Crop Science 40:1079-1087 (2000)
© 2000 Crop Science Society of America
CROP ECOLOGY, PRODUCTION & MANAGEMENT
Influence of Paired-Row Spacing and Fertilizer Placement on Yield and Root Diseases of Direct-Seeded Wheat
R.James Cooka,
Bonnie H. Ownleyb,
Hao Zhangc and
Dan Vakochc
a Dep. of Plant Pathology, Washington State Univ., P.O. Box 646430, Pullman, WA 99164-6430 USA
b Entomology and Plant Pathology Dep., Institute of Agriculture, The Univ. of Tennessee, Knoxville, TN 37996 USA
c Program in Statistics, Washington State Univ., Pullman, WA 99164 USA
rjcook{at}wsu.edu
 |
ABSTRACT
|
|---|
Root diseases, namely take-all caused by Gaeumannomyces graminis (Sacc.) Arx & D. Olivier var. tritici J. Walker, Rhizoctonia root rot caused by R. solani Kühn AG8, and Pythium root rot caused by several Pythium Pringsh. species, become yield-limiting to wheat (Triticum aestivum L.) and barley (Hordeum vulgare L.) in the Inland Pacific Northwest when these crops are grown without adequate rotation, especially in direct-seed (no-till) systems. We tested whether yields of wheat could be increased by (i) pairing the rows rather than spacing them uniformly, so as to keep the crop canopy open longer into the growing season, and (ii) placing fertilizer within rather than between the rows, so as to make nutrients more accessible to diseased roots. All experiments were done in fields with a recent history of intensive wheat. Direct seeding was used to further intensify pressure from root diseases, and soil fumigation and/or stubble burning were used in some experiments to reduce pressure from root diseases. Mixed effect models were used to analyze the yield results, with location (observation) treated as the random effect. When all yield data for four winter wheat experiments were used, and spacing, fumigation, and their interaction were the fixed effects in the model, spacing and fumigation but not their interaction were significant. When the yield data were limited to fertilizer placed within each row (directly beneath the seed), spacing, fumigation, and their interaction all were significant. However, when the yield data were limited to fertilizer placed between the rows, or to yields from plots without straw on the soil surface, neither spacing, fumigation, nor their interaction were significant. Similar results were obtained in a spring wheat experiment where fumigation, row spacing, and fertilizer placement were tested in all combinations. Moreover, the amount of Rhizoctonia root rot, take-all, or both were less in response to paired-row spacing in two winter wheat experiments. The higher yields can be explained by the microenvironmental benefits of a more open canopy with paired rows together with the concentration of soil disturbance and row cleaning with paired shanks (presumably resulting in more warming and drying of the top soil where the pathogens are most active), and placement of fertilizer directly beneath each seed row, all of which have the potential to reduce the effects of root diseases.
Abbreviations: PCFS, Palouse Conservation Field Station
|
INTRODUCTION
|
|---|
ROOT DISEASES, namely take-all caused by Gaeumannomyces graminis var. tritici, Rhizoctonia root rot caused by R. solani AG8, and Pythium root rot caused by several Pythium species, have become major limiting factors to yields of wheat and barley in the Inland Northwest (Cook, 1992; Cook and Haglund, 1991; Cook and Veseth, 1991; Cook et al., 1987). The climate of this region, being suited almost exclusively to cool-season crops, is ideal for wheat and barley but not for many broadleaf crops needed to diversify and lengthen the rotation. About half of the annually cropped area (annual precipitation >400–450 mm) is planted every third or fourth year to a cool-season pulse crop, canola (Brassia napus L. or B. rapa L.), but even these rotations include wheat or barley 2 in every 3 or 3 in every 4 yr. Results with soil fumigation used as a research tool showed that the average yield response to fumigation was 70, 22, and 7%, respectively, for winter wheat grown in the same field every year, every other year, or every third year (Cook, 1990; Cook and Veseth, 1991). The drier parts of the region are managed in a winter wheat–fallow or winter wheat, spring cereal, fallow rotation, but the trend is away from fallow because of its contribution to erosion (Papendick, 1998) and because of the inefficiency of producing only one crop every 2 yr. Direct seeding (no-till) is making it possible to extend annual cropping into the drier areas (Schillinger et al., 1999), but seeding wheat directly into wheat or barley stubble has also been shown to exacerbate each of take-all (Moore and Cook, 1984), Rhizoctonia root rot (Pumphrey et al., 1987; Weller, 1988), and Pythium root rot (Cook et al., 1980). No useful sources of genetic resistance to these diseases have been available to the wheat and barley breeding programs, and seed-treatment chemicals provide protection of germinating seeds but little or no protection of the roots (Cook and Veseth, 1991). Take-all declines in severity after many years of monoculture wheat (Raaijmakers et al., 1997), but the rhizosphere microorganisms responsible for this natural biological control are apparently specific for take-all and provide no known control of either Rhizoctonia or Pythium root rot (Weller, 1988). Stubble burning can provide significant control of these diseases (Cook and Haglund, 1991), apparently because a bare-black soil surface favors greater warming and drying of the top few centimeters of soil where the pathogens are most active. However, stubble burning is not a long-term option for this region, both because of its detrimental effects on soil quality (Cook and Veseth, 1991; Rasmussen et al., 1989), and societal concerns for its effect on air quality.
