Agronomy Journal Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 8 January 2009
Published in Agron J 101:131-139 (2009)
DOI: 10.2134/agronj2008.0014x
© 2009 American Society of Agronomy
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 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 Google Scholar
Google Scholar
Right arrow Articles by Robinson, A. P.
Right arrow Articles by Santini, J. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Robinson, A. P.
Right arrow Articles by Santini, J. B.
Agricola
Right arrow Articles by Robinson, A. P.
Right arrow Articles by Santini, J. B.
Related Collections
Right arrow Soybean
Right arrow Crop Growth and Development

Analysis of High Yielding, Early-Planted Soybean in Indiana

Andrew P. Robinsona,*, Shawn P. Conleyb, Jeffrey J. Voleneca and Judith B. Santinia

a Dep. of Agronomy, Purdue Univ., 915 West State St., West Lafayette, IN 47907-2054
b Dep. of Agronomy, 1575 Linden Drive, Univ. of Wisconsin, Madison, WI 53706

* Corresponding author (arobinson{at}purdue.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
A trend toward early planting of soybean [Glycine max (L.) Merr.] in Indiana results in higher yield, but the limit to which a positive response to early planting occurs has not been evaluated. Our objective was to determine how early planting affects yield components and seed composition of indeterminate soybean planted in late March through early June in Indiana. Three cultivars (Pioneer brand 92M61, Becks brand 321NRR, and Becks brand 367NRR) were sown at six planting dates (late March through early June) in West Lafayette, IN, in 2006 and 2007. Across cultivars, yield in 2006 ranged between 4.24 to 4.43 Mg ha–1 at the planting dates from late March to mid-May, and decreased to 3.36 and 3.56 Mg ha–1 at later planting dates. In 2007, yield ranged from 4.21 to 4.44 Mg ha–1 for the 10 April, 30 April, and 9 May planting dates. Yield was reduced at the late March and early June plantings and ranged from 3.85 to 3.99 Mg ha–1. Path analysis revealed that pods m–2 had the greatest impact on yield, but seed mass was also an important constituent. Mean oil concentration decreased approximately 12 g kg–1 as planting was delayed in both years. In 2006, average seed protein concentration varied by planting date. In 2007, mean protein concentration increased 14 g kg–1 as planting was delayed. Delaying planting until late May or early June altered seed composition slightly, but significantly reduced yield. Planting in April or early May is an effective management strategy to increase soybean yield in Indiana.

Received for publication July 10, 2008.
   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
PLANTING DATE is perhaps the most important and least expensive cultural consideration that impacts soybean yield. Today, two-thirds of Indiana farmers plant soybean 1 to 3 wk earlier than they did 10 yr ago (Conley and Santini, 2007). In 2005, 27% of soybean growers began planting in April, whereas 68% began planting in May (Conley and Santini, 2007). Current recommendations for soybean are to plant from mid-May through June (Wilcox and Frankenberger, 1987); however, farmers believe that earlier planting results in greater yield (Conley and Santini, 2007). To date, few studies have addressed the effect of late-March through April planting dates in the Midwest.

Recommendations for soybean planting date have changed since nonforage uses of soybean began in the United States. In one of the earliest-known planting date studies, seed yield was highest when planting dates were from mid-June through early July (Mooers, 1908). Mooers did report that some cultivars yielded well when planted in April and May, but results were inconsistent. By the mid-20th century, recommended planting dates for soybean ranged from 1 to 23 May in the Midwest (Torrie and Briggs, 1955). Experience over the past 27 yr indicates that sowing from late April through May is best because planting in June or July results in substantially lower yield (De Bruin and Pedersen, 2008; Wilcox and Frankenberger, 1987). Other findings revealed no difference in yield, or a yield reduction when soybean was planted in early compared with late May (Grau et al., 1994).

Planting date impacts soybean growth characteristics. Early planting of indeterminate soybean cultivars result in more nodes (Wilcox and Frankenberger, 1987) and a greater numbers of pods and seeds (Pedersen and Lauer, 2004). These yield component changes are linked to extended vegetative and reproductive development (Wilcox and Frankenberger, 1987) during R1 (Fehr and Caviness, 1977) through R8 soybean growth stages (Wilcox and Frankenberger, 1987) in early vs. late-planted soybean. Late-planted soybean often has a higher floral abortion rate (Heitholt et al., 1986), but heavier seed, which can compensate for reduced seed number (Spaeth and Sinclair, 1984). A shorter daylength can also decrease growth stage length (Kantolic and Slafer, 2001) and increase seed mass (Morandi et al., 1988).

The period from R5 through R6 is a critical stage in soybean development. Drought stress during R5 through R6 is associated with fewer pods m–2 and low seed yield (Foroud et al., 1993). Temperature and soil moisture during the reproductive period also control seed composition (Dornbos and Mullen, 1992; Gibson and Mullen, 1996). Planting early can stimulate early initiation of R5 and lengthen the duration of the R5 through R6 period (Bastidas et al., 2008; Wilcox and Frankenberger, 1987). When planted early, R5 through R6 begins in warmer weather compared with soybean planted in late May or early June.

Delaying planting from late April or early May to June or July usually results in higher seed protein concentration (Kane et al., 1997), although Bastidas et al. (2008) reported an inconsistent effect of planting date on protein concentration. Oil and protein concentration can change according to maturity group (Yaklich et al., 2002) and cultivar (Bastidas et al., 2008), but environmental conditions seem to have the greatest effect.

Temperature during reproductive growth changes seed composition. Gibson and Mullen (1996) found that changes in protein and oil concentrations were most responsive to temperatures during R5 to R8. However, others report that temperatures 20 to 40 d before maturity were the most influential on seed composition (Wilcox and Cavins, 1992). High temperature during R5 through R6 increases oil concentration and generally decreases protein concentration (Dornbos and Mullen, 1992). Available moisture during R5 through R6 also is an important factor. As soil moisture declines during R5 through R6, protein concentration rises and oil concentration falls (Dornbos and Mullen, 1992; Foroud et al., 1993).

Planting in late March rarely occurs in central Indiana, because soils are wet and cold, and because of the threat of a killing frost. In West Lafayette, growing degree days (base temperature = 10°C) typically start accumulating on 23 March (30 yr mean). Previous research has not explored the effect of planting in late March or early to mid-April on yield and seed composition in the Midwest (De Bruin and Pedersen, 2008; Wilcox and Frankenberger, 1987). Although previous studies show optimal yield was attained when soybean was planted in late April or early May, these investigators did not evaluate earlier planting dates. Our hypothesis was that soybean planted in April and early May will produce a greater yield because of an increase in reproductive development time but seed protein concentration would decrease and oil concentration would increase. The objective of our study was to quantify the effects of planting date on soybean yield components, seed protein, and oil concentration.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
We conducted field experiments in 2006 and 2007 at the Purdue University Agronomy Center for Agricultural Research and Education near West Lafayette, IN (40°48' N, 86°99' W). Soil type was Drummer (fine-silty, mixed, mesic, Typic Haplaquoll) in 2006 and Chalmers (fine-silty, mixed, mesic, Typic Haplaquoll) in 2007. All fertility and pest management practices were conducted according to Purdue University recommended practices. Soil tests indicated adequate nutrient availability. Each year, soybean followed maize (Zea mays L.).

The experimental design was a split-plot arrangement with four blocked replicates. The whole-plot factor was planting date (30 March, 13 and 27 April, 10 and 30 May, and 6 June in 2006; and 27 March, 10 and 30 April, 9 May, and 1 and 7 June in 2007). Corresponding days of the year were 89, 103, 117, 130, 150, and 157 in 2006, and 86, 100, 120, 129, 152, and 158 in 2007. The subplot factor was cultivar: Pioneer brand 92M61 (Maturity Group 2.6), Becks brand 321NRR (Maturity Group 3.2), and Becks brand 367NRR (Maturity Group 3.7). Becks 321NRR served as the appropriate maturity group cultivar for the location of the study. Soybean was planted no-till in 38-cm rows at 370,000 plants ha–1. Plots measured 58 m long by 2.3 m wide (six rows).

Soil samples were collected for bulk density and water retention measurements at the completion of planting. Gravimetric soil moisture was taken each week from planting until physiological plant maturity from three blocks. Each planting date was sampled at depths of 0 to 5, 5 to 10, and 10 to 20 cm (Fig. 1 , 2 ). Soil water content was calculated on a mass basis.


Figure 1
View larger version (45K):
[in this window]
[in a new window]

  		Fig. 1. Daily average air temperature, precipitation, and cumulative precipitation during the 2006 soybean growing seasons, and gravimetric soil moisture for six planting dates (PD) during the growing season. Gravimetric soil moisture was measured at 0 to 5, 5 to 10, and 10 to 20 cm from planting to physiological maturity. Field capacity (FC) and permanent wilt point (PWP) are shown on each graph. Growth stages are provided for each cultivar for each planting date. The top row is Becks 367NRR (Maturity Group 3.7), middle row is Becks 321NRR (Maturity Group 3.2), and bottom row is Pioneer 92M61 (Maturity Group 2.6). The vegetative growth stages were similar among cultivars and were averaged.

 

Figure 2
View larger version (67K):
[in this window]
[in a new window]

  		Fig. 2. Daily average air temperature, precipitation, and cumulative precipitation during the 2007 soybean growing seasons, and gravimetric soil moisture for six planting dates (PD) during the growing season. Gravimetric soil moisture was measured at 0 to 5, 5 to 10, and 10 to 20 cm from planting to physiological maturity. Field capacity (FC) and permanent wilt point (PWP) are shown on each graph. Growth stages are provided for each cultivar for each planting date. The top row is Becks 367NRR (Maturity Group 3.7), middle row is Becks 321NRR (Maturity Group 3.2), and bottom row is Pioneer 92M61 (Maturity Group 2.6). The vegetative growth stages were similar among cultivars and were averaged.

 
Precipitation and air temperature records were obtained from a weather station located less than 2 km from the site. Vegetative and reproductive growth stages were recorded three times a week from 10 arbitrarily selected plants. We considered a plot to have reached a certain stage of development when at least half of the plants were at that stage (Fehr and Caviness, 1977). Seed was harvested from the four inner rows (a 1.5-m swath) for a length of 16 m from each plot at R8 using a Massey 8XP plot combine. Seed moisture and yield was recorded and yield was adjusted to 130 g kg–1 moisture.

Oil and protein concentrations were determined from machine harvested seed by near-infrared reflectance spectroscopy at the Purdue University Grain Quality Laboratory in 2006 and at the University of Wisconsin Grain Quality Laboratory in 2007. Both laboratories utilized the same laboratory techniques and calibration equations.

Hand harvesting was used to obtain yield components and plant growth characteristics. Plants were randomly harvested at R8 using a meter square quadrat. Ten plants were arbitrarily selected from the hand-harvested sample to determine the following yield components: main stem nodes m–2, main stem reproductive nodes m–2, pods reproductive node–1, seeds pod–1, seed mass (grams per 100 seeds), pods m–2, and percentage reproductive nodes. Seed mass was calculated similarly to the method of Board and Modali (2005), except we counted 300 seeds by hand in place of using an automatic seed counter. All other yield components were calculated according to the methods described by Board and Modali (2005).

Analysis of variance was conducted using the SAS GLM procedure (SAS Institute, 2004). Each year was analyzed separately due to heterogeneity of error variances between years for all response variables (two-tailed F test, P ≤ 0.01). Planting date and the planting date x cultivar sums of squares were partitioned into orthogonal polynomial contrasts. Mean separation tests for cultivar, planting date, and the planting date x cultivar interaction were performed using the least significant difference (P ≤ 0.05). Yield, yield components, and oil and protein concentration regression models were developed based on significant orthogonal polynomial contrasts. In most cases, a single regression model using the cultivar x planting date means was selected, producing a single r2 value. Linear-plateau models were considered as indicated by the data. When linear-plateau models were selected, the regression models and r2 values were determined for each cultivar separately. Comparisons of treatment means between years were performed by t tests (P ≤ 0.05).

Because the yield components that affect yield are often correlated, we used a three-way (pods m–2, seed mass, seeds pod–1) path analysis to determine the direct and indirect effects of these yield components on seed yield across years and cultivars. A bidirectional model was used because multiple yield components can form at the same time in indeterminate soybean. The ratio of the path coefficients provides an estimate of the relative influence of each yield component on yield.

Protein and oil concentration were related to temperature during R5 through R7. The mean daily minimum temperature, mean average temperature, mean daily maximum temperature, and growing degree days (base = 10°C) during R5, R6, and R7, as well as R5 through R6, R5 through R7, and R6 through R7 were also considered as predictors of seed composition. The REG procedure in SAS was used for regression analysis. Linear and quadric regressions were performed on mean values of seed composition traits. Model selection was based on model significance, maximization of the coefficient of determination (r2), and an assessment of the residuals.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
To test the impact of the earliest possible planting date on yield, we started planting in late March. The precipitation and temperature patterns were different in 2006 and 2007 (Fig. 1, 2). The temperature during May 2007 was above average. Temperature patterns differed greatly between years from July through September. This timing coincided with the R3 through R7 growth stages of soybean. In 2006, sufficient moisture was available for plant growth throughout the season. However, in 2007, 0- to 20-cm soil moisture levels declined below the permanent wilt point at critical growing periods in July and August because of long intervals with little or no rainfall and high temperatures. Reproductive growth stages varied in length by planting date and cultivar.

There were significant effects for planting date, cultivar, and planting date x cultivar interactions (Table 1 ). The effect of planting date and cultivar on yield components and growth characteristics were significant in each year except for seeds pod–1, main stem reproductive node m–2, and protein concentration. In 2007, a greater number of planting date x cultivar interactions occurred than in 2006.


View this table:
[in this window]
[in a new window]

  		Table 1. Levels of significance for yield components and plant growth characteristics for three soybean cultivars planted at six dates. Years were analyzed separately.

 
Yield was influenced by planting date in both years (Fig. 3 ). The influence of planting date varied among the cultivars, but the magnitude of the interaction was small. The influence of planting date on yield was explained by linear-plateau segmented models, linear models, or quadratic models. Averaged across cultivars, 2006 yield from plants seeded on Day 157 was only 76% of yield from plants seeded on Day 89. Yield declined rapidly when planting was delayed beyond the middle of May. Yield responded differently to planting date in 2007. Averaged across cultivars, the highest yield was from plants seeded on Day 120. Yield from plants seeded on Day 158 was only 87% of yield from plants seeded on Day 120, and yield from plants seeded on Day 86 was only 90% of yield from plants seeded on Day 120. Becks 367NRR had a similar trend in both years, having high yield at early planting dates. Across planting dates and cultivars yield was not different between years.


Figure 3
View larger version (34K):
[in this window]
[in a new window]

  		Fig. 3. Regression of yield (Mg ha–1) on planting date (day of year) for three soybean cultivars during 2006 and 2007. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7). Regression models were developed based on significant orthogonal polynomial contrasts, or as linear-plateau models when means for early planting dates were not different within a cultivar. Error bars represent the least significant difference for comparing cultivar means within each planting date, P ≤ 0.05.

 
Our yield results are similar to others showing that early planting, in late April or early May, increases yield when compared with later planting dates (De Bruin and Pedersen, 2008; Wilcox and Frankenberger, 1987). Our data supports the hypothesis that soybean yield would increase at early planting dates. Increased yield from early planting was mainly driven by increased pods m–2.

Pods m–2 responded to planting date in both years, except for Becks 367NRR in 2006 (Fig. 4 ). In 2006, pods m–2 decreased linearly as planting was delayed. Averaged across cultivars, pods m–2 from plants seeded on Day 157 were only 66% of pods m–2 from plants seeded on Day 103. The decline in pods m–2 for Becks 321NRR was less than Pioneer 92M61. In 2007, the response to planting day differed among the cultivars. The magnitude of the interaction was small. The relationship of pods m–2 to planting date was best explained by a quadratic response. Averaged across cultivars, pods m–2 from plants seeded on Day 158 was only 86% of pods m–2 from plants seeded on Day 120, and pods m–2 from plants seeded on Day 86 was 94% of pods m–2 from plants seeded on Day 120. Across planting dates and cultivars, pods m–2 was greater in 2006 than 2007.


Figure 4
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 4. Regression of pods m–2 (no.) on planting date for three cultivars during 2006 and 2007. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7). Regression models were developed based on significant orthogonal polynomial contrasts. Error bars represent the least significant difference for comparing cultivar means within each planting date, P ≤ 0.05.

 
The difference in yield between years for the first two planting dates was likely due to weather. In 2006, moisture was adequate. However, in 2007 the effect of dry soil during R5 and R6 (Fig. 2) reduced pods m–2 in Pioneer 92M61 and Becks 321NRR that had been planted on Days 86 and 100. Becks 367NRR was less affected by the dry soil, perhaps because it was a later maturity group than the other two cultivars (Fig. 4). The R5 and R6 growth stage for Becks 367NRR started later and lasted longer, which allowed this cultivar to benefit from late rain during a majority of R6. Early planting may promote high yield by lengthening the growing season, but this would depend on sufficient precipitation.

Early planting allowed the reproductive period to start earlier, when days were longer and light was more intense, than when planting later in the season (Fig. 1, 2). This can contribute to increased yield (Cooper, 2003). Planting in late March can be a risk, as we observed with high yield in 2006 and lower yield in 2007. Weather conditions throughout the season can have a large impact on soybean planted in late March, but cultivars of late maturity groups can minimize the effects of short periods of little or no rainfall because reproductive periods are longer.

Planting on Days 150 through 158 resulted in lower yield because the growing period from planting to physiological maturity was substantially reduced compared with early planting dates (Fig. 1, 2). Similarly, Wilcox and Frankenberger (1987) show days to maturity are reduced in late compared with early planting dates. We found that high temperatures during vegetative development shortened intervals between vegetative and reproductive growth stages. Hesketh et al. (1973) found that increased temperature decreases days between vegetative growth stages and increases days between reproductive growth stages, but their work was in the greenhouse with one cultivar.

Averaged across cultivars, seed mass increased for the last three planting days. However, seed mass of Becks 367NRR changed little in both years as planting was delayed (Fig. 5 ). Seed mass of Becks 321NRR and Pioneer 92M61 increased linearly in 2006 and 2007. Each cultivar had a different overall seed mass with Becks 321NRR > Becks 367NRR > Pioneer 92M61. Averaged across planting dates and cultivars seed mass was greater in 2007 than 2006.


Figure 5
View larger version (28K):
[in this window]
[in a new window]

  		Fig. 5. Regression of seed mass (g 100 seeds–1) on planting date for three cultivars during 2006 and 2007. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7). Regression models were developed based on significant orthogonal polynomial contrasts. Error bars represent the least significant difference for comparing cultivar means within each planting date, P ≤ 0.05.

 
Seeds pod–1 were not influenced by planting date (data not shown). In 2006, averaged across planting dates and cultivars seeds pod–1 averaged 2.3 and in 2007 averaged 2.6.

Main stem nodes m–2 responded to planting date in each year (Fig. 6 ). The variation in main stem nodes m–2 was best explained by a quadratic equation. Across cultivars, the fourth planting date had the greatest number of main stem nodes m–2 in each year. Across planting dates and cultivars, main stem nodes m–2 were greater in 2006 than 2007.


Figure 6
View larger version (34K):
[in this window]
[in a new window]

  		Fig. 6. Regression of main stem nodes m–2 (no.) on planting date for three cultivars during 2006 and 2007. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7). Regression models were developed based on significant orthogonal polynomial contrasts. Error bars represent the least significant difference for comparing cultivar means within each planting date, P ≤ 0.05.

 
The percentage of reproductive nodes on the main stem declined linearly as planting was delayed (data not shown). In 2006, across cultivars the percentage of reproductive nodes on the main stem decreased from 76.2 to 67.9% from Planting Day 89 to Day 157 and in 2007 decreased from 70.6 to 63.4% from Planting Day 86 to Day 158. Pioneer 92M61 had the highest percentage of reproductive nodes on the main stem in both years.

We observed that node number affected reproductive node number (data not shown), and reproductive node number affected pods reproductive node–1 (data not shown). This agrees with previous studies in determinate soybean (Board and Harville, 1993). A positive linear relationship exists between main stem node number and flower number. As flower number increases, pods m–2 increases (Egli, 2005). On average, we found that plants from earlier sowing dates had a higher percent of reproductive nodes, but had fewer main stem reproductive nodes. This was partially offset by more pods node–1.

Pods m–2, seed mass, and seeds pod–1 are all associated with yield, but it is often difficult to determine the influence of each yield component. Path analysis is used by plant breeders to distinguish relative importance of yield components. The ratio of the path coefficients is used to estimate the relative contribution of each yield component on yield. Averaged across years and cultivars, the three-way path analysis revealed that pods m–2 was generally the most important yield component, followed by seed mass and then seeds pod–1 (Table 2 ) for the first four planting dates (Day 86 through Day 130). The direct effect of pods m–2 was 1.2 to 3.3 times greater than the effect of seeds pod–1. The direct effect of seed mass was 1.5 to 2.8 times greater than the effect of seeds pod–1. For the last two planting dates (Days 150 to 158), pods m–2 became less important as seed mass contributed more to yield.


View this table:
[in this window]
[in a new window]

  		Table 2. Three-way path coefficient analysis of direct and indirect effects of planting date on soybean yield components averaged across years and cultivars. Ratio of path coefficients given to show the relative impact of each yield component.

 
We found that a wide range of planting dates caused plants to differentially compensate for yield by changing the importance of specific yield components. Others have found that pods m–2 have the greatest effect on yield (Singer et al., 2004). Seeds m–2 was a poor predictor of yield in our study (data not shown), but others have shown seed number to be a good predictor of yield (De Bruin and Pedersen, 2008).

Seed composition response to planting date differed among cultivars. Average oil concentration was 185 g kg–1 and ranged from 177 to 189 g kg–1 in 2006. The mean oil concentration was 206 g kg–1 and ranged from 199 to 211 g kg–1 in 2007 (Fig. 7 ). Oil concentration decreased as planting was delayed. Predicted oil concentration was generally highest at the second planting date in both years and decreased as planting was delayed. In both years oil concentration of Becks 321NRR responded more to planting date than did oil concentration in the other two cultivars. In both years, Becks 321NRR had high oil concentrations at the first planting date and relatively low seed oil concentrations by the last planting. By comparison, Becks 367NRR consistently had the lowest oil concentrations and these were relatively insensitive to planting date. When planted early, Pioneer 92M61 had intermediate (2006) to high (2007) oil concentrations, and exhibited little response to planting date. Across planting dates and cultivars oil concentration was greater in 2007 than 2006.


Figure 7
View larger version (29K):
[in this window]
[in a new window]

  		Fig. 7. Regression of oil (g kg–1) on planting date for three cultivars during 2006 and 2007. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7). Regression models were developed based on significant orthogonal polynomial contrasts. Error bars represent the least significant difference for comparing cultivar means within each planting date, P ≤ 0.05.

 
In 2006, mean protein concentration was 359 g kg–1 and ranged from 355 to 363 g kg–1; whereas in 2007, the mean protein concentration was 339 g kg–1 and ranged from 332 to 346 g kg–1 (Fig. 8 ). In 2006, protein concentration was inconsistent among cultivars and planting dates. In 2007, mean protein concentration generally increased as planting was delayed, although it declined from Planting Day 86 to Day 100. In both years, Becks 367NRR had the highest protein concentration and Pioneer 92M61 had the lowest protein concentration. In both years, protein concentration increased for Becks 367NRR as planting was delayed. In 2006, Becks 321NRR was relatively insensitive to planting date, but in 2007 it increased as planting date was delayed. Across planting dates and cultivars, protein concentration was greater in 2006 than 2007.


Figure 8
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 8. Regression of protein (g kg–1) on planting date for three cultivars during 2006 and 2007. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7). Regression models were developed based on significant orthogonal polynomial contrasts. Error bars represent the least significant difference for comparing cultivar means within each planting date, P ≤ 0.05.

 
Our results agree favorably with other researchers, who show a decrease in oil concentration and a general increase in protein concentration as planting is delayed (Bastidas et al., 2008; Kane et al., 1997). They indicated that the differences in the environment, namely temperature, causes seed composition to change.

Regression analysis indicated that the mean daily maximum temperature during R6 best explained the variation in oil concentration (Fig. 9 ). An F test indicated that the regression lines for the 2 yr were not different (P ≥ 0.05); therefore, years were combined. Oil concentration increased as the mean daily maximum temperature increased. The quadratic equation between oil concentration and mean daily maximum temperature during R6 was significant (P ≤ 0.0001, r2 = 0.92).


Figure 9
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 9. Relationship between protein concentration and oil concentration and mean daily maximum temperature (°C) during R6. Cultivars were Pioneer brand 92M61 (P92) (Maturity Group 2.6), Becks brand 321NRR (B321) (Maturity Group 3.2), and Becks brand 367NRR (B367) (Maturity Group 3.7), years were combined (n = 36).

 
The variation in protein concentration was best described by a linear relationship with the mean daily maximum temperature during R6 (Fig. 9). An F test indicated that the regression lines for the 2 yr were not different (P ≥ 0.05); therefore, data were combined across years. Protein concentration declined linearly as mean daily maximum temperature increased during R6. For every 1°C increase in mean daily maximum temperature, protein concentration decreased by 4.5 g kg–1. In 2006, oil decreased 5.1 g kg–1 for every 10 g kg–1 increase in protein (P = 0.032, r2 = 0.26). In 2007, oil decreased 9.6 g kg–1 for every 10 g kg–1 increase in protein (P ≤ 0.0001, r2 = 0.76).

Seed composition in our experiment was found to be affected by temperature in a similar manner to the response observed by Gibson and Mullen (1996) during R5 to R8. Howell and Cartter (1958) confirm oil concentration is most responsive to a short period of high temperature during R5 and R6. Temperature directly affects seed composition, but the effect of the time and duration of temperature is less understood.

Protein and oil concentrations compensate for one another. Thus, it is difficult to obtain high concentrations of both protein and oil. Early planting could increase both protein and oil concentrations in areas where mean temperatures during R5 through R6 are between 25 and 31°C (Piper and Boote, 1999).


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Soybean yield declined when planting was delayed until late May or early June because there were fewer pods m–2. Pods m–2 was the most important yield component for late-March through early May planting. Both pods m–2 and seed mass were both important for yield in late May and early June planting. Variation in seeds pod–1 had little effect on differences in yield. An increase in seed mass did not sufficiently compensate for the loss in pods m–2 and seeds pod–1 to maintain yield at the latter planting dates. Planting in late March proved risky in Indiana, but cultivar choice can minimize risk. Planting soybean in April or early May increased yield by increasing pods m–2, which was a result of an interaction of the number of reproductive nodes plant–1 and the number of pods node–1. Our results demonstrate that indeterminate soybean benefited from early planting in Indiana.

Planting date changes oil and protein concentration, because temperature during R6 differed by planting date. Early planting increased oil concentration and yield in soybean when the mean daily maximum temperature during R6 was highest. Planting in late May or early June resulted in higher protein concentration than planting in March, April, or early May. This was a result of lower mean daily maximum temperature during R6.


   ACKNOWLEDGMENTS
 
We would like to acknowledge the Indiana Soybean Alliance and the Indiana Crop Improvement Association for funding and support.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





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 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 Google Scholar
Google Scholar
Right arrow Articles by Robinson, A. P.
Right arrow Articles by Santini, J. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Robinson, A. P.
Right arrow Articles by Santini, J. B.
Agricola
Right arrow Articles by Robinson, A. P.
Right arrow Articles by Santini, J. B.
Related Collections
Right arrow Soybean
Right arrow Crop Growth and Development

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science 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