Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 August 2005
Published in Crop Sci 45:1800-1808 (2005)
© 2005 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Heitholt, J. J.
Right arrow Articles by Eason, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Heitholt, J. J.
Right arrow Articles by Eason, R.
Agricola
Right arrow Articles by Heitholt, J. J.
Right arrow Articles by Eason, R.
Related Collections
Right arrow Soybean
Right arrow Crop Systems
Right arrow Plant and Environment Interactions

CROP ECOLOGY, MANAGEMENT & QUALITY

Planting Configuration x Cultivar Effects on Soybean Production in Low-Yield Environments

J. J. Heitholta,*, J. B. Farra and R. Easonb

a 17360 Coit Road, Dallas, TX 75252
b 1509 Aggie Drive, Beaumont, TX 77713-8530

* Corresponding author (j-heitholt{at}tamu.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
Texas soybean [Glycine max (L.) Merr.] growers generally use row spacings ranging from 15 to 102 cm and seeding rates that provide 25 to 62 plants m–2. However, the interaction of cultivars with planting configuration is not well understood. The objectives of this research were to measure the yield response of selected cultivars to seeding population and row spacing in low-yielding environments and to relate morphology to this response. In north Texas, on a highly calcareous clay soil with 10 maturity group (MG) III cultivars (where 2-yr average yield was 1.22 Mg ha–1) and three seeding populations (25–75 m–2), only one cultivar, AG3702, showed consistent yield increases at 50 and 75 m–2. The row-spacing tests were conducted in two north Texas environments where 14 MG IV cultivars varying in leaf size were grown in 36- vs. 71-cm rows and two environments in southeast Texas where 21 cultivars varying in height and maturity (MG IV– MG VIII) were grown (18- vs. 76-cm rows). In north Texas, yields were not significantly different between 71-cm rows (1.31 Mg ha–1) and 36-cm rows (1.02 Mg ha–1). Likewise, in southeast Texas, yield in 18-cm rows (1.49 Mg ha–1) was similar to yield in 76-cm rows (1.46 Mg ha–1). Cultivar x row spacing effects on yield were not significant and leaf size and plant height did not affect the yield response to row spacing. In general, our results demonstrated that management x genotype interactions within the range of these cultivars and management practices were not important in these low-yielding environments and possibly were overwhelmed by the severity and timing of drought stress.

Abbreviations: DAP, days after planting • MG, maturity group • PAR, photosynthetically active radiation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
HOT AND DRY CONDITIONS during either soybean seedling growth or pod filling occur frequently in both north Texas and southeast Texas. Until genetic advances provide more tolerant genotypes, growers must be aware of and employ management practices that minimize the effects of stress. Soybean growers in these regions may have the opportunity to adjust planting date so that reproductive growth avoids drought and heat during July and August (Bowers, 1995; Heatherly, 1999). Unfortunately, wet and cool conditions in the spring often force relatively late planting dates. Choice of a planting configuration (row spacing and seeding rate) and cultivar are management decisions that growers can control. However, the selection of row spacing, cultivar, and seeding rate (hereafter called seeding population) is often made without knowing whether the cultivar is suited for the chosen configuration (i.e., whether cultivar x planting configuration interactions are significant).

In regions where temperatures are cooler than Texas, research has identified some seeding population x cultivar interactions. For example, Elmore (1998) found that yield of a determinate cultivar grown in Nebraska increased as seeding population increased from 11 to 35 seeds m–2 but the yield of an indeterminate cultivar was unaffected. Ball et al. (2000a)(2000b) found that an indeterminate MG IV cultivar (Asgrow 4922) exhibited yield increases in Arkansas as plant population density increased from approximately 10 to 100 plants m–2; a determinate MG IV cultivar (Manokin) also exhibited a positive response at these densities but its response was less sensitive than Asgrow 4922. In north Texas, the interest in growing late commercial MG III (relative maturity 3.5 to 3.9) cultivars has increased, but their response to seeding population has not been well characterized.