The best tool for management of Rhizoctonia root rot, and possibly other root diseases of wheat and barley seeded directly into stubble, is to spray volunteer cereals and grass weeds with a burn-down herbicide well in advance of planting the next crop and while the plants are still small (Roget et al., 1987; Smiley et al., 1992). This can greatly limit the amount of inoculum of the pathogens available to infect the next crop. Early elimination of the green bridge, as this practice is known among growers, has greatly increased the yields of direct-seeded cereals in the Inland Northwest, especially direct-seeded spring cereals. The 7- to 8-mo break between harvest in early August and planting a spring cereal the following March or April can provide a significant rotation benefit provided the field is also kept clean of volunteer and grass weeds. Rotations to spring cereals also provide one of the best tools for management of winter annual grasses such as downy brome (Bromus tectorum L.) and jointed goat grass (Aegilops cylindrica Host.) in this region (Young et al., 1994). However, green bridge management provides little control of root disease of winter wheat planted directly into the stubble of a winter or spring cereal, since the pathogen inoculum produced on volunteer cereals and grass weeds is minor compared with that present in the fresh crown and root tissue left in the soil from the crop harvested only 2 to 3 mo earlier. Moreover, growers cannot achieve effective green bridge control for spring cereals in years when the volunteer develops too late to be sprayed in the fall or not enough time can be allowed between spraying and planting in the spring. Other management techniques are needed to reduce the effects of these diseases while permitting intensive cereals and direct seeding and without depending on stubble burning.
Among the many experimental and commercial drills tested, modified, or abandoned for direct seeding in the Inland Northwest, one locally developed drill known as the Yielder (Yielder Co., Spokane, WA) has proved particularly successful on the basis of its ability to plant through heavy residue, track properly on the steep hillsides characteristic of much of the Inland Northwest, and produce consistently good yields (Papendick, 1984). A unique feature of this drill is its paired-row configuration, whereby rows are planted in pairs spaced 100 to 120 mm apart by means of heavy-duty, double-disk openers staggered one in front of the other, with nitrogen, phosphorus, and sulfur (NPS) fertilizer injected as a deep band through another heavy-duty, double disk unit positioned between each pair of seed openers. This configuration leaves a 300- to 450-mm space between the pairs of rows, depending on specifications or modifications made by the owner. Planting configurations that result in a more open canopy structure have been shown to reduce leaf diseases (Blad et al., 1978; English et al., 1989), but whether the more open canopy created by a paired-row configuration can increase yields through control of root diseases has not been tested. The Yielder paired-row design was developed to reduce the number of deep-banding units while still placing fertilizer below and no more than 50 to 60 mm to one side of each seed row. Roget et al. (1996) showed that the tillage provided by a tine in front of each opener provided significant control of Rhizoctonia root rot of direct-seeded wheat in Australia. This raises the possibility that the shanks or double-disks used for deep-banding fertilizer might also provide the kind of soil disturbance shown by Roget et al. (1996) to help in the management of Rhizoctonia root rot. Fertilizer placement within rather than between the rows could be useful if this increased access of diseased roots to plant nutrients (Cook and Veseth, 1991). This study was undertaken to test the effects of a paired-row configuration and position of the fertilizer shanks on yields of direct-seeded wheat in fields where yields are limited by root diseases.
