HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (8) Citing Articles via Google Scholar Google Scholar Articles by Bowers, G. R. Articles by Santini, J. B. Search for Related Content PubMed Articles by Bowers, G. R. Articles by Santini, J. B. Agricola Articles by Bowers, G. R. Articles by Santini, J. B. Related Collections Soybean Other Crop Management Seed Production

SOYBEANS

Row Spacing in the Early Soybean Production System

^a Novartis Seeds, Inc., P.O. Box 729, Bay, AR 72411 USA
^b Louisiana State Univ. Red River Res. Stn., P.O. Box 8550, Bossier City, LA 71113 USA
^c Univ. of Arkansas Coop. Ext. Serv., 2201 Brookwood Dr., Little Rock, AR 72203 USA
^d Dep. of Agronomy, 1150 Lilly Hall, Purdue Univ., West Lafayette, IN 47907 USA

glenn.bowers{at}seeds.novartis.com

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
In the southern USA, determinate soybean grown in row spacings (RS) of 50 cm or less generally produce higher yields than that grown in RS of 75 to 100 cm. However, the Early Soybean Production System (ESPS) commonly employs indeterminate cultivars in environments with limited rainfall. This study was conducted to determine whether RS affects seed yield in the ESPS. Twenty-one field experiments were conducted from 1984 to 1997 at sites in Arkansas, Louisiana, and Texas to determine the effect of RS on seed yield in the ESPS. Row spacings of 80 and 40 cm were compared in seven tests; of 75, 50, and 25 cm in six tests; of 75 and 25 cm in one test; of 100, 50, and 25 cm in four tests; and of 100 and 25 cm in three tests. One to five Maturity Group (MG) III and IV indeterminate cultivars were planted in a given test. Based on similarity of treatments and experimental designs, each field test was assigned to one of 10 environmental clusters (EC). Five EC showed a significant (17–36%) yield increase as RS narrowed. Of the other five EC, three had nonsignificant yield increases, one had virtually no yield change, and one had a nonsignificant yield decrease as RS narrowed. Yield was most responsive to RS when the total July and August rainfall ranged from 100 to 270 mm. These results suggest that narrow rows (40 cm) should be used to optimize yields from ESPS plantings in the midsouthern USA.

Abbreviations: ANOVA, analysis of variance • EC, environmental cluster • Env, environment • ESPS, Early Soybean Production System • MG, Maturity Group • RS, row spacing(s)

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
MATURITY GROUP V TO VIII determinate soybean grown in narrow RS (generally 50 cm or less) produces higher yield than soybean grown in wide RS (75 to 100 cm) in the southern USA (Beatty et al., 1982; Carter and Boerma, 1979; Ethredge et al., 1989; Oriade et al., 1997; Parker et al., 1981). However, Heatherly (1988) found that yield increases associated with narrow rows were inconsistent over years.

Planting date does not seem to alter soybean's response to RS. Narrowing RS has resulted in increased seed yield when planting occurred at later than optimum times (Beatty et al., 1982; Boquet, 1990; Parker et al., 1981; Parvez et al., 1989). In fact, yield increases associated with narrow rows may be greater from late planting dates than from optimum dates (Board et al., 1990; Boerma and Ashley, 1982; Boquet et al., 1982). Narrow rows have also yielded more than wide rows when planting was early (April) (Beatty et al., 1982; Boquet et al., 1982).

Rainfall and soil moisture mediate the effect of RS on soybean yield. Taylor (1980), in a year of below-average rainfall, found no yield differences among soybean planting systems with 25-, 50-, 75-, and 100-cm RS. Yield tended to increase as RS decreased in years of average rainfall, but the differences were not significant. In an above-average rainfall year, however, a soybean planting system with 25-cm RS outyielded the 100-cm RS by 17%. Alessi and Power (1982) reported that, in 2 yr of a 4-yr study, total water use was greatest and average soybean yields were least from a 15-cm RS compared with 45- or 90-cm RS. Their data suggested that planting in 15-cm rows enhanced water use prior to flowering. They concluded that, in extreme drought situations, this enhanced early-season water use left less water available for seed development. Seed yields were reduced accordingly. Under less severe drought stress, however, RS had no effect on soybean yield.

