Fiber quality data from five upland cotton (Gossypium hirsutum L.) genotypes, which were grown at College Station, TX during 2001 and 2002, were subjected to diallel and generation means analyses to determine the potential for improvement of fiber length and to determine the inheritance of length distribution data. Four near-long staple (NLS) upland cotton genotypes and one short-staple genotype were crossed in all combinations, excluding reciprocals. Estimates of general (GCA) and specific combining ability (SCA) for fiber length based on Griffing’s diallel Model I, Method 4 were calculated for high volume instrumentation (HVI) upper-half mean (UHM) fiber length and advance fiber information system (AFIS) mean fiber length by weight (FLw), mean fiber length by number (FLn), upper quartile length by weight (Uqlw), fiber length distribution cross entropy (using 3 different standard or check distributions - CEA, CEB, and CEC), fiber length distribution kurtosis (FLwKurt), and fiber length distribution skewness (FLwSkew) for FLw. Across environments, GCA effects were significant for fiber length measurements of UHM, FLw, FLn, Uqlw, and SFCw and distribution measurements of CEA, CEB, FLwKurt, and FLwSkew. On the basis of GCA effects, TAM 94L-25 was the best parent to be used in a cross to improve upland fiber length, while Acala 1517-99 was the parent of choice to improve distribution among the 4 parents tested. The inheritance of AFIS fiber length measurements and distribution data was estimated using parents, F1, F2, and backcross generations. The magnitude and significance of the estimates for non-allelic effects in the parental combinations suggest that epistatic gene effects are present and important in the basic mechanism of AFIS fiber length and length distribution inheritance for the populations studied. Gene effects and variances for all AFIS fiber length and distribution data measurements were inherited differently in different environments and specific parental combination, suggesting environmentally specific mechanisms. Developing genotypes with enhanced fiber length and an optimal fiber length distribution should be a priority to improve spinning performance and product quality of U.S. upland cotton.

The tobacco cyst nematode [Globodera tabacum solanacearum (Miller & Gray, 1972) Behrens, 1975] is an important pathogen affecting flue-cured tobacco (Nicotiana tabacum L.) in Virginia, North Carolina, and Maryland. The resistant cultivars Coker 371 Gold and Kutsaga 110 were evaluated during 1999 and 2000 in the greenhouse to determine the mode of inheritance of resistance to the tobacco cyst nematode (TCN). Each cultivar was crossed to the susceptible cultivar K 326 and F1 progeny were backcrossed to each parent. Plants from each parent and F1, F2, BC1Ps, and BC1Pr progeny were evaluated for TCN resistance. Six-week-old transplants were inoculated with 6000 TCN eggs from crushed cysts. Eight weeks after inoculation, a 1-g sample of fibrous root was stained and vermiform, swollen, pyriform, and adult nematodes were counted. The number of cysts and eggs per 400 cm3 of soil were counted from each transplant. Generation means analyses were performed. Additive and dominance gene action play an important role in resistance to TCN in Coker 371 Gold and Kutsaga 110. F2 generation data from the Coker 371 Gold cross fit a 3:1 (resistant:susceptible) segregation ratio and BC1Ps generation data fit a 1:1 segregation ratio, indicating that resistance to TCN is conferred by a single dominant gene. A continuous range of variation was observed among the F2 progeny for the K 326 X Kutsaga 110 cross, indicating resistance in Kutsaga 110 is quantitative. TCN resistance in Coker 371 Gold and Kutsaga 110 may be derived from different sources. / Master of Science