Study of Gene Action for Yield and Yield Components in Gossypium hirsutum L.

MATERIALS AND METHODS

The experiment was conducted in Randomized Complete Block Design (RCBD) at the Department of Plant Breeding and Genetics, Agricultural University Faisalabad during 1998-99. The direct and reciprocal crosses were made of four upland cotton varieties viz. Stone ville-213, HG-6-1-N, Fregobract-83 and AG-hirsutum-87. Data was collected for the characters, GOT%, Seed index and Seed cotton yield/plant, and was statistically analyzed through analysis of variance technique (Steel and Torrie, 1960) to determine the difference among F₁ hybrids and parents. For further genetic analysis diallel cross technique developed by Jinks (1955; 1956) was used. The information on gene action was inferred by plotting the variance (Vr) of each array against its covariance (Wr).

Results and Discussions

The mean performance of parents and F₁ hybrids is presented in Table 1 to 3. The mean square values of the characters are presented in Table 4, which showed that the differences between genotypes were highly (P≥ 0.01) significant indicating the existence of genetic variability among genotype for the traits studied.

Ginning out turn percentage: The analysis of variance for ginning out turn percentage (Table 4) indicated that the differences among the genotypes were significant. The values of variance and covariance graph shown in Fig. 1, also display that the regression line cuts the Wr axis below the origin. This signifies the over dominance type of gene action. As the regression line deviates significantly from a unit slop the expression appeared to be contaminated with non-allelic interaction. From the position of array points on the regression line it was shown that AG-hirsutum-87 being the most proximate to the origin had most dominant genes while the HG-6-1-N possesses the recessive genes for this character.

Fig. 1:	Wr/Vr graph for ginning outtrun percentage

Fig. 2:	Wr/Vr graph for seed index

Fig. 3:	Wr/Vr graph for yield of seed cotton

Table 1:	Diallel for ginning out turn percentage

Table 2:	Diallel for seed index (g)

Table 3:	Diallel for seed cotton yield/plant (g)

Table 4:	Mean squares for GOT %, seed index and seed cotton yield/plant
** = Significant at 0.01 level

It is clear from Table 1, that Stone ville-213 displayed 35.65 ginning out turn percentage but showing the highest array mean value (38.39) suggesting that Stone ville-213 was good general combiner and within the array the cross Stone ville-213 x Fragobract-83 showed highest value (39.84) which exhibited that this cross is best specific combiner for this trait. The results of variance and covariance are in accordance with the results of BingTang et al. (1993). They reported over dominance type of gene action whereas, Singh et al. (1982) and Keerio et al. (1995) explained additive type of gene action with partial dominance for this character.

Seed index: The analysis of variance for seed index is presented in Table 4, which indicates that the difference among the genotypes were significant. The variance and covariance graph for the character seed index is depicted in Fig. 2, which showed that the regression line intercepted the Wr axis below the origin, indicating the over dominance type of gene action. As the regression line deviates significantly from unit slop there appears some non-allelic interactions in this respect. The variety Fregobract is close to the origin which showed maximum dominant genes, whereas the Stone ville-213 have maximum recessive genes.

In Table 2, the variety AG-hirsutum-87 showed the mean 6.1 g but in the array it carries the highest mean (7.30 g) value which showed that it is good general combiner for this trait. Within the array the cross of Stone ville-213 x AG-hirsutum-87 had the highest value (8.28g) suggesting that the cross is specific combiner in this respect. The results are in conformity with the results of Larik et al. (1997), whereas, the Singh et al. (1982) reported additive type of gene action.

Seed cotton yield/plant: The analysis of variance for seed cotton yield/plant is presented in Table 4, indicates that the differences among the genotypes were highly significant. The Fig. 3 showed that the regression line passes the Wr axis above the origin which revealed the additive type of gene action with partial dominance. As the regression line deviates significantly from a unit slope there appears some interaction of nonallelic genes in this respect. The variety AG-hirsutum-87 is nearer to the origin indicating the presence of maximum dominant genes whereas, HG-6-1-N is far from the origin demonstrating that this variety has maximum recessive genes for this character. Results further (Table 3) showed that Stone ville-213 had highest array mean value (107.69) which suggested that Stone ville-213 is good general combiner for this trait and the cross. Stone ville-213 and HG-6-1-N had the highest value (127.07), thus it indicates that it is better specific combiner as compared to other crosses. Azhar et al. (1983) and Nadre et al. (1984) have also reported similar results from their studies. However, Keerio et al. (1995) reported over dominance type of gene action.