Differences in soybean cultivar morphology and how they affect the response to row spacing is also a concern. Shibles and Green (1969) were among the first to hypothesize that a shorter, less bushy morphology is likely preferable for narrow rows, whereas a tall, bushy morphology may be best suited for wide rows. In southeast Texas, cultivars from MG IV to MG VIII are grown and tall phenotypes are often chosen by growers. However, whether tall cultivars are best suited for wide rows or whether short cultivars are best suited for narrow rows has not been thoroughly addressed in this region.

In addition to plant height, leaf size in the upper canopy is another possible, yet under-investigated characteristic that may affect yield response to row spacing. Large-leaf types might be better suited for wide rows, whereas small-leaf types might be better suited for narrow rows. Metz et al. (1984) using only 35-cm rows in Iowa found that smaller individual leaves in the upper canopy gave a yield advantage, but in North Carolina, Wells et al. (1993) using 48- and 96-cm rows found that smaller upper-canopy leaves reduced canopy photosynthetically active radiation (PAR) interception and subsequent yield. Unless further studies are conducted to help clarify how cultivars with different phenotypes respond to row spacing and seeding rate, chances of achieving improved yield of soybean in a low-yield environment will remain poor. Therefore, the objectives of this research were: (i) to characterize the effects of seeding population on yield, leaf angle, and plant circumference of MG III and IV cultivars; (ii) to characterize the performance of soybean cultivars varying in height and leaf morphology to row spacings narrower and wider than 50 cm; and (iii) to determine how morphological traits are related to yield in low-yielding environments.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 
Seeding Population x Cultivar Trials (Prosper 2000–2001)
At Prosper, TX (33°12' N lat), 10 MG III cultivars (Asgrow AG3702, Eagle Seed ES3901, Midwest Premium Genetics MPV398NRR, and seven Syngenta cultivars, S30-P6, S32-Z3, S33-N1, S34-A9, S34-B2, S38-T8, and S39-D9) were planted 7 Apr. 2000 and 15 May 2001 at populations of 25, 50, and 75 seeds m–2 in 36-cm rows on a Houston Black clay (fine, smectitic, thermic Udic Haplustert) with pH 8.3. Soil tests indicated that no fertilizer was needed. Plots were 6.0 by 3.1 m with eight rows. Plant population density was determined at approximately R1 (Fehr and Caviness, 1977) by counting plants in two separate 1-m segments of each plot. Plant height was measured at growth stage R6. At growth stage R8, 3.35 m of the center four rows of all plots were harvested after end trimming to determine yield (adjusted to 13% moisture). The design of the experiment was a randomized complete block with a factorial arrangement of cultivars and seeding populations with four replicates.

Seeding Population x Row Spacing x Cultivar Trials (2002–2003)
At Prosper (site described above), the experiment included three cultivars (AG 3702, Garst D399, and Deltapine DP4344), four row spacings (18, 36, 54, and 71 cm), and two seeding populations (31 and 44 seeds m–2). Two replicates were planted 13 Apr. 2002 but rain fell and replicates 3 and 4 were planted 18 and 19 April, respectively. The planting date in 2003 was 21 April. The remaining nine MG III cultivars from 2000 and 2001 that did not respond consistently to seeding population were dropped because of their low yield and the two later maturing cultivars were added. A randomized complete block design with a split-plot arrangement was used. Row spacing was the main plot and the seeding population–cultivar combinations were subplots and four replicates were used. Plant population density was determined at approximately R1. Because of poor emergence in 2003, only plots in two replicates could be used for measurements of canopy morphology. On 12 June 2002 and on 13 June 2003 (growth stages R3 to R5) at approximately 1000 h to 1200 h, the petiole angle of two leaves (third uppermost) per plot was measured using a protractor (with 0° being totally erectophile/vertical orientation and 90° being totally planophile/horizontal). On 19 June 2002 and 19 June 2003 at approximately 1000 to 1200 h, the circumference of one plant per plot was measured by sequentially lowering a series of circular and slightly flexible plastic rings around the plant and recording the circumference of the smallest ring that encompassed but did not touch any part of the plant. The adjacent plants were held back during the circumference measurement and the distance of their stems from the measured plant at ground level was recorded for possible use as a covariate. Yield data were obtained in 2002 as described above but not in 2003 because of the poor emergence described earlier.

