Creating an Alternative Soybean Gene Pool For High Yield

Average soybean yields from 1924 to 2010.

Soybean yield in Illinois has increased an average of 0.46 bushel per acre per year since 1924, an increase that has been remarkably consistent over 86 years. In 1924, we harvested 115,000 acres with an average yield of 12 bushels/acre. In 2010, we obtained the highest average yield of 51.5 bushels/acre, exceeding the previous high of 50.0 bushels per acre in 2004. In terms of soybean breeding, this linear trend begins before scientific soybean breeding was established and continues through the application of all of the modern field and laboratory technology, including transgenic varieties. The rate of yield increases has not kept pace with the investment of resources.

Despite all the changes that have occurred, the narrow genetic base of soybean production in Illinois has changed little over the decades. Just 12 ancestral lines accounted for 85% of the genes in the lines released in Illinois between 1947 and 1988, and the last of these major ancestors was added to the current gene pool in 1954. Although varieties developed by commercial companies occupy most of the soybean production area, the pedigree data show that the change to proprietary varieties has not changed the genetic base. This lack of diversity could be one reason why we are unable to improve the rate of yield improvement. Less than 0.5% of the soybean germplasm available in the USDA Soybean Germplasm Collection housed at the University of Illinois has made any contribution to current soybean varieties. The objective of our research is to develop high-yielding experimental lines derived from exotic germplasm that can provide new genetic diversity to enrich the commercially used gene pool and increase the rate of yield increase in future varieties.

Agronomic data for experimental lines.

Over the past 30 years we have evaluated experimental lines derived from several hundred soybean introductions that have not previously been used in U.S. soybean breeding.We currently have more than 100 of these exotic lines in the pedigrees in our active breeding program, and over 50 of the exotic parents are in pedigrees of experimental lines in advanced regional tests this year. Since 1997 we have released 21 high-yielding experimental lines derived from 30 exotic parents. LG04-6000 has the Chinese variety Jilin 15 as a great-grandparent. Released by the Jilin Academy of Agricultural Sciences in 1978, Jilin 15 has a pedigree that traces to three Chinese landraces. In the USDA regional tests in 2007 and 2008, LG04-6000 significantly exceeded the yield of the best variety in the test by 8%.

One of the parents of LG06-5798 is an experimental line (LG00-3372) that we released in 2003. LG00-3372 has a pedigree that is 100% exotic germplasm from China. The other parent is the high-yielding variety, LD00-3309, released by the University of Illinois. In the 2010 USDA regional tests at 15 locations, LG06-5798 was the highest-yielding entry, with a yield nearly 7% greater than LD00-3309, the highest-yielding variety in the test . This provides strong evidence that the genes from previously unused Chinese germplasm are able to significantly improve the yield of high-yielding U.S. varieties.

More recently we have also been using wild soybean as a parent. All research shows that wild soybean has much more genetic diversity than soybean, but because it is a vine with very small seeds, it has been very difficult to use as a parent in breeding programs. In regional testing in 2010, LG07-2309, with a pedigree that is 12% wild soybean, equaled the yield of the best variety in the test. The development of experimental lines derived from exotic germplasm that equal or exceed the yield of the best varieties offers soybean breeders a resource that can add important new genetic diversity for high yield.We are actively working with soybean breeders in both the private and public sectors to encourage the use of these lines in their variety development programs.


Randall Nelson

Randall Nelson
USDA-ARS Research Geneticist
Professor of Crop Sciences
217-244-4346
Randall.Nelson@ars.usda.gov