| Huge heterosis exists in rice Indica-japonica subspecies hybrids F1. How to utilize the heterosis in indica-japonica subspecies hybrids rice efficiently is a topic of general interest in the field of hybrid rice breeding in the world. In view of the breeding practice, deriving pedigree of indica-japonica subspecies hybrids rice and its genetic basis are of theoretical importance and guiding significance in breeding for elite indica-japonica subspecies hybrids rice. In this paper, 18 parents and 39 deriving pedigrees were used to reveal the relationship between indica-japonica differentiation and heterosis, and the relationship between heredity diversity and pedigree. The improving effects of yield traits in different breeding stages and genetic effects of yield traits and their interaction by environment were also analyzed. The result would provide scientific foundation for further effective utilization of deriving pedigree of indica-japonica subspecies hybrids rice and hybrid rice combination. The main experimental results are as follows.1. Differentiation of indica-japonica rice subspecies hybrids deriving pedigree, including 18 parents, 39 deriving pedigrees and 2 control varieties, were determinated by using 58 ILP markers (Intron Length Polymorphism Markers in rice) and Cheng's Index method. The discrimination value Di and component index of Indica-japonica were selected as qualitative index to distinguish the amplified fragment polymorphism and classify types of indica-japonica. The results show that indica-japonica differentiation existed at the detection sites in the genomic DNA from materials. no indica type (or indica type) was detected with 100% component index. Of 18 parents, 4 japonica, 5 japonica-clinious, 8 indica-clinous and 1 indica were detected to contain 79.69%, 32.08%, 10.94%, and 6.25% of average component indexes of japonica respectively. In 39 deriving pedigrees of rice subspecies indica-japonica, 1 japonica, 11 japonica-clinious, 20 indica-clinous and 7 indica were detected to have 68.70%, 24.62%, 12.92%, and 4.76% of average component indexes of japonica respectively. The component indexes of Nippornbare and 9311, which belong to japonica and indica, were 95.83% and 4.17% respectively. The indexes of japonica rice cultivar C418 and Minghui 502 were 31.25% and 10.42% respectively, but indica cultivar 9308 and Minghui 63 were 33.33% and 12.50% respectively. Further analysis indicated that cultivar Shanyou 63 hybrid, which were derived from interbred from cultivar Minghui 63, had the largest plantation area in the world the heterosis between rice subspecies indica and japonica. There were also significant correlations between component index and Cheng's index, and the classification of indica and japonica was similar to the dermination of shape index. Therefore, it's reasonable to use component and determination index of indica or japonica obtained from ILP markers, in order to determinate the differentiation of indica and japonica. It's vital to utilize the useful genes from rice subspecies indica-japonica. 2. Studies on genetic diversity and genetic relatives of indica-japonica subspecies rice hybrid deriving pedigrees and the parents was conducted by using 221 SSR and pedigree analysis. The results showed that the indexes of genetic diversity, such as polymorphic marker number, polymorphic marker percentage (PPB), effective number of alleles (Ne), observed number of alleles (Na), Shannon diversity index (H) et al., from parents were higher than those from deriving pedigrees, the forth generation deriving pedigree was higher than the second pedigree. And the order of indica and japonica differentiation from high to low is japonica-clinious, indica-clinious, japonica, indica. The deriving pedigrees and parents were divided into 5 clusters by SSR markers, and then the third cluster divided into 4 sub-clusters. There was a significant difference between different indica-japonica types of deriving pedigrees or parents. The cultivars, which were closely relative to cultivar Minghui 63, were gathered at the second sub-cluster, and the third cluster. It's suggested that most deriving pedigrees could significantly improve the genetic basis of three line restorer. The results of population genetic structure indicates that with the improvement of deriving pedigrees, genetic component was increased and the component ratio of indica or japonica deriving pedigrees tends to be reasonable. Pedigree allele method was used to analyze the transfer of genetic materials from parents in deriving generation. The results showed that the genetic contribution rations of ancestors, such as cultivar Ketan Nangka, Jiananjin, 02428 and Minghui 72, to the first deriving generation were 15.09%, 21.32%, 22.78% and 34.36% respectively, the second generation were 6.05%, 7.09%, 7.98% and 13.41%, respectively, for the third were 3.51%, 4.04%, 4.56% and 7.92% respectively, for the forth were 3.31%, 3.23%, 3.68% and 7.55% respectively. The order of genetic contribution rations of ancestor parents to deriving generations from large to small is cultivar Minghui 72, 02428, cultivar Jiananjin, Ketan Nangka. Each parent including the specific genetic loci was also passed to the deriving offspings, and the genetic contribution to deriving pedigrees decreased with the increase of the deriving generations. Cultivar 92gk729 kept 10 SSR specific loci from ancestor parents after genetic reconstruction, the 5 loci of which could be stably passed to deriving generations with 23.21%, 35.71%, 82.14%, 64.29% and 41.07%. The forth deriving generation inherited 2.21 loci from parents on average. The cluster results of cultivar Ketan Nangka, Jiananjin and 02428 were similar. Parents and their deriving pedigrees, but had a significant difference from cultivar Minghui 72. The results suggested that the different genetic background of deriving pedigrees and helped to understand the genetic structure differentiation of deriving pedigrees.3. Using 4 indica or japonica sterile line and 54 deriving pedigrees and their parents, 216 cross combinations were generated with incomplete diallel cross design (NC-II), so that the relation between genetic differences of biparental indica-japonica and heterosis of yield characters could be discussed. 96 cross combinations, from 8 three-line and tow-line CMS and 12 rice subspecies differentiation with significant differences, were selected in order to research on the application sphere of genetic differences of parental incica-japonica. 