Published online 17 November 2005
Published in Agron J 97:1647-1657 (2005)
DOI: 10.2134/agronj2005.0095
© 2005 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
Economic Analysis
Economic Evaluation of Soybean Fungicide Seed Treatments
Paul Scott Poaga,
Michael Poppa,*,
John Rupeb,
Bruce Dixona,
Craig Rothrockb and
Carol Bogerb
a Dep. of Agricultural Economics and Agribusiness, 217 Agriculture Building, Univ. of Arkansas, Fayetteville, AR 72701
b Dep. of Plant Pathology, 217 Plant Science Building, Univ. of Arkansas, Fayetteville, AR 72701
* Corresponding author (mpopp{at}uark.edu)
Received for publication April 1, 2005.
 |
ABSTRACT
|
|---|
The effects of fungicide seed treatments on seeding rate, location, simulated rainfall at emergence, time of planting, and seed quality were analyzed for soybean [Glycine max (L.) Merr.] in this study. Variation in plant emergence allowed estimation of economically optimal seeding rates and partial returns (PR = Gross revenue – Seed cost) across seed treatment options. Study results proved a single seed treatment to be superior across most study conditions. In fact, a comparison of optimally treated to untreated seed revealed that a seemingly insignificant input in terms of cost (<$8.65 ha–1) enhanced profitability by an average of $43.71 ha–1 in this study. Using high rather than low quality treated seed increased producer returns by an average of $64.27 ha–1. Seeding rate recommendations need to be viewed with the precaution that added seed may be low cost insurance against lesser-than-expected survival rates. For the cultivar Hutcheson (MG V), planting in May compared with April and June provided better yields using less seed on average. Finally, as the planting season progressed, replanting plant population density thresholds decreased.
Abbreviations: FL, artificial flood • LOC, location • PM, planting in April or June compared to May • PPD, plant population density 4 wk after planting • PR, partial returns • ROS, rate of survival of a seed until 4 wk past planting • SQ, seed quality • ST, seed treatment • YR, experimental trial year
|
INTRODUCTION
|
|---|
THIS RESEARCH examined the cost effectiveness of seed treatments for managing seedling diseases. Seedling diseases are caused by numerous seed and soilborne pathogens, the most common of which are Pythium spp., Fusarium spp., and Rhizoctonia solani. Seedling diseases can lead to less than optimal plant populations and reduced plant vigor, which in turn can lead to reduced yields and higher weed control cost. In 2003, Arkansas farmers produced 1.17 million ha of soybean that produced 
3 million Mg. Seedling diseases caused an estimated 5% yield loss in 2003 (Koenning, 2004) and thus led to an estimated $32.4 million loss to producers in Arkansas.
Pathogens that cause seedling diseases on soybean have been characterized, but no feasible instrument exists to economically test for the presence of pathogens or to predict resulting yield impacts at the time of planting. Further, pathogen interactions with soil edaphic factors often make the seed treatment decision one associated with a lot of uncertainty and as a result some producers view fungicide seed treatment as inefficient. For this reason, a comprehensive study that identifies the economic feasibility of various seed treatments, protecting against different groups of pathogens under various resource conditions over time, may aid producers in making a more informed seed treatment decision before planting. Also, since seed cost and associated technology fees have made seed cost a greater percentage of operating costs (Lambert and Lowenberg-DeBoer, 2003), analyses surrounding seeding rate and replanting decisions for soybean are becoming more important to producers.
In this study the effects of four different seed treatments: Fludioxonil (Maxim), Carboxin–Tetramethylthiuram disulfide–Metalaxyl (Stiletto), Metalaxyl (Allegiance), and Carboxin + PCNB (Vitavax + PCNB) were compared with a control with no seed treatment.1 This allowed for the assessment of the importance of different pathogens based on the fungicides applied to seed (for example, Allegiance targets Pythium spp., Vitavax + PCNB targets Rhizoctonia solani, Maxim targets Rhizoctonia solani and Fusarium spp., while Stiletto is a broad spectrum fungicide that targets all three groups of organisms).
The specific objectives of the study were to: (i) estimate the relationship between seed treatment, plant population density (PPD) 4 wk after planting, and yield; (ii) calculate economically optimal seeding rates for different environmental conditions under investigation in this study; (iii) determine optimal seed treatment recommendations to producers that may face similar conditions analyzed in this study; and (iv) present economic PPD threshold levels for reseeding.
|
MATERIALS AND METHODS
|
|---|
Experimental Description
The data for this study were obtained from a 3-yr experiment at Stuttgart, AR (34°27' N lat; 091°24' W long), on Crowley silt loam [fine, smectitic, thermic Albaquultic Hapludalfs], a shallow silt loam soil that has poor water holding capacity; and Keiser, AR (35°39' N lat; 90°4' W long), on a Sharkey silty clay [very fine, smectitic, thermic Chromic Epiaquerts], a deep clayey soil with good water holding capacity. Aside from soil and location differences, other parameters examined included (i) seeding rate (100 or 200 seeds per 6.1-m row spaced 0.76 and 0.97 m apart, which translates to approximately 39.3 vs. 78.5 and 30.9 vs. 61.7 kg ha–1 of seed at Stuttgart and Keiser, respectively); (ii) seed quality (high quality seed or low quality seed by artificially altering seed quality to reflect planting of older seed and or seed that was improperly stored or physically damaged2); (iii) seed treatment (untreated, Stiletto, Maxim, Vitavax + PCNB, or Allegiance); (iv) planting month (April, May, or June3); and (v) simulated rainfall post planting (untreated or applying a manual flood to a depth of 2.5 cm when seed was cracking the soil surface and keeping the flood level at that point for 24 h). The experimental design was a randomized complete block with seed treatment and seed quality arranged factorially for each set of year, location, planting month, flooding, and seeding rate combinations. All plots were 3.05 m by 6.1 m for Stuttgart and 3.86 m by 6.1 m for Keiser. Yield observations, adjusted to 13% moisture, were taken from the two middle rows of four row plots to avoid border effects. Seedbed preparation, fertilizer, herbicide, and irrigation regiments were the same across all plots and in accordance with University of Arkansas cooperative extension recommendations (Ashlock, 1999). Table 1 summarizes data collection problems.
To avoid problems associated with seasonal or cyclical price effects, a 1994–2003 10-yr average price of $0.217 kg–1 less $0.006 kg–1 for marketing and transportation costs was used in this study (USDA-NASS, 2004). A seed price of $1.06 kg–1 for high quality, glyphosate resistant seed was considered representative of Arkansas conditions in 2004 as most soybean hectareage in Arkansas is grown using this technology (Windham and Marshall, 2004a). The Hutcheson variety, a common soybean variety in Arkansas, is not glyphosate-resistant and hence no significant seed treatment x seed variety interactions were assumed (i.e., different varieties would react to seed treatment and PPD in a similar fashion). Further, this study used a soybean seed count of 5500 seeds kg–1 and the seed price for low quality seed was set at 50% of high quality seed.
Since the cost of different seed treatments was relatively minor compared with other production costs (cost of seed treatment in 2003 was $0.068 kg–1 for Allegiance, $0.044 kg–1 for Vitavax + PCNB, $0.039 kg–1 for Maxim, and $0.080 kg–1 for Stiletto), the economic impact of the different seed treatments is mainly driven by yield effects.4 Since yields are expected to be mainly a function of plant densities or PPD as in previous studies (Ball et al., 2000a, 2000b; Wiley and Heath, 1969), the analysis focused on determining optimal PPD and indirectly optimal seeding rate for making seed treatment recommendations by location, planting month, and rainfall or soil moisture expectations past planting.
Model Estimation
Two response functions were estimated to solve for the optimal PPD and associated yield required for the economic analysis. The first response function was used to estimate seed establishment or the rate of survival (ROS) of a seed until 4 wk past planting. This was performed to estimate an average ROS for a range of seeding rates (rather than the two levels used in the experiment) across the study conditions. In turn, this allowed estimation of the marginal cost of an established plant at 4 wk past planting.5 The second response function modeled the yield response to PPD. Taking the first derivative of this yield function enabled determination of marginal physical product required to determine the revenue of an added established plant and hence economically optimal seeding rates.
The specification of these two equations was as follows:
 | [1] |
where ROS = PPD/SR was PPD in plants per 6.1-m row divided by the seeding rate (SR = 100 or 200 seeds per 6.1 m row), SQ was a zero or one dummy variable for seed quality (0 = untreated, high quality seed, 1 = seed treated to lessen seed quality), ST was a set of four zero or one dummy variables to compare seed treatment to the control without seed treatment, YR were two zero or one dummy variables for experimental trial year, FL was a dummy variable to control for artificial flood condition post planting, PM were two zero or one dummy variables for planting in April or June compared with May, and LOC was a zero or one dummy variable to adjust for differences in location. The baseline for the rate of survival equation was high quality, untreated seed planted in May 2003 at Keiser, AR, under nonflooded conditions.
The yield response function was specified as follows:
 | [2] |
where Y is the soybean yield achieved and the remaining variables and base line conditions are as defined in Eq. [1].
Economic Analysis
Profit Maximizing Plant Population Densities
For each of the experimental conditions modeled in this study the estimated ROS was used to determine the marginal cost of PPD (left side of Eq. [3]). The marginal cost of PPD, which is constant across different levels of SR, was subsequently equated to the marginal value product of PPD (right side of Eq. [3] and is hypothesized to decline as PPD increases) to determine the economically optimal PPD as follows:
 | [3] |
where w is the cost per seed (adjusted for seed treatment) and P is the soybean price. In other words, the optimal seeding rate is the one where added revenue from an extra plant no longer exceeds the cost of adding that plant.6 Since ROS = PPD/SR, the optimal seeding rate is as follows:
 | [4] |
The optimal PPD was then used to determine the estimated yield under various conditions using results from Eq. [2]. In turn, this estimated yield x soybean price – seed cost allowed calculation of PR to various seed treatments under the study conditions. Seed treatments could then be compared on the basis of estimated PR with the seed treatment consistently exhibiting highest PR chosen as the optimal seed treatment.
Because the ROS observations are between 0 and 100%, Eq. [1] was estimated using the two-limit Tobit maximum likelihood estimation technique (Greene, 2002) as linear least squares would have lead to estimated ROS beyond the feasible range. Eq. [2] was estimated using linear least squares. For both equations, data were pooled across treatments and all two-way interactions of independent variables were included.7 The White's heteroskedasticity consistent covariance matrix option available in Eviews v. 2.0 was used to correct for heteroskedasticity8 (Maddala, 2001; Hall et al., 1995). Nonlinear responses were hypothesized for PPD and yield (Ball et al., 2000a, 2000b; Duncan, 1986; Wiggans, 1939; Wiley and Heath, 1969; M. Popp et al., unpublished, 2005). Therefore, quadratic and square root functional forms of the yield response function were tested using Ramsey's Reset test for misspecification bias also available in Eviews v 2.0 (Studenmund, 1992; Hall et al., 1995).
Replanting Thresholds
Essentially, the question of replanting, if faced with a lesser-than-expected survival rate due to disease, soil crusting, flooding, etc., requires the establishment of replanting PPD thresholds for the initial planting. Figure 1
shows both revenue (Yield x Price) and partial returns (Yield x Price – Seed cost) functions for a hypothetical April and May planting date as a function of PPD. Subtracting replanting costs9 from expected R*May at the optimal PPD for a May planting (A), determines the PPD threshold (TApril) for the April planted crop (follow points A, B, C, to TApril). In other words, the producer replants if expected PR from replanting in a subsequent month are greater than PR expectations with a less than adequate PR associated with PPD < TApril for the initial planting.
Replanting charges can vary considerably since (i) seed suppliers may waive the technology fee on the second pass of seed; (ii) crop insurance programs exist that can be used to recover all or part of replanting charges (USDA Federal Crop Insurance Corp., 2005); (iii) different methods of preparing the field for replanting exist; and (iv) much uncertainty exists about the need for or success of replanting (e.g., spatial differences in PPD within a field may make the replanting decision difficult, weed control from herbicide activity from the initial planting may prevent adequate weed control for the second attempt, and seedbed condition may have deteriorated especially for late-season replanting). For these reasons, this study includes two levels of replanting charges to approximate the above situations. A low estimate of replanting charges is presented where seed is replaced at no cost (crop insurance payment is collected and technology fee is waived) whereas the high estimate includes all direct costs of replanting including the second pass of seed. This is intended to reflect a spectrum of producer choice where replanting would be triggered (i) relatively quickly with the low cost replanting scenario or (ii) only rarely to portray the situation where a producer hesitates to replant due to the cost and uncertainty of replanting success.
|
RESULTS
|
|---|
Model Estimation
The final estimates of the seed establishment equation (Eq. [1]) are presented in Table 2. The model accounted for approximately 73% of the variation in the rate of survival over a wide range of environmental conditions and their interactions. Many of the independent variables exhibited statistically significant responses either by themselves or as interactions. The signs on the coefficients were as expected (a reduction in survival rate with lower quality seed and excessive rainfall as well as improvements in seedling survival with seed treatment). Table 3 highlights estimated survival rates (ROS) for each location, year, soil moisture (flood or nonflood), seed quality, planting month, and seed treatment. Ignoring interaction effects, Stiletto had the most beneficial impact on survival, followed by Allegiance, Maxim and Vitavax + PCNB (see coefficients in Table 2). It is noteworthy that treatment of seed did not guarantee higher ROS compared with untreated seed. Seed treatment appeared to be less effective under flooded conditions while it enhanced seedling survival for low quality seed.
The yield response function to PPD shown in Table 4 exhibited similar explanatory power as Eq. [1]. By contrast, the coefficient estimates associated with seed treatment and seed quality were insignificant (with the exception of ST x A). This suggested that seed treatment and seed quality had little effect on yield once seedlings were established or that soybean plants can compensate for varying growing and plant conditions throughout the growing season (Ball et al., 2000a; Wiley and Heath, 1969). Highly significant coefficients on PPD in conjunction with expected signs supported a good model fit. While the Ramsey Reset statistic suggested some misspecification bias, the statistics for the square root functional form were superior to those of the quadratic functional form and similar to observations made by Popp et al. (unpublished, 2005).
Economic Analysis
Profit Maximizing Plant Population Densities
Using coefficients from Tables 2 and 4, the optimal economic PPD, resultant seeding rate recommendations, and expected yields for each of the study conditions are presented in Tables 5 through 8. The tables list estimated yields and partial returns using the optimal seeding rate for the different locations, year, seed treatments, seed quality, and planting month.10 Averaging partial returns across study conditions by seed treatment alternative resulted in Allegiance showing the highest partial returns at $541.20 ha–1 followed by Stiletto ($518.91 ha–1), Maxim ($515.74 ha–1), untreated seed ($502.84 ha–1), and Vitavax + PCNB ($502.50 ha–1). Partial return differences between seed treatment options and the untreated seed ranged anywhere from –$67.14 to $230.97 ha–1 (avg. $43.71 ha–1) in favor of seed treatment. This suggests that the choice of seed treatment is not trivial.
Table 9 summarizes the information of Tables 5 through 8 by presenting the likelihood of choosing the optimal seed treatment on the basis of highest partial return by location (Keiser both and Stuttgart both), by seed quality (high quality and low quality), and overall. These statistics suggest that Allegiance is likely the best seed treatment option since partial returns are highest using this seed treatment under most circumstances. Further, in cases where Allegiance was not optimal, the next best choice exceeded Allegiance by an average11 of $4.05 ha–1 for untreated, $26.93 ha–1 for Vitavax + PCNB, and $19.35 ha–1 for Stiletto. This is the average regret the producer would have experienced for conditions where Allegiance was not optimal. The only location and seed quality combination where Allegiance would not have been optimal on average is for Keiser using low quality seed. Under this condition choosing Allegiance incorrectly led to average partial returns that were $16.04 ha–1 less than those observed for using Stiletto.
Since a deviation from Allegiance as the optimal treatment in Table 9 was only close for Stiletto, the results are deemed very robust across study conditions. Further, the large difference in partial returns across treatments suggested that potential seed treatment cost changes would not affect the outcome of this research.
Since Allegiance appeared to outperform the other seed treatments except at Keiser using low quality seed (Stiletto), Table 10 reports estimated yield, seeding rate, and partial returns achieved using these recommended seed treatments. For corresponding planting months and years, a direct comparison of partial returns by seed quality revealed that producers should attempt to use high quality seed. Partial returns using high quality seed averaged $64.27 ha–1 higher than partial returns using low quality seed. Further, use of high quality seed was always superior to low quality seed by a difference ranging anywhere from $9.34 to $231.74 ha–1.
University of Arkansas soybean production budgets for 2004 report operating cost requirements of approximately $518.93 ha–1 ($197.69 of the $518.93 is cash rent) beyond seed cost to grow soybean in the short run (Windham and Marshall, 2004b, 2004c; Hill et al., 2003). This threshold is met for a majority of the time (i) when using high quality seed; (ii) more so at Keiser than Stuttgart; and (iii) more so in May compared with April or June.
The optimal seeding rates from Table 10 suggested that planting in May required less seed to produce optimal yields. This month's average seeding rate across location, soil moisture conditions, year, and seed quality was 54.98 (kg ha–1) with a range of 40.39 to 80.78 (kg ha–1). Seeding rate recommendations were also generally higher for flooded conditions compared with nonflooded conditions. Since the rates vary by year, location, planting month, and soil moisture stress, no specific seeding rate recommendations can be made. If a decision maker treats the average seeding rate across years observed in this study as a recommendation, the shape of the soybean yield response function to PPD suggests the following (see the rapid decline in PRApril below PPD*April compared with the less pronounced decline at PPD > PPD*April in Fig. 1): (i) seeding at a higher than recommended rate affords some insurance against lesser-than-expected seed survival at relatively low cost (additional seed generates additional yield which offsets the cost of the added seed); (ii) seeding at a lesser-than-recommended rate, however, is expected to lead to a comparatively large negative yield impact, which is offset to a lesser extent via seed cost savings. A producer may thus be advised to be liberal when picking an appropriate seeding rate as presented in Table 10.
Replanting Thresholds
Table 11 shows the PPD thresholds for April and May plantings (using both low and high replanting charges) below which replanting in a subsequent month becomes profitable using high quality seed treated with Allegiance. Of note, in most cases, were large differences in profit maximizing PPD and the replanting PPD thresholds especially when the replanting costs are born entirely by the producer. Also, May replanting PPD thresholds as a percentage of optimal PPDs were lower than those of April. This was likely a reflection of the relatively low partial returns associated with June plantings and may in part be due to the chosen maturity group of the soybean variety in the trial (i.e., matching soybean maturity to planting month may impact the results shown here). Overall, producers may need to learn that it is more profitable to have a field with a less than desired plant population that will eventually do well, compared with replanting to show a more aesthetically pleasing soybean field. This may be especially true if replanting is associated with a partial loss of weed control (i.e., the first crop's herbicide application prevents timely application of weed control for the replanted crop) and therefore much greater uncertainty about crop success exists with replanting (Wasilas et al., 1990).
View this table:
[in this window]
[in a new window]
|
Table 11. Reseeding plant population density (PPD) thresholds for April and May plantings in plants per 1-m length of row using allegiance.
|
|
|
DISCUSSION
|
|---|
This study summarizes findings of soybean seed treatment studies conducted at Keiser and Stuttgart under varying environmental conditions from 2001 to 2003. The conditions reported in these trials are expected to be representative of Arkansas soybean production.
The study suggested that a producer should plant in May, using high quality seed treated with Allegiance to maximize partial returns. The fungicide seed treatment decision is robust across location, planting month, and soil moisture condition as well as possible changes in seed treatment cost. A partial return comparison of optimally treated to untreated seed revealed that a seemingly insignificant input choice, costing less than $8.65 ha–1, can enhance profitability by an average of $43.71 ha–1 across all study conditions.
Producers were also advised to view seeding rate recommendations with the precaution that added seed may provide low-cost insurance against lesser-than-expected seedling survival. The opposite, planting at lesser than recommended rates, is costly, however, as yield losses may be much greater than seed cost savings.
The use of low quality seed should be avoided as partial returns are always inferior compared with returns associated with high quality seed. If low quality seed has to be used, the recommendation of Allegiance holds for Stuttgart, but changes to Stiletto at Keiser. Producers should therefore not use seed from the previous production year and are encouraged to ensure proper seed storage to avoid excessive heat exposure to seed for late plantings.
The replanting analysis indicated economical PPD thresholds for replanting that declined as a percentage of optimal PPDs with progress in the planting season. Part of this result may be a function of using the same variety of soybean regardless of planting date. Especially when a high replanting charge is used to reflect producer costs and/or uncertainty, the replanting thresholds were quite low and suggested that producers, interested in profit maximization, may need to lower their aversion to having a field that begins with a sparser PPD.
This study did not account for effects of soybean maturity group as well as effects of additional changes in seeding rate (within row spacing as well as distance between rows). For shorter-season varieties (not analyzed in this study), earlier seeding dates as well as narrower rows would likely affect reseeding thresholds. Also, careful monitoring of production costs (post planting) with different seeding dates would likely show irrigation and perhaps herbicide application differences across planting dates. These issues were not accounted for in this study.
|
ACKNOWLEDGMENTS
|
|---|
The authors appreciate the support from the Arkansas Soybean Promotion Board as well as the University of Arkansas, Division of Agriculture on this project. Additional thanks go to the agricultural experiment station staff at Keiser and Stuttgart, AR.
|
NOTES
|
|---|
1 Although many combinations of different application rates and seed treatment combinations can be used, they were not examined in this study. 
2 Seed was manually aged using Nanayakkara's (2001) procedure by exposing seed to 40°C for 14 d at 12.9% moisture.
3 Planting dates in 2001 were 16 May in Stuttgart only. In 2002 at both Keiser and Stuttgart the planting dates were 15 April, 16 May, and 15 June. Finally, the planting dates for 2003 in Keiser were 18 April, 27 May, and 9 June as well as 16 April, 1 June, and 23 June at Stuttgart.
4 Preliminary analysis on yield effects of seed treatments using cumulative density functions generated using Schlaifer's (1959) methods showed that seed treatments had yield effects. More detailed stochastic dominance analyses were not deemed appropriate, given a somewhat unbalanced data set (Poag, 2005).
5 Given these restrictions, the marginal cost of an added established plant is assumed to be constant across initial seeding rates.
6 Alternatively, the producer is expected to maximize partial returns: PR = P x Y(PPD) – w x SR. Differentiating PR with respect to SR yields, P x 
Y/PPD x PPD/SR – w, where PPD/SR = ROS, and, setting PR/SR = 0 leads to P x Y/PPD = w/ROS.
7 Inclusion of additional interaction terms would have introduced additional multicollinearity while adding little additional explanatory power.
8 Heteroskedasticity was tested using White's test (Eq. [2]) and found to be significant at the 0.001 level.
9 Replanting costs were estimated to be $14.63 ha–1 using tillage to incorporate the first crop. The second pass of seed (also treated and at the optimal seeding rate for the subsequent month) was added to the above so that total direct replanting costs encompass seed, labor, fuel, repair, and maintenance for required planting (standard planter) and tillage (field cultivator) equipment (Windham and Marshall, 2004a, 2004b, 2004c). The replanting cost estimate does not include ownership costs of equipment; those costs are considered fixed and would not change whether planting and tillage equipment are used once or twice for a crop during a production season.
10 Additional information showing estimated seeding rates, yields, and partial return comparisons to untreated seed for all seed treatments is available from the authors upon request. Seed cost can be calculated by: Seeding rates x Seed cost.
11 Note that this average is based on 2, 2, and 11 of 50 possible study treatment combinations for untreated, Vitavax + PCNB, and Stiletto, respectively.
|
REFERENCES
|
|---|
- Ashlock, L. 1999. Arkansas soybean handbook. Misc. Publ. 197. Univ. of Arkansas Coop. Ext. Serv., Little Rock, AR.
- Ball, R.A., L.C. Purcell, and E.D. Vories. 2000a. Optimizing soybean plant population for a short-season production system in the southern USA. Crop Sci. 40:757–764.[Abstract/Free Full Text]
- Ball, R.A., L.C. Purcell, and E.D. Vories. 2000b. Short-season soybean yield compensation in response to population and water regime. Crop Sci. 40:1070–1078.[Abstract/Free Full Text]
- Duncan, W.G. 1986. Planting patterns and soybean yields. Crop Sci. 26:584–588.[Abstract/Free Full Text]
- Greene, W. 2002. Econometrics modeling guide. Vol. 2. Limdep Version 8.0. Econometrics Software, New York.
- Hall, R.E., D.M. Lilien, G. Sueyoshi, R. Engle, J. Johnston, and S. Ellsworth. 1995. EViews: User guide. Quantitative Micro Software, Irvine, CA.
- Hill, J., M. Popp, and P. Manning. 2003. Focus group survey results: Typical Arkansas crop producer production and marketing practices. Res. Rep. 971. Univ. of Arkansas Exp. Stn., Little Rock, AR.
- Koenning, S. 2004. Proc. of the Southern Soybean Disease Workers: Southern United States Soybean Disease Loss Estimate for 2003, St. Louis, MO. 15–16 Feb. 2004. Southern Soybean Disease Workers, North Carolina State Univ., Raleigh, NC.
- Lambert, D.M., and J. Lowenberg-DeBoer. 2003. Economic analysis of row spacing for corn and soybean. Agron. J. 95:564–573.[Abstract/Free Full Text]
- Maddala, G.S. 2001. Introduction to econometrics. 3rd ed. John Wiley & Sons, New York.
- Poag, P.S. 2005. Economics of Arkansas Soybean seed treatments. M.Sc. thesis. Dep. of Agric. Econ. and Agribusiness, Univ. of Arkansas, Little Rock, AR.
- Nanayakkara, R. 2001. Influence of soybean cultivar, seed quality, and temperature on seed exudation and Pythium disease development. Ph.D. diss. Dep. of Crop, Soil and Environ. Sciences, Univ. of Arkansas, Little Rock, AR.
- Schlaifer, R. 1959. Probability and statistics for business decisions: An introduction to managerial economics under uncertainty. McGraw-Hill, New York.
- Studenmund, A.H. 1992. Using econometrics a practical guide. 2nd ed. Harper Collins Publ., New York.
- USDA Federal Crop Insurance Corp. 2005. Soybean loss adjustment standards handbook. Available at www.rma.usda.gov/FTP/Publications/directives/25000/pdf/05_25440-1H.pdf (accessed 23 May 2005; verified 24 Aug. 2005). USDA, Washington, DC.
- USDA-NASS. 2004. Arkansas Statistical Office: Soybean acreage, yield, production, price and value. Available at www.nass.usda.gov/ar/fildsoys.PDF (accessed 5 Feb. 2004; verified 24 Aug. 2005). USDA-NASS, Washington, DC.
- Wasilas, B.L., M.A. Jacob, and G.E. Pepper. 1990. Stand reductions, replanting and offset row effects on soybean yield. J. Prod. Agric. 3:120–123.
- Wiggans, R.G. 1939. The influence of space and arrangement on the production of soybean plants. Agron. J. 31:314–321.[Free Full Text]
- Wiley, R.W., and S.B. Heath. 1969. The quantitative relationships between plant population and crop yield. Adv. Agron. 21:281–322.
- Windham, T.E., and J. Marshall. 2004a. Soybeans, border irrigated following wheat, loamy soils, Roundup Ready. Enterprise budget AG-786–12–03. Univ. of Arkansas Coop. Ext. Service, Little Rock, AR.
- Windham, T.E., and J. Marshall. 2004b. Soybeans, flood irrigated following rice, loamy soils. Enterprise budget AG-789-12-03. Univ. of Arkansas Coop. Ext. Service, Little Rock, AR.
- Windham, T.E., and J. Marshall. 2004c. Soybeans, furrow irrigated, loamy soils. Enterprise budget AG-792-12-03. Univ. of Arkansas Coop. Ext. Service, Little Rock, AR.
TOP
ABSTRACT
INTRODUCTION