Boquet et al. (1982) reported that, in a very dry year, reducing RS from 100 to 50 cm increased yield, but reducing RS further to 25 cm resulted in yield similar to that obtained at the 100-cm RS. Gebhardt and Minor (1983) partially attributed the lack of yield response to narrow RS in their study to limited rainfall. Devlin et al. (1995) found that, at high-yielding sites (>3.3 Mg ha^-1), seed yields were higher in 20-cm than in 76-cm RS. If drought stress resulted in reduced seed yields, yields were greater from 76-cm RS than from 20-cm RS. Similar results were reported from another study (Graterol et al., 1996). Soybean in 25-cm RS had no yield advantage over those in 76-cm RS in a year with yield-limiting conditions. In a year with above-average rainfall but unfavorable rainfall distribution, there also were no yield differences between the narrow and wide rows. However, in a year with no yield-limiting conditions and with a larger amount of total water available for soybean plants at early reproductive periods, soybean in narrow rows yielded more than that in wide rows. Elmore (1998) reported that the seed yield of indeterminate cultivars in both rainfed and irrigated fields and determinate cultivars with irrigation was greatest in 51-cm RS and least in 25- and 76-cm RS. In contrast, determinate cultivars in rainfed conditions yielded similarly in 51- and 76-cm RS and less in 25-cm RS. Elmore concluded that producers with rainfed fields where drought stress is expected could optimize yields by planting indeterminate cultivars in 51-cm RS and determinate cultivars in either 51- or 76-cm RS.

Twin-row planting systems have been compared with single-row planting systems as another method of improving seed yield. Graterol et al. (1996) reported that soybean in the twin-row system had no yield advantage over the conventional single-row system in a year with yield-limiting conditions. However, in a year with no yield-limiting conditions, twin-row planting systems offered yield advantages over a single-row planting system.

Drought stress during seed formation and development in soybean is mainly responsible for reduced seed yield in the midsouthern USA (Reicosky and Heatherly, 1990). In many areas of this region, temperatures are higher and rainfall is lower in July and August than in May and June. The conventional practice in the midsouthern USA involves planting MG V, VI, VII, and VIII cultivars during May and June. Under this system, determinate soybean sets pods and fills seed during August and September; thus, the plant's period of highest moisture demand occurs during a period with a high probability of soil moisture deficit.

The ESPS was developed as a method of drought avoidance for soybean production in the midsouthern USA (Bowers, 1995; Heatherly, 1996). The ESPS involves planting early-maturing cultivars (MG III, IV, and V) in April, when soil moisture is more available. These plants mature in late August and early September, thereby avoiding drought stress that normally occurs during critical reproductive development stages. Only one study (Savoy et al., 1992) examined the effect of RS in the ESPS. `Williams 82' soybean was planted in April in both 36- and 102-cm RS and in irrigated and nonirrigated plots. Row spacing had no effect on yield in either year of this 2-yr study. However, relatively high nonirrigated yields (3.39 Mg ha^-1 in 1988 and 4.23 Mg ha^-1 in 1989) and a lack of significant response to irrigation suggest that their test environments did not suffer from drought stress.

The ESPS is favored in environments in the midsouthern USA that are prone to drought stress. Previous research, using both determinate and indeterminate cultivars planted in May and June, indicated that RS had little effect on seed yield in environments with moderate drought stress. In fact, narrowing RS actually decreased seed yield under very dry conditions. This regional study was designed to determine whether seed yields from the ESPS could be enhanced by reducing RS.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Wide vs. Narrow Rows
Twenty-one field experiments were conducted from 1984 to 1997 in Arkansas, Louisiana, and Texas. Planting dates, RS, and cultivars for each specific environment (site-year) are listed in Table 1 . All three field sites in Texas were on a Houston Black clay (fine, smectitic, thermic Udic Haplustert). The Miller County, AR, site was on an Oklared fine sandy loam (coarse-loamy, mixed, active, calcareous, thermic Typic Udifluvents), the St. Francis County, AR, site was on a Calhoun silt (fine-silty, mixed, active, thermic Typic Glossaqualfs), and the Bossier City, LA, site was on a Latanier clay (clayey over loamy, smectitic, thermic Oxyaquic Hapluderts).

