© 2009 Plant Management Network.
Impact of Cropping Sequences and Alternative Hosts on Take-all Management of Winter Wheat in Arkansas
Eugene A. Milus, Richard D. Cartwright, and Craig S. Rothrock, Department of Plant Pathology, University of Arkansas, Fayetteville, AR 72701; Merle Anders, Rice Research and Extension Center, University of Arkansas, Stuttgart, AR 72160; and Nathan Slaton, Department of Crop, Soil, and Environmental Sciences, University of Arkansas, Fayetteville, AR 72701
Milus, E. A., Cartwright, R. D., Rothrock, C. S., Anders, M., and Slaton N. 2009. Impact of cropping sequences and alternative hosts on take-all management of winter wheat in Arkansas. Online. Plant Health Progress doi:10.1094/PHP-2009-0512-02-RS.
Cultural practices are the principle means for managing take-all of wheat caused by Gaeumannomyces graminis var. tritici. This research identified cropping sequences that can be used to manage take-all in Arkansas. For dryland fields where the opportunity to grow rotational crops is limited, summer fallow was the best option for managing take-all. For irrigated fields, rotation out of wheat for at least one year reduced incidence and severity of take-all, and rice was the most effective rotational crop. Summer fallow or a rice crop was more detrimental to survival of take-all inoculum compared to corn or soybean. Reductions in inoculum were associated with elevated soil temperatures during the summer in fallow fields and with soil anoxia in flooded rice fields. Managing grassy weeds is important during rotations out of wheat. Rescuegrass was the most susceptible grassy weed to G. graminis var. tritici in this study. Although Italian ryegrass was only moderately susceptible, it likely plays a major role in maintaining inoculum because of its wide distribution and large population size.
Take-all, caused by the soilborne fungus Gaeumannomyces graminis (Sacc.) Arx & D. Olivier var. tritici J. Walker (Syn. Ophiobolus graminis Sacc.), is the most important root disease of wheat (Triticum aestivum L.) worldwide (8). Symptoms occur throughout the growing season (4), but soil temperatures of 5 to 15°C are most favorable for initial infection (5). Initial symptoms include root lesions with dark brown to black vascular discoloration and associated black ectotrophic hyphae on the root surface. Symptoms are most apparent after the heading stage and appear as patches of stunted plants with few tillers, and these plants usually die prematurely giving rise to "whiteheads" that can be seen from a distance (4).
Take-all has been difficult to manage because there are no resistant cultivars or effective fungicides (13). Numerous cultural practices have been shown to reduce incidence and severity of take-all, but cultural practices that are effective in one area may not be feasible or effective in other areas. Consequently, cultural practices for managing take-all are often specific for each unique production region (8). In Arkansas, soft red winter wheat is grown on approximately 400,000 ha annually, and is the only commercially-viable cool-season crop except for a small amount of oat grown mostly for horse feed. Typically, wheat is planted in October and harvested in early June. A dryland or irrigated soybean crop can be planted immediately after wheat harvest and harvested immediately before wheat is planted again in the fall. In fields with irrigated summer crops, wheat is grown as a rotational crop after corn, cotton, soybean, or flooded rice. In the mid 1990s, take-all caused noticeable yield losses in many fields, especially in fields that had been double-cropped to wheat and soybean for several years. Wheat grown after rice was observed to be only slightly affected by take-all.
G. graminis var. tritici survives in the absence of a host as a saprophyte on previously colonized host tissue (16). Thus, crop rotations may allow time for infested crop residue to decompose. However, grassy weeds or volunteer wheat during the rotation period have been implicated in the survival of the pathogen. G. graminis var. tritici has a wide host range with at least 402 susceptible host species (11). Brooks (3) examined 15 grass species for their ability to perpetuate inoculum of G. graminis var. tritici and concluded that host susceptibility was associated with the ability to serve as an inoculum source. Cotterill and Sivasithamparam (7) determined that grass-infested legume crops increased the inoculum density and infectivity of G. graminis var. tritici and the severity of take-all in subsequent wheat crops. The effects of cropping sequences and alternative hosts on take-all in Arkansas have never been investigated, but could be very important in survival of the pathogen when wheat is not grown. The objectives of this research were to determine the effects of cropping sequences on take-all, the effects of summer crops on survival of take-all inoculum, and the relative susceptibility of local grassy weeds.