Leaf Size x Row Spacing Trials (A, Dallas 2000; B, Prosper)
In Dallas, TX (32°51' N lat), and Prosper, 14 MG IV cultivars varying in leaf size were planted 21 Apr. 2000 and 16 May 2001, respectively, in 36- and 71-cm rows. The 14 cultivars were chosen on the basis of visual ratings of upper leaf size obtained in a 1999 north Texas variety trial. The cultivars were Croplan 466RR, Croplan 4792, DP4750RR, DP4690, Garst D485, Garst 4882, Garst 4888RR, HY4540, MFA4477, MPV457, NK3474, SG468RR, Taylor 488RR, and Triumph TR4718. Plots were 5.9 m long and 12 rows wide (36-cm rows) and 6 rows wide (71-cm rows). The seeding population was 50 seeds m–2 and the soil type was also a Houston Black clay with a 1% slope (Dallas) or no slope (Prosper). At approximately R5, five fully expanded leaves from the third uppermost main stem node were harvested from each plot, and leaf blade area, petiole length, blade width, and blade length were determined. Area was determined with a LiCor Model 3000 Leaf Area Meter (LiCor, Inc., Lincoln, NE). In 2000, the cultivars were classified as either small leaf (<83 cm2) or large leaf (≥83 cm2), but in 2001 leaf blade area was significantly affected by row spacing and by a row spacing x cultivar interaction. Therefore, blade areas were categorized as small (<50 cm2) or large (≥50 cm2) for narrow rows, whereas categories for wide rows were small (<58 cm2) and large (≥58 cm2). Canopy closure was visually ranked on three dates during development by estimating the ratio of canopy width to canopy plus bare ground width. Plant height was determined at R6, and after end trimming to 3.0 m at R8, seed were harvested from the center four rows of 36-cm row plots and the center two rows of 71-cm row plots. For both trials, the experimental design was a randomized complete block with a split-plot arrangement. Row spacing was the main plot and cultivar was the subplot and three replicates were used.

Plant Height x Row Spacing Trials (Beaumont 2000–2001)
In 2000, 23 cultivars from MGs IV, V, VI, VII, and VIII were planted 26 May 2000 at the Research and Extension Center in Beaumont, TX (29°57' N lat), at a seeding population of approximately 35 seeds m–2. The cultivars were chosen on the basis of yield and plant height results of a 1999 variety trial in Beaumont so that tall, medium, and short cultivars were present for all MGs when possible. Cultivars were: MG IV: AP 4980RR, DP5989, DK 4868RR, MPV437RR; MG V: Garst 5961RR, DK 5995, MPV537RR, SG 597; MG VI: DP 6880RR, SG 678RR, TsB 93-1083; MG VII: APX 9711RR, H 7550RR, Haskell, Sharkey, TsB 92-3986, TsB 95-4656, TsB 95-5641; MG VIII: Cook, DPX 8574RR, H 8001, Perrin, TsB 93-2220. Row spacings were 18 and 76 cm. Canopy closure was determined as described previously on two dates during development in 2000. In 2001, 19 cultivars (the same as 2000 except for the inclusion of Travis and the exclusion of DK 5995, H8001, MPV537RR, SG597, and Perrin) were planted 17 May 2001 as previously described. In June 2001, the site was flooded for 2 d by a tropical storm, but by 25 June the crop appeared to be growing acceptably. In both years, plots were 6.0 by 3.0 m for 76-cm rows and 6.0 by 1.5 m for 18-cm rows. Height and yield data (no end trimming) were collected at R6 and R8, respectively. A randomized complete block with a split-plot arrangement was used. Row spacings were main plots and cultivars were sub-plots and three replications were used.

Cropping History, Implements, Weather Data, and Statistical Analysis
Previous cropping history for all Prosper, TX, sites was winter wheat (Triticum aestivum L.) followed by a 9-mo fallow period. At Dallas, land was fallow for several years, and at Beaumont soybean followed soybean. At the two north Texas sites (Prosper and Dallas), a seven-unit grain drill planter was used for each row spacing with tubing diverted to selected units to achieve the 36- and 71-cm spacings. At Beaumont a conventional row planter for 76-cm rows and a seven-row grain drill set at 18-cm row width were used. Weeds were controlled using preplant incorporated and post-emergence herbicides and insects were controlled as needed using post-emergence insecticides.