Mature plant and pod heights were recorded just prior to harvest. Plant height (mean distance from the ground to the terminal node on the main stem) and pod height (mean distance from the ground to the first node on the main stem with pod attached) was measured on four randomly selected plants in the center two rows of each plot. All plots were end-trimmed prior to harvest. The middle 4.4 m of the two center rows for the 80-cm RS and the four center rows for the 40-cm RS were harvested. A sample of 100 seed from each plot was weighed to determine seed size.

The experiments at Miller County, AR, in 1993 and 1994 and at St. Francis County in 1994 and 1995 had a factorial arrangement of treatments in a randomized complete block design. The Miller County experiments in 1995 and 1996 were planted in a randomized complete block design with a split-plot arrangement of treatments. Row spacings were main plots and cultivars were split plots. Treatments were replicated four times in all experiments. Each experimental unit was 6.1 m long and consisted of 6 rows for the 75-cm RS, 8 rows for the 50-cm RS, and 16 rows for the 25-cm RS. At the Miller County site in 1993 and 1994 and at the St. Francis County site in 1994 and 1995, seed of each cultivar were planted at the rate of 321000 viable seed ha^-1 in all three RS. Seed of each cultivar were planted at the rate of 296000 viable seed ha^-1 in all three RS at the Miller County site in 1995 and 1996. On the basis of soil test recommendations, fertilizer was applied prior to planting at a rate of 0–0–55.6 kg ha^-1 (N–P–K) at the 1993 Miller County site and at a rate of 0–19.7–37.4 kg ha^-1 at the 1995 St. Francis County site. Weed control was maintained by preplant incorporation of trifluralin and imazaquin in all tests. Postemergence application of imazaquin and sethoxydim {2-[1-(ethoxyimino)butyl]-5-[2-(ethylthio)propyl]-3-hydroxy-2-cyclo-hexen-1-one} was used at the 1995 St. Francis County site and fomesafen {5-[2-chloro-4-(trifluoromethyl)phenoxy]-N-(methylsulfonyl)-2-nitro-benzamide} and fluazifop {(±)-2-[4-[[5-(trifluoromethyl)-2-pyridinyl]oxy]phenoxy]propanoic acid} were used at the 1995 and 1996 Miller County sites for additional weed control. Permethrin [(3-phenoxyphenyl)methyl(±)-cis,trans-3-(2,2-dichloroethenyl)-2,2-dimethyl-cyclopropanecarboxylate] was applied for stink bug [Nezara viridula (L.), Acrosternum hilare (Say), and Euschistus servus (Say)] control at the Miller County site in 1993, 1994, and 1995. Flood irrigation was used to supplement rainfall at the St. Francis County site in 1994 and 1995, with 40 to 50 mm of water being applied once in July and once in August each year. All plots were end-trimmed prior to harvest. The middle 4.9 m of the three center rows for the 75- and 50-cm RS and of the six center rows for the 25-cm RS were harvested.

A second study was conducted at the Miller County site in 1996. This test was planted in a randomized complete block design with a split-plot arrangement of treatments. Cultivars were main plots, RS were split plots, and treatments were replicated four times. Each split plot was 61 m long and consisted of 16 rows for the 75-cm RS and 48 rows for the 25-cm RS. Seed of each cultivar were planted at 67.4 kg ha^-1 for both RS. The middle 45.7 m of the 8 center rows for the 75-cm RS and the 24 center rows for the 25-cm RS were harvested. Mature plant height was measured, as described previously, just prior to harvest in all Arkansas tests.