|
Materials and methods
|
|---|
Locations and Designs of Field Experiments
The sites were in fields with a recent history of either continuous wheat or wheat-lentil (Lens culinaris Medik) (2-yr) rotation (Table 1)
, to assure adequate pressure from root diseases. The locations and field histories were (i) the Washington State University Dry Land Research Station near Lind, WA, Adams Co. (244 mm annual precipitation), on a site cropped annually for 19 consecutive years to sprinkler-irrigated wheat; (ii) a commercial field west of Pendleton, OR, Umatilla Co. (300 mm annual precipitation), on a site cropped for eight consecutive years to sprinkler-irrigated wheat; (iii) a field west of Colfax, WA, Whitman Co. (450 mm annual precipitation), on a site cropped without irrigation (dryland) for eight consecutive years to wheat; (iv) the Palouse Conservation Field Station (PCFS) near Pullman, WA, Whitman Co. (500 mm annual precipitation), on a site cropped for eight consecutive years to dryland direct-seeded wheat; and (v) a site near Fairfield, WA, Spokane County (550 mm annual precipitation), cropped in a 2-yr dryland winter wheat–spring lentil rotation for 4 yr, following several years of Kentucky bluegrass (Poa pratensis L.). The experiments on these five sites were carried out during the years of 1987–1993 as part of a larger research program on direct-seed cropping systems. The Fairfield, Pendleton, and Colfax sites were on commercial farms in cooperation with the growers, who were also the owners.
View this table:
[in this window]
[in a new window]
|
Table 1 Locations, cropping history, years of planting, and treatments used to investigate the influence of paired-row spacing and fertilizer placement on yields of intensively cropping wheat
|
|
Winter wheat yields at Lind, Pendleton, and Colfax the year before our study were between 5.5 and 7.0 Mg ha-1. Since the wheat variety in each case was a semi-dwarf, with a harvest index approaching 0.5, each site presented a layer of straw and standing stubble estimated at up to 7.0 Mg ha-1. Planting directly into and through this much fresh straw is possibly the single greatest challenge for direct seeding wheat after wheat in this region. This problem was solved at both the Pendleton and Colfax sites where the cooperating growers burned the residue before we direct-seeded winter wheat. At Lind, we used two shallow passes with an off-set disk, which left about half the residue on the soil surface, and then planted winter wheat. The Pullman site was less problematic, since (i) this site was planted to spring wheat the previous year (which produces less residue) and (ii) it was not direct seeded until the following spring (to spring wheat), which provided more time for the surface residue to become brittle and manageable. The Fairfield site where lentils had been grown the previous year was nearly bare, because of the sparse amount of residue left by this cool-season pulse crop. We added wheat straw (an estimated 4–5 Mg ha-1) to 2.4- by 8-m plots at both the Colfax (burned) and Fairfield (lentil) sites, immediately after planting, so as to test the effects of paired-row spacing both with and without straw on the soil surface at these two locations.
Planting and Harvest
All experiments were planted with a custom-built, 2.4-m-wide plot drill equipped with a Winterstiger cone (Salt Lake City, UT) driven by a gear box so as to deliver a predetermined quantity of seed equally to eight Acra-Plant openers (Garden City, KS) with eight 2-cm-wide steel-tipped fertilizer shanks in front of the openers to inject liquid nitrogen, phosphate, and sulfate (NPS) at the time of planting (Cook and Haglund, 1991). The openers were fastened to one tool bar at the rear of the drill so as to plant eight rows spaced either uniformly at 30 cm apart (30/30), or as four pairs of rows with 18 cm between the paired rows and 42 cm between the pairs of rows (18/42). Eighteen centimeters was the narrowest spacing allowed by this equipment. Four fertilizer shanks were fastened on the front and four on the middle tool bar so they could be staggered to facilitate straw clearance at the time of planting. The shanks were kept at a constant depth but moved horizontally so as to provide fertilizer 5 to 6 cm below the depth of seed either within each seed row (within) or between two rows (between). The flow-rate of liquid NPS was doubled (2x) when one fertilizer band was shared by two rows. Thus, regardless of row spacing and fertilizer placement, the depth and rate of seeding and fertilizer application remained constant across all treatments and sites.
For the first two experiments at Lind and Fairfield, we had a shank within each uniformly spaced row and between every pair of rows. For the experiment at Pendleton, we had a shank with each row whether spaced uniformly or paired. For the experiments at Colfax and Pullman, we used all combinations of between and within the uniformly spaced and paired rows.
All plots were eight rows wide and 8 m long, and fertilized with NPS applied as a solution of urea-ammonia nitrate (320 g N kg-1) plus ammonium phosphate and ammonium thiosulfate based on a soil test and yield goal for each site (Cook and Haglund, 1991; Cook and Veseth, 1991). The actual rates of N ranged between 90 and 150 kg ha-1, with P at 20 or 40 kg ha-1, and S at 10 kg ha-1. Weed control was maintained with commercially available herbicides applied pre-plant [glyphosate; N-(phosphonomethyl) glycine] and post-plant [for wild oats (diclofop; methyl 2-[4-(2,4-dichlorophenoxy)phenoxy]propanoate) and broadleaf weeds [thifensulfuron + 2,4-D; (4-chloro-2-methylphenoxy)acetic acid and (2,4-dichlorophenoxy)acetic acid, respectively] according to accepted agronomic practices and recommended rates. Yields were determined by harvesting the two center pairs of rows in paired-row plots (with one pair of rows left on each side for borders) or five center rows from each uniform-row plot (with one row on one side and two rows on the other side left as borders). The four winter wheat experiments were replicated four times, whereas the one spring wheat experiment was replicated only three times because of the limited area for this experiment.
