niles and Freckman 1998). As consumers, bacterivorous and fungivorous nematodes influence microbial population dynamics and thus decomposition processes, and contribute directly to nutrient mineralization (Ingham et al. 1985). In contrast, phytoparasitic nematodes are recognized as important pests in many agricultural systems, and several cause significant yield loss in corn (Todd 1989, Todd and Oakley 1996).

Bacillus thuringiensis (Berliner) is a naturally occurring rod-shaped soil-borne bacterium found worldwide (Deacon 2001, Ostlie et al. 1997). This bacterium produces aproteinaceous parasporal crystalline inclusion during sporulation. Upon ingestion by insects, this crystalline inclusion is solubilized in the midgut, releasing proteins called delta-endotoxins. These proteins (protoxins) are activated by midgut proteases, and the activated toxins interact with the larval midgut epithelium to cause a disruption in membrane integrity that ultimately leads to insect death (Knowles 1994, Gill et al. 1992).

Genes coding for the production of some Bt toxins that have insecticidal activities have been transferred to plants, which are known as Bt crops (Ostlie et al. 1997). Current commercial Bt corn hybrids for European corn borer are good examples of Bt crops (Steffey and Gray 1999). During the 6-yr period from 1996 to 2001, the global area of transgenic crops increased >30-fold, from 1.7 million ha in 1996 to 52.6 million ha in 2001. In the United States, 35.7 million ha (68% of the global total) of transgenic crops are grown, and 9.8 million ha of that area is transgenic corn (ISAAA 2002). Planting Bt corn hybrids resistant to corn rootworm is being tested as a method to control the larval stage of this pest. These hybrids express Cry3Bb1 toxin and are expected to only directly impact chrysomelids and possibly related taxa (EPA 2003).

One of the environmental concerns about genetic modification technology in plants is the possible effect on nontarget organisms (National Academy of Sciences 2000). Saxena et al. (1999) reported that a Cry1Ab Bt toxin was released from corn plants into the rhizosphere soil in root exudates from Bt corn for European corn borer. They also mentioned that the Bt toxin released in soil from roots during growth of a Bt corn crop would add to the amount of toxin introduced into soil from pollen during taselling as a result of the incorporation of plant residues after harvesting the crop. Tapp and Stotzky (1998) observed that the bound state of the Bt toxin persisted for up to 234 d. Active Bt toxins could persist and remain insecticidal in soil as a result of binding to humic acids (Crecchio and Stotzky 1998) and clays (Saxena et al. 2002). This persistence could pose a hazard to nontarget organisms and enhance the selection of toxin-resistant target species. However, Saxena and Stotzky (2001) found that the Bt toxin (Cry1Ab) released from root exudates and biomass of Bt corn had no apparent effect on earthworms, nematodes, protozoa, bacteria, and fungi in soil. Sayaboc et al. (1975) found that B. thuringiensis sprays applied on rice stubble just before soil incorporation had no effect on predatory and saprophagous mite populations.

Bt corn for rootworm control, to date, is not commercially available, but several companies are testing different Bt events in the field. Data from preliminary experiments have demonstrated that some of the new Bt events can be very effective against corn rootworm larvae, and the Bt plants have well-developed root systems compared with non-Bt insecticide-treated plants. Obryckiet al. (2001) mentioned that Bt corn is a good control measure against several important corn pests, but its potential risks and effects on the environment need to be evaluated. The purpose of this study was to evaluate the effect of Bt corn expressing the Cry3Bb1 protein for corn rootworm control on several soil-inhabiting nontarget organisms, including microarthropods and nematodes.

Materials and Methods

Field Sites and Experimental Design, 2000

Bt and non-Bt corn hybrids were planted at eight locations in Kansas. Those locations included replicated experiments at four locations (Manhattan, Scandia, Byron, and Sublette) and unreplicated experiments at four locations paired in treatments (Sedgwick and Pratt, Oakley and Hoxie), in which each location served as a replication. The presence of Cry3Bb1 toxin in Bt corn plants was tested with a simple protein expression test (enzyme-linked immunosorbent assay (ELISA) strip) specific for the Cry3Bb1 toxin (provided by Monsanto Company, St. Louis, MO). The Manhattan experiment had four treatments in a randomized complete block design with three replications. Each treatment was planted in six rows 9.14 m long in a 76.2-cm row spacing. Treatments were: Bt corn (MON 853) CRW +seed treatment (confidential, Monsanto Company), isoline, Btcorn (MON 863) CRW, and Bt corn (MON 863) CRW +seed treatment. The Scandia experiment had five treatments in a randomized complete block design with four replications. Each treatment was planted in four rows 9.14 m long in a 76.2-cm row spacing. Treatments were: Bt corn (MON 862) CRW, Bt corn (MON 853) CRW +seed treatment, isoline 1 +Force 3G (tefluthrin) at 46.7 g/100 m of row, isoline 1, and Bt corn (MON 863) CRW. The Sublette experiment had four treatments in a randomized complete block design with three replications. Each treatment was planted in four rows 7.62 m long in a 76.2-cm row spacing. Treatments were: Bt corn (MON 863), Bt corn (MON 863) +Gaucho (imidacloprid) seed treatment at 60 g/cwt, isoline +Furadan (carbofuran) at 2.38 liters/ha, and isoline. The Byron experiment had five treatments in a randomized complete block design with two replications. Each treatment was planted in four rows 30.48 m long in a 76.2-cm row spacing. Treatments were: Bt corn (MON 863) R ×670, R ×670, Bt corn (MON 863) R ×740, R ×740 +Furadan (carbofuran) at 2.38 liters/ha, and R ×740. The four unreplicated experiments were paired in treatments (Sedgwick and Pratt, Oakley and Hoxie). Every experiment had Bt corn hybrids planted in plots adjacent to its non-Bt corn isoline. Each plot was four rows 30.48 m long in a 76.2-cm row spacing. Each of the Sedgwick and Pratt experiments had three treatments: Bt corn (MON 863), isoline, and isoline +Furadan (carbofuran) at 2.38 liters/ha. Each of the Oakley and Hoxie experiments had three treatments: R ×670, R ×670 +(MON 863), and R ×670 +Furadan (carbofuran) at 2.38 liters/ha

Field Sites and Experimental Design, 2001

Corn hybrids were planted at three locations in Kansas. One location was at Manhattan, and two were at Scandia The Manhattan experiment had four treatments in a randomized complete block design with four replications. Each plot was 20 rows 15.24 m long in a 76.2-cm row spacing. Treatments were: Bt corn (MON 863) +seed treatment, isoline +seed treatment, isoline +Force 3G (tefluthrin) at 46.7 g/100 m of row, and isoline. There were two field sites at the Scandia location (referred to as Scandia 1 and Scandia 2). Each site had seven treatments in a randomized complete block design with four replications. Each plot was four rows 9.14 m long in a 76.2-cm row spacing. Treatments were: Bt corn (MON 863) CRW +seed treatment, isoline +seed treatment, isoline +Force 3G (tefluthrin) at 46.7 g/100 m of row, isoline +Prescribe (imidacloprid) seed treatment at 1.34 mg ai/kernel, isoline +Clothianidin seed treatment at 1.25 mg ai/kernel, isoline +Lorsban 15G (chlorpyrifos) at 74.7 g/100 m of row, and isoline +Aztec 2.1G (tebupirimofos and cyfluthrin) at 62.5 g/100 m of row.

Tullgren-Type Extraction, 2000–01

The collembolans, mites, and a variety of small insects collectively known as microarthropods can be sampled in soil cores or litterbags and extracted by a heat gradient apparatus such as Tullgren or Berlese funnels (Coleman and Crossley 1996). In the current study, the high-gradient Tullgren-type extractor and the sampling tool described by Crossley and Blair (1991) were used. In 2000, samples were taken from eight locations (Manhattan, Scandia, Byron, Sublette, Sedgwick, Pratt, Oakley, and Hoxie). The different sites represented a variety of soil types, rainfall, temperature, and other environmental conditions in Kansas. Plots at Scandia, Sublette, Pratt, Sedgwick, Oakley, Hoxie, and Byron were irrigated; the Manhattan site was nonirrigated. In 2001, samples were taken from three locations, one in Manhattan and two in Scandia (Scandia 1 and 2). Five treatments were sampled at each of the two Scandia locations. Plots at Scandia 1 and 2 were irrigated, while Manhattan was nonirrigated. Soil samples were taken on three occasions in the corn-growing season: early season (corn was in the two-leaf stage), mid-season (corn was in the silking stage), and late season (corn was at physiological maturity), from each location in 2000 and 2001. A total of 8–10 samples per treatment was taken from each location at each sampling date. The sampling tool included a split corer that held a sleeve (5-cm-long sleeves cut from 5-cm-i.d. aluminum pipe) to contain one soil sample. Soil samples (5 cm diameter ×5 cm deep) were taken from the root system area (≈2 cm away from the stalk base) with a simple half twist of the tool. Samples were then wrapped in aluminum foil and placed in a portable cooler for transport to the laboratory. Samples were processed on the day they were collected. Four extractors were used, and each extractor held 20 samples at a time. Mites and Collembola were sorted and identified under a dissecting stereoscope (Nikon SMZ1500). Mites were identified to the suborder level (Prostigmata, Mesostigmata, and Oribatei) (Evans 1992, Krantz 1970, McDaniel 1979). Most keys used for identifying soil mites deal with mature mites, and the identification of some immature stages is very difficult. Therefore, a category called Others was used for all of the unidentifiable mites. The category Totals combined the counts of all mites and collembolans and served as a general indicator of the abundance of soil microarthropods in samples. Voucher specimens of soil mites (Prostigmata, Mesostigmata, and Oribatei) were submitted to the Insect and Prairie Arthropod Research Museum at Kansas State University.

Nematode Extraction, 2000-2001

Nematodes were extracted from soil and corn roots that were taken from one of the Scandia sites in 2000 and 2001. Root nematodes were extracted using the method described by Todd and Oakley (1996). Nematode soil densities were determined from soil samples (5 cm deep) taken from the root system area (≈2 cm away from stalk base) with a shovel. A total of 8 –10 samples per treatment was taken from each location. Nematodes were extracted from 100-cm3 subsamples of soil using centrifugal-flotation procedure (Jenkins 1964). In 2000, soil and root samples were taken from two treatments: Bt corn (MON 863) and its isoline on one occasion when corn was at physiological maturity. Numbers of phytoparasitic Helicotylenchus (Spiral), Pratylenchus (Lesion), Tylenchorhyncus (Stunt), Hoplolaimus (Lance) in soil, and Pratylenchus in root were recorded. In 2001, soil and corn root samples were taken on two occasions (silking stage and postharvest). Numbers of phytoparasitic (Tylenchorhyncus, Pratylenchus), fungivorous (Tylenchidae, Aphelenchina), bacteriovorous (Rhabditidae, Cephalobidae), and omnivorous (Dorylaimida) taxa were recorded.

Statistical Analyses

An analysis of variance (ANOVA) was carried out. For the Tullgren-type extraction data, analyses were performed using the PROC MIXED procedure of SAS statistical software (Littell et al. 1996, SAS Institute 2001). In each season, the three sampling times of each location were analyzed as repeated measures over time. Means were compared with the LSMEANS procedure of SAS. For the nematode extraction data, analyses were performed using the PROC GLM, and means were compared with the least significant difference procedure of SAS.

