Abstract
The study was undertaken for selection of maize inbred lines through morpho-molecular approaches for crossing, identification of promising hybrids through combining ability and heterosis study and finally assessing their stability and adaptability. The experiments were conducted at Regional Agricultural Research Station (RARS) Hathazari, Chittagong, Molecular Laboratory of Plant Breeding Division, BARI, Gazipur and other four different locations i.e., RARS at Jessore, RARS at Rahmatpur, RARS at Ishwardi and Jamalpur during December 2011 to May 2014. Fifty maize inbred lines from different countries were evaluated for 11 morpho-agronomic traits. The studied fifty maize inbred lines varied significantly for yield and yield attributes. The highest variation was found in grain yield among the inbred lines. Based on D2 values, 50 inbred lines were grouped in 5 clusters. The inter cluster D2 values exhibited high range of diversity among the inbred lines. All inter-cluster distance were larger than all intra-cluster distances. From 5 clusters, 22 inbred lines were selected based on their per se performances for molecular analysis to examine the variability of maize inbred lines using SSR markers. A total of 187 alleles were detected at 18 SSR loci among 22 maize inbred lines with an average of 10.4 alleles per-microsatellite locus. The highest number of alleles per locus of each genotype was detected using SSR primer set phi026, showing 17 alleles. The lowest allele number per locus among the homologous chromosomes was observed using SSR primer set p-umc1292, phi074 and phi090 showing a total of 6 alleles per genotype. The highest level of gene diversity value (0.92) was observed in locus phi026 and the lowest of gene diversity value (0.77) was observed in locus p- umc1292 with a mean diversity of 0.862. The frequency of the most common allele at each locus ranged from 0.14 (bnlg371) to 0.36 (p-umc1292). On an average, 0.22 of the 22 selective maize inbred lines shared common major allele at any given locus. The polymorphic information content (PIC) values ranged from 0.74 to 0.92 with an average of 0.85. PIC value revealed that phi026 and nc013 were considered as the best marker for 22 selective maize inbred lines followed by bnlg1124. The UPGMA clustering also generated seven genetic clusters with similarity coefficient of 12.5%. From seven clusters two sets of inbred lines viz. in set I eight inbred lines and in set II seven inbred lines were selected and crossed in a half diallel fashion. From set I the GCA effects suggest that parents E38 and E36 are excellent general combiner for earliness, E21, E36 and E25 for desirable height, E19 for good husk cover, E19 and E20 for long ear, E38 and E20 for big ear girth, E19 and E25 for number of row, E20 for kernel per row, E38 for 1000 kernel weight and E38 and E20 for grain yield. The crosses from set I having significant and positive economic heterosis for grain yield (heterosis estimated over the commercial hybrid, 999) are E21×E20, E25× E38, E20-E14, E19×E20, E38×E20, E38-E14, E36×E14 and E21×E36 couple with positive SCA effects for yield and yield contributing traits like ear length, ear girth, number of rows per ear, number of kernel per row and 1000-kernel weights and negative SCA effects for the traits like, days to tasseling, silking and maturity, plant and ear height; and poor husk cover. From set II the parent E34 is an excellent general combiner for earliness, E39 and E43 for desirable height, for good husk cover E16, for long ear E16 and E17, for big ear girth E28 and E39, for number of row E39, for kernel per row E39 and E16, for 1000 kernel weight E34 and for grain yield E16 and E39 inbred lines could be used as donor partners for obtaining high yield and desirable traits. The crosses from set II having significant and positive economic heterosis for grain yield (heterosis estimated over the commercial hybrid, 999) are E34×E16, E39×E17, E43×E17, E39×E16, E28×E43, E28×E35, E39×E35, E34×E28, E16-E17, E34-E17 and E35×E16 couple with positive SCA effects for yield and yield contributing traits. Moreover, some traits showed negative SCA effects. From experiment 3 & 4, 21 high yielder crosses (F1) were selected for stability study. The combined analysis of variance over locations revealed significant difference among the genotypes (crosses) and environments (locations) for all the characters. G x E interaction was significant for all the traits suggesting that genotypes interacted significantly with the changing environments. Considering stability parameter and AMMI biplot, the crosses T3 (E20-E14), T5 (E38×E14), T8 (E34-E16), T9 (E39×E17), T13 (E39×E35), T15 (E16×E17), T18 (E28×E16) and T19 (E35×E16) are found more stable with higher yield. These crosses may be selected for commercialization after verifying at farmer's field by large plot observation trial.