Assessment of Wheat Root Diseases
Fumigation treatments were applied once, just prior to planting, at the Lind, Colfax, and Fairfield sites, and twice, just prior to planting (1990) and 1 yr earlier (1989), at the PCFS site. The PCFS site with plots fumigated in the previous year had been cropped to spring wheat during 1989, with the fumigated plots carefully marked so they could be precisely relocated and replanted without refumigation in the year of our study. The method of fumigation was previously described (Cook and Haglund, 1991). Briefly, each area selected was fumigated with methyl bromide applied at 50 g m-2 under clear, 4- or 6-mil plastic tarp just before planting. The edges of plastic tarp were sealed in a shallow trench dug by hand around the perimeter of each area. Fumigated plots were 8 m long, to accommodate the full length of one plot when planted and, as multiples 2.4 m wide, to accommodate full eight-row drill strips, depending on the design of the experiment. Plots were planted 2 to 3 d after fumigation.
The severity of take-all and Rhizoctonia root rot on wheat was assessed at the Lind and Pendleton sites. At the Lind site, winter wheat stubbles with accompanying roots were removed to 12- to 15-cm depth from random locations (representing five 15-cm lengths of row) in each plot immediately after harvest, bulked for each plot, and the roots washed free of soil. Stubbles representing individual plants were separated by gently pulling them apart; the subcrown internodes were used to identify individual plants. The roots representing 25 plants picked randomly from the bulked samples from each plot were then scored on a scale of 0 to 5 for take-all, with 1 representing one or two diseased roots (<10%), 2 representing up to 25% of the roots with take-all, 3 representing up to 50% of the roots with take-all, 4 representing up to 75% of the roots with take-all, and 5 representing >75% of the roots with take-all. The original bulked sample of stubble was then remixed and 25 were again chosen randomly and the roots scored on a scale of 0 to 5 for Rhizoctonia root rot with the same scale. Take-all lesions are black and typically cover several centimeters of root length, whereas Rhizoctonia lesions are brown and girdle or pinch-off the root.
At the Pendleton site, plants were dug from each of five random locations (15-cm length of row per sample) in the border rows of each plot on 14 May (late tillering), 29 May (early stem extension), and 25 June (early dough). Plants from the five locations in each plot were bulked, washed free of soil, and then individual plants were selected randomly for the assessment. The total number of roots and the numbers of roots with take-all and Rhizoctonia root rot were counted for seminal roots on each mainstem and crown roots on each mainstem and tiller. These numbers were then used to calculate the percentage of seminal and crown roots with each of the two root diseases per plant. In addition, adult plants were rated on 2 July for incidence of white heads (plants ripened prematurely because of take-all) on a scale of 1 to 5, with 1 representing a trace to 1%, 2 representing 1 to 5%, 3 representing 5 to 10%, 4 representing 10 to 25%, and 5 representing >25%.
Analysis of Data
Fumigation was used for main plots at the Lind, Colfax, Fairfield, and PCFS sites, with straw added back to the soil surface, row spacing, and fertilizer placement as subplots in factorial experiments (Table 1). Mixed effect models were then used to analyze the wheat yields with location (observation) treated as the random effect (Hocking, 1996). By treating locations as random samples from all possible samples (locations), conclusions from the mixed effects models are not restricted to the fixed locations but rather, apply to all locations across the region. The mixed model analysis allows us to look across a progression of independent experiments and draw conclusions not possible by analyzing any single experiment by itself.
The experiment with spring wheat (PCFS) was examined as a 2 by 2 by 3 factorial in a split-plot design using the GLM Procedure of PC SAS (SAS Institute Inc., 1996) with two row-spacings (paired and uniform), two fertilizer shank positions (within and between rows), and three soil treatments [fumigated with methyl bromide the year of planting (1990), the previous year (1989), and not fumigated]. The row-spacing and shank-position factors were main plots and soil treatments were subplots.
