Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Crop Genetics Research » Research » Publications at this Location » Publication #189838


item Meredith Jr, William

Submitted to: National Cotton Council Beltwide Cotton Conference
Publication Type: Proceedings
Publication Acceptance Date: 12/10/2005
Publication Date: 1/3/2006
Citation: Meredith Jr, W.R. 2006. Obsolete conventional vs. modern transgenic cultivar performance evaluations. National Cotton Council Beltwide Cotton Conference. p. 836-844.

Interpretive Summary: The biggest change in cotton breeding in the last decade has been to go from 100% conventional varieties to 82.6% genetically engineered (transgenic) varieties in the US and 97.6% in Mississippi. Conventional varieties have, for over a century, been produced by crossing one variety with another, advancing the progenies from the cross for several generations, and then selecting progenies with the hope that new combinations of genes will come together and produce a new superior variety. This process usually takes from eight to ten years. Transgenic varieties are produced by taking a single transgene, crossing it to a successful variety, selecting for the transgenic trait and repeating the process several times. This process is called backcrossing and takes about five years. Transgenes are produced by genetic engineering by taking a gene from some non-cotton source, and identifying that gene so that it can be backcrossed into a successful variety. In the last 10 years, transgenic cotton varieties have decreased losses due to worms and weeds, reduced the number of pesticides, and made crop management simpler and more efficient. However, in this study the transgenics did not result in any gain in yield or fiber quality. This is after 15 years, on average, of intense cotton breeding. Competition with man-made fibers and foreign cotton competition demands, if the US cotton industry is to survive, in the future both higher yields and fiber quality are needed. The lack of genetic progress for yield and fiber quality is not due to the transgenes themselves, but is due to the almost total reliance on the backcross method to produce new transgenic varieties.

Technical Abstract: The percentage of acres planted to transgenic varieties has increased from zero in 1995 to 82.6% and 97.6% in 2005 for the US and Mississippi, respectively. Most of the transgenic varieties have been produced by the backcross method. The backcross method transfers a single gene trait into an established variety with expectations that the new transgenic variety will have about the same yield and fiber quality characteristics as the original variety plus the “added value” trait. Yields in Mississippi have been increasing for the last 25 years at about 6.2 lbs lint/acre per year. How much of this increase is due to breeding and especially to transgenics? The objective of this study was to compare the yield and fiber quality performance of obsolete varieties produced by conventional breeding with the new transgenic varieties. Seventeen conventional varieties which were released for commercial use from 1936 to 1995 and twelve transgenic varieties whose release dates ranged from 1996 to 2003 were grown in variety tests near Stoneville, MS. There were three locations in the 2004 test and two locations in 2005. Within a year, all tests were grown on different soil types. Planting dates varied from April 18 to May 20. The experimental design was a randomized complete block with six replications for yield and three for yield components and fiber traits. Plot size was two 40 inch rows, 51 feet long. Standard crop management was practiced on all plots. Two harvests with a mechanical plot picker were made on all plots. The regression of yield on year of variety releases from 1936 to 1995 was 6.5 lbs acre/year but no net progress was made after 1995. A comparison of nine conventional varieties whose release date was 1974 to 1995 was made with the twelve transgenic varieties for yield and fiber traits. The average yield for the conventional varieties was 1321 lbs lint/acre and that for the transgenics was 1323 lbs per acre. There was no difference or trends in the average length, uniformity, HVI strength, and micronaire detected between the conventional and transgenic variety groups. There were significant differences within both groups for all characteristics. A comparison was made between the five transgenics that constituted over 76% of the Midsouth and Mississippi acreage for 2004, and 2005 with five obsolete conventional varieties. The average lint yield of the transgenics was 1365 lbs lint/acre, and the average for the five obsoletes was 1393 lbs lint/acre. Similar comparisons for fiber length was 1.157 inches and 1.144 inches, for uniformity ratio was 84.1 and 84.2, for HVI strength was 29.8 and 30.2, and for micronaire was 4.68 and 4.74 for transgenics and obsoletes, respectively. Considering that all but one of the twelve transgenics had been produced by the backcross method, one should not expect them to be superior in performance to their recurrent parents. Some of the conventional varieties used in these tests as recurrent parents for transgenics were DP 5415, DP 50, ST 474, and SG 125. Lack of genetic progress for yield and fiber quality is not due to the transgenes per se, but is due to almost total reliance on the backcross breeding method to produce new transgenics. The results clearly show that continuous strong dependence on the backcross method to improve Midsouth yield and fiber quality will not keep pace with gains made by competing countries who use the backcross method sparingly.