Published online 22 January 2007
Published in Crop Sci 47:19-24 (2007)
© 2007 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
PLANT GENETIC RESOURCES
Reniform Nematode Resistance in Upland Cotton Germplasm
David B. Weavera,*,
Kathy S. Lawrenceb and
Edzard van Santena
a Dep. of Agronomy & Soils, Auburn Univ., Auburn, AL 36849-5412
b Dep. of Entomology and Plant Pathology, Auburn Univ., Auburn, AL 36849-5412
* Corresponding author (weavedb{at}auburn.edu)
 |
ABSTRACT
|
|---|
Cotton (Gossypium spp.) is attacked by parasitic nematodes including the reniform nematode (Rotylenchulus reniformis Linford and Oliveira). Options for management of reniform nematode are limited. No cultivars of upland cotton (G. hirsutum L.) have genetic resistance. Our objectives were to evaluate the USDA G. hirsutum collection for reaction to parasitism by R. reniformis, and determine the value of measurement of eggs (reproduction) or vermiform stages (nematode survival) as an indicator of nematode resistance. In groups of 50, accessions were evaluated in the greenhouse, using single plants in four replicates. Accessions were planted in sterile soil and inoclated with a mixture of R. reniformis isolates. After 60 d, soil populations of vermiform nematodes were determined, and eggs were extracted from the root system and counted. Paymaster ‘PM 1218’ was included as a check in every experiment. Out of 1973 accessions with at least one replication, none showed high levels of resistance. Seven accessions had lower population development than PM 1218 after repeated evaluations. Results indicated egg counts and vermiform counts were correlated, but not closely. Egg counts were higher and more variable than vermiform counts. While some accessions showed levels of resistance that might be useful in cotton improvement, evaluation remains difficult and introgression of genes for reniform nematode resistance remains a long-term breeding objective.
|
INTRODUCTION
|
|---|
UPLAND COTTON is subject to attack by a variety of plant-parasitic nematodes, including the reniform nematode. The reniform nematode was first identified as a cotton pathogen in 1940 (Smith, 1940), recognized as a serious problem a few years later (Jones et al., 1959), and has been increasing in importance since that time (Lawrence and McLean, 1995). This nematode is widespread throughout the cotton belt, from North Carolina to New Mexico (National Cotton Council, 2005) with particularly dense populations in areas of intense cotton production. Reniform nematode is known to affect cotton primarily through reductions in yield, boll size, and lint percentage (Cook et al., 1997; Jones et al., 1959). Plants can be stunted and may respond poorly to inputs such as irrigation and fertilization (Birchfield and Jones, 1961). Due to a lack of resistance in adapted upland cotton cultivars (Robinson et al., 1999, 2004), strategies for management of reniform nematode in cotton are limited primarily to nematicides and crop rotation. While sources of genetic resistance and immunity exist within Gossypium and related species (Yik and Birchfield, 1984), success in finding resistance within G. hirsutum has been limited (Robinson and Percival, 1997; Robinson et al., 2004). Germplasm lines have been released that reportedly have some level of resistance to reniform nematode (Jones et al., 1988). However these lines were released primarily because of their resistance to southern root-knot nematode [Meloidogyne incognita (Kofoid & White) Chitwood] and the source of resistance to R. reniformis in these germplasm releases is unclear.
Genetic options using current cultivars have been limited to use of tolerance. Cook et al. (1997) tested several advanced breeding lines, including one of the germplasm lines released by Jones et al. (1988) in a field infested with R. reniformis and M. incognita race 3. He demonstrated that some lines yielded well in untreated plots compared to fumigated plots. However, both nematode species were able to reproduce freely on these lines in growth chamber experiments and did not differ in reproduction from that observed in the reniform-susceptible control. Thus Cook et al. (1997) concluded that much of the superior performance of these lines could be attributed to tolerance. Usery et al. (2005) also evaluated cultivars for tolerance to reniform nematodes and found that the expression of tolerance was dependent on location. Tolerance can be an attractive alternative to producers because it minimizes yield loss, but it has some major drawbacks. It does nothing to suppress nematode populations. In fact tolerance to reniform in cotton has been shown to increase nematode populations, as the species reproduces freely on tolerant types (Cook et al., 1997; Robinson et al., 2004). Tolerance is also a limited option for cotton breeders. It is utilized primarily through the identification of tolerant cultivars after their development. Tolerance is difficult to evaluate in breeding populations, and the physiological and genetic basis of tolerance is largely unknown.