Data on the incidence of roots with take-all and Rhizoctonia root rot and incidence of white heads caused by take-all likewise were analyzed with the GLM Procedure of PC SAS. Following significant F tests, main effects and interactions were further examined with Least Significant Difference (LSD) tests.
|
Results
|
|---|
Winter Wheat Yields
Row spacing and fumigation both were significant when all data for winter wheat yields were used, and spacing, fumigation, and their interaction were the fixed effects in the model (Table 2 ; Fig. 1A)
. Row spacing, fumigation, and their interaction all were significant when the data were limited to cases where fertilizer placement was within the row (Table 2; Fig. 2A)
. In contrast, neither spacing, fumigation, nor the interaction were significant when the data were limited to cases where fertilizer placement was between the rows (Table 2; Fig. 3A)
. Likewise, neither spacing, fumigation, nor the interaction were significant when the analysis was limited to yield data from plots without straw, either because of burning (without returning straw) or left naturally bare by the lentil residue (Table 2; Fig. 1C). Spacing and fumigation had a significant effect on yield only when fertilizer was placed below the seed and within the row (Fig. 2) and wheat straw was present on the soil surface (Table 2; Fig. 1B).
View this table:
[in this window]
[in a new window]
|
Table 2 Level of significance (P-values) for treatment effects and their interactions on grain yields for winter wheat experiments
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1 Box plots of the yields of soft white winter wheat direct-seeded at the indicated locations with (+) and without (-) soil fumigation in rows spaced uniformly (U) at 30 cm apart or in pairs (P) spaced 18 cm apart with 42 cm between the paired rows, (A) all treatments, (B) only those treatments with wheat straw on the soil surface, and (C) only those treatments with bare soil. White bars are the yield means and brackets are the yield extremes
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 2 Box plots of the yields of soft white winter wheat in response to all fertilizer in a deep band below the seed and direct-seeded at the indicated locations with (+) and without (-) soil fumigation in rows spaced uniformly (U) at 30 cm apart or in pairs (P) spaced 18 cm apart with 42 cm between paired rows, (A) all treatments, (B) only those treatments with wheat straw on the soil surface, and (C) only those treatments with bare soil. White bars are the yield means and brackets are the yield extremes
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 3 Box plots of the yields of soft white winter wheat in response to all fertilizer in a deep band between every two rows and direct-seeded at the indicated locations with (+) and without (-) soil fumigation in rows spaced uniformly (U) at 30 cm apart or in pairs (P) spaced 18 cm apart with 42 cm between paired rows, (A) all treatments, (B) only those treatments with wheat straw on the soil surface, and (C) only those treatments with bare soil. White bars are the yield means and brackets are the yield extremes
|
|
The mixed effect model used to analyze these results treats years and locations as random samples and provides a novel approach by which to explore data statistically. Thus, while our experimental designs did not include fertilizer placement and surface residues across all locations, fumigation, and spacings, the boxplots clearly revealed the interations between fertilizer placement and surface residues with the other three factors. The only way to confirm these interactions was to test subsets of the data as shown in Table 2.
Soil fumigation caused an uncharacteristic yield depression at the Fairfield site. Wheat plants appeared more lush and vigorous in the fumigated plots during the tillering and stem extension stages of plant development at Fairfield, but starting at heading and continuing through grain fill, the leaves on plants in the fumigated plots became progressively more blotched and necrotic. The leaf symptoms resembled those produced by leaf rust, but was confined to the fumigated plots and presumably was due to an injurious effect of residual bromide left by the fumigation treatment. Similar but milder symptoms of bromide injury were evident at the Colfax site but not at the Lind site.
Spring Wheat Yields
Yields were higher in response to both fumigation and pairing the rows in the spring wheat study (Fig. 4)
with no evidence of injury to the spring wheat caused by fumigation either the year of planting or the previous year. Spring wheat yields averaged 5.8 Mg ha-1 in plots fumigated in the year of planting, 4.4 Mg ha-1 in plots fumigated the previous year, and 3 Mg ha-1 in plots not fumigated (Table 3)
. In addition, the main effects of soil treatment (fumigation), row spacing, and shank position each were significant (P < 0.01) for yield (Table 3). The two-way interactions of shank position and row spacing, and shank position and fumigation treatment also were significant for yield
. Across all treatments, yields were higher with paired-row (5.0 Mg ha-1) than with uniform-row spacing (3.8 Mg ha-1) and when shanks were positioned within each row (4.9 Mg ha-1) compared to between the rows (3.9 Mg ha-1). With fertilizer shanks positioned within the row, yields were higher with paired- than uniform-row spacing (Table 3).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 4 Box plots of the yields of soft white spring wheat on the Palouse Conservation Field Station, Pullman, in plots fumigated just prior to direct-seeding, one year earlier, or not fumigated and planted in rows spaced uniformly (U) at 30 cm apart or in pairs (P) spaced 18 cm apart with 42 cm between the paired rows. White bars are the yield means and brackets are the yield extremes
|
|
View this table:
[in this window]
[in a new window]
|
Table 3 Effect of the interactions of shank position with row spacing, and with soil treatment on yield of Penawawa spring wheat at Pullman, WA, in 1990 on a site cropped for the ninth consecutive year to nonirrigated, direct-drilled wheat
|
|
Effects on Root Diseases
At the Lind site, the row-spacing effect was significant
for severity of Rhizoctonia root rot, with an average root disease rating of 2.3 for plants from paired-row plots and 3.5 for plants from uniform-row plots. There were no differences for the take-all ratings at the Lind site, with average ratings of 2.2 for plants from both paired- and uniform-row treatments. At the Pendleton site, the percentage of plants with one or more diseased roots was high already at the time of the first (14 May) sampling, namely 87% for take-all and 96% for Rhizoctonia root rot for both paired- and uniform-row spacing. These percentages were about the same or only slightly higher by the third (25 June) sampling at 98% for take-all and 95% for Rhizoctonia root rot and no difference between paired and uniform-row spacing. Likewise the percentage of roots with these two diseases were about the same and increased at about the same rate over the three sampling times whether the rows were paired or spaced uniformly. By the third sampling date, and for each of paired and uniform row spacing, 42% of the seminal roots had take-all and 37% had Rhizoctonia root rot. Of the crown roots formed in the mainstem, 25 and 23% had take-all lesions for paired and uniform-row spacing, respectively, and 10 and 9% had Rhizoctonia root rot for paired and uniform row spacing, respectively (NS). Of the crown roots formed on tillers, these percentages were 24 and 22 for take-all and 13 and 14 for Rhizoctonia root rot. An effect of paired-row spacing on disease development only became apparent as the plants matured and began to show whiteheads. The incidence of white heads caused by take-all was 2.5 and 3.0, respectively, for paired- and uniform-row spacing, respectively, and the volumetric weight of the harvested grain was 78.3 and 77.8 kg hL-1, for paired- and uniform-row spacing, respectively, both significant at
. In other words, the premature blight caused by root and crown rot and expressed as white heads also resulted in more shriveled grain, with less damage where rows were paired than when spaced uniformly.
|
Discussion
|
|---|
These results point to a yield (or a test weight) advantage for wheat under pressure from root diseases if the rows are paired (18/42 cm) than when spaced uniformly (30/30 cm), but suggest further that the response is dependent on situations where (i) the soil surface is covered with wheat straw, rather than bare, and (ii) the fertilizer is placed directly below the seed (within the seed row) rather than at the same depth but between the rows. The fumigation effect was also significant for the winter wheat yields when the soil surface was covered with straw but not when bare. This confirms earlier results of Cook and Haglund (1991) who concluded that root disease pressures are greater for wheat grown with a layer of surface residue compared to sites with residue eliminated by burning. Paired-row spacing may offer an alternative to stubble burning as a means to improve the growth and yield of direct-seeded wheat where root diseases are yield-limiting. The significant interaction between spacing and fumigation when the data were limited to yields produced with fertilizer within the rows (directly beneath the seed) suggests that some root disease pressure existed even in the fumigated plots and that highest yields were obtained in the straw-covered plots by the combination of fumigation and paired-row spacing. The one exception was at the Fairfield site where fumigation produced an artifact owing to the apparent injurious effect of the methyl bromide treatment on adult plants.
Similar benefits of paired-row spacing and in-row fertilizer placements were evident in the spring wheat experiment carried out on the PCFS; yields were significantly higher with paired-row spacing compared with uniform row spacing, and when fertilizer was placed within the rows compared with between the rows. Again, root disease pressure was yield-limiting in this experiment based on the incrementally higher yields in response to fumigation 1 yr prior to planting and just prior to planting. As further evidence that ready access to nutrients can help offset the damage caused by root disease, the yields of spring wheat in plots fumigated one year prior to planting were similar to those in plots fumigated the year of planting where fertilizer was placed within the row, but were no better than yields in the nonfumigated check where fertilizer was placed between the rows.
Documentation of effects of variables such as row spacing and fertilizer placement on wheat root diseases by direct inspection of the roots is complicated by the fact that the root diseases occur in various mixtures on the same plants and even on the same roots. Thus, the incidence of roots with take-all lesions is potentially underestimated on plants with roots pruned off by Rhizoctonia root rot, and alternatively, the amount of Rhizoctonia root rot is potentially underestimated on plants with roots with take-all. Furthermore, there is no satisfactory way to quantify the destruction of fine rootlets and root hairs by Pythium on the scale required for field-plot research, yet this kind of damage is typical on wheat in the Inland Northwest (Cook et al., 1987). There is also the complication whereby the reduction in the amount of one root disease opens the way for greater damage from one or both of the other two root diseases (Weller, 1988). Nevertheless, the measurements showed that Rhizoctonia root rot was suppressed by pairing the rows at Lind, and take-all was suppressed (determined on the basis of lower white head ratings and higher volumetric weight of grain) by pairing the rows at Pendleton.