During 1989 to 1991, the Louisiana experimental design was a randomized complete block with a split-plot arrangement of treatments and three replications. Cultivars were main plots and RS were split plots. Each split plot was 52.4 m long and consisted of 8 rows for the 100-cm RS and 32 rows for the 25-cm RS. The Louisiana Cooperative Extension Service recommends higher seeding rates for narrow RS (Boquet, 1996). Seed of each cultivar were planted at 67.4 kg ha^-1 for the 100-cm RS and 84.2 kg ha^-1 for the 25-cm RS. Weed control was maintained by preplant incorporation of metolachlor in 1989 and 1990 and by preplant incorporation of alachlor [2-chloro-N-(2,6-diethylphenyl)-N-(methoxymethyl)acetamide] and imazaquin in 1991. In 1991, methyl parathion [0,0-dimethyl-(0-p-nitrophenyl)phosphorothioate] was applied for stink bug control.

During 1993 and 1994, the Louisiana experimental design was a randomized complete block with a split-plot arrangement of treatments with four replications. Row spacings were main plots and cultivars were split plots. In 1993, each split plot was 52.4 m long and consisted of 8 rows for the 100-cm RS, 16 rows for the 50-cm RS, and 32 rows for the 25-cm RS. In 1994, each plot was 27.4 m long and consisted of 12 rows for the 100-cm RS, 24 rows for the 50-cm RS, and 48 rows for the 25-cm RS. Seed of each cultivar were planted at 50.2 kg ha^-1 for the 100-cm RS, 56.1 kg ha^-1 for the 50-cm RS, and 78.6 kg ha^-1 for the 25-cm RS. Weed control was maintained by preplant incorporation of alachlor and imazaquin. Methyl parathion was applied in 1993 for stink bug control and thiodicarb {dimethyl-N,N-[thiobis[(methylimino)carbonyloxy]]bis-(ethanimidothioate)} was applied in 1994 for the control of soybean looper [Pseudoplusia includens (Walker)].

During 1996 and 1997, the Louisiana experiments were planted in randomized complete block designs with four replications. Plot size and seeding rates were the same as described for the 1994 study. Weed control was maintained by preplant incorporation of trifluralin and imazaquin. Thiodicarb was applied both years for the control of soybean looper.

Irrigation and fertilizers were not applied to any of the Louisiana test sites. Plant height was measured just prior to harvest in 1996 and 1997. Plots in Louisiana were not end-trimmed. The entire length of the two center rows of the 100-cm RS, the four center rows of the 50-cm RS, and the eight center rows of the 25-cm RS were harvested.

Hand weeding provided additional weed control at all test sites described here. Plots in all environments were harvested with a plot combine. Harvested seed were weighed and adjusted to 130 g kg^-1 moisture content for yield determination.

Based on similarity of treatments and experimental designs, each of the 21 field tests were assigned to one of 10 EC (Table 1). Missing data were estimated using analysis of covariance within the specific whole-plot treatment level and error degrees of freedom were adjusted (Steel and Torrie, 1980, p. 427). Examination of the data indicated transformation was not required (Box et al., 1978). Each site-year data set was analyzed by analysis of variance (ANOVA), for either a split-plot (14 cases) or factorial (7 cases) design, using the ANOVA procedure of SAS (SAS Inst., 1989). Environments (site-year) were considered fixed effects, where the environments were considered to be unique based solely on the rainfall amount received during the growing season (Table 2) . Row spacing and cultivar also were considered fixed effects. Replications (nested within environments) were considered random effects. Bartlett's test was used on both Error a and Error b mean squares to determine if error variances were homogeneous across environments within an EC. Based on the majority of cases (32 of 46), it was determined that the error variances were homogeneous. Within each EC, analyses of variance were performed for split-plot designs (six cases) repeated across environments. An F-test was performed on Error a to determine if error terms could be pooled (P > 0.25). Based on a majority of cases (16 of 23), it was determined that the error terms could be pooled. All treatment and environment x treatment effects were tested with the pooled error (a + b). Seed yield from the widest and narrowest row spacings within each environment were compared using Fisher's protected LSD for each EC where row spacing or row spacing x environment was significant.

