برآورد پارامترهای ژنتیکی در برنج با استفاده از روش‌های مختلف دای آلل گریفینگ

Six Iranian rice cultivars (Binam, Domsiyah, Shahpasand, Sepidrud, Khazar and Valed 46) were crossed in 1989 in the Iranian Rice Research Institute in Rasht, Iran in a full-diallel design. The F1 progenies together with parents were transplanted in a CRBD in the 5 x 0.75 m plots at plant density of 25 x 25 cm (60 plantlets per plot) in 3 replications. Part of this research was published in 1994 as a half-diallel design and the data of full-diallel, using Griffing approach, is subject of this paper. The analysis of variance showed significant differences among genotypes. The mean of six parents, 15 crosses and 15 reciprocal crosses were analyzed using the four diallel crossing systems of the Griffing approach. The SS of genotypes were separated into GCA for each parent and SCA for each cross using corresponding formula. The additive variance (VA) and dominance variance (VD) values were estimated using the table of variance analysis. The heritability (h2ns) was also estimated from VA and VD variances. The simple variance analysis of “grain yield per plant” and other agronomic traits using each of the 4 diallel crossing methods of Griffing showed significant differences (P < 1%) among genotypes, indicating a sufficient genetic potential of the investigated genotypes. The GCA variance analysis of the lines was significant for all the characteristics, indicating the importance of additive variance (VA) by inheritance of these traits. Using Griffing’s diallel methods 1 and 3, SCA variances for all investigated traits showed significance, whereas in the methods 2 and 4, traits “1000 grain weight” and “tiller per plant” were not significant according to SCA. This also indicates the importance of dominant variance (VD) in most of the traits, except for “1000 grain weight” and “tiller per plant”. The differences among the reciprocal crosses in diallel methods 1 and 3 were also examined where for all the characteristics (except for “deaf grains per panicle” and “length to breadth ratio of brown rice grain”) significant differences were observed, suggesting the possibility of cytoplasmic effect of mother line on the reciprocal crosses. The heritability (h2ns), which indicates the ratio of additive variance (VA) to phenotypic variance (VP), was estimated to be equal to zero due to the absence of additive variance for grain per panicle and the number of days from transplanting to full maturity of grain. The heritability for other characteristics was estimated high or low according to additive variance. For example in all 4 diallel methods the heritability estimation for length to breadth ratio of brown rice grain was relatively high (65 – 71%) whereas for “panicle length” and “grain yield per plant” was relatively low (13 – 48%). The correlation among genetic parameters (VA, VD, D, h2ns ) were generally high and significant.