Determining Effects of Cropping Sequences on Take-all
Seven commercial fields in eastern Arkansas with a history of take-all were selected for this study. Within each field, a 60 × 60 m block was selected and located in relation to a fixed object. All sampling was done within this area to reduce the variability typically found in large fields. Each time a field was assessed, 100 samples from the block were taken on 6-m centers in a 10 × 10 grid pattern. Growers were responsible for all production decisions and inputs.
To determine the severity of take-all near wheat maturity in each field, the percentage of whiteheads was visually estimated in sample areas (approximately 0.5-m-square) and 100 clumps of wheat were dug from the same areas. Two plants were selected at random from each clump, and roots were washed free of soil and rated for take-all severity on Shipton’s (15) 0 to 4 scale (0 = no lesions, 1 = lesions on < 25%, 2 = lesions on 25 to 50%, 3 = lesions on 51 to 75%, and 4 = lesions on > 75% of roots). Data for whiteheads and root ratings were plotted in bar charts to show the percentage of sample areas and plants, respectively, in various classes.
During the 3-year study, take-all severity was monitored in the following fields: (A) one field with wheat/irrigated soybean double-cropping every year; (B) and (C) two dryland fields with wheat/summer fallow every year; (D, E, and F) three fields with double-cropped wheat/irrigated soybean in 1998 and 2000, and flooded rice in 1999; and (G) one field double-cropped to wheat/non-irrigated soybean in 1998 and 2000, and non-irrigated soybean in 1999. Take-all severity increased during the study in the field with wheat/irrigated soybean double-cropping every year, whereas severity decreased in the two dryland fields with wheat/summer fallow. Take-all severity decreased dramatically from the first to second wheat crops in the three fields with double-cropped wheat/irrigated soybean in 1998 and 2000 and flooded rice in 1999. Take-all severity as measured by whiteheads decreased in the field with double-crop wheat/non-irrigated soybean in 1998 and 2000 and non-irrigated soybean in 1999, but almost half of the plants still had root lesions (Figs. 1 and 2).
Determining Susceptibility of Grassy Weeds
Seeds of cool and warm-season grasses in and around wheat fields were collected in May and September 1999, respectively, and planted in growth chamber tests. Inoculum of G. graminis var. tritici was grown on autoclaved oat grain (13), air-dried, and coarsely ground in a blender. The fraction passing through a 2.83-mm sieve and retained on a 0.71-mm sieve was used. Two grams of prepared inoculum were mixed with 150 cc of blasting sand and placed in an 8-cm-square pot. Seeds of the grasses and a wheat check (Pioneer cultivar 2548) were placed on the surface of the infested sand. Depending on the size of the seeds, seeds were covered with 2 to 5 mm of sand. Cool-season grasses were evaluated after incubation for 4 week at 18°C. Warm season grasses were germinated at 22°C for 2 week because they did not germinate well at 18°C and then incubated for 4 week at 18°C. Plants were grown using a 14-h photoperiod and were fertilized once after emergence using Peter’s 20-20-20 (0.5 g/liter). The design of each experiment was a randomized complete block with three replications, and the experiments were done twice. At the end of the incubation period, roots of the seedlings were washed free of sand and examined using a dissecting microscope. Roots from 10 random seedlings per pot were rated for take-all symptoms on the 0 to 4 scale, and a weighted average severity was calculated. The weighted average severity was calculated by multiplying the number of root systems in each category by the value of that category and dividing the sum by the total number of root systems. Also, roots of the 10 seedlings were rated for colonization by ectotropic hyphae (0 = none, 1 = light, 2 = moderate, and 3 = heavy), and values for the 10 seedlings in each pot were averaged before analysis using SAS proc GLM (SAS Institute Inc., Cary, NC).