Take-all is a progressive disease owing to the fact that the pathogen spreads as mycelium from root to root and eventually into the tiller bases (crown) where diseased tissue constricts the flow of water to the tops and causes premature ripening (white heads). In contrast, virtually every root infected by R. solani AG8 represents a separate infection from primary inoculum in the soil, and the pathogen does not typically infect tiller bases to cause white heads. While pairing the rows had no effect on the percentage of roots infected by either pathogen, the fact that a paired-row benefit was detected as fewer prematurely ripened plants indicates that take-all developed more slowly where rows were paired compared to rows spaced uniformly.
Many microenvironmental variables with potential to influence root disease epidemiology come into play with the combination of different row spacings and shank positions for placement of fertilizer. All three root diseases are favored by cool, moist soil conditions, and therefore any treatment with potential to accelerate warming and drying of the top layer of soil where these pathogens are active can be expected to retard root disease development. With paired-row spacing, the more open canopy longer into the growing season compared with uniform-row spacing could limit pressure from root diseases because of a microenvironment effect, specifically, greater warming and/or drying of the top few centimeters of soil. Such differences in the crop-canopy are well known to affect the severity of above-ground diseases (Blad et al., 1978; English et al., 1989). The loosening of soil with shanks placed within the seed row and shifting of crop residue to one side or the other of each seed row by the fertilizer shanks would also contribute to soil warming and drying. Roget et al. (1996) showed for direct-seeded wheat in South Australia that the severity of Rhizoctonia root rot could be reduced by a pointed tine positioned so as to loosen the soil below the seed and within the seed row at the time of planting. The combination of pairing the fertilizer shanks as well as the seed rows has the added benefit of intensifying the potential for soil disturbance and row cleaning in a zone planted to paired rows with fertilizer placed directly beneath each row.
Our results do not distinguish between benefits of soil disturbance within the seed row and the benefits of fertilizer placed within rather than between the rows of wheat. Data from other field experiments with spring barley and winter wheat (R.J. Cook, 1991, unpublished data) suggest that placement within easy access of roots, especially for the relatively immobile phosphorus, is more critical when roots are diseased. Past research on optimal position of the fertilizer band relative to the seed at planting has taken into account the potential for root damage caused by the fertilizer (Allred and Ohlogogge, 1964) but not the role of disease in limiting ability of roots to reach the fertilizer. It is logical that healthy roots will be more efficient in the uptake of mineral nutrients than roots pruned by Rhizoctonia, stripped of root hairs and fine rootlets by Pythium, and/or severely rotted due to take-all.
Our results suggest that growers should use a row spacing that will leave the soil surface exposed to warming and drying longer into the growing season, together with shanks or other tools designed to provide fertilizer within each row and below the seed at the time of planting. Wide row spacings have the added advantage of better straw clearance, but have the disadvantages of more safe sites for weeds to grow and plant densities too low to make maximum use of available water. Schillinger et al. (1999) showed for an area with 300 mm of annual precipitation that simply widening a uniform row spacing to 405 mm can reduce the yield potential by lowering the number of spikes per unit area. Paired-row spacing is an obvious compromise; such configurations allow for a more open canopy without compromising plant density. Furthermore, wheat plants potentially will be more competitive with weeds where fertilizer is placed directly beneath each seed row rather than between the rows or broadcasted on the soil surface (Cook and Veseth, 1991). It is important to recognize, however, that these situations apply to fields where root diseases are yield-limiting, and that 3- and 4-yr crop rotations can lower the pressure from root diseases, thereby permitting more flexibility in row spacing, residue management, and fertilizer placement.
|
ACKNOWLEDGMENTS
|
|---|
We are grateful to Arnold Saxton (Statistical Computing Services, Institute of Agriculture, The University of Tennessee) for advice and assistance with statistical analysis of data, and the McGregor Company, Colfax, WA, for taking soil samples and providing soil test data for determination of nitrogen, phosphorus, and sulfur. We also thank Oded Ziv of the Volcani Institute, Israel, for his suggestions for this project. This research was supported by the USDA Agricultural Research Service and the Washington Wheat Commission.
|
NOTES
|
|---|
Research supported by the USDA-ARS and funds provided by the Washington Wheat Commission.
