Development and Comparison of Three Selection Indices for High Yield Potential and Resistance to Phymatotrichum Root Rot in Cotton

C.G. Cook and K.M. El-Zik


 
ABSTRACT

Phymatotrichum root rot (caused by Phymatotrichum omnivorum [Shear] Dug.) is one of the major constraints to cotton production in the Blacklands of Texas and in parts of Arizona and New Mexico. Twelve cotton genotypes were selected and evaluated in 1987 and 1988 at the Phymatotrichum root rot (PRR) nursery at Temple, and at PRR-free nurseries at College Station and Corpus Christi, Texas. Eleven genotypes were from the multi-adversity resistance (MAR) cotton improvement program, and Lankart LX571 was a non-MAR cultivar. Percentage of diseased plants was recorded at four and five weeks after initial symptoms. Plots were harvested and lint yields determined at crop maturity. Significant differences were observed among genotypes for percentage of diseased plants at both four and five weeks after initial symptoms in 1987 and 1988. The genotypes MLB2BENH-1-85, BLLCABS-1-86 and CABUCD3H-1-86 consistently had lower percentages of diseased plants than did the genotypes CDP37HCIH- 1-86 and C4HUGBEH-1-2-86. Combined over years at Temple, the MAR genotypes CABUCD3H-1-86, BLLCABS-1- 86, and C4HUGBEH-1-2-86 produced significantly higher yields than Lankart LX571.

In order to identify and combine both resistance to P. omnivorum and high yield potential under conditions, three selection indices were developed an compared. The values of index I for a genotype were calculated using lint yield of the genotype, mean lint yield of all genotypes, percentage of plants of the genotype exhibiting disease symptoms, and mean percentage of disease symptoms for all genotypes. Index II was based on disease incidence four and five weeks after initial symptoms and lint yield of the genotype under PRR and PRR-free conditions. Values for selection index III were calculated with a weighted equation; lint yield and disease incidence were weighted equally to maintain comparisons for both yield potential and resistance levels.

Each of the three indices has merit for evaluating germplasm for both levels of resistance to PRR and high yield potential. In comparing the three indices, it appears that selection indices I and III have the best merit for evaluating cotton germplasm for the desired traits. The major drawbacks of index II were the inability to test differences among genotypes statistically and the tendency of the index to select for a single trait. Each index also indicated that CABUCD3H-1-86 and BLLCABS-1-86 were superior to other genotypes in both yield and resistance. The results confirm that differences exist among genotypes for resistance levels and yield potential under PRR conditions. Utilizing indices that select simultaneously for both traits should provide a tool for genetic gains in cotton improvement breeding programs.



Reprinted from Proceedings: 1989 Beltwide Cotton Research Conferences pp. 33 - 36
©National Cotton Council, Memphis TN

[Main TOC] | [TOC] | [TOC by Section] | [Search] | [Help]
Previous Page [Previous] [Next] Next Page
 
Document last modified Sunday, Dec 6 1998