41 ILP markers were used to detect the indica-japonica differentiation. The results revealed that the proportion index of japonica of CMS lines cultivar Chuenjiang 12A, Guangkang 13A, K17A, Zhong 9A, Jin 23A and II-32A was 92.50% (japonica), 9.76% (indica-clinious), 9.76% (indica-clinious), 5% (indica), 7.32% (indica), 7.69% (indica), respectively. The proportion index of japonica of TGMS lines cultivar Peiai 64S, Guangzhan 63S, and SE21S were 24.39 % ( japonica-clinious), 12.50 % ( indica-clinious) and 7.32 % (indica), respectively. Effects of genetic differences of parental indica-japonica on the heterosis of yield traits were diverse for the different materials, but had same trends in different places. Genetic differences of parental indica-japonica had considerable influence on the transgressive heterosis of setting grains per panicle and setting grain percentage, but little influence on the effective panicles and 1000-grain weight. The result of research on the application sphere of genetic differences of parental incica-japonica showed that the character of grain weight per plant was the lowest negative heterosis when the difference of restorer to sterile lines was -5%. When the difference from -5% to 0, the positive heterosis increased, When the difference from 0 to 20%, the positive heterosis was much more than 20%, but the positive heterosis decreased from 20% to 25%. The hybrid rice with strong heterosis could be screened from the restorer lines, which were indica-clinious and japonica-clinious type, when the difference from 0 to 10% and 10% to 20%, respectively. The application range of genetic difference of the parental indica-japonica in two-line hybrid rice was much wider than in three-line hybrid rice. It could be easy to select high yield combinations from the difference of 7.7% to 15%.4. On the basis of application range, the effects of genetic difference from parental indica-japonica on the heterosis of stalk and photosynthetic characters were further analyzed. The results indicated that there was a significantly or extremely negative correlation between the component index differences of parents and lodging index of heterosis, but no significant correlation between bending moment and stem ellipse section area of heterosis in all cross combinations. There was a significant correlation between the component index differences of parents with breaking resistance and stem wall section area of transgressive heterosis. The component index of parents was significantly correlated with the photosynthetic characters at full heading stage of tansgreesive or mid-parent heterosis, but at maturity there was a positive correlation.5. By using 8 hybrid lines of subspecies rice and 8 three-line and tow-line sterile lines, two generations of parent and F1 hybrid were selected by using incomplete diallel cross design (NC-â…¡), so that the variation trends of phenotypic value of 9 yield characters, genetic effects and heterosis in different deriving pedigrees was analyzed with additive-dominance genetic model. The results of phenotypic value showed that grain weight per plant increased along with the increase of deriving generations. And the change of the second and the forth deriving generation was not significant, but both of them increased much more than the first deriving generation. The total seed per panicles and the setting seeds per panicles of parents and F1 aggrandized, but plant height in the second deriving pedigrees and the third deriving pedigrees changed little, but increased much more than the first deriving generation. The small change of heading stage between deriving generations indicates that transgressive heterosis has good improving effect on controlling growth period of F1. Genetic effect analysis reckoned that additive variance of total seeds per panicles and setting seeds per panicles increased of phenotypic variance, but dominant variance decreased, and the environmental effect was not obvious. The additive variance of growth period and 1000-seeds weight of phenotypic variance in the forth deriving generation was larger than those in the first and second deriving generation, and dominant variance of phenotypic variance was low. The results of heterosis analysis showed that growth period of population average and transgressive heterosis changed less in different deriving generations, and most were negative heterosis. The total seeds per panicles, setting seeds per panicles and plant height decreased, and most were positive heterosis.6. By using 8 hybrid strains of the forth deriving generation and 8 three-line and tow-line sterile lines, two generations of parent and F1 hybrid were selected with incomplete diallel cross design (NC-â…¡), so that the genetic effects and heterosis of yield characters under different types of environments was analyzed by using additive-dominance genetic model, and the value in use of hybrid strains was estimated. The results show that 8 characters were controlled by the main genetic effect, and the main genetic effect variance was from 54.9% to 91.3% of genetic variance except for grain weight per plant. The main genetic effect variance of characters, such as heading stage, effective panicles per plant, total seeds per panicles, setting seeds per panicles and 1000-seeds weight, mainly was the additive variance, whose main genetic effects were 76.8%, 70.9%, 94.3%, 84.95 and 85.8%, respectively. The characters of them could be stably inherited by selection. The main genetic effect variance of characters, such as grain weight per plant, setting percentage, plant height and panicle length, mainly was the dominant variance, which was 68.8%, 78.8%, 56.3% and 61.3% of main genetic effect, respectively. These 4 characters could be utilized by heterosis. The results of heterosis analysis showed that there were significant or extremely significant positive heterosis in 7 characters, except effective panicles per plant and heading stage. The grain weight per plant was the biggest (18.5%), and 1000-grain weight was the smallest, only 3.2%. The stability of population average heterosis of effective panicles, 1000-grain weight and plant height was better than the other 6 characters. There were significant negative population transgressive heterosis in the 6 characters, except for grain weight per plant, plant height and panicle length. The stability of population trangressive heterosis of plant height, heading stage and effective panicle per plant was better than other 6 characters. Analysis of breeding value for breed strains indicated that cultivar Minghui 413 could increase effect for the total seeds per panicles, panicles length and grain weight per plant in hybrid progeny, cultivar Minghui 