Weather data were collected by weather stations located on the Beaumont, Dallas, and Prosper stations within 1 km of the plots. However, when malfunctions occurred at Prosper, weather data were obtained from the National Weather Service records available from McKinney, TX (10 km from Prosper). Rainfall during reproductive growth for the north Texas sites (four seeding population studies) indicated a consistently low rainfall during crop maturation whereas southeast Texas (Beaumont) had more rain (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Rainfall for periods during reproductive growth in three Texas locations used for seeding population and row spacing studies.

 
Analysis of variance was performed on all data sets using SAS (SAS Institute Inc., Cary, NC). Because of year x treatment interactions (seeding population studies), each year was analyzed separately. For the four row-spacing studies, each environment was analyzed separately. A LSD for mean separation to compare any two treatment means in the 2000 and 2001 seeding population studies was calculated using the error mean square and t value based on error degrees of freedom. For split-plot studies, ANOVA significance using rep x main effect as error was used to separate main effects and standard LSD equations were used to separate sub-plot means within and across main effects. For regressions, SAS Proc Reg was used and model coefficients with P values greater than 0.05 were dropped.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 SUMMARY
 REFERENCES
 
Seeding Population x Cultivar (Prosper 2000–2001)
There was a strong linear relationship between seeding population and resulting plant population density at R1 with density = 5.10 + 0.82(seeding population), R2 = 0.91 in 2000 and density = –14.5 + 1.38(seeding population) – 0.008(seeding population)2, R2 = 0.80 in 2001. The yield of AG3702 responded to a seeding population up to 75 seeds m–2 in 2000 and to 50 seeds m–2 in 2001 (Table 2). Yield of S30P6 responded to 50 seeds m–2 in 2000 but not in 2001. Yield of MPV398RR and S38-T8 responded to 50 seeds m–2 in 2001 but not in 2000. Cultivar and the cultivar x seeding population interaction effects on yield were significant (P < 0.05) in each year except for the interaction in 2000 (P = 0.09). Seeding population effects were not significant in 2000 (P = 0.12) or 2001 (P = 0.08).


View this table:
[in this window]
[in a new window]
  		Table 2. Effect of seeding population on midseason plant population and seed yield of 10 Maturity Group III soybean cultivars at Prosper, TX, in 2000 and 2001.

 
Seeding Population x Row Spacing x Cultivar (Prosper 2002–2003)
Grain yields in 2002 were only affected by cultivar and not by row spacing, seeding population, or any interactions (data not shown). Yields of DP4344 (1.90 Mg ha–1) and Garst D399 (1.72 Mg ha–1) were significantly greater than AG3702 (1.31 Mg ha–1). Petiole angle averaged across years was unaffected by row spacing, but the leaves of AG7302 were oriented more horizontally (39.5° from vertical, P < 0.05) than the more vertically oriented Garst D399 (33.5°) and DP4344 (32.0°). Averaged across years, plant circumference was unaffected by seeding rate and cultivar but increased as row spacing increased (Fig. 1).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. Soybean plant circumference as a function of row spacing for six cultivar x seeding rate combinations. Data are averaged across years. With data averaged across cultivar and seeding rate, regression was significant (P = 0.02). PC and RS represent plant circumference and row spacing, respectively.