Twin vs. Single Rows
Four additional field experiments were conducted from 1985 to 1988 at Blossom, TX. The objective of these experiments was to compare the effects of single versus twin rows on seed yield, plant and pod height, and seed size. Each year, the test site was bedded into 75-cm row centers. A single row was planted on the top of each bed for the single-row treatment. Two rows spaced 15 cm apart were planted on the top of each bed for the twin-row treatment. The experimental design was the same as described above for the other Texas experiments. Cultivars included Crawford, Franklin, JMS 4982, Pioneer 4880, and Union. Each split plot was 5.8 m long and consisted of four beds. Seed of each cultivar were planted at 322800 viable seed ha^-1 in both row designs. On the basis of soil test recommendations, fertilizer was applied prior to planting at a rate of 0–12.7–46.5 kg ha^-1 (N–P–K) in both 1987 and 1988. Fertilizer was not used in 1985 or 1986. Weed control was maintained by preemergence application of metolachlor and imazaquin. Cultivation and hand weeding provided additional weed control. Irrigation was not used.

All plots were end-trimmed prior to harvest. Mature-plant and pod heights were recorded. The middle 4.4 m of the two center beds were harvested with a plot combine. Harvested seed were weighed and adjusted to 130 g kg^-1 moisture content for yield determination. A sample of 100 seed from each plot was weighed to determine seed size.

The data from each year were analyzed by analysis of variance for a split-plot design using the ANOVA procedure of SAS (SAS Inst., 1989). Years, planting system, and cultivar were considered fixed effects. Replications (nested within years) were considered random effects. Bartlett's test was used on both Error a and Error b mean squares to determine if error variances were homogeneous across years. In all cases (12 of 12), it was determined that the error variances were homogeneous. Therefore, data were combined over years for analysis. Using the approach previously described, it was determined that error terms could be pooled.

Treatment effects mentioned in the text are significant at P 0.05 unless specifically indicated. The Fisher protected LSD at the 0.05 level of probability was used for mean separation tests.

	Results and discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 
Plant height was measured in eight EC (data not shown). The environment x RS interaction was significant for plant height in EC-3 and EC-10. In EC-3 there were significant differences between RS in Environment (Env)-F but not in Env-G. In Env-F, plant height in the 80-cm RS (73 cm) was greater than in the 40-cm RS (62 cm). In EC-10, there were significant differences among RS in Env-V but not in Env-W. In Env-V, plant height in the 25-cm RS (120 cm) was greater than in the 50-cm RS (110 cm), which was greater than in the 100-cm RS (94 cm). Row spacing main effect was significant for plant height only in EC-1, where plant height was 75 cm in the 80-cm RS and only 71 cm in the 40-cm RS. The effect of RS on plant height was inconsistent and not large enough to be agronomically important.

Pod height was measured in four EC (data not shown). The environment x RS interaction was significant for pod height in EC-1 with a difference between RS occurring in only one of the three environments. In Env-C, pod height was greater in the 40-cm RS (13 cm) than in the 80-cm RS (10 cm). No other significant differences were detected for pod height. In general, RS had no measurable effect on pod height.

Seed weight was measured in three EC (data not shown). There was a significant environment x RS interaction in EC-1. Within this EC, seed produced from the 80-cm RS were heavier than from the 40-cm RS in Env-A (118 vs. 104 mg seed^-1, respectively) and Env-B (97 vs. 92 mg seed^-1, respectively). The opposite effect of RS on seed weight was found for Env-C where seed from the 40-cm RS were heavier than seed produced in the 80-cm RS (177 vs. 166 mg seed^-1, respectively). Row spacing significantly affected seed weight in EC-3, where seed from the 80-cm RS were heavier than seed from the 40-cm RS (104 vs. 98 mg seed^-1, respectively). In the few cases were RS significantly affected seed weight, the changes in seed weight were small and did not account for the differences in yield discussed below.

In the Louisiana studies, different seeding rates were used for the different RS. Seeding rate could confound the effect of RS on yield. However, the trends and conclusions (discussed later) drawn from the Louisiana data were equivalent to those from the Arkansas and Texas data where seeding rate was held constant as RS changed. Therefore, plant population does not appear to have affected the results obtained in this study.

