RESEARCH The Use of Root Gall Ratings to Determine High Risk Zones in Cotton Fields Infested by Meloidogyne incognita J. A. Wrather,* W. E. Stevens, E. D. Vories, T. L. Kirkpatrick, J. D. Mueller, and A. Mauromoustakos ABSTRACT Farmers growing cotton (Gossypium hirsutum L.) need a reliable, accurate, and inexpensive method for mapping areas of potentially high risk from root-knot nematodes (RKN) [Meloidogyne incognita (Kofoid & White) Chitwood] within individual fields for site-specific application of nematicides. Evaluation of postharvest cotton roots for galling severity due to RKN may be an alternative to soil analysis for nematodes for developing these maps. The main objectives of this study were to determine the relationship between yield of cotton and root galling severity the year before planting, and the estimated costs per hectare for rating root galling severity compared with that of conventional soil sampling using a 15-m grid spacing. There was a significant negative correlation between root galling severity in October and cotton yield the next 2 yr, indicating galling severity may be a useful indicator of the potential threat of RKN to crop performance for more than 1 yr. The estimated costs for assessing galling severity, $183 ha–1, were much less than for soil analysis for nematodes, $968 ha–1. Unfortunately, maps based on galling severity will only be useful guides for site-specific application of nematicides if RKN is the only economically important cotton parasitic nematode present. More accurate and less expensive ways of sampling for RKN are needed to identify within-field areas where the risk of nematode-induced yield loss is high. J.A. Wrather and W.E. Stevens, Univ. of Missouri-Delta Center, Portageville, MO 63873; E.D. Vories, USDA-ARS, Portageville, MO; T.L. Kirkpatrick, Univ. of Arkansas, Fayetteville, AR 71801; J. D. Mueller, Clemson Univ., Blackville, SC 29817; A. Mauromoustakos, Agriculture Statistics Lab., Univ. of Arkansas, Fayetteville, AR 72701. Received 3 Feb. 2010. *Corresponding author (wratherj@missouri.edu). Abbreviations: HWCI, half width confidence interval; J2, nematode second-stage juvenile; Pf, nematode juvenile population at harvest; Pi, nematode juvenile population at planting; RKN, root-knot nematodes. F armers in Missouri harvested about 129,000 ha of cotton annually from 2004 to 2008, all from the southeast area of the state. The majority (97.4%) of production was from four counties: New Madrid, Pemiscot, Dunklin, and Stoddard. Lint yields averaged 1126 kg ha–1 each year during this period. Cotton yield in Missouri would have been greater if not for southern RKN. Southern root-knot and other parasitic nematodes were found in Missouri cotton fields, and the presence and geographic distribution of nematode pests of cotton in Missouri have been described (Wrather et al., 1992). Reniform nematodes, Rotylenchulus reniformis (Linford & Oliveira), and lance nematodes, Hoplolaimus galeatus (Cobb) Thorne, were present in only 3 and 2% of the Missouri cotton fields sampled, respectively. The population density of both nematodes near cotton harvest was never >10 juveniles per 100 cm 3 soil. Southern RKN were found in 30% of the cotton fields surveyed and nematode distribution was spatially aggregated within each field. The estimated cotton yield loss in Missouri due to RKN averaged 3.6 million kg of lint each year from 2004 to 2008 (Wrather and Sweets, 2009). The value of this yield loss was about $5.25 million per year. In the United Published in Crop Sci. 50:2575–2579 (2010). doi: 10.2135/cropsci2010.02.0054 Published online 27 Sept. 2010. © Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher. CROP SCIENCE, VOL. 50, NOVEMBER– DECEMBER 2010 2575 States, M. incognita reduced cotton yield approximately twice as much as all other nematode parasites of this crop, and the biology and management of this pest has recently been reviewed (Koenning et al., 2004). Farmers have tolerated cotton yield loss due to RKN because they had few effective nematode control options. Resistant cotton cultivars are not available and cultural practices, such as crop rotation, planting winter cover crops, and minimum tillage systems have not been helpful for management of this nematode (Koenning et al., 2004). At-plant applications of the nematicide aldicarb (Temik) and preplant applications of the nematicide 1,3-dichloropropene (Telone II) have been the most commonly-used strategy for mitigation of RKN-induced crop loss. Farmers generally apply a uniform rate of nematicide across entire fields to protect cotton against nematodes. However, field-wide application of nematicides is generally rather inefficient because the nematodes are not uniformly distributed within most fields (Monfort et al., 2007; Wheeler and Kaufman, 2003; Wrather et al., 2002). The expense and environmental risk of field-wide application of nematicides could be reduced if these products were applied only to areas within fields where infestations are sufficiently high to warrant nematode control. Cotton farmers need a reliable, accurate, and inexpensive method for determining the potential threat of RKN to cotton within individual fields. Grid-maps of RKN distribution can be developed for each field through nematode analysis of soil samples collected postharvest. However, relatively small grids (0.1-ha spacing) may be necessary, and the expense and feasibility of timely completion of sampling and conducting appropriate laboratory assays can be prohibitive (Wrather et al., 2002). Evaluation of cotton roots for RKN galls at harvest is the most diagnostic characteristic of an M. incognita infection (Goodell et al., 1996; Koenning et al., 2004) and may be an alternative to soil analysis for nematodes for mapping RKN distribution in fields. Root galling provides an immediate visual indication both of the distribution of the nematodes within the field and the severity of the nematode-induced crop damage without the need for soil sampling and laboratory analysis. The objectives of this project were (i) to determine the relationship between yield of cotton and soil population density of M. incognita juveniles at planting (Pi) and postharvest soil population density of M. incognita juveniles (Pf ) and root galling severity the year before planting, (ii) to determine the percent of roots that should be examined for galling within a field site or grid to estimate galling severity with 95% confidence of true galling severity, (iii) to determine if grid sampling at 15-m spacing for root galling severity was close enough to develop an interpolated map of root galling severity, and (iv) to compare the estimated costs per hectare for rating root galling severity with that of conventional soil 2576 sampling on grids for determining potential for damage to the subsequent crop. MATERIALS AND METHODS A 3.6-ha portion of a field near Hornersville, MO, was selected as the study site. The soil was a fine-silty, mixed, thermic, Aeric Ochraqualfs and was 67% sand, 13% silt, and 20% clay. The field had been planted to cotton from 1991 through 2001, was known to be infested with RKN, and the RKN distribution was spatially aggregated. The study site was subdivided into 156 grid points, 15 m spacing, soon after cotton harvest in October 2001. At each grid point, 10 soil cores (2.5 cm diameter × 20 cm deep) were arbitrarily collected from a 2 m 2 area. The soil cores were composited, and nematodes were extracted from a 250-cm 3 subsample by semiautomatic elutriator and centrifugation (Barker, 1978). Plant parasitic nematodes were identified to genus. Secondstage juveniles ( J2) of RKN were assumed to be M. incognita because of the history of cotton in the field and because other species of Meloidogyne do not reproduce on cotton (Koenning et al., 2004). A portion of each soil sample was analyzed for percentage sand, silt, and clay. Ten arbitrarily selected cotton plants were dug from the 2-m 2 area at each grid point in October 2001 and 2007, and the roots of each were evaluated for galling severity due to RKN using a 1 to 6 rating system where 1 = no galls and 6 = 100% of the total root system galled (Barker, 1978). The upper tap root and lower stem of each plant were split and examined for discoloration due to Verticillium and/or Fusarium wilt (Bell, 2001; Colyer, 2001). Twelve of the 156 grid points were selected for plots because of similarity in soil percentage sand, silt, and clay, and differences in cotton root gall severity in October 2001. The galling severity at these grid points ranged from 1 to 6, with the average severity of 1 to 2 in four of the plots, 3 to 4 in four plots, and 5 to 6 in four plots. The plots were one row wide (0.97-m row spacing) and 6 m long, and were established at the same location in both 2002 and 2003. Nematicides were not applied to these plots. The field was planted to DPL 451 (Delta Pine and Land Company, Scott, MS) on 10 May 2002 and 12 May 2003. A professional consultant surveyed the crop weekly during the growing season for pests, and insecticides and herbicides were applied based on the survey results. The consultant also made suggestions to the farmer about application of fertilizer, growth regulators, defoliants, and irrigation. Soil samples for preplant (Pi) nematode counts were collected from plots immediately before planting in May 2002 and 2003. Soil samples for harvest (Pf ) nematode counts were collected from plots immediately after harvest in October 2002. Soil samples were collected and processed as previously described. All cotton plants (90–120 plants) were collected from each plot after harvest in October 2002 and evaluated for root galling severity as previously described. The true galling severity for a plot was assumed to be the average galling severity for all plants in the plot. Seed cotton from each plot was collected by hand harvest in 2002 and 2003, and yield of seed cotton per hectare was calculated. To test whether a 15-m sampling grid was sufficiently close for creating interpolated maps of root galling severity, WWW.CROPS.ORG CROP SCIENCE, VOL. 50, NOVEMBER– DECEMBER 2010 geostatistical analysis of root galling severity was performed for the October 2001 sampling date (Wrather et al., 2002). For this analysis, a semivariagram was created that showed semivariation (half the variation) between pairs of points (y axis) relative to the distance between those points (x axis). Typically, a number of pairs at the same distance are averaged. If the semivariation for the various separation distances examined is approximately the same, then the interpretation is that information obtained at any grid point is independent of the information at any other grid point. If semivariation increases with increasing separation distance, the interpretation is that the grid spacing is close enough to capture spatial dependence information between the points. Spatial dependence of gall severity at this site was evaluated using SAS software (SAS Institute, Cary, NC). When spatial dependence between sample points was evident, linear, spherical, and exponential models were tested and the best-fitting model was identified using least-squares procedure selecting the model with the greatest R 2. To determine with 95% confidence of the true galling severity the percentage of plant roots that should be examined for galls at a site, a Monte Carlo simulation program was used with SAS to determine how close different random sample sizes were to the true mean of galling severity at a site (mean rating of all plant roots for galls at a site). The SAS PROC CORR procedure was used to determine the correlations between seed cotton yield and Pi, Pf, and root gall severity. To compare the estimated costs per hectare for rating root galling severity with that of conventional soil sampling on grids for determining potential for damage to the subsequent crop, some known and some assumed values were used. We knew the University of Missouri Nematode Diagnostic Laboratory charged $20 per sample for nematode analysis. We assumed consultants would charge $2 per soil sample for collection, $50 h–1 for crop consultant labor, and they would examine about 10% of plant roots for galling at a site. RESULTS Other than Meloidogyne, only Paratichodorus spp. and Xiphinema spp. were found in this site, but the population density of these nematodes never exceeded 30 juveniles per 250 cm3 soil at any sampling date (data not shown). The economic importance of these genera on cotton has not been determined (Koenning et al., 2004), and they are not considered damaging to cotton in Missouri (Wrather et al., 1992). Meloidogyne incognita was found in most but not all plots. The soil population density of M. incognita J2 in samples collected ranged from 0 to 227 per 250 cm 3 soil in May 2002 and 2003 and from 0 to 1200 per 250 cm 3 soil in samples collected in October 2001 and 2002. Root galling severity due to RKN ranged from 0 to 6 in October 2001 and 2002. Discoloration of tissue due to wilt diseases was not observed. There was no significant correlation between either Pi or Pf soil population density of M. incognita and cotton yield in 2002 or 2003 (Table 1), indicating that nematode population density was not a useful indicator of the potential threat of RKN to crop performance. There was, CROP SCIENCE, VOL. 50, NOVEMBER– DECEMBER 2010 Table 1. Pearson correlation coefficients between 2002 and 2003 seed cotton yield and Meloidogyne incognita soil population density at harvest (Pf) 2001 and 2002 and at planting (Pi) 2002 and 2003.† Soil population Pearson correlation coefficient Probability Pi 2002 P f 2001 Cotton yield 2002 0.09727 –0.21119 0.7636 0.5100 Pi 2003 P f 2001 P f 2002 Cotton yield 2003 0.17043 –0.27371 –0.32872 0.5964 0.3893 0.2968 † N = 12. however, a significant negative correlation between root galling severity in October 2001 and cotton yield in both 2002 and 2003, and between galling severity in October 2002 and yield in 2003 (Table 2), indicating galling severity may be a useful indicator of the potential threat of RKN to crop performance. The half width confidence interval (HWCI) for an estimate of root galling severity to true galling severity declined as the percentage of plant roots examined increased (Table 3). For example, a galling estimate using a 1 to 6 severity scheme at a location would be ±2.58 of true galling severity if 3% of the roots were examined and ±0.30 of true galling severity if 50% of roots were examined. Using a 15-m sampling grid, root gall severity in October 2001 was too variable between adjacent grid points to observe any spatial dependence, that is, the galling severity at any grid point was independent of the severity at any other grid point. The same grid points were sampled for root galling severity after the cotton harvest in October 2007, and severity was again too variable between adjacent grid points to observe any patterns in severity distribution. The mean galling severity October 2001 was 1.93, and the severity ranged from 1.7 to 2.1 with 95% confidence. Root galling severity at most grid points was less in October 2001 than October 2007 when the mean gall severity was 3.66 and ranged from 3.4 to 3.8 with 95% confidence. The cost of collecting and analyzing soil samples for nematode juveniles on a 15-m grid was estimated to be $968 ha–1 (44 samples ha–1 at $2 per sample for collection Table 2. Pearson correlation coefficients between 2002 and 2003 seed cotton yield and Meloidogyne incognita caused cotton root gall severity determined October 2001 and 2002.† Gall severity Pearson correlation coefficient Probability 2001 Cotton yield 2002 –0.68079 0.0148 2001 2002 Cotton yield 2003 –0.60220 –0.67605 0.0383 0.0158 † N = 12. WWW.CROPS.ORG 2577 Table 3. Half width confidence interval (HWCI) for gall severity rating at the 95% confidence for percent of plant population sampled at a location.† % of plant population sampled 3 5 10 25 50 HWCI 2.58 1.22 0.74 0.43 0.30 † These results are based on the postharvest root galling severity rating for all cotton plants from each plot (about 90 –120 plants per plot) in October 2002. and $20 per sample for nematode analysis at the University of Missouri Nematode Diagnostic Laboratory). Consultants currently charge Missouri cotton farmers about $2 per sample to collect soil. Conversely, the estimated cost of analysis of 10% of plant roots for galling severity on a 15-m grid was $183 ha–1 (44 samples ha–1 at $4.00 per sample). This estimate was based on expected time needed to evaluate plant galling severity by a consultant (about 5 min to walk from site to site, dig roots, and examine roots for gall severity) and $50 h–1 for crop consultant labor. The cost for consultants may be higher or lower in other states. DISCUSSION The variation in M. incognita population density among grid points at this site was similar to that reported by others (Wrather et al., 2002) and was likely due to multiple factors. Although the overall soil percentages of sand, silt, and clay were similar in the soils at this site, other edaphic or environmental factors, such as percentage of very coarse, coarse, fi ne, and very fine sand may have varied among grid points and influenced the development of M. incognita. These data were not collected for this study. These soil characteristics can influence the ecology and distribution of M. incognita (Koenning et al., 1996; Prot and Van Gundy, 1981; Robinson et al., 1987). The absence of a significant correlation between Pi, Pf, and cotton yield was probably because the J2 population densities at these times did not accurately reflect the actual damage potential of RKN to cotton. Detection of M. incognita juveniles in the spring is difficult because winter mortality of eggs and J2 can be very high (Barker and Imbriani, 1984), and cool soil (about 16–19°C) slows egg hatch. Farmers in the upper Mississippi delta region must plant cotton in late April and early May for best yield (Wrather et al., 2008), and the soil temperatures during this period are generally 15 to 20°C. Eggs of M. incognita represent a major portion of the overwintering nematode population (Jeger et al., 1993; Starr and Jeger, 1985), and accurate techniques for detection of eggs or egg masses free in soil have not been developed (Starr, 1993). The soil population of J2 at harvest may not reflect the actual damage potential of RKN to cotton because of extreme variability in hatch of eggs over time at harvest. The population density of M. incognita J2 in soil varied significantly among 2578 biweekly sample dates from early October to late November in the upper Mississippi delta region (P. Donald, USDAARS, 2008, personal communication). This variability was probably the reason for poor correlation between Pf and yield of subsequent cotton crops in this study. The results of this study are the first to show that assessment of root galling severity due to RKN is an alternative to soil analysis for Pi or Pf for developing field maps showing areas of potential high risk to RKN. Galling severity ratings were more reliable for estimating M. incognita actual damage potential to cotton than soil population density of J2 in this study. The estimated costs for assessing galling severity, $183 ha–1, were much less than for assessing soil population density of J2, $968 ha–1. The results of an exam of plant roots for galling severity were available immediately compared with waiting sometimes weeks for the results of soil analysis for J2 population, so maps of RKN density based on galling severity could be available immediately after harvest for use by farmers for site-specific application of nematicides such as Telone II. In addition, our study implies that maps of M. incognita based on galling severity may be effective indicators of areas of potentially high risk for at least 2 yr after they are developed. As a result, the costs for assessing galling severity may be amortized over 2 yr so the costs will be less than $183 ha–1 each year. More work is needed to determine the correlation between galling severity and cotton yield the third and subsequent years following the year ratings are collected. The estimate of cost for developing a root galling severity rating at a grid point was based on an exam of 10% of plants. Clearly, the HWCI and the cost for developing a root galling severity rating at a grid point based on an exam of 50% of plants would be greater than an exam of 10% of plants. In this study, the greatest expense for estimating galling severity was the time to dig roots and examine them for galls. For a rating scale of 1 to 6, a HWCI ≥ 0.5 (i.e., a confidence interval wider than 1 rating point) may not provide the precision necessary for site specific application of nematicides. Therefore, interpolating the values in Table 3 suggests that a minimum of 22% of the plant roots should be examined for galling for a reliable estimate. Other work is needed to determine the accuracy of galling severity estimates most suitable and this may vary among fields due to soil characteristics. Information about the relationship between galling severity at harvest and subsequent cotton crop yield differences between cotton treated preplant with Telone II and not treated may be useful for understanding the accuracy of galling severity estimates needed. The consultant using this system must consider the costs of analysis and accuracy needed when determining the percent of plant roots to examine at a grid point. A 15-m spacing between grid points for analysis of soil for juveniles and analysis of roots for galling severity was not close enough to provide evidence of spatial dependence between points with any statistical confidence. Geostatistics WWW.CROPS.ORG CROP SCIENCE, VOL. 50, NOVEMBER– DECEMBER 2010 may provide accurate maps of soil juveniles or root galling severity, but grid spacing may need to be closer than is economically practical. Interpolated maps will be necessary for site specific application of variable rate nematicides, but application of multiple variable rates of nematicides may not be essential for adequate management of RKN. A map showing hot spots of RKN may be sufficient for guiding site specific application of single rate nematicides (Evans et al., 2002). Unfortunately, maps based on galling severity will only be useful guides for site-specific application of nematicides or other remediation tactics if M. incognita is the only economically important cotton parasitic nematodes present. Soil analysis for juveniles of all plant parasitic nematodes will be necessary if the presence of other nematodes is known, or suspected. In a mixed population of M. incognita and Rotylenchulus reniformis, the reniform nematodes might be managed by rotating corn (Zea mays L.) with cotton or by planting R. reniformis resistant soybean cultivars in rotation with cotton (Koenning et al., 2004). The population of reniform nematodes could be reduced below the threshold so that site-specific application of nematicides for RKN management could be guided with distribution maps based on root galling severity. More accurate and less labor-intensive and expensive ways to identify within-field areas where the risk of nematode-induced yield loss is high are needed (Starr et al., 2007). An alternative to soil sampling for elucidating cotton-parasitic nematode distribution and quantifying population density may be the use of galling ratings, particularly in fields where M. incognita is the only economic nematode present, or in regions or soil types where the probability of this nematode being the only economically significant nematode is high. Although the correlation between nematode density distribution and soil type has received considerable attention (Noe and Barker, 1985; Monfort et al., 2007), relationships between nematode population densities and other factors that could be important within-field are not known. Emerging precision technology including mapping soil electrical conductivity, remote images of cotton growth, and remote images of bare soil may be very useful in detecting variations in nematode population density within fields and should be explored in the search for a more effective and less expensive means of assessment of plant damaging populations of cotton parasitic nematodes. Acknowledgments This research was supported in part by the Missouri Agriculture Experiment Station, USDA Initiative for Future Agriculture and Food Systems Grant 00-52103-9648, and Cotton Incorporated Projects 05-628MO and 07-968MO. The authors thank Cory Cross, Joyce Elrod, and Ronnie Bateman for technical assistance and God for guidance. References Barker, K.R. 1978. Determining nematode population responses to control agents. p. 114–127. In E.I. Zehr et al. (ed.) Methods for CROP SCIENCE, VOL. 50, NOVEMBER– DECEMBER 2010 evaluating plant fungicides, nematicides, and bactericides. American Phytopathological Society, St. Paul, MN. Barker, K.R., and J.L. Imbriani. 1984. Nematode advisory programs— Status and prospects. Plant Dis. 68:735–741. Bell, A.A. 2001. Verticillium wilt. p. 29–30. In T.L. Kirkpatrick and C.S. Rothrock (ed.) Compendium of cotton diseases. 2nd ed. The American Phytopathological Society, St. Paul, MN. Colyer, P.D. 2001. Fusarium wilt. p. 27–28. In T.L. Kirkpatrick and C.S. Rothrock (ed.) Compendium of cotton diseases. 2nd ed. The American Phytopathological Society, St. Paul, MN. Evans, K., R.M. Webster, P.D. Halford, A.D. Barker, and M.D. Russell. 2002. Site-specific management of nematodes-pitfalls and practicalities. J. Nematol. 34:194–199. Goodell, P.B., M.A. McClure, P.A. Roberts, and S.H. Thomas. 1996. Nematodes. p. 105–107. In P.A. Rude (ed.) Integrated pest management for cotton in the western region of the United States. Publ. 3305. Div. of Agric. and National Res. Univ. of California, Davis. Jeger, M.J., J.L. Starr, and K. Wilson. 1993. Modeling winter survival dynamics of Meloidogyne spp. eggs and juveniles with loss of egg viability. J. Appl. Ecol. 30:496–503. Koenning, S.R., T.L. Kirkpatrick, J.L. Starr, J.A. Wrather, N.R. Walker, and J.D. Mueller. 2004. Plant–parasitic nematodes attacking cotton in the United States. Plant Dis. 88:100–113. Koenning, S.R., S.A. Walters, and K.R. Barker. 1996. Impact of soil texture on the reproductive and damage potential of Rotylenchulus reniformis and Meloidogyne incognita on cotton. J. Nematol. 28:527–536. Monfort, W.S., T.L. Kirkpatrick, C.S. Rothrock, and A. Mauromoustakos. 2007. Potential for site-specific management of Meloidogyne incognita in cotton using soil textural zones. J. Nematol. 39:1–8. Noe, J.P., and K.R. Barker. 1985. Relation of within-field spatial variation of plant-parasitic nematode population densities and edaphic factors. Phytopathology 75:247–252. Prot, J.C., and S.D. Van Gundy. 1981. Effect of soil texture and the clay component on migration of Meloidogyne incognita second-stage juveniles. J. Nematol. 13:213–216. Robinson, A.F., C.M. Heald, S.L. Flanagan, W.H. Thames, and J. Amador. 1987. Geographical distributions of Rotylenchulus reniformis, Meloidogyne incognita, and Tylenchulus semipenetrans in the lower Rio Grande Valley as related to soil texture and land use. Ann. Appl. Nematol. 1:20–25. Starr, J.L. 1993. Recovery and longevity of egg masses of Meloidogyne incognita during simulated winter survival. J. Nematol. 25:244–248. Starr, J.L., and M.J. Jeger. 1985. Dynamics of winter survival of eggs and juveniles of Meloidogyne incognita and M. arenaria. J. Nematol. 17:252–256. Starr, J.L., S.R. Koenning, T.L. Kirkpatrick, A.F. Robinson, P.A. Roberts, and R.L. Nichols. 2007. The future of nematode management in cotton. J. Nematol. 39:283–294. Wheeler, T.A., and H.W. Kaufman. 2003. Relationship of aerial broad band reflectance to Meloidogyne incognita density in cotton. J. Nematol. 35:48–57. Wrather, J.A., T.L. Niblack, and M.R. Milam. 1992. Survey of plantparasitic nematodes in Missouri cotton fields. Suppl. J. Nematol. 24:779–782. Wrather, J.A., B.J. Phipps, W.E. Stevens, A.S. Phillips, and E.D. Vories. 2008. Cotton planting date and plant population effects on yield and fiber quality in the Mississippi Delta. J. Cotton Sci. 12:1–7. Wrather, J.A., W.E. Stevens, T.L. Kirkpatrick, and N.R. Kitchen. 2002. Effects of site-specific application of aldicarb on cotton in a Meloidogyne incognita-infested field. J. Nematol. 34:115–119. Wrather, A., and L. Sweets. 2009. Cotton nematodes in Missouri: Your hidden enemies. Extension Publ. G4259. Univ. of Missouri, Columbia. WWW.CROPS.ORG 2579