 
Leaf Size x Row Spacing Trials A and B
Leaf traits (area, petiole length, blade width, and length) were affected by row spacing x cultivar interactions in 2001 but not in 2000 (data not shown). In 2000, Hyperformer HY4540 exhibited the largest leaf area (100 cm2 leaf–1), whereas Deltapine DP4750 exhibited the smallest (63 cm2 leaf–1). In 2001, Garst D485 (64 cm2) had the largest individual leaf area in narrow rows and Garst 4882 (34 cm2) had the smallest. In wide rows, HY4540 (67 cm2) and MFA4477 (67 cm2) had the largest leaf area and DP4690 (44 cm2) had the smallest. In 2000, leaf areas were similar between row spacings but in 2001 leaf area in wide rows was 28% greater than narrow rows (58 vs. 48 cm2, P = 0.05). As expected, there was greater canopy closure in the 36-cm rows than in the 71-cm rows in 2000 (Fig. 2A). In 2001, the 36-cm rows had greater canopy closure at 50 d after planting (DAP) and earlier, but differences thereafter were not significant (Fig. 2B). The large-leaf cultivars had greater canopy closure than small-leaf cultivars in 2000 but not 2001.



View larger version (33K):
[in this window]
[in a new window]
  		Fig. 2. Effect of third uppermost trifoliolate leaf size and row spacing (36 vs. 71 cm) on canopy closure of seven small- and seven large-leaf soybean cultivars grown at (A) Dallas, TX, 2000; and (B) Prosper, TX, 2001. Bars indicate LSD (0.05).

 
Grain yields in 2000 were similar between row spacings (1.64 Mg ha–1 for 71-cm rows and 1.33 Mg ha–1 in 36-cm rows). Likewise, yield was similar in wide rows (0.99 Mg ha–1) and narrow rows (0.71 Mg ha–1) in 2001. In 2000, yield for cultivars grown in 36-cm rows was negatively correlated with maturity (r = –0.64, P ≤ 0.05), but in 2001 there was no correlation between yield and maturity at either row spacing. Yields in 2001 were extremely low because of late planting and lack of rain during reproductive growth (Table 1). The cultivar x row spacing interaction effect on yield was not significant in either year (P = 0.12). In 2000, plants were similar in height between row spacings (84 vs. 82 cm, for 71- and 36-cm rows, respectively) and in 2001, a similar result occurred (58 vs. 52 cm for 71- and 36-cm rows, respectively). There was no relationship between yield and individual leaf size at either row spacing (Fig. 3A, 3B).



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 3. Yield as a function of leaf area (third uppermost trifoliolate) for 14 soybean cultivars grown in 36- and 71-cm rows at two north Texas locations: (A) Dallas, 2000; (B) Prosper, 2001. No significant relationships were found.

 
Plant Height x Row Spacing Trials (Beaumont 2000–2001)
At 35 DAP in 2000, the short (≤63 cm) cultivars had greater canopy closure than the tall (≥72 cm) cultivars at both row spacings with medium height cultivars falling in between (Table 3). At 49 DAP, differences were only apparent in 76-cm rows. There was no significant linear relationship between yield and height in either year (Fig. 4A, 4B). The row spacing x cultivar interaction effect on yield was not significant in either year although a strong relationship between yield in wide rows and yield in narrow rows was only found in 2000 (Fig. 5A, 5B). Rainfall during reproductive growth was greater than in north Texas (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 3. Effect of soybean cultivar classification as tall (≥72 cm), medium (64–71 cm), and short (≤63cm) and row spacing on canopy closure in Beaumont during the 2000 season.

 


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between yield and plant height of 23 soybean cultivars grown in two row spacings at Beaumont, TX, in (A) 2000 and (B) 2001. No statistically significant relationship was found.

 


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 5. Relationship between wide-row (76 cm) and narrow-row (18 cm) soybean yield at Beaumont in (A) 2000 and (B) 2001.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
SUMMARY
 REFERENCES
 
Responses to Seeding Rate
Yield of only one of the 10 MG III cultivars, AG3702, consistently increased when seeding population increased from 25 to 50 seeds m–2 or to 75 seeds m–2. In the follow-up study, petioles of AG3702 exhibited a slightly more horizontal orientation than the two later maturing, better adapted cultivars. AG3702 was also the earliest maturing and lowest yielding of the three cultivars used in the 2002 test. Therefore, it is possible that the petiole orientation response of AG3702 compared with Garst D399 and DP4344 was related to its maturity or its lesser adaptability to north Texas. Plant circumference did not seem to be linked to yield response to seeding population although we were able to document the expected increase in plant circumference in wider rows.

