Page Banner

United States Department of Agriculture

Agricultural Research Service


item Montgomery, Kevin
item Pollak, Linda
item Goggi, Susan
item Abel, Craig
item Williams, William
item Xu, Wenwei

Submitted to: Corn Conference Interregional Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 2/11/2004
Publication Date: 2/12/2004
Citation: Montgomery, K., Pollak, L.M., Goggi, S., Abel, C.A., Williams, W.P., Xu, W. 2004. Yield of two sets of hybrids and the grain composition of the parents and hybrids with associated agronomic traits. Corn Conference Interregional Proceedings. Available:

Interpretive Summary:

Technical Abstract: Two approaches to selection of extremes in grain chemical composition of corn were investigated. In the first (Elite Hybrid Separation), 230 near-commercial hybrids were evaluated by NIR. Twenty were chosen, including Bt and non-Bt hybrids as well as 3 with average grain composition. Replicated yield trials were conducted in 2002 and 2003, respectively. Companion trials were conducted in both years in five locations for disease and insect evaluations. Additionally, each of the five locations generated self pollinated samples of the hybrids for chemical composition. In 2003, there were significant differences among entries and locations for grain yield and grain moisture; grain composition demonstrated significant differences among entries and to a much smaller degree, among locations. Frequently, hybrids with opposite expression in a trait (% oil, for example) had very similar yields. Consequently, very few correlations between grain composition traits and yield were significant, either on an individual or across location basis. Hybrids with extremes in expression (high or low) for grain composition frequently had common parents. Hybrids with opposite expression (high and low) infrequently had common parental contributions. Thus, parental contributions may be additive and non-additive for a given trait. Previous work identified a commercial hybrid (H-2390) with elevated grain protein levels. Two breeding crosses were developed, each the equivalent of a BC1 population representing the female and male of H-2390. For the second approach (Extreme Inbred Development), inbreds were developed for expression of high grain protein from these populations. In 2001, selection was practiced among and within S2 lines for agronomic type, followed by selection for grain protein on an individual ear basis. All selected ears had protein contents higher than their respective recurrent parents. Ten females and eleven males were chosen for further testing. These preliminary hybrids were tested in 2002 in replicated yield trials along with companion grain composition and seed quality studies. Selections based on yield and grain composition further reduced the number of selections to 3 lines x 2 sisters for females and 4 lines x 2 sisters for males, respectively. A 6 x 8 Design II (all females x all males) plus 2 checks was conducted 2003 at 7 locations. Companion studies for grain composition and seed quality were conducted at 2 locations similar to 2002. Sister lines behaved similarly for yield, moisture and protein content. Significant progress was demonstrated for protein percentage on an inbred basis. Progress for improved protein composition in the hybrids was more difficult to demonstrate, indicating non-additive parental contributions in the hybrid. Selection for protein led to very low yielding hybrids, when compared to the relevant check hybrid. Screening for diverse chemical composition in the grain of hybrids appears to be a more effective short term approach than selecting for extreme expression in the parents.

Last Modified: 10/17/2017
Footer Content Back to Top of Page