JWSS - Journal of Water and Soil Science

---

Diallel crosses among 6 Avena sativa L. and A. sterilis L. lines and introductions were used to evaluate the validity of the assumptions for the genetic model. Number of days to pollination, plant height at pollination and at maturity, as well as grain and stem protein percentages were evaluated. According to Griffing's method 1 the reciprocal mean squares for all the traits under study were significant. But based on Hayman's analysis, the maternal effects for all the traits studied were not significant. Therefore, reciprocal means were used to evaluate the validity of the absence of multiple alleles, linkage and epistasis effects using regression of W_r on V_r and analysis of variance for W_r + V_r and W_r - V_r. Based on the results of the 3 methods, the genetic models for plant height at maturity and days to pollination were unbiased. Therefore, Hayman's analysis was used to estimate the genetic parameters for these traits. For plant height at pollination and stem protein percentage, significant and nonsignificant differences of regression slope from one and zero (Ho: β = 1 and Ho: β = 0) were detected, respectively. For these traits the biasedness of the model was removed after elimination of one parent from the diallel table. Grain protein percentage was not analyzed as it necessitated the elimination of 2 parents from the diallel table. In general, regression of W_r on V_r compare to W_r + V_r and W_r - V_r analysis of variances showed to be a more valid inductive method for testing the accuracy of the genetic model assumptions. Also, results of the Hayman and Jinks analysis, both when the assumptions are valid and when not, showed that genetic parameters are affected by the biasedness of the model, and that different estimates will be obtained. The Griffing's method was less affected by the unbiasedness of the model than Hayman and Jinks method. Therefore, it seems that preliminary testing for validity of the assumptions is necessary in Hayman and Jinks genetic model.

---

Seven Iranian and foreign cultivars of rice were crossed in a half-diallel. Parents with F2-generations were transplanted in a randomized complete blocks design with 3 replications and evaluated for length to breath ratio of paddy and brown rice, head rice and crust, grain yield, percent of head rice and broken rice, amylose content, gelatinization temperature and gel consistency. The ANOVA table showed significant differences among genotypes, general combining ability effects for all traits and specific combining ability effects for length to breath ratio of paddy and brown rice, percent of head rice and broken rice, so representing additive and non-additive effects in genetic control of the traits involved, respectively. The graphical analysis of the results showed partial dominance for length to breath ratio of paddy and brown rice, grain yield, amylose content, gelatinization temperature and gel consistency. Because of additive gene effects, genetic advance from selection would be high for these traits. The genetic control of brown rice weight is complete dominance, and for weight of head rice and rice crust weight over dominance involved. So, genetic advance by selection for these traits would be poor. In the genetic control of higher length to breath ratio of paddy and brown rice, brown and head rice weight, grain yield and amylose content recessive genes are involved. Domsiah for higher to breath ratio of paddy and brown rice and Sepidrud for brown and head rice weight had the most recessive genes. IR28 for higher grain yield and Sepidrud for less crust weight had the most recessive genes.

---

Half diallel crosses of eight bread wheat cultivars were used to estimate the genetic parameters and types of genetic control for yield and yield components, and to determine the effects of environmental factors on these parameters. Parents and their F2 hybrids were grown in a randomized block design under irrigated and stressed (terminal drought) conditions. For all traits, except for number of spikelets per spike and number of grain per spike, the effect of environment (irrigated v.s stressed) and genotype by environment interaction were significant. The mean square of GCA for all characters was significant in both environments, except for number of fertile tillers in the irrigated condition and grain yield in the stressed conditions. The mean square of SCA was significant for plant height only in the irrigated condition, while for number of fertile tillers and grain yield it was significant in both conditions. The high significant ratios of GCA to SCA mean square indicated the importance of additive genetic effects on controlling plant height, spike length number of grain per spike in both environments and number of spikelets per spike in the stressed condition. The parameter estimation revealed partial dominance gene actions for plant height, spike length and grain per spike in both environments, and number of spikelets per spike in stressed environment, whereas over-dominance gene actions were observed for the other traits in one or both conditions.

---

Half diallel crosses of nine Iranian wheat cultivars were used to estimate the general and specific combining ability effects and other genetic parameters related to yield and its components. Parents with F2 generations were evaluated at Research Farm, College of Agriculture, University of Shahrekord, in a randomized complete block design with 3 replications. Partitioning of genotypes mean squares to GCA and SCA mean squares based on Griffing’s method 2 and mixed model indicated high significant differences among cultivars in their GCA values for all traits. There were significant differences among crosses in their SCA values for all traits except for flag leaf length, main spike length, plant height and grain yield per plant. The high ratio of GCA to SCA mean squares implied the importance of additive gene effects in the appearance of flag leaf length , main spike length, plant height and grain yield per plant. For flag leaf width, grain number per main spike and peduncle length , the contributions of additive gene effects and for peduncle weight, the contributions of non- additive gene effects were higher. According to the estimates of average degree of dominance, the gene actions for flag leaf length, flag leaf width, plant height, grain yield per plant, main spike length, peduncle length and peduncle weight were partial dominance. The GCA effects indicated that Sefid – Aliabad cultivar was the suitable general combiner for flag leaf length , flag leaf width, plant height , grain yield per plant , main spike length and peduncle length.

---

Six rice cultivars were crossed in half diallel design to evaluate their GCA and SCA in 2005. In the following year, parents and their progenies were grown in a randomized complete block design with three replications, and 10 traits were measured. The analysis of variance showed significant differences (p≤0.01) between cultivars. Furthermore, general combining ability (GCA) and specific combining ability (SCA) for parents and hybrids were significant. Therefore, additive and non-additive gene effects on controlling traits were demonstrated. According to the analysis based on the second and fourth Griffing methods, additive gene effects were more than non-additive gene effects on controlling growth period, plant height, panicle length, number of panicles/plant and brown rice length, while other studied traits were more controlled by non-additive gene effects. Comparison of the second and fourth Griffing methods showed that the proportions of additive and non-additive variances in two methods were different. Moreover, GCA and SCA resulting from the two methods for several traits such as vegetative and productive growth period, plant height, number of filled grains/panicle and grain yield were significant. Therefore, it could be concluded that using parental generations in the second Griffing method may cause biased estimate of the GCA and SCA variances. Thus, using the fourth Griffing method is more suitable than the other methods in providing time, cost and facilities, and is recommended as an applicable method.

---

In order to study gene action in rice for traits related to quality (gelatinization temperature, gel consistency and amylose content), four varieties of rice (Sang-e-Tarrom, Gerdeh, IRRI2 and IR229) were investigated. Ten different generations including P1, P2, F1, RF1, BC1, RBC1, BC2, RBC2, F2 and RF2 were evaluated using generation mean analysis. In generation mean analysis, one of non-allelic interaction components, [i], [j]1, [j]2, [l]1, [l]2, [l] was significant indicating the genetic model of these characters were described by additive-dominance model with non-allelic gene interaction (except for gelatinization temperature trait in Sang-Tarrom × Gerdeh cross). The cross IRRI2×IR229 showed duplicate epistasis for gel consistency trait. Cytoplasmic effects and interactions between cytoplasmic and nuclear effects in two crosses were significant for amylose content and gel consistency traits. The estimation of narrow and broad-sense heritability for two crosses were 0.77 to 0.99 and 0.05 to 0.93, respectively. The predominantly additive nature of the genetic variability was further revealed by the variance components. Component D was detected significant in all the crosses. The covariance component and , however, showed indirectly that dominance contributed significantly to variability at the variance level. Therefore, according to the obtained results, selection can be effectively done in later segregation generations for gel consistency and amylose content and in early generation for traits.