The environment x RS interaction was significant for seed yield only in EC-1 (Table 3) . Within this EC, yield from the 40-cm RS was greater than that from the 80-cm RS for Env-B and Env-C (2.21 vs. 1.36 and 2.17 vs. 1.71 Mg ha^-1, respectively). In Env-A, yield from the 40-cm RS (0.89 Mg ha^-1) was not significantly different from that from the 80-cm RS (1.11 Mg ha^-1). Row spacing main effect was significant for seed yield in EC-1, EC-2, EC-4, EC-8, and EC-9 (Tables 3–5) . In all five cases, the narrow RS yielded significantly more than the wide RS (Table 6) . In another three EC, the narrow RS yielded numerically (nonsignificantly) more than the wide RS (Table 6). These three EC were EC-5 (P = 0.10), EC-7 (P = 0.07), and EC-10 (P = 0.26). In EC-6 there was very little numerical difference among the seed yields obtained from the three RS (P = 0.69) (Table 6). Only EC-3 had numerically (nonsignificant, P = 0.06) greater seed yield from the wide versus the narrow RS (Table 6).

View larger version (17K): [in this window] [in a new window]   Fig. 1 Soybean seed yield response to row spacing and total rainfall (including irrigation) received during late season (July–August) in 21 test environments from 1984 to 1997. Data points are indicated by both a letter (environment) and number (row spacing in cm). Capital letters for an environment indicate that the yields of the two row spacings shown for that environment are significantly different (Fisher's protected LSD, P 0.05). The environment codes are, for Blossom, TX: A, 1985, B, 1987, and C, 1988; for Port Lavaca, TX: D, 1984, and F, 1986; for Corpus Christi, TX: E, 1985, and G, 1986; for Miller County, AR: H, 1993, J, 1994, N, 1995, O, 1996 test 1, and P, 1996 test 2; for St. Francis County, AR: K, 1994, and M, 1995; and for Bossier City, LA: Q, 1989, R, 1990, S, 1991, T, 1993, U, 1994, V, 1996, and W, 1997

In low rainfall environments, wide RS was reported to yield more than narrow RS (Alessi and Power, 1982; Devlin et al., 1995). In the present study, the wide RS yielded numerically but not significantly (P = 0.06) more than the narrow RS only in EC-3 (Table 6). Environmental Cluster 3 contained the two driest environments in this study (note rainfall for July and August, Table 2). There was a total of 0 mm rainfall for both months at Env-G and only 39 mm for Env-F. Devlin et al. (1995) reported that wide RS yielded more than narrow RS under low-yield conditions. In our studies, the three lowest-yielding environments (Env-A, Env-F, and Env-G) were the only environments where wide RS yielded numerically more than narrow RS (Fig. 1).

In this study, yield was most responsive to RS when late-season rainfall (plus irrigation) ranged from 100 to 270 mm (Fig. 1). The narrow RS yielded more than the wide RS in 9 of the 11 environments having late-season rainfall in this range. Where late-season rainfall was greater than 270 mm, the narrow RS yielded more than the wide RS in only one of the six environments. When late-season rainfall was less than 100 mm, narrow RS yielded more than wide RS in only one of four environments. Wide RS never yielded significantly more than narrow RS in this study. Our results do not agree with previous reports (Devlin et al., 1995; Graterol et al., 1996; Taylor, 1980) that found the greatest benefits from narrow RS in environments with above-average rainfall and without yield-limiting conditions.

In the only other published study describing the effect of RS on seed yield in the ESPS, Savoy et al. (1992) evaluated Williams 82 for 2 yr at one location and reported test means for seed yield of 3.55 Mg ha^-1 in 1988 and 4.26 Mg ha^-1 in 1989. These yield levels indicate high-yielding test environments. Moreover, they reported no significant increase in seed yield due to irrigation. Under high-yielding conditions with no apparent drought stress, they found no significant yield differences between narrow and wide rows. Our data from multiple environments support those reported by Savoy et al. (1992).

