Baskheti, D.C.Budhlakoti, Vartika2021-08-092021-08-092021-02https://krishikosh.egranth.ac.in/handle/1/5810171657The present investigation aimed to estimate selection parameters for various characters, heterosis and combining ability for various quantitative and quality characters and to determine the role of epistasis in the genetic control of grain yield, yield component traits and quality characters. The study comprised of two separate but related experiments. The material for Experiment-I, comprised of 36 all possible crosses (excluding reciprocals) among 9 diverse genotypes of rice viz., Pusa Basmati 4, Pusa Sugandh 5, Pusa Basmati 6, Pant Sugandh Dhan 15, Pant Sugandh Dhan 17, Pant Basmati 1, Pant Basmati 2, Pusa Basmati 1 and Taraori Basmati. The parents along with their 36 F1's including one standard variety (Improved Pusa Basmati 1) were evaluated in RBD with three replications during kharif 2018-19 at N.E.B.C.R.C, GBPUA&T, Pantnagar. The data on 25 characters were analyzed following Griffing (1956) method 2, model I, to estimate GCA and SCA variances. Heterosis was estimated over better parent (Heterobeltiosis) and Standard variety as suggested by Hayes e t a l. (1955). Six generations of five crosses viz.; Pusa Basmati 4 × Pant Basmati 1 (Family A), Pusa Sugandh 5 × Pant Basmati 1 (Family B), Pusa Sugandh 5 × Pusa Basmati 1 (Family C), Pusa Basmati 6 × Pant Basmati 2 (Family D) and Pant Basmati 2× Pusa Basmati 1 (Family E) selected from above diallel set, formed experimental material for Experiment-II. The material was evaluated in Compact Family Block design with 3 replications. The data were subjected to generation mean analysis by joint scaling tests to test the adequacy of different genetic models and to estimate various gene effects (Cavalli, 1952). Results of ANOVA showed that mean squares due to GCA and SCA were highly significant for all the characters. The magnitude of former was higher than later for all characters. Relative estimates of 𝜎2D and 𝜎2H for various characters revealed predominance of additive gene action for the expression of all the characters except 1000 grain weight, harvest index, hulling percent, milling percent, head rice recovery and length/breadth ratio after cooking, which also gains ground from higher value of predictability ratio (>1) and lower value of degree of dominance (<1) for all the characters except the above mention 6 traits.The narrow sense heritability (h2 n) was high for 17 characters including grain yield and moderate estimates for 5 characters viz., number of grains per panicle, harvest index, hulling percent, panicle length and 1000 grain weight; and low for milling percent, head rice recovery and length/breadth ratio after cooking. A summon of general combining ability showed that overall parent Pusa Sugandh 5 emerged as the best donor parent as it exhibited significant and favourable GCA effect for most (i.e., 21 characters) of the yield and quality traits under study. So, this line can be utilized to create a population with favourable genes for grain quality and yield component traits. The second-best general combiner is Pusa Basmati 6 for nineteen, followed by Pusa Basmati 1 for sixteen, Pusa Basmati 4 for twelve, Pant Basmati 2 for ten, Pant Basmati 1 for nine, Pant Sugandh Dhan 15 for seven, Pant Sugandh Dhan 17 for six and Taraori Basmati for six characters. Considering all the characters under study, the best six crosses on the basis of results of SCA effect were Pant Basmati 2 × Pusa Basmati 1, Pusa Sugandh 5 × Pusa Basmati 1, Pusa Sugandh 5 × Pant Basmati 1, Pusa Basmati 6 × Pant Basmati 2, Pusa Basmati 4 × Pusa Basmati 6 and Pusa Basmati 4 × Pusa Basmati 1. Further, it was revealed from the present investigation that there was no consistency in GCA status of the parents involved in the crosses having significant SCA effects. The superior crosses identified for grain yield per plant are having G×G, G×A, G×P and P×P GCA parents. The evaluation of heterosis over better parent (BPH) as well as standard variety (SVH) revealed that the magnitude of heterosis was in general low for quality traits as compared to grain yield per plant and yield component traits. For grain yield per plant, 15 crosses over better parent (5.74 to 56.23%) and 13 over standard variety (7.20 to 40.18%) manifested significant desirable heterosis. Of these twelve crosses displayed heterosis over both better parent as well as standard variety for grain yield per plant and also showed desirable BPH and SVH for yield component traits and various quality traits. Five potential crosses namely Pusa Sugandh 5 × Pusa Basmati 6, Pusa Sugandh 5 × Pant Basmati 1, Pusa Sugandh 5 × Pusa Basmati 1, Pusa Basmati 6 × Pant Basmati 2, Pant Basmati 2 × Pusa Basmati 1 could be identified for grain yield per plant, yield component traits and quality traits based on the magnitude of improvement in grain yield per se over better parent and check variety along with significant desirable SCA for grain yield, yield component traits and quality traits. GCA status of both the parents was good for these crosses. Thus, these crosses are amenable for improvement through conventional breeding procedures and can also be exploited by heterosis breeding. The results Experiment II showed that there existed significant differences among progenies in all the families. The families comprising of five crosses six generations each also showed significant variability for all the characters barring panicle length. Adequacy test of different models for characters showed that an epistatic model was adequate for all the characters in most of the families. Additive–dominance (3PM) model was found adequate in some of the families for various characters viz.; 3 characters for Family-A, 6 characters for Family-B, 1 character each for Family-C and Family-D and 6 characters for Family-E. Direct selection may be effective for these traits in case of the above mentioned crosses. In respect of epistatic effect, duplicate epistasis was predominant for all the traits except for biological yield in Family-D, hulling percent in Family-A and milling percent in Family-A which showed complimentary epistasis. In the case of good × good general combiners there are possibilities of complementary epistatic interaction acting in the direction of additive effects of the good combiners. The crosses would be utilized for target trait through single plant selection in segregating generations. Results showed significance of additive components besides duplicate epistasis which suggest chances of recovering transgressive plants effect for the characters governed by non additive gene actions and epistasis through recurrent selection methods. Exploitation of duplicate epistasis in self-pollinatedcrops like rice, can be done through biparental mating in early generation followed by selection besides repeated back crossing.EnglishThesis