Received for publication August 4, 1999.
|
REFERENCES
|
|---|
- Allred S.E., Ohlogogge A.J. Principles of nutrient uptake from fertilizer bands. VI. germination and emergence of corn as affected by ammonia and ammonium phosphate. Agron. J. 1964;56:309-315.
- Blad B.L., Steadman J.R., Weiss A. Canopy structure and irrigation influence on white mold disease and microclimate of dry edible beans. Phytopathology 1978;68:1431-1437.[ISI]
- Cook R.J. Diseases caused by root-infecting pathogens in dryland agriculture. In: Stewart B.A., ed. Advances in soil science, vol. 13. New York: Springer-Verlag, 1990:214-239.
- Cook R.J. Wheat root health management and environmental concern. Can. J. Plant Pathol. 1992;14:76-85.
- Cook R.J., Haglund W.A. Wheat yield depression associated with conservation tillage caused by root pathogens in the soil not phytotoxins from the straw. Soil Biol. Biochem. 1991;23:1125-1132.
- Cook R.J., Veseth R.J. Wheat health management. St. Paul, MN: APS Press, 1991.
- Cook R.J., Sitton J.W., Haglund W.A. Influence of soil treatments on growth and yield of wheat and implications for control of Pythium root rot. Phytopathology 1987;77:1192-1198.
- Cook R.J., Sitton J.W., Waldher J.T. Evidence for Pythium as a pathogen of direct-drilled wheat in the Pacific Northwest. Plant Dis. 1980;64:1061-1066.
- English J.T., Thomas C.S., Marois J.J., Gubler W.D. Microclimate of grapevine canopies associated with leaf removal and control of Botrytis bunch rot. Phytopathology 1989;79:395-401.
- Hocking R.R. Methods and applications of linear models. New York: John Wiley & Sons, Inc, 1996.
- Moore K.J., Cook R.J. Increased take-all of wheat with direct-drilling in the Pacific Northwest. Phytopathology 1984;74:1044-1049.
- Papendick R.I. Changing tillage and cropping systems: Impacts, recent developments, and emerging research needs. In: Hass G., ed. The optimum tillage challenge. Saskatoon, SK: Univ. Sask. Printing Services, 1984:1-19 Proc. Sask. Inst. of Agrologists Update Series. 6–8 Nov. 1984..
- Papendick, R.I., 1998. Farming with the wind. Best management practives for controlling wind erosion and air quality on the Columbia Plateau croplands. Wash. State Univ., College of Agriculture and Home Econ. Misc. Pub. No. MISCO208.
- Pumphrey R.V., Wilkins D.E., Hane D.C., Smiley R.W. Reduction of Rhizoctonia root rot (bare patch) of winter wheat. Plant Dis. 1987;71:125-127.
- Raaijmakers J.M., Weller D.M., Thomashow L.S. Frequency of antibiotic producing Pseudomonas spp. in natural environments. Appl. Environ. Microbiol. 1997;63:881-887.[Abstract]
- Rasmussen, P., H. Collins, and R.W. Smiley. 1989. Long term management effects on soil productivity and crop yields in semi-arid regions of eastern Oregon. Ore. State Univ. Columbia Basin Agric. Res. Ctr., Station Bulletin 675.
- Roget D.K., Venn N.R., Rovira A.D. Reduction of Rhizocotonia root rot of direct-drilled wheat by short-term chemical fallow. Aust. J. Exp. Agric. 1987;27:430-536.
- Roget D.K., Neate S.M., Rovira A.D. The effect of sowing point design and tillage practice on the incidence of Rhizoctonia root rot, take-all, and cereal cyst nematode. Aust. J. Exp. Agric. 1996;36:683-693.
- SAS Institute Inc. SAS/STAT7 Software: Changes and Enhancements through Release 6.11. Cary, NC: SAS Institute, Inc, 1996.
- Schillinger W.F., Cook R.J., Papendick R.I. Increased cropping intensity for dryland with no-till barley. Agron. J. 1999;91:744-752.[Abstract/Free Full Text]
- Smiley R.W., Ogg A.G., Cook R.J. Influence of glyphosate on Rhizoctonia root rot, growth, and yield of barley. Plant Dis. 1992;76:937-942.
- Weller D.M. Biological control of soilborne plant pathogens in the rhizosphere with bacteria. Annu. Rev. Phytopathology 1988;26:379-407.[ISI]
- Young F.L., Ogg A.G., Jr., Papendick R.I., Thill D.C., Alldredge J.R. Tillage and weed management affect winter wheat yields in an integrated pest management system. Agron. J. 1994;86:147-154.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