There were significant environment x planting system interactions for all of the traits measured in the study comparing the single- and twin-row planting systems (Table 8 and data not shown). Plant and pod heights were generally greater in the twin rows than in the single rows (data not shown). However, the differences were significant only in 1988 for plant height (70 vs. 60 cm) and in 1986 for pod height (13 vs. 9 cm). Planting system affected seed weight in only 1 yr (data not shown). In 1985, seed produced in the single rows (114 mg seed^-1) were heavier than those from the double rows (106 mg seed^-1). There were significant differences between planting systems for yield in 3 of the 4 yr of the study (Table 9) . Yield was greater from the twin rows than from the single rows in 1985 and 1988, with the reverse being true in 1986. Late-season rainfall also affected yield response to planting system in this study. The twin-row planting system produced a yield advantage in the 2 yr receiving the most late-season rainfall (Fig. 2) . Graterol et al. (1996) found that twin-row planting systems yielding more than single-row systems only in environments without yield limitations.

View larger version (14K): [in this window] [in a new window]   Fig. 2 Soybean seed yield response to planting system and total rainfall received during late season (July–August) at Blossom, TX, from 1985 to 1988. Data points are indicated by both a letter (planting system) and number (year). Capital letters for planting system indicate that the yields of the two row configurations shown for that year are significantly different (Fisher's protected LSD, P 0.05). The row configuration codes are S, single row planted on the top of each bed for the single-row treatment; and T, two rows spaced 15 cm apart planted on the top of each bed for the twin-row treatment. Beds were on 75-cm centers

Received for publication June 25, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion REFERENCES

 

• Alessi J., Power J.F. Effects of plant and row spacing on dryland soybean yield and water-use efficiency. Agron. J. 1982;74:851-854.[Abstract/Free Full Text]
• Beatty K.D., Eldridge I.L., Simpson A.M., Jr. Soybean response to different planting patterns and dates. Agron. J. 1982;74:859-862.[Abstract/Free Full Text]
• Board J.E., Harville B.G., Saxton A.M. Narrow-row seed-yield enhancement in determinate soybean. Agron. J. 1990;82:64-68.[Abstract/Free Full Text]
• Boerma H.R., Ashley D.A. Irrigation, row spacing, and genotype effects on late and ultra-late planted soybeans. Agron. J. 1982;74:995-999.[Abstract/Free Full Text]
• Boquet D.J. Plant population density and row spacing effects on soybean at post-optimal planting dates. Agron. J. 1990;82:59-64.[Abstract/Free Full Text]
• Boquet D.J. Row spacings and plant population density. In: Honeycutt J., ed. Louisiana soybean handbook. Pub. 2624. Baton Rouge: La. Coop. Ext. Ser., La. State Univ, 1996:90-92.
• Boquet D.J., Koonce K.L., Walker D.M. Selected determinate soybean cultivar yield responses to row spacings and planting dates. Agron J. 1982;74:136-138.[Abstract/Free Full Text]
• Bowers G.R. An early soybean production system for drought avoidance. J. Prod. Agric. 1995;8:112-119.
• Box G.E.P., Hunter W.G., Hunter J.S. Statistics for experimenters: An introduction to design, data analysis, and model building. New York: John Wiley & Sons, 1978.
• Carter T.E., Jr., Boerma H.R. Implications of genotype x planting date and row spacing interactions in double-cropped soybean cultivar development. Crop Sci. 1979;19:607-610.
• Devlin D.L., Fjell D.L., Shroyer J.P., Gordon W.B., Marsh D.H., Maddux L.D., Martin V.L., Duncan S.R. Row spacing and seeding rates for soybean in low and high yielding environments. J. Prod. Agric. 1995;8:215-222.
• Elmore R.W. Soybean cultivar responses to row spacing and seeding rates in rainfed and irrigated environments. J. Prod. Agric. 1998;11:326-331.
• Ethredge W.J., Jr., Ashley D.A., Woodruff J.M. Row spacing and plant population effects on yield components of soybean. Agron. J. 1989;81:947-951.[Abstract/Free Full Text]
• Gebhardt M.R., Minor H.C. Soybean production systems for claypan soils. Agron. J. 1983;75:532-537.[Abstract/Free Full Text]
• Graterol Y.E., Elmore R.W., Eisenhauer D.E. Narrow-row planting systems for furrow-irrigated soybean. J. Prod. Agric. 1996;9:546-553.
• Heatherly L.G. Planting date, row spacing, and irrigation effects on soybean grown on clay soil. Agron. J. 1988;80:227-231.[Abstract/Free Full Text]
• Heatherly L.G. Yield and germinability of seed from irrigated and nonirrigated early- and late-planted MG IV and V soybean. Crop Sci. 1996;36:1000-1006.[Abstract/Free Full Text]
• Oriade C.A., Dillon C.R., Vories E.D., Bohanan M.E. An economic analysis of alternative cropping and row spacing systems for soybean production. J. Prod. Agric. 1997;10:619-624.
• Parker M.B., Marchant W.H., Mullinex B.J., Jr. Date of planting and row spacing effects on four soybean cultivars. Agron J. 1981;73:759-762.[Abstract/Free Full Text]
• Parvez A.Q., Gardner F.P., Boote K.J. Determinate- and indeterminate-type soybean cultivar responses to pattern, density, and planting date. Crop Sci. 1989;29:150-157.[Abstract/Free Full Text]
• Reicosky D.C., Heatherly L.G. Soybean. In: Stewart B.A., Nelson D.R., eds. Irrigation of agricultural crops. Agron. Monogr. 30. Madison, WI: ASA, CSSA, and SSSA, 1990:639-674.
• SAS Institute. 1989. SAS/STAT user's guide. Version 6, 4th ed. Vol 2. SAS Inst., Cary, NC.
• Savoy B.R., Cothren J.T., Shumway C.R. Early-season production systems utilizing indeterminate soybean. Agron. J. 1992;84:394-398.[Abstract/Free Full Text]
• Steel R.G.D., Torrie J.H. Principles and procedures of statistics, 2nd ed New York: McGraw-Hill, 1980.
• Taylor H.M. Soybean growth and yield as affected by row spacing and by seasonal water supply. Agron. J. 1980;72:543-547.[Abstract/Free Full Text]