Results

Tullgren-Type Extraction, 2000

Soil microarthropods (Prostigmata, Mesostigmata, Oribatei, and Collembola) were counted from soil samples collected at Manhattan, Scandia, Sublette, Byron, Sedgwick, Pratt, Oakley, and Hoxie. At all locations, Prostigmata and/or Oribatei constituted the majority of the Acarina identified. With one exception, there was no significant difference among treatments in numbers of soil microarthropods. Collembola were present in moderate to low numbers at most locations, and numbers of this group were more variable. At Manhattan, no significant difference occurred among treatments in number of Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =4.20; df =3,6; P =0.0638; F =2.65; df =3,6; P =0.1432; F =0.71; df =3, 6; P =0.5781; F =0.29; df =3,22; P = 0.8311; F =4.03; df =3, 6; P =0.0691, respectively) (Table 1). However, a significant difference occurred among treatments in the number of Prostigmata (F =6.11; df =3, 6; P =0.0296), in which Bt corn (MON 853) with seed treatment had significantly more prostigmatid mites than the isoline and the other treatments. At Scandia, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =1.11; df =4, 42; P = 0.3654; F =0.60; df =4, 42; P =0.6643; F =0.61; df =4, 12; P =0.6623; F =0.07; df =4, 42; P =0.9898; F =2.25; df =4, 42; P =0.0796; F =1.02; df =4, 42; P = 0.4094, respectively) (Table 1). At Sublette, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =1.00; df =3, 21; P =0.4137; F =1.87; df =3, 6; P =0.2363; F =1.44; df =3, 6; P =0.3220; F =0.33; df =3, 21; P =0.8015; F =1.87; df =3, 6; P =0.2351; F=2.05;df =3, 6;P =0.2091, respectively) (Table 1). At Byron, no significant difference occurred among treatments in number of Prostigmata, Oribatei, others, and total number of microarthropods (F =3.04; df =4, 4; P =0.1531; F =2.41; df =4, 14; P =0.2074; F = 1.84; df =4,4; P =0.2843; F =4.46; df =4,4; P =0.0885, respectively) (Table 1). However, a significant difference occurred in the number of Mesostigmata and Collembola (F =3.34; df =4,14; P =0.0407; F =10.0; df =4, 14; P =0.0005, respectively). Plots of one Bt corn hybrid had significantly fewer mesostigmatid mites than its isoline. There were significantly more collembolans in corn plots treated with Furadan than in plots of Bt corn and its isoline. At Sedgwick and Pratt, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =0.18; df =2, 2; P =0.8500; F =0.85; df =2, 2; P =0.5398; F =0.38; df =2, 8; P =0.6969; F =0.07; df =2, 8; P =0.9373; F =1.17; df =2, 8; P = 0.3593; F =0.16; df =2, 8; P =0.8510, respectively) (Table 1). At Oakley and Hoxie, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =0.19; df =2, 2; P =0.8382; F =0.35; df =2, 2; P =0.7391; F =0.42; df =2, 2; P =0.7033; F =1.03; df =2, 8; P = 0.3988; F =3.94; df =2, 2; P =0.2025; F =0.31; df =2, 2; P =0.7607, respectively) (Table 1).

View this table:
Table 1.

Tullgren-Type Extraction, 2001

Results of arthropods in soil samples collected at three locations (Manhattan, Scandia 1 and 2) in 2001 were in agreement with the results obtained in 2000. At Manhattan, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =0.89; df =3, 32; P =0.4567; F =1.71; df =3, 9; P =0.2345; F =0.39; df =3, 9; P =0.7641; F =1.82; df =3, 33; P =0.1624; F =1.18; df =3, 33; P =0.3330; F =0.61; df =3,9; P =0.6264, respectively) (Table 2). At Scandia 1, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =0.90; df =4, 42; P = 0.4703; F =1.27; df =4, 42; P =0.2959; F =0.83; df =4, 42; P =0.5136; F =1.33; df =4, 12; P =0.3149; F =0.35; df =4, 42; P =0.8451; F =1.09; df =4, 42; P = 0.3730, respectively) (Table 2). At Scandia 2, no significant difference occurred among treatments in number of Prostigmata, Mesostigmata, Oribatei, Collembola, others, and total number of microarthropods (F =0.29; df =4, 42; P =0.8806; F =1.14; df =4, 42; P =0.3501; F =0.50; df =4, 42; P =0.7354; F =0.79; df =4,12; P =0.5512; F =1.57; df =4, 42; P =0.2011; F =0.78; df =4,12; P =0.5572, respectively) (Table 2).

View this table:
Table 2.

Nematode Extraction, 2000-2001.

In 2000, no significant difference occurred between Bt corn (MON 863) and its isoline in number of Helicotylenchus (Spiral), Pratylenchus (Lesion), Tylenchorhyncus (Stunt), Hoplolaimus (Lance) in soil, and Pratylenchus in roots (F =5.63; df =1, 3; P =0.0982, F =1.26; df =1,3; P =0.3434; F =1.40; df =1,3; P =0.3217; F =0.58; df =1, 3; P =0.5012; F =1.16; df =1, 3; P =0.3607, respectively) (Table 3). In 2001, nematodes were extracted on two occasions. On the first occasion, no significant difference occurred among treatments in number of Tylenchorhyncus, Pratylenchus in roots, Tylenchidae, Aphelenchina, Rhabditidae, Cephalobidae, and Dorylaimida (F =0.51; df =6, 21; P =0.79; F =2.07; df =6, 21; P =0.11; F =1.59; df =6, 21; P =0.21; F =0.20; df =6, 21; P =0.97; F =2.08; df =6, 21; P = 0.11; F =1.48; df =6, 21; P =0.24; F =0.64; df =6, 21; P =0.70, respectively) (Table 4). However, a significant difference occurred in number of Pratylenchus in soil among treatments (F =3.66; df =6, 21; P = 0.01). Isoline +Lorsban had significantly more Pratylenchus than isoline +ST, isoline +Force, isoline +Prescribe. No significant difference occurred among the Bt corn (MON 863), its isoline, and the other treatments. On the second occasion, no significant difference occurred between Bt corn (MON 863) and its isoline in number of Tylenchorhyncus, Pratylenchus in soil, Pratylenchus in roots, Tylenchidae, Aphelenchina, Rhabditidae, Cephalobidae, and Dorylaimida (F =1.47; df =1, 3; P =0.31; F =0.17; df =1, 3; P =0.71; F =0.48; df =1, 3; P = 0.54; F =9.00; df =1, 3; P =0.06; F =3.35; df =1, 3; P =0.16; F =1.00; df =1, 3; P =0.39; F =0.86; df =1,3; P =0.42; F =0.27; df =1,3; P =0.64, respectively) (Table 5). These results were in agreement with those obtained in 2000.

View this table:
Table 3.
View this table:
Table 4.
View this table:
Table 5.

Discussion

Most of the published research on the effect of Bt corn on soil fauna is on corn hybrids expressing the Cry1 proteins, which target lepidopteran insects. Studies with Bt corn expressing the Cry1Ab protein for European corn borer control have suggested that roots release Bt toxins into the soil (Saxena et al. 1999) and that they persist there (Tapp and Stotzky 1998). In this study, the corn had a different protein (Cry3Bb1) that targets coleopteran insects.

Corn was planted in eight locations in 2000 and three locations in 2001. One site (Manhattan) was nonirrigated, while the rest were irrigated. Soil mites and Collembola were sampled on three occasions per season in 2000 and 2001. Nematodes were sampled once in 2000 and twice in 2001. In general, there were no significant differences in numbers of Prostigmata, Mesostigmata, Oribatei, and Collembola in soil from Bt corn or its isoline. There was no significant difference in the nematode groups sampled. The fact that no significant differences occurred in numbers of soil mites and Collembola in the irrigated and the non-irrigated sites suggests that soil moisture was not affecting the fate or activity of the Cry3Bb1 protein in soil.

There are several possible explanations for these results. It is likely that toxins from the Bt corn roots in this study are not toxic to the groups of arthropods sampled or that the Bt toxin is not released into soil or is not present in lethal concentrations. The results of the current study are in agreement with several other studies. For example, Yu et al. (1997) found that the Bt toxins Cry1Ab and Cry1Ac had no effects on Folsomia candida Willem (Collembola: Isotomidae) and Oppia nitens Koch (Acari: Orbatidae). Similarly, Sims and Martin (1997) found no toxicological risk from Cry1Ab, Cry1Ac, Cry2A, and Cry3A to two species of Collembola (F. candida and Xenylla grisea Axelson). As far as we know, this is the first report on the effects of Cry3Bb1 protein on nematodes. The results of different studies on effects of Bt corn on nontarget organisms support the conclusion of Wolfenbarger and Phifer (2000) that the risks and the benefits of genetically engineered organisms vary in time and space on a case-by-case basis.

Studies on the effect of Bt corn on soil mites, Collembola, and nematodes elucidate the relationship between a specific Bt corn and natural soil fauna. In general, the results of this study demonstrated that Bt corn expressing the Cry3Bb1 protein for corn root-worm control does not have deleterious effects on the nontarget microarthropods and nematodes sampled. Further studies on soil after repeated growing of Bt corn in the same field for several years are needed.

Acknowledgments

We give special thanks to George Milliken for his help in the statistical analyses. We also thank Jennifer Bieber and Brooklyn Cleveland for the technical assistance. We are grateful to John Reese for reviewing the manuscript and to Monsanto Company for providing the corn hybrids. This is contribution 03-257-J from the Kansas Agriculture Experiment Station.

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions{at}oup.com

References Cited

View Abstract