Of the cool-season and warm-season grassy weeds tested, rescuegrass and prairie wedgegrass had root lesion ratings that were not significantly different from wheat (Table 1). Cheatgrass, Italian ryegrass, and giant foxtail had intermediate ratings. As evidenced by ectotrophic hyphae of G. graminis var. tritici, these grasses plus Japanese brome were colonized to a level similar to that of wheat. The other grasses had low ratings for root lesions and the presence of ectotrophic hyphae, but all grasses supported some colonization by G. graminis var. tritici.
Table 1. Susceptibility of inoculated, cool-season and warm-season grassy weeds to Gaeumannomyces graminis var. tritici in growth chamber experiments as measured by ratings for root lesions and ectotrophic hyphae.
x 0 = none, 1 = < 25%, 2 = 26 to 50%, 3 = 51 to 75%, and 4 = >75% of the roots with take-all lesions.
y 0 = none, 1 = light, 2 = moderate, and 3 = heavy colonization of roots by ectotrophic hyphae.
z Values within a column of cool-season or warm-season grasses followed by the same letter(s) are not significantly different by a LSD test at P = 0.05.
Effect of Summer Crop on Inoculum Survival
To obtain natural inoculum of G. graminis var. tritici for the experiment, crowns of plants killed by take-all were collected from Arkansas during late May 2000 and from Oklahoma during late May 2001. Soil was removed, and stems were cut off at the crown. Crowns were air-dried for 16 h in an air-conditioned room at 20°C, after which the remaining soil was removed. To determine the effect of ambient field conditions on survival of inoculum in each crop, the crowns were placed in sample holders constructed of 15.2-cm-diameter PVC pipe cut to 5-cm lengths and covered with 1.4-mm-mesh nylon screen glued over each end. Thirty grams of infested crowns were placed in each sample holder, and the remaining space was packed with sieved (3.36-mm mesh) silt loam soil that had been air-dried for approximately 20 h.
To simulate summer conditions under various summer cropping systems, sample holders were buried in field plots that were part of a cropping systems study initiated during 1999 at the Rice Research and Extension Center, Stuttgart, AR. Plots managed as no-till fallow, tilled fallow, flooded rice, irrigated corn, and irrigated soybean were selected, and each treatment was replicated in two plots. Crops were managed using standard management and fertility practices (1). On 2 June 2000 and 5 June 2001, three sample holders and a temperature recorder were buried in each of the ten plots with the top of the sample holder 5 cm below the soil surface. In plots with a crop, the sample holders and temperature recorder were buried between rows. The corn, rice and soybean crops were at the 10-leaf, 5-leaf (first tiller), and emerged growth stages, respectively in 2000, and at the 10-leaf, 5-leaf, and 4-leaf growth stages, respectively in 2001. As a reference standard, three sample holders and a temperature recorder were buried in a tub of air-dried silt loam soil and kept in an air-conditioned room at 18°C. One sample holder was recovered from each plot after 60 days, and the remaining two sample holders were recovered after 90 days.
To determine the effects of each cropping system on inoculum survival, inoculum was recovered from each sample holder by sieving. All plant material retained on a 3.36-mm mesh sieve and not readily identified as fresh roots from a summer crop was considered as recovered inoculum. Soil was removed and inoculum was air-dried as described above, and the inoculum was weighed. For samples taken from field plots after 90 days, the weights of inoculum recovered from the two sample holders per plot were averaged, and the inoculum from both sample holders was combined for bioassays.
To determine the ability of the recovered inoculum to cause disease, it was coarsely ground in a coffee grinder, and the fraction passing through a 2.83 mm-sieve and retained by a 0.71-mm sieve was used. Inoculum from each field plot in 2000 and 2001 was tested in a bioassay at rates of 20 and 40 g/m², and inoculum for the 90-day assay in 2001 was tested at 40 and 80 g/m². To conduct the bioassay, 12,000 cc of vermiculite moistened with 3 liters of tap water was placed in plastic tubs (30 cm × 40 cm × 18 cm deep) with drain holes in the bottom. Each tub was divided in half with a sheet of acetate in order to accommodate two treatments per tub. Inoculum aliquots of 1, 2, and 4 g (corresponding to 20, 40 and 80 g/m², respectively) from each field plot was mixed with 1,750 cc of blasting sand, moistened with 0.25 liter of tap water, and placed in a 2.5-cm deep layer on top of the vermiculite in one half of a tub. One hundred wheat seeds (cv. Pocahontas) were planted 1 cm deep on a uniform grid spacing in each half of the tub. Seeded tubs were covered loosely with a piece of aluminum foil to retain moisture, and incubated in a dark growth chamber at 18°C. After 5 days, the foil was removed, and lights were programmed for a 14-h photoperiod. Plants were fertilized 7 days after planting using Peter’s 20-20-20 (0.5 g/liter).
To evaluate the bioassay, seedlings were removed from tubs 4 week after planting, and roots were washed free of sand and vermiculite. The root system of each seedling was rated for take-all severity on the 0 to 4 scale using a dissecting microscope. A sample of lesions was plated on the GGT3 selective medium (9) to verify that the lesions were caused by G. graminis var. tritici. A weighted average severity was calculated for the seedlings to determine the inoculum potential. The inoculum potential of the crown material from each plot was calculated by multiplying the weight of recovered inoculum by the weighted average severity.
Soil temperatures at a depth of 5 cm during summer months in tilled fallow and no-till fallow plots were similar and were higher than soil temperatures under corn, soybean or rice crops during 2000 and 2001 (Fig. 3). In 2000, soil temperatures were ≥ 35°C for 6 or more hours on 33 days in tilled fallow plots and on 19 days in the no-till fallow plots but not on any days in plots planted to soybean, corn, or rice. In 2001, soil temperatures were ≥ 35°C for 6 or more hours on 54 days in tilled fallow plots and on 59 days in the no-till fallow plots but not on any days in plots planted to soybean, corn, or rice. Precipitation from 1 June to 15 September totaled 148 mm with 7 days receiving ≥ 10 mm in 2000 and totaled 197 mm with 12 days receiving ≥ 10 mm in 2001.
In 2000, an average of 70% of the original crown weight across treatments was lost to decomposition after 60 days, but there were no significant differences among treatments (Table 2). After 90 days, crown weights in the two fallow treatments decreased to approximately 50% of their weights after 60 days, but crown weights for the other treatments were similar to their weights after 60 days. Crown weights were highest in the control and rice treatments, intermediate in the corn treatment, and lowest in the soybean and two fallow treatments. Infectivity of inoculum in bioassays of recovered crowns and inoculum potential at 60 and 90 days was consistently high in the control, intermediate in corn and soybean, and low in the rice, tilled fallow, and no-till fallow treatments. Of 24 root lesions from seedlings in the bioassay that were plated on GGT3 medium, G. graminis var. tritici was isolated from 20 of them, indicating that visual evaluations using a dissecting microscope accurately identified take-all lesions.
Table 2. Weight of crown tissue recovered after 60 and 90 days, take-all severity caused by standardized amounts of the crown tissue when used as inoculum in bioassays, and inoculum potential after being buried in soil under different cropping systems at Stuttgart, AR, in early June 2000 and 2001.
In 2001, there was less decomposition of crowns at both 60 and 90 days than in 2000 (Table 2). Crown weight at 60 days was reduced by decomposition in all treatments except rice. More decomposition occurred in the soybean, tilled fallow, and no-till fallow treatments than in the control and corn treatments. At 90 days, there were no significant differences (P = 0.05) among treatments for weight of recovered crowns, but the control and rice treatments had the highest weights as in 2000. The ability of inoculum from all treatments to cause take-all at both 60 and 90 days was much less than in 2000, and there were no differences among treatments for infectivity or inoculum potential for the 90-day sample. The low levels of take-all in 2001 may be a result of inoculum originating from severely diseased plants with crowns in a more advanced stage of decomposition which allowed only limited saprophytic survival of the pathogen in soil. This explanation is supported by results of Rothrock and Cunfer (14) who reported that take-all severity in the field was lower following wheat that was killed by take-all early in the previous season than following less susceptible small grains that were not killed by take-all.
Conclusions and Recommendations
The results of this study indicate that take-all can be managed successfully under both irrigated and dryland cropping systems in Arkansas. For irrigated fields, rotation out of wheat for at least one year was the best management option. Rice was the most effective crop for reducing inoculum potential in the survival study, and incidence and severity of take-all was greatly reduced in the three wheat fields that followed rice. These results support observations in China that take-all declines where wheat is grown following rice (Don Huber, personal communication). Soil oxygen levels are low in flooded rice fields (12), and this likely contributed to the slow decomposition of wheat residue and low survival of inoculum buried in rice fields. Although soybean and corn crops reduced inoculum potential compared to the control in the survival study, these rotational crops had inoculum potentials that were more than 100 times greater than for rice.
For non-irrigated fields, continuous wheat with summer fallow was the best management option. Take-all severity measured by both percentage of whiteheads and root ratings declined during the course of this study in both fields with this cropping system. Summer-fallowed soils were hotter than soils cropped to corn, rice, or soybean, and high soil temperatures with intermittent precipitation for extended periods of time likely were responsible for the decline in inoculum potential. These results agreed with those of Cotterill and Sivasithamparam (6) who concluded that both soil moisture and high temperature were needed to inactivate take-all inoculum because inoculum survived all temperature treatments if the soil was dry. These results also agree with those of Bockus et al. (2) who concluded that thermal inactivation of inoculum in summer-fallowed soil was an important component of take-all management in Kansas, where inoculum survived better and take-all was more severe if soil was shaded by crop debris or summer crops. They concluded that temperatures ≥ 35°C for 6 h on 12 days was sufficient to inactivate take-all inoculum. The two fallow treatments in this study averaged 26 and 56 days with temperatures ≥ 35°C for at least 6 h in 2000 and 2001, respectively. Summer fallowing should be very effective in Arkansas because of the high number of days with soil temperatures above 35°C for at least 6 h.
The success of any rotation for managing take-all likely will depend on eliminating volunteer wheat and grassy weeds during winters out of wheat production because inoculum was able to survive the period when rotational crops were grown. Environmental conditions from October through May are likely to be most favorable for infection and colonization by G. graminis var. tritici. Large differences in susceptibility of local grassy weeds were found in this study, but most grassy weeds appear able to assist the survival of the pathogen. Although Italian ryegrass was only intermediate in susceptibility to take-all, it is the most important weed problem in wheat in Arkansas. This weed can be found in most fields that are cropped to wheat and can be at high densities in many of these fields. Italian ryegrass also was implicated as an important alternative host in Great Britain (3) and Brazil (10), and it likely is the most important grassy weed for survival of G. graminis var. tritici in Arkansas because of its wide distribution and large population size. All of the cool-season grasses tested in this study were reported by Nilsson (11) as hosts for G. graminis var. tritici.
Of the warm-season grasses tested, yellow foxtail was reported as a host, barnyardgrass and large crabgrass were listed as non-hosts, and the others were not listed (11). All of the tested warm-season grasses supported some colonization, and giant foxtail appeared to be more susceptible than others. However, these warm-season grasses likely are less important alternative hosts than the cool-season grasses because of their lower susceptibility and because soil temperature during most of their growing season is warmer than optimal for infection by G. graminis var. tritici.
We thank Dennis Chivers, John Dozier, Billi Fletcher, Dennis Haigwood, Jack Hardin (deceased), and Mike Weaver for allowing this research to be conducted on their farms; Mark Browning, Randy Chlapecka, Lazaro English, Quinton Hornsby, Charlie Parsons, Jeremy Ross, and Brannon Thiesse for assistance with the field plots; Amal deSilva, Peter Rohman, Chris Weight, and Scott Winters for technical assistance; Dr. Dick Oliver for weed identification; Marci Milus for manuscript preparation; and the Arkansas Wheat Promotion Board for financial support.
1. Anders, M. M, Windham, T. E., Moldenhauer, K. A. K., Bacon, R. K., McNew, R. W., Cartwright, R. D., and Gibbons, J. W. 2002. Helping Arkansas rice farmers exploit market opportunities by improved use of soybean, wheat, and corn in rice rotations. Pages 165-172 in: B.R. Wells Rice Research Studies 2001. R. J. Norman and J.-F. Meullenet, eds. Univ. of Arkansas Agr. Exp. Station, Fayetteville, AR.
2. Bockus, W. W., Davis, M. A., and Norman, B. L. 1994. Effect of soil shading by surface residues during summer fallow on take-all of winter wheat. Plant Dis. 78:50-54.
3. Brooks, D. H. 1965. Wild and cultivated grasses as carriers of the take-all fungus (Ophiobolus graminis). Ann. Appl. Biol 55:307-316.
4. Clarkson, J. D. S., and Polley, R. W. 1981. Diagnosis, assessment, crop-loss appraisal and forecasting. Pages 251-269 in: Biology and Control of Take-all. M. J. C. Asher and P. J. Shipton, eds. Academic Press, New York.
5. Cook, R. J. 1981. The effect of soil reaction and physical conditions. Pages 343-352 in: Biology and Control of Take-all. M. J. C. Asher and P. J. Shipton, eds. Academic Press, New York, NY.
6. Cotterill, P. J., and Sivasithamparam, K. 1987. Intermittent wetting of soils at high temperature reduces survival of the take-all fungus. Plant Soil 103:289-291.
7. Cotterill, P. J., and Sivasithamparam, K. 1988. Importance of the proportion of grassy weeds within legume crops in the perpetuation of Gaeumannomyes graminis var. tritici. Plant Pathol. 37:337-343.
8. Hornby, D. 1998. Take-all Disease of Cereals: A Regional Perspective. CAB Int'l., Cambridge, UK.
9. Juhnke, M. E., Mathre, D. E., and Sands, D. C. 1984. A selective medium for Gaeumannomyes graminis var. tritici. Plant Dis. 68:233-236.
10. Monterroso, L., and Juan, V. 2002. Importancia de la maleza Lolium multiflorum (L.) Como fuente de inoculo del pietin del trigo en rotacion con avena. Fitopathologia 37:149-155.
11. Nilsson, H. E. 1969. Studies of root and foot rot diseases of cereals and grasses. I. On resistance to Ophiobolus graminis Sacc. Ann. Agr. Coll. Swed. 35:275-807.
12. Ponnamperuma, F. N. 1972. The chemistry of submerged soils. Adv. Agron. 24:29-96.
13. Rothrock, C. S. 1988. Effect of chemical and biological treatments on take-all of winter wheat. Crop Prot. 7:20-24.
14. Rothrock, C. S., and Cunfer, B. M. 1991. Influence of small grain rotations on take-all in a subsequent wheat crop. Plant Dis. 75:1050-1052.
15. Shipton, P. J. 1972. Take-all in spring-sown cereals under continuous cultivation: disease progress and decline in relation to crop succession and nitrogen. Ann. Appl. Biol. 71:33-46 .
16. Shipton, P. J. 1981. Saprophytic survival between susceptible crops. Pages 295-316 in: Biology and Control of Take-all. M. J. C. Asher and P. J. Shipton, eds. Academic Press, New York, NY.