This article has been cited by other articles:

				J. J. Heitholt, J. B. Farr, and R. Eason Planting Configuration x Cultivar Effects on Soybean Production in Low-Yield Environments Crop Sci., August 1, 2005; 45(5): 1800 - 1808. [Abstract] [Full Text] [PDF]

				L. G. Heatherly and J. R. Smith EFFECT OF SOYBEAN STEM GROWTH HABIT ON HEIGHT AND NODE NUMBER AFTER BEGINNING BLOOM IN THE MIDSOUTHERN USA Crop Sci., September 1, 2004; 44(5): 1855 - 1858. [Abstract] [Full Text] [PDF]

				R. J. Kratochvil, J. T. Pearce, and M. R. Harrison Jr. Row-Spacing and Seeding Rate Effects on Glyphosate-Resistant Soybean for Mid-Atlantic Production Systems Agron. J., July 1, 2004; 96(4): 1029 - 1038. [Abstract] [Full Text] [PDF]

				L. G. Heatherly, S. R. Spurlock, and C. D. Elmore Row Width and Weed Management Systems for Early Soybean Production System Plantings in the Midsouthern USA Agron. J., September 1, 2002; 94(5): 1172 - 1180. [Abstract] [Full Text] [PDF]

				D. L. Holshouser and J. P. Whittaker Plant Population and Row-Spacing Effects on Early Soybean Production Systems in the Mid-Atlantic USA Agron. J., May 1, 2002; 94(3): 603 - 611. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (8) Citing Articles via Google Scholar Google Scholar Articles by Bowers, G. R. Articles by Santini, J. B. Search for Related Content PubMed Articles by Bowers, G. R. Articles by Santini, J. B. Agricola Articles by Bowers, G. R. Articles by Santini, J. B. Related Collections Soybean Other Crop Management Seed Production