For most of the MG III cultivars in the present study, the optimal density of approximately 20 plants m–2 agrees with other reports (Ramseur et al., 1984; Boquet, 1990; Hicks et al., 1990; Wells, 1991; Elmore, 1998) that high seeding populations were not necessarily advantageous in low-yield environments. Devlin et al. (1995) found little yield response to seeding populations of 12 to 70 seeds m–2 in 76-cm rows for four of five sites across Kansas, where seed yield ranged from 0.3 to 4.0 Mg ha–1. In contrast, for that same Kansas study, yields increased in response to these same seeding populations when grown in 20-cm rows in four of the five sites. In three full-season mid-Atlantic environments, Kratochvil et al. (2004) found stands between 20 and 27 plants m–2 to produce optimal yield with some cultivars, but nearly 40 plants m–2 were needed to optimize yield for other cultivars.

In relatively high yield environments, Parvez et al. (1989) in Florida and Ball et al. (2000a)(2000b) in Arkansas reported stronger yield responses to plant population densities than to the 20 plants m–2 in the present study. Yield in those studies was strongly correlated with maximum PAR intercepted, crop growth rate, and inversely related to the time required to reach canopy closure (Ball et al., 2000b). Positive responses to seeding populations were not only observed in high-yield environments but also were found in low-yield environments when the yield-reducing stress occurred early in the season (Holshouser and Whittaker, 2002). Many researchers (we among them) might argue that increasing density from 10 to 30 plants m–2 will increase yield when conditions that compromise vegetative growth (e.g., poor soil, late planting, earlier maturity, early-season flooding stress, early-season water deficit, low solar insolation, and disease) are present. However, if water first becomes limiting after flowering, high plant population densities will exacerbate a reduction in canopy photosynthesis and may result in less yield than moderate plant densities (Taylor, 1980; Reicosky et al., 1985). The timing of water deficit in our north Texas environments (Table 1) is likely to have played a role in the lack of yield response to plant population density even though we used relatively early maturing cultivars for this region as did Edwards et al. (2003) in Arkansas. Consequently, except for the response of AG3702 in our study, it appears that most MG III cultivars are unlikely to show a positive yield response to a seeding populations of 50 seeds m–2. We do not know if seeding populations of 50 seeds m–2 would be useful in our region under conditions where water stress is alleviated before the early stages of seed filling or if seeding populations between 25 and 50 seeds m–2 would be advantageous.

Responses to Row Spacing
Yield of soybean grown in wide rows (71 cm) was as high or higher than in narrow rows (36 cm) in our north Texas studies (Dallas and Prosper) where dry conditions occurred during reproductive growth in each of the 3 yr. Likewise, wide rows (76 cm) in southeast Texas (Beaumont) yielded essentially the same as narrow rows (18 cm) where moisture was less limiting. In the north Texas region of the present study, little useful rain fell between 61 and 105 DAP. Therefore, we believe that our narrow-row canopies reduced the available soil moisture to a greater extent than did the wide-row canopies, a hypothesis already proposed by others (Taylor, 1980; Alessi and Power, 1982; Frederick et al., 1998). Our lack of row spacing response agrees with Taylor (1980), Weaver and Wilcox (1982), Gebhardt and Minor (1983), Savoy et al. (1990), and Devlin et al. (1995) but conflicts with Cooper (1981), Board et al. (1990a), Boquet (1990), Egli (1994), Board and Harville (1996), Oriade et al. (1997), Bullock et al. (1998), Robinson and Wilcox (1998), Nelson and Renner (1999), Heatherly et al. (2001), and Manning et al. (2001) who showed row spacings of 50 cm or less outyielded wider spacings. It should be noted that several of the studies reporting a narrow-row yield increase were conducted where moisture did not limit yield or where row spacings of {approx}100 cm (often consider too wide) were compared with narrower spacings.

The competitiveness of the wide-row yields in the present study compared with narrow rows might appear to conflict with Bowers et al. (2000), a study conducted in regions near our north Texas sites. However, the narrow-row advantages found in Bowers et al. (2000) were found primarily in environments with between 100 and 250 mm of late-season rain. In the present study, our late-season rain was 74 mm or below in three of four environments (Table 1) suggesting that our results were in agreement with Bowers et al. (2000). Many other factors used in our north Texas row spacing studies, such as a relatively high seeding population of 50 m–2 (a rate sometimes used by growers) might have intensified the narrow-row moisture stress conditions mentioned earlier. Row spacings such as 50 cm, which were not thoroughly tested in our studies, have been very successful in other regions (e.g., Beatty et al., 1982; Heatherly et al., 2001).

We found similar rankings between wide-row yield and narrow-row yield and similar results were reported by Weaver and Wilcox (1982) and Board et al. (1996). We did not find a consistent differential response to row spacing because of plant height. Ethredge et al. (1989) reported that the cultivar Essex, which exhibited "a narrow profile and upright branches" had 15% greater yield in 25-cm rows than in 75-cm rows whereas yield of the taller ‘Deltapine 105’ increased only 5%. Weaver et al. (1991) reported that taller indeterminate genotypes with little branching may have a yield disadvantage in wider rows, at least when planted late or in a low-yield environment, as compared with shorter determinate genotypes with prominent branching.

Other interrelated factors such as leaf size, canopy closure, crop growth rate, and branching are often discussed when interpreting row spacing results. Although we did not find that smaller leaf size was related to row-spacing response, Metz et al. (1984) demonstrated that small leaflet types were advantageous in narrow rows when 40 F5–derived MG II lines were tested in Iowa. Our finding that large leaf size increased canopy closure (Fig. 2A, 2000 only) is supported by Wells et al. (1993) that reported lanceolate leaflets (small leaf area) reduced PAR interception during vegetative growth of isolines derived from the Tracy-M background (MG VI). In that study, yield was closely linked to PAR interception. Those Tracy-M, small-leaf types were associated with greater yield, especially in the later maturing lines. Board et al. (1990a) showed narrow-row yield enhancement was related to PAR interception, and that narrow row–induced yield increase at late planting was due to increased branch dry matter (Board et al., 1990b).


   SUMMARY
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
REFERENCES
 
Seeding population and the cultivar x seeding population interaction had little effect on yield except for the cultivar AG3702. Seeding population did not affect leaf orientation or individual plant circumference, but as expected, plant circumference increased as row spacing increased. Although some cultivars are exceptions, our results indicated that factors other than cultivar (soil type, row spacing, seed costs) are likely to govern selection of a seeding population for MG III cultivars in the north Texas. We tested cultivars varying in upper leaf traits, height, and maturity, yet no cultivar exhibited a preferential response to either narrow or wide rows. Although we expected the shorter cultivars and small leaf types to yield less in wide rows (>50 cm) and to respond better to narrow rows than the taller cultivars, we observed little relationship between the canopy traits and yield or yield response to row spacing. Taller plants are desired for many growers in north, central, and southeast Texas, but this preference may relate more to harvest efficiency and less to minor differences in grain yield. As with seeding rate, our results in these regions of Texas do not support basing cultivar selection on row spacing or vice versa.


   ACKNOWLEDGMENTS
 
The authors thank Novartis (Syngenta), Monsanto, and Eagle Seed for supplying seed, Nitragin, Inc. for supplying the inoculant, and the Texas Soybean Board for funding the research. We thank Robert Alexander, Damie Jean Barber, Stephanie Eaker, Jeffrey Stowers, and Russell Sutton for technical assistance. The authors also thank Drs. G.R. Bowers, D.B. Egli, C.T. Mackown, and the three anonymous reviewers for their helpful suggestions.

Received for publication October 13, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 SUMMARY
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
J. L. De Bruin and P. Pedersen
Effect of Row Spacing and Seeding Rate on Soybean Yield
Agron. J., May 7, 2008; 100(3): 704 - 710.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Heitholt, J. J.
Right arrow Articles by Eason, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Heitholt, J. J.
Right arrow Articles by Eason, R.
Agricola
Right arrow Articles by Heitholt, J. J.
Right arrow Articles by Eason, R.
Related Collections
Right arrow Soybean
Right arrow Crop Systems
Right arrow Plant and Environment Interactions

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome