Crop Science 41:993-998 (2001)
© 2001 Crop Science Society of America
CROP BREEDING, GENETICS & CYTOLOGY
Genetic Progress in Soybean of the U.S. Midsouth
Ali Ustun*,a,
Fred L. Allenb and
Burton C. Englishc
a Blacksea Agric. Res. Institute, P.O. Box 39, Samsun, Turkey
b Univ. of Tennessee, Dep. of Plant and Soil Sciences, P.O. Box 1071, Knoxville, TN 37901
c Univ. of Tennessee, Dep. of Agricultural Economics and Rural Sociology, P.O. Box 1071, Knoxville, TN 37901
* Corresponding author (ustunali{at}hotmail.com)
 |
ABSTRACT
|
|---|
Yield increase in crops has occurred due to plant breeding and improved production and management techniques. The amount of increase due to plant breeding in soybean in the southern USA has not been determined. The objective of this study was to evaluate the genetic improvements in soybean yield and other important agronomic traits. ‘CNS’ and ‘S-100’ were chosen as representative ancestral lines from the 1940s. ‘Ogden’ and ‘Lee’, ‘Hill’ and ‘Essex’, and ‘TN 5-85’, and ‘Hutcheson’ were released in early 1950s, late 1950s and early 1970s, and late 1980s, respectively. The experiments were conducted in Knoxville Experiment Station in Knoxville, TN and at the Ames Plantation near Grand Junction, TN, for 3 yr, and at the Milan Experiment Station in Milan, TN, and at the Highland Rim Station in Springfield, TN, for 2 yr. Results showed that cultivar improvement increased soybean yield 14 kg ha-1 per year. Yield increase over time was linear. This linear increase demonstrates that a yield plateau in the U.S. Midsouth has not been reached in soybean [Glycine max (L.) Merr.]. Most of the increase has come via genetically related elite by elite parent crosses. However, infusion of exotic germplasm into elite germplasm can often promote increases in yield, as in the development of Hutcheson. Current cultivars showed better yield stability and more response to favorable growing conditions than ancestral lines. Genetic improvement resulted in shorter plant height and decreased lodging. Oil content was increased until the 1980s, but then started to decrease. However, because of correlated effects, protein content started to increase after the 1980s.
|
INTRODUCTION
|
|---|
ONE WAY to increase yield significantly is through genetic manipulation of crops. Soybean breeders have been able to increase the seed yield and improve other traits such as lodging, protein content, and oil content. Soybean yield increased from 1350 to 2250 kg ha-1 in the last 60 yr in the USA (FAO, 1998). This increase cannot be attributed to only cultivar improvement, because management practices have also played a significant role in yield increase.
Accurate estimates of genetic progress realized over time for a given trait is a difficult task. The number of genotypes and choice of cultivars for experimentation may change the estimate. Cox et al. (1988) suggested that evaluation of cultivars in common environments provided the most direct estimate of breeding progress. Slafer et al. (1994) stated that an experiment for this purpose should satisfy the following conditions: (i) experiments must be conducted under field conditions, (ii) measurements must be made on plots, not on single plants, and (iii) cultivars released at different times must be compared simultaneously.
In the north central region of the USA, the contribution of soybean breeding to yield increase was between 10 and 21 kg ha-1 per year from 1902 to 1977 in maturity groups 00 through IV (Specht and Williams, 1984). However, when they summarized only cultivars from hybridizations after 1939, the average annual gain was 12 kg ha-1. Williams and Specht (1979) concluded, based on the same study, that 79% of the yield increase in the northern states could be attributed to genetic improvement. Wilcox et al. (1979) compared five cultivars and ancestral lines from both maturity groups II and III. They found that the cultivars gave 25% more seed yield compared to ancestral lines. Protein content decreased over time, but this was accompanied by an increase in oil content. Regression coefficients for cultivars were similar in estimation of yield stability.
Luedders (1977) tested 21 first- and second-cycle soybean cultivars to estimate genetic improvement. It was estimated that the increase in yield for the first and second cycles was 26 and 16%, respectively. He showed that lodging resistance was improved substantially in commercial cultivars in maturity groups II, III, and IV. During this time, only a slight increase in plant height occurred.
Boerma (1979) tested 18 southern cultivars from maturity groups VI, VII, and VIII. He reported significant yield increases from 1942 to 1973. This increase was 0.7% (13.7 kg ha-1) per year. This yield increase was associated with increasing pod number. Seed size and number of seeds per pod did not show significant change over time. Plant height was decreased in maturity group VIII but did not change in groups VI and VII.
Karmakar and Bhatnagar (1996) compared cultivars developed from 1969 to 1993 in India. They found that the annual genetic gain was 22 kg ha-1. Voldeng et al. (1997) tested 41 soybean cultivars from maturity groups 0, 00, and 000 to document the genetic improvement of soybean in Canada. These cultivars were released between 1934 and 1992. They found an annual yield increase of 0.5% (9.3 kg ha-1 yr-1) until 1976, and after 1976 it was 0.7% (13 kg ha-1 yr-1). In their study, it was observed that lodging tolerance was increased without any decrease in plant height. Yield stability of cultivars was similar over time.
In the U.S. Midsouth, the soybean acreage has decreased since the 1970s, although yield improvements have occurred in soybean. It is not known how much of the yield increase has been caused by genetic improvement or advances in technology or growing soybean in better environments. Similar studies have been conducted for northern soybean cultivars. In the south, a similar study was conducted by Boerma (1979), but it included cultivars from maturity groups VI to VIII. There has not been any investigation for maturity group V and VI soybean which is grown in the U.S. Midsouth. This study was conducted to estimate in the U.S. Midsouth (i) the contribution of public soybean breeding programs to yield increase, and (ii) what changes occurred in other important agronomic traits.
|
MATERIALS AND METHODS
|
|---|
Two ancestral lines and six soybean cultivars representing different eras of development were chosen for this study. Ancestral lines, and first, third and fifth-generation cultivars were each represented by two cultivars that have been produced on significant acreage in southern states (Table 1). First generation cultivars were developed by crossing ancestral lines. Crossing of the first generation cultivars produced second generation cultivars and so on (Allen and Bhardwaj, 1987).
The experiments were conducted in a randomized complete block design with three replications. Experiments were carried out at the Knoxville Experiment Station in Knoxville, TN and the Ames Plantation near Grand Junction during 1996 to 1998, and the Milan Experiment Station in Milan, TN, and the Highland Rim Experiment Station in Springfield, TN, during 1997 and 1998. The soil type at Knoxville, Ames Plantation, Milan, and Springfield was a Sequatchie silt loam (fine-loamy, siliceous, thermic, Humic Hapludults), Loring silt loam (fine-silty, mixed, active, thermic Oxyaquic Fragiudalfs), Collins silt loam (coarse-silty, mixed, active, acid, thermic Aquic Udifluvents), and Dickson silt loam (fine-silty, siliceous, semiactive, thermic Glossic Fragiudults), respectively. Cultivars were seeded in four rows 6.1 m long with a 76 cm spacing between rows. Experiments were planted in mid-May with the exception of Milan 1998 which was seeded in June.
Maturity, plant height, lodging, protein and oil content, and yield were recorded. Maturity was recorded as the date when approximately 95% of the pods had reached their mature pod color. Plant height was measured at maturity as being the distance between soil surface and the apex of the main stem. Protein and oil were analyzed with a NIR food and feed analyzer at the USDA-ARS , National Center for Agricultural Utilization Research, Peoria, IL. Protein and oil contents were expressed on a seed dry weight basis. Lodging was scored on a 1 to 5 scale; 1 being all plants erect and 5 being 95+% of plants prostrate. Yield data were taken from the two center rows of each plot after 50 cm from the ends of the rows were discarded. Yields were adjusted to 130 g kg-1 moisture content.
SAS Procedures MIXED (SAS, 1998) and GLM were run to obtain analysis of variance for different traits and Procedure REG was used to estimate regression parameters. In the mixed model, cultivars and locations were designated as fixed factors and years and blocks as random factors. In combined analysis of variance, each experiment was treated as one environment. In regressions for the genetic trends, least square means of agronomic traits were used as the dependent variable while experiment means were used as the independent variable. Stability analyses were carried out by using experiment means as environmental indices (Eberhart and Russell, 1966). The mean yield of each cultivar in each experiment was regressed on the environmental index.
|
RESULTS AND DISCUSSION
|
|---|
Cultivars differed from each other significantly for all traits. Significant differences were also found between environments which was the pooled effects of years and locations. Cultivar by environment interactions were also significant (Table 2).
View this table:
[in this window]
[in a new window]
|
Table 2. Mean squares (MS) and degrees of freedom (df) for the combined analyses of variance of yield, protein content, and oil content for soybean cultivars/lines representing different eras in the U.S. Midsouth.
|
|
Yield has been the major objective in most soybean breeding programs. This study demonstrates that significant yield increases were achieved by soybean breeders in the U.S. Midsouth. The mean yield of TN 5-85 and Hutcheson (Fifth generation cultivars) was 32.3, 20.0, and 16.6% higher than that of ancestral lines, and first and third generation cultivars, respectively (Fig. 1). Soybean yield was increased from 1972 to 2609 kg ha-1 from ancestral lines to fifth generation cultivars. The dramatic increase in yield occurred with the development of Essex. Essex in the southern states could be considered as a milestone in the soybean breeding history of southern USA.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1. Yield increase of soybean ancestral lines/cultivars representing different eras in the midsouthern states of the USA. **Significantly different from zero (P  0.01).
|
|
The annual gain was 14 kg ha-1 and has been maintained (Fig. 1). The rate of gain in this study is lower than that reported by Karmakar and Bahatnagar (1996), but it is consistent with the rate of Luedders (1977), Wilcox et al. (1979), Boerma (1979), and Specht and Williams (1984). Frederick et al. (1990)(1991) compared two old (‘Manchu’ and ‘Dunfield’) and two modern (‘Williams 82’ and ‘Clark 63’) soybean cultivars under drought stress and irrigated conditions. Although older cultivars were found to be more efficient in conserving water under drought, the yield increase from old cultivars to modern cultivars was 31 and 9% under irrigated and drought conditions, respectively. In general the annual genetic gain for yield in the southern and northern USA is similar.
It has been claimed (Waddington et al., 1986; Schmidt, 1984; and Schmidt and Worrall, 1984) that genetic improvement has reached the yield plateau, as in wheat (Triticum aestivum L.). This study shows no indication for a yield plateau in soybean. The yield level of the cultivars evaluated in this study increased progressively over generations. Most of the cultivars were derived from genetically related elite by elite parent crosses. However, Hutcheson has the pedigree, V-68-1034 x Essex. V-68-1034 was derived from a York x PI 71506 cross (Buss et al., 1988). The pedigree and yield of Hutcheson is a good example of genetic progress from an elite by elite cross, following the earlier infusion of plant introduction germplasm into one of the elite parent lines.
In low-yielding environments, Hutcheson and Essex gave the highest yield and Ogden gave the lowest in these conditions. The mean of cultivar generations was significantly different in both low- and high-yielding conditions. However, yield differences were more pronounced in higher-yielding environments. Hutcheson was the most desirable cultivar for yield and yield stability, because it had higher or equal yield in any given environment than the others (Fig. 2). In higher-yielding environments, newer cultivars (Hutcheson, TN 5-85, and Essex) were superior to older cultivars.
When yield stability of cultivars in this experiment were examined, the apparent difference was between cultivars and ancestral lines. Ancestral lines had a coefficient less than one, while cultivars had coefficients equal to one or greater than one. The slopes of CNS and TN 5-85 were significantly different from one while those of S-100 and Essex were on the border for significance. Essex and TN 5-85 were more responsive to better growing conditions than other cultivars and lines. On the other hand, ancestral lines, CNS and S-100, were the least responsive to change in environmental conditions. Wilcox et al. (1979) and Voldeng et al. (1997) obtained results that conflict with our findings for yield stability. The reason for this conflict might be different cultivars and test locations in those studies. A coefficient less than one reflects that the genotype is less responsive as the environment changes from low yielding to high yielding.
Protein and oil content are both very important traits (Burton, 1985 and 1991, and Pantalone et al., 1996), but they are negatively correlated with each other and negatively correlated with yield. This correlation is not very strong. It seems that oil content increased at the expense of protein until 1980s, when oil content reached around 19.5% in the southern cultivars. The protein content of cultivars showed a different trend compared to the oil content. Although the difference among cultivars and ancestral lines was significant for protein content (Table 2), the slope of protein change over time was not significant. S-100 had the highest protein content; whereas Hill and Hutcheson had the lowest protein content (P < 0.05, Fig. 3). For oil content, Hutcheson and Lee had the highest value while Ogden, Hill, Essex, and TN5-85 yielded almost the same percent oil content and CNS produced the lowest oil content. The largest difference in oil content was between ancestral lines and cultivars (P < 0.05, Fig. 3). Wilcox et al. (1979) showed that oil content in soybean over time was increased, but protein content was decreased. Considering that their study was conducted before 1980, our results and their results are consistent for the same period. In these two traits, negative correlations and low genetic variation for selection seem to limit the genetic progress. Although the negative correlation exists between two traits, it is possible to create a cultivar with relatively high protein and oil content such as Essex.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3. Changes in protein and oil content of soybean cultivars/ancestral lines representing different eras in U.S. Midsouth. *Significantly different from zero (P 0.05).
|
|
Over generations, maturity shifted towards earlier genotypes. From ancestral lines to the fifth generation cultivars, maturity was shortened by 10 d. In the mid-south, early cultivars often have less weather-related risks compared to late cultivars, and they also give producers flexibility to adjust labor. The decrease in days to maturity did not occur at the expense of seed yield.
Plant height and lodging are interrelated in most cases. In the U.S. Midsouth, plant height decreased until the 1970s, but it began to increase after that (Fig. 4). This might be due to increase in lodging resistance in taller plants. When sufficient success has been achieved for lodging, a good strategy would be to increase plant height, leading to higher yield potential. Soybean breeders first selected shorter plants to reduce lodging. Plant height averaged 97 cm in ancestral lines, and it was approximately 80 cm in the fifth generation cultivars. Plant height was slightly increased in cultivars tested by Luedders (1977) and Voldeng et al. (1997). Their material consisted of maturity groups 000 to IV. Within the same environment early cultivars tend to have shorter plants. Therefore, plant height may not be associated with lodging in early cultivars. In northern material, less emphasis might have been given to plant height compared to southern soybean material. Boerma (1979) showed that plant height was decreased only in maturity group VIII. Lodging showed a significant linear decrease over time. Lodging was decreased over time in soybean (Luedders, 1977; Voldeng et al., 1997). Plant height and lodging are the major plant traits that affect harvest losses. Data from this research and other studies imply that breeders spent some effort to increase yield by avoiding losses during mechanical harvesting.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 4. Changes in plant height of soybean cultivars/ancestral lines representing different eras in U.S. Midsouth. *Significantly different from zero (P 0.05).
|
|
|
CONCLUSIONS
|
|---|
Yield in crops is the major target in most breeding programs. The sustainability of yield increase by cultivar improvement is of interest. Genetic progress achieved over time should be estimated to plan further plant breeding strategies in soybean. In this study, average yield of the cultivars tested was approximately 27% higher than ancestral lines and 20% higher than first generation cultivars. The linear increase in yield over time emphasizes the steady genetic progress that soybean breeders have made in developing cultivars for the mid-south. Soybean yield increase over cultivar generations showed that a yield plateau has not been reached in the U.S. Midsouth. When yield stability of ancestral lines and cultivars from different generations were compared, ancestral lines were different from cultivars and less responsive to better growing conditions than cultivars. Newer cultivars produced higher yield in both low- and high-yielding environments.
Genetic improvement of soybean in the U.S. Midsouth has led to earlier, shorter plants which are more tolerant to lodging compared to taller plants. Breeding for oil and protein content at the same time represents a paradox. This dilemma between two negatively correlated traits has been shown in this research. For specific needs, high protein or high oil cultivars can be developed at the expense of the other. However, breeders have succeeded at maintaining the protein and oil content within desirable limits while increasing yield.
Received for publication November 12, 1999.
|
REFERENCES
|
|---|
- Allen, F.L., and H.L. Bhardwaj. 1987. Genetic relationships and selected pedigree diagrams of North American soybean cultivars. The University of Tennessee, Agric. Exp. Stn., Knoxville, TN. Bull. 652.
- Boerma, H.R. 1979. Comparison of past and recently developed soybean cultivars in maturity groups VI, VII and VIII. Crop Sci. 19: 611–613.[Abstract/Free Full Text]
- Burton, J.W. 1985. Breeding soybeans for improved protein quantity and quality. p. 361–367. In R. Shibles (ed.) World soybean research conference III. Westview Press, Boulder, CO.
- Burton, J.W. 1991. Development of high-yielding high-protein soybean germplasm. p. 108–117. In R.F. Wilson (ed.) Designing value-added soybeans for markets of the future. AOCS, Champaign, IL.
- Buss, G.R., H.M. Camper, Jr., and C.W. Roane. 1988. Registration of ‘Hutcheson’ soybean. Crop Sci. 28:1024–1025.[Free Full Text]
- Cox, T.S., J.P. Shroyer, Liu Ben-Hui, R.G. Sears, and T.J. Martin. 1988. Genetic improvement in agronomic traits of hard red winter wheat cultivars from 1919 to 1987. Crop Sci. 28:756–760.[Abstract/Free Full Text]
- Eberhart, S.A., and W.A. Russell. 1966. Stability parameters for comparing varieties. Crop Sci. 6:36–40.[Abstract/Free Full Text]
- FAO, 1998. FAOSTAT. Agriculture data, available at http://apps.fao.org/page/collections?subset=agriculture (verified 7 Feb. 2001).
- Frederick, J.R., J.T. Woolley, J.D. Hesketh, and D.B. Peters. 1990. Seed yield and agronomic traits of old and modern soybean cultivars under irrigation and soil water-deficit. Field Crops Res. 27: 71–82.
- Frederick, J.R., J.T. Woolley, J.D. Hesketh, and D.B. Peters. 1991. Water deficit development in old and new soybean cultivars. Agron. J. 82:76–81.
- Karmakar, P.G., and P.S. Bhatnagar. 1996. Genetic improvement of soybean varieties released in India from 1969 to 1993. Euphytica 90:95–103.
- Luedders, V.D. 1977. Genetic improvement in yield of soybeans. Crop Sci. 17:971–972.[Abstract/Free Full Text]
- Pantalone, V.R., J.W. Burton, and T.E. Carter, Jr. 1996. Soybean fibrous root heritability and genotypic correlations with agronomic and seed quality traits. Crop Sci. 36:1120–1125.[Abstract/Free Full Text]
- SAS, 1998. SAS Online Doc. V7-0. SAS Institute Inc., Cary, NC.
- Schmidt, J.W. 1984. Genetic contributions to yield gains in wheat. In W.R. Fehr (ed.) Genetic contributions to yield gains of five major crop plants. CSSA Spec. Publ. 7. CSSA, Madison, WI.
- Schmidt, J.W., and W.D. Worrall. 1984. Trends in yield improvement through genetic gains. Proc. 6th Int. Wheat Genet. Symp. 6:691–700.
- Slafer, G.A., E.H. Satorre, and F.H. Andrade. 1994. Increase in grain yield in bread wheat from breeding and associated physiological changes. p. 1–68. In G.A. Slafer (ed.) Genetic improvement of field crops. Marcel Dekker, Inc., New York.
- Specht, J.E., and J.H. Williams. 1984. Contribution of genetic technology to soybean productivity-retrospect and prospect. p. 49–74. In W.R. Fehr (ed.) Genetic contributions to yield gains of five major crop plants. CSSA Publ. 7. CSSA, Madison, WI.
- Voldeng, H.D., E.R. Cober, D.J. Hume, C. Gillard, and M.J. Morrison. 1997. Fifty eight years of genetic improvement of short-season soybean cultivars in Canada. Crop Sci. 37:428–431.[Abstract/Free Full Text]
- Waddington, S.R., J.K. Ransom, M. Osmanzai, and A. Saunders. 1986. Improvement in the yield potential of bread wheat adapted to Northwest Mexico. Crop Sci. 26:698–703.[Abstract/Free Full Text]
- Wilcox, J.R., W.T. Schapaugh, Jr., R.L. Cooper, W.R. Fehr, and M.H. Niehaus. 1979. Genetic improvement of soybeans in the Midwest. Crop Sci. 19:803–805.[Abstract/Free Full Text]
- Williams, J.H., and J.E. Specht. 1979. A perspective on yield advances attributable to soybean variety development. p. 81. In Agronomy abstracts. ASA, Madison, WI.
TOP
ABSTRACT
INTRODUCTION
Significantly different from one (P