Although success at identifying G. hirsutum genotypes with good resistance to R. reniformis has been limited, genetic resistance remains a very attractive management option. Several attempts have been made to identify resistance sources within G. hirsutum. Yik and Birchfield (1984) tested 110 primitive accessions of G. hirsutum and identified three that supported a statistically lower population than the susceptible control, but were later found not to be statistically different from the control in another study (Robinson and Percival, 1997). This has led researchers to look to other Gossypium species as sources of resistance. Several accessions of long-staple cotton (G. barbadense), including TX-110, TX-1347, and TX-1348, were identified as resistant and confirmed in later experiments (Yik and Birchfield, 1984; Robinson and Percival, 1997; Robinson et al., 2004). Although G. hirsutum and G. barbadense are cross compatible, the successful transfer of any genes from one species to the other via interspecific crosses has proven to be difficult due to chromosome pairing issues during meiosis (Jiang et al., 1999). Robinson et al. (2004) further evaluated G. barbadense (907 accessions) and G. hirsutum (1866 accessions) for resistance to either or both reniform and southern root-knot nematode. He used a two-tiered approach, with initial screening based on the observation of a single inoculated plant in the greenhouse. Based on the initial screening data, 34 accessions of G. hirsutum were identified that supported poor reproduction of reniform nematode and were advanced to the second tier, a replicated growth chamber experiment. This experiment revealed that many of the accessions that had supported low populations in the greenhouse screenings were able to support populations that were not different from the susceptible check ‘Deltapine 16’ when evaluated in the growth chamber. However six accessions (TX-25, TX-1828, TX-1860, TX-748, TX-1586, and TX-2469) were identified as being moderately resistant because they still supported from 26 to 34% of the reniform population supported by the susceptible control. One G. barbadense accession, GB-713 supported only 3% of the nematode population supported by the susceptible control. Thus, a high level of resistance to R. reniformis in G. hirsutum remains to be discovered.
Evaluation for resistance to reniform nematode in cotton is difficult. For other nematode species such as root-knot nematode, researchers can often rely on disease symptoms, such as a galling index, to measure resistance (Shepherd, 1974; Robinson et al., 1999, 2004; Zhang et al., 2006). No such symptoms develop in plants infected with R. reniformis. According to Robinson (1999) root systems infected with R. reniformis can appear more or less normal unless viewed under magnification, even though aboveground symptoms observed in the field are well documented. Hence most greenhouse or growth chamber evaluation schemes have involved either measurement of egg production as an indicator of nematode reproduction, determination of nematode feeding and survival as indicated by a final vermiform nematode population count, or both. These are usually then expressed as a percentage of a susceptible check. Yik and Birchfield (1984) measured eggs per gram of root and expressed resistance level as a percentage of eggs per gram of root produced on a susceptible check. Robinson and Percival (1997) also used egg production as a criterion to evaluate 46 primitive accessions of G. hirsutum from Mexico for resistance and also counted vermiform nematodes recovered per pot at the end of the experiment. They added both numbers together and divided by the initial inoculum population deriving a multiplication factor. Later Robinson et al. (1999) used this same technique to evaluate resistance among cotton cultivars grown in the United States since 1950. Parameters related to symptoms, such as galling indices used to differentiate genotypic response to M. incognita, tend to be much less variable than measurements related to nematode population or reproduction (Zhang et al., 2006). Measurements of population are more time consuming than the observation of symptoms. Together, both these factors have hampered efforts to evaluate the G. hirsutum germplasm collection on a systematic basis. Extraction of eggs from roots also destroys the plant and makes progeny testing difficult. Also, these types of data tend to be highly variable and seasonally influenced. Robinson et al. (2004) noted a high correlation between the variance and the mean for nematode population density over several experiments for both a susceptible and a moderately resistant genotype. He also found that nematode populations tended to be higher during the summer months than during the winter. Reniform population frequency distributions tend not to be normal and are usually skewed. Log transformations tend to normalize the data (Robinson, personal communication, 2005).
Objectives of our research were to evaluate the USDA primitive G. hirsutum collection for reaction to R. reniformis and determine if a relationship exists between egg production and vermiform number as an indicator of resistance.
|
MATERIALS AND METHODS
|
|---|
Upland cotton accessions were obtained from R. G. Percival, curator of the USDA cotton collection. We evaluated germplasm from the "TX" list, which only includes genotypes that have never been U.S. cultivars. A total of 2102 accessions are listed (USDA, 2006). Of these, 14 are listed as "dead" (TX70A, TX119A, TX618, TX721, TX1104, TX1198, TX1312, TX1313, TX1428, TX1637, TX1720, TX1732, TX2101, and TX2108) bringing the total accessions available to 2088. The R. reniformis population used for screening is a combination of populations originally isolated from naturally infested cotton fields located in Alabama, Arkansas, Louisiana, and Mississippi. The nematodes were increased on Paymaster ‘PM 1218 BRR’ (PM 1218) in the greenhouse.
Accessions were screened in groups of 50. PM 1218 was included in each set as a check. It was selected as a control based on performance in limited field trials that indicate this cultivar performs well in fields infested with R. reniformis (Sciumbato et al., 2005). Thus our objective was to use a control that was not necessarily the most susceptible genotype available, but one that has a better level of tolerance or resistance than other adapted germplasm. Accessions were germinated on 26 by 39 cm germination paper for 72 h at 22°C. Many accessions had hard or impermeable seed coats and were difficult to germinate even after soaking in warm water. As a result we were not able to evaluate the full complement of 50 accessions in every set. To overcome this problem, seed coats of all accessions were cut on the cotyledonary (blunt) end before germination. As a result, germination was greatly improved. One seedling with 1 to 2 cm radicle was planted in a 150 mL cone-tainer (Ray Leach Cone-tainers, Washougal, WA; 2.5-cm diameter by 20-cm depth) filled with soil (85% sand, 9% silt, and 6% clay with 0.3% OM and pH of 6.0). The soil was sterilized by autoclaving twice at 121°C and 103.4 kPa for 2 h on two consecutive days. Each accession was inoculated with R.reniformis at a population level of 1000 juveniles and vermiform adults per 150 mL of soil. Tests were arranged on a greenhouse bench with supplemental lighting (1000-W metal halide lamps with a 15-h photoperiod) in a randomized complete block design with four replications. Pots were watered twice daily and fertilized weekly using Peter's 20–10–20 water-soluble fertilizer (Buddies Plant Food, Ballinger, TX). Sixty days after inoculation, nematodes were extracted from the soil using a modified Baermann funnel technique. The entire root system was gently removed from the cone-tainer, encased in a 30 by 30 cm Kimwipe, and positioned on a 15 by 15 cm pliable screen covering a 470-mL plastic funnel. Rubber tubing was attached to the stem of the funnel and pinched with a clamp. Water was added to the funnel covering the root system and evaporated water was replaced as needed. After 72 h, 100 mL of water was drained from the tubing and sieved through a 25-µm pore sieve to collect and count the vermiform nematodes.
Rotylenchulus reniformis eggs were extracted from the root system immediately following removal from the Baermann funnel. The soil was gently washed from the roots. Eggs were extracted by shaking the root system for 4 min in a 0.6% NaOCl solution (Hussey and Barker, 1973). The freed eggs were collected by washing through a 75-µm pore sieve nested on a 25-µm pore sieve. Eggs collected on the 25-µm pore sieve were rinsed with tap water. Eggs were stained by boiling for approximately 10 s in a 10% solution red food coloring. Samples were allowed to cool to room temperature, drained, and resuspended in 20 mL of water for counting.
As indicated in the introduction, accession means for reniform counts and associated variances are often highly correlated (Robinson et al., 2004), indicating that the underlying distribution is not a Gaussian but a Poisson distribution (Schabenberger and Pierce, 2002). Recent advances in computational ability have led to the development of software procedures that enable users to model such non-Gaussian responses in experiments with a designed structure based on generalized mixed models developed by Nelder and Wedderburn (1972). In the case of an underlying Poisson distribution, fixed effects are modeled through a natural log-linear link function but random effects are assumed to be normally distributed. We used the recently published SAS procedure GLIMMIX to analyze the data (http://support.sas.com/rnd/app/papers/glimmix.pdf; verified 23 Oct. 2006) with accessions as the fixed effect and blocks as a random effect. The initial analysis indicated an over-dispersion problem (variance rising faster than the mean), as indicated by a large ratio of the Pearson 
2 divided by degrees of freedom for error, when Poisson was used as the underlying distribution. This was addressed by using the negative binomial distribution as the underlying distribution to model fixed effects, still with a natural log link function, as well as employing the residual option to introduce a scale parameter. This resulted in ratios close to 1.0, indicating that the variability was modeled properly and there was no indication of residual overdispersion. Least squares means calculated are natural logs of the counts. Egg and vermiform least squares means were standardized by dividing each by the corresponding value for the PM 1218 check from the appropriate set. Thus all accessions were evaluated on their performance relative to the control, PM 1218, regardless of the experimental set in which they were evaluated.
A multilevel approach was used in which accessions were initially evaluated in 49 separate experiments (sets). Some accessions may have been evaluated more than once during this initial screening due to poor germination or other problems involved with getting complete replication in a set. After all accessions had been evaluated at least once in the initial screening, the two accessions with the numerically lowest ratios [loge(accession count)/loge(PM 1218 count)] for eggs and the two accessions with the numerically lowest ratios for vermiform counts were advanced to a second level of evaluation, consisting of three nonidentical sets (sets 50, 51, and 52). Methods were the same as the initial series of evaluations, except data was collected on vermiform nematodes only. Eggs were not counted, as this is a laborious procedure, and this was deemed to be no better measure of resistance than vermiform counts. The top performing accessions (relative to PM 1218) from the second evaluation were reevaluated yet a third time in two identical sets (sets 53 and 54) based on numbers of vermiform nematodes. Data from these two identical sets was combined and Dunnett's test was used to separate the backtransformed means. During the second level of evaluation, it was noticed that one accession, TX1419, was heterogeneous for pollen color and other phenotypic traits, and was thus divided into two subgenotypes, TX1419 (normal yellow pollen) and TX1419dp (dark yellow pollen).
|
RESULTS AND DISCUSSION
|
|---|
We were able to germinate and attempt to screen 2066 of the 2088 total accessions. Of these, 1973 accessions had data from at least one replication. Complete (balanced) data were obtained on 1603 accessions. Data on the remaining accessions were either incomplete (one or more missing replications) or no data were obtained either due to lack of seed germination or plant death following germination or infection and feeding by nematodes. Of the 49 sets evaluated in the initial screening, data from set 20 were discarded because of low egg and vermiform counts for the control, PM 1218 (Fig. 1
). Vermiform populations on PM 1218 were somewhat cyclic, probably due to seasonal influences, ranging from 232 (set 31) to 21 437 (set 1), excepting set 20 which had a vermiform count for PM 1218 of 52. The average vermiform number across all sets for PM 1218 was 2597. Vermiform nematode numbers for PM 1218 were consistent (Fig. 1), and nematodes were able to feed freely on the control in every experiment except set 20. Entries that were originally evaluated in set 20 were later evaluated as a group in set 45. Across all accessions, vermiform numbers ranged from 32 (set 49) to 31 801 (set 39).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1. Reniform nematode vermiform numbers (loge for PM 1218 as control) determined in 54 screening experiments. Sets 1–49 were the initial screening, sets 50–52 were the first rescreening, and sets 53 and 54 were the second rescreening. Vertical bars indicate 95% confidence intervals.
|
|
Egg numbers were greater than vermiform numbers across all experiments and all accessions (Fig. 2
). For the control, egg numbers ranged from a low of 708 (set 35) to a high of 185 561 (set 23), while egg numbers across all accessions ranged from 2 (set 28) to 648 900 (set 1). Phenotypes of accessions, when standardized to the control, were normally distributed throughout the entire G. hirsutum collection (Fig. 3
) for both egg and vermiform numbers. The majority of the accessions showed standardized values between 0.8 and 1.2 for both eggs and vermiform numbers, indicating that most of the accessions are similar to the adapted PM 1218 in their reaction to R. reniformis. Using either egg or vermiform counts, no G. hirsutum accessions were found to have levels of resistance previously reported in G. barbadense and other Gossypium spp. (Yik and Birchfield, 1984; Robinson and Percival, 1997; Robinson et al., 2004).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 2. Frequency distribution of eggs and vermiform (loge) for all accessions in original evaluations (sets 1–49).
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 3. Frequency of log values for vermiform and egg counts of accessions standardized as loge(accession count)/loge(PM 1218 count) in original evaluations (sets 1–49).
|
|
After standardization to the control, there was no direct relationship between egg counts and vermiform counts across all accessions (Fig. 4
). Although loosely correlated (r = 0.40, P < 0.001), the high amount of variability we observed for both eggs and vermiform numbers decreases the likelihood that a better correlation could ever be obtained without significantly increasing the number of replications. Although eggs can come only from vermiforms, using vermiform numbers to predict the number of eggs would not be reliable. Therefore, as a measure of resistance, there appears to be no advantage of one over the other. However, counting eggs is more time and labor intensive than counting vermiform numbers, because of the egg extraction step. Also, egg numbers tend to be higher than vermiform numbers and, because of the relationship between mean and variance, they are more variable. Therefore, once we had evaluated all the accessions initially in the first screenings (sets 1 through 49) using both egg and vermiform numbers, we discontinued measuring egg production and performed the second and third screenings based on vermiform counts only.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 4. Relationship between vermiform and egg ratios standardized as loge(accession count)/loge(PM 1218 count) for all experiments in original evaluations (sets 1–49).
|
|
Some of the observed variability within accessions may be attributable to genetic heterogeneity within accessions. Variability within the control PM 1218 was also large, however, and as a cultivar PM 1218 should be relatively homogeneous. Measuring genetic heterogeneity within these accessions was beyond the scope of this study, but Robinson and Percival (1997) previously evaluated only one plant per accession for reniform and root-knot nematode resistance based on their observation of phenotypic homogeneity within accessions.
It is worthwhile to look at the performance of the six accessions previously identified by Robinson et al. (2004) as having a moderate level of resistance to reniform nematode, based on final nematode populations in growth chamber experiments (TX25, TX1828, TX1860, TX748, TX1586, and TX2469). Only TX1828 showed a sufficient level of resistance to merit further evaluation after the initial screening, with a standardized egg ratio of 0.83 and vermiform ratio of 1.02 and ranking second within its set for egg ratio. TX1586 also had a relatively low standardized egg ratio of 0.78, but ranked 10th within its set and was not considered for rescreening. The other accessions had standardized values for egg counts ranging from 0.96 to 1.17, indicating that they were not largely different from PM 1218 in their reaction as measured by egg production. This could be partly due to differences in nematode populations used in these inoculation experiments or the use of a different control for comparison. We attempted to reevaluate TX1828 but were not able to collect data due to lack of germination or plant death following germination. Both TX1586 and TX1828 probably merit reevaluation at some point.
Eight accessions were evaluated in the final repeated evaluations (sets 53 and 54). Seven were found to have levels of nematode reproduction that were significantly lower than PM 1218 (Table 1). For Table 1, least squares means obtained by the GLIMMIX procedure were backtransformed to the original scale and 95% confidence intervals calculated. These confidence intervals generally are not symmetric around the mean as would be the case for normally distributed data. TX245, TX378, TX500, TX1419, TX1472, TX1565, and TX1765 all supported vermiform populations that were smaller than those on PM 1218. Separation of TX1419 into two groups in the latter stages of evaluation based on phenotypic differences seemed to make little difference in the performance of either type (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Vermiform counts for accessions averaged over two identical sets (four replications per set) in the final evaluation (sets 53 and 54).
|
|
Yik and Birchfield (1984) proposed arbitrary levels of resistance based on egg production per gram of root as percentage of control as: 0% = immune, 1 to 10% = highly resistant, 11 to 25% = resistant, and 26 to 40% = moderately resistant. Using percentage of vermiform numbers compared to the control and applying this scale, accessions supporting low populations in our experiments would at best be classified as only moderately resistant, falling in the general range of 20 to 35% compared to the PM 1218 control. Thus, a high level of resistance remains to be discovered among accessions of G. hirsutum germplasm.
Geographic origins of the moderately resistant accessions evaluated in the final screening are diverse (Table 2). Three originated in North America (Mexico), one from Central America (Guatemala), one from South America (Brazil), two from the Caribbean (Guadeloupe), and one from Africa (Sudan). It is possible that these accessions might carry different genes for resistance, thus allowing for recombination to occur leading to the development of genotypes with higher levels of resistance. However, the heritability and gene action of resistance in these accessions is yet to be determined. Determining the inheritance of resistance is going to be difficult due to the large amount of phenotypic variation involved in reniform nematode resistance studies.
View this table:
[in this window]
[in a new window]
|
Table 2. Race, geographic origin, and phenotypic traits for G. hirsutum accessions supporting low populations of R. reniformis.
|
|
|
CONCLUSIONS
|
|---|
We have identified seven accessions of upland cotton (TX245, TX378, TX500, TX1419, TX1472, TX1565, and TX1765) that consistently support lower populations of reniform nematode than the control, PM 1218. Nematode reproduction (egg numbers) and nematode feeding and survival (vermiform numbers) are only loosely correlated, and both have large amounts of variation proportional to their means. Measurement of egg production has the disadvantage of being destructive to the plant, while measurement of nematode numbers allows the plant to be returned to the soil and grown to maturity. This opens the possibility of using selected plants for crossing and progeny testing, necessary steps in any crop improvement program.
|
NOTES
|
|---|
Research supported in part by a grant from Cotton Inc.
Received for publication February 28, 2006.
|
REFERENCES
|
|---|
- Birchfield, W., and J.E. Jones. 1961. Distribution of the reniform nematode in relation to crop failure of cotton in Louisiana. Plant Dis. Rep. 45:671–673.
- Cook, C.G., A.F. Robinson, and L.N. Namken. 1997. Tolerance to Rotylenchulus reniformis and resistance to Meloidogyne incognita race 3 in high-yielding breeding lines of upland cotton. J. Nematol. 29:322–328.[ISI]
- Hussey, R.S., and K.R. Barker. 1973. A comparison of methods of collecting inocula of Meloidogyne spp., including a new technique. Plant Dis. Rep. 57:1025–1028.
- Jiang, C., P.W. Chee, X. Drave, P.L. Morrell, C.W. Smith, and A.H. Paterson. 1999. Multilocus interactions restrict gene introgression in interspecific populations of polyploid Gossypium (cotton). Evolution Int. J. Org. Evolution 54:798–814.
- Jones, J.E., J.P. Beasley, J.I. Dickson, and W.D. Caldwell. 1988. Registration of four cotton germplasm lines with resistance to reniform and root-knot nematodes. Crop Sci. 18:199–200.
- Jones, J.E., L.D. Newsom, and E.L. Finley. 1959. Effect of the reniform nematode on yield, plant characters, and fiber properties of upland cotton. Agron. J. 51:353–356.[Abstract/Free Full Text]
- Lawrence, G.W., and K.S. McLean. 1995. Distribution of nematodes. p. 196. In D.A. Richter and J. Armour (ed.) Proc. Beltwide Cotton Conf., Memphis, TN. 4–7 Jan. 1995. Natl. Cotton Counc., Memphis, TN.
- National Cotton Council. 2005. 2001 Reniform nematode distribution. [Online]. Available at www.cotton.org/tech/pest/nematode/distributions.cfm?bw=1&type=Reniform (verified 23 Oct. 2006). Natl. Cotton Counc., Memphis, TN.
- Nelder, J.A., and R.W.M. Wedderburn. 1972. Generalized linear models. J. Royal Stat. Soc. (A) 135:370–384.[CrossRef]
- Robinson, A.F. 1999. Cotton nematodes. p. 595–615. In C.W. Smith and J.T. Cothren (ed.) Cotton: Origin, history, technology, and production. John Wiley & Sons, New York.
- Robinson, A.F., A.C. Bridges, and A.E. Percival. 2004. New sources of resistance to the reniform (Rotylenchulus reniformis Linford and Oliveira) and root-knot (Meloidogyne incognita (Kofoid & White) Chitwood) nematode in upland (Gossypium hirsutum L.) and Sea Island (G. barbadense L.) cotton. J. Cotton Sci. 8:191–197.
- Robinson, A.F., C.G. Cook, and A.E. Percival. 1999. Resistance to Rotylenchulus reniformis and Meloidogyne incognita race 3 in the major cotton cultivars planted since 1950. Crop Sci. 39:850–858.[Abstract/Free Full Text]
- Robinson, A.F., and A.E. Percival. 1997. Resistance to Meloidogyne incognita race 3 and Rotylenchulus reniformis in wild accessions of Gossypium hirsutum and G. barbadense from Mexico. J. Nematol. 29:746–755.[ISI]
- Schabenberger, O., and F.J. Pierce. 2002. Contemporary statistical models for the plant and soil sciences. CRC Press, Boca Raton, FL.
- Sciumbato, G.L., S.R. Stetina, and L.D. Young. 2005. Tolerance of popular cotton varieties to the reniform nematode. p. 150. In P. Duggar and D. A. Richter (ed.) Proc. Beltwide Cotton Conf. New Orleans, LA. 5–7 Jan. 2005. Natl. Cotton Counc., Memphis, TN.
- Shepherd, R.L. 1974. Transgressive segregation for root-knot nematode resistance in cotton. Crop Sci. 14:872–875.[Abstract/Free Full Text]
- Smith, A.L. 1940. Distribution and relation of meadow nematode, Pratylenchus pratensis to Fusarium wilt of cotton in Georgia. Phytopathology 30:710.
- [USDA] USDA-ARS, National Genetic Resources Program. 2006. Germplasm Resources Information Network (GRIN) [Online]. Available at www.ars-grin.gov/cgi-bin/npgs/html/tax_site_acc.pl?COT%20Gossypium%20hirsutum (verified 23 Oct. 2006). Natl. Germplasm Resour. Lab., Beltsville, MD.
- Usery, S.R., K.S. Lawrence, G.W. Lawrence, and C.H. Burmester. 2005. Evaluation of cotton cultivars for resistance and tolerance to Rotylenchulus reniformis. Nematropica 35:121–134.
- Yik, C.P., and W. Birchfield. 1984. Resistant germplasm in Gossypium species and related plants to Rotylenchulus reniformis. J. Nematol. 16:146–153.
- Zhang, J., C. Waddell, C. Sengupta-Gopalan, C. Potenza, and R.G. Cantrell. 2006. Relationships between root-knot nematode resistance and plant growth in upland cotton: Galling index as a criterion. Crop Sci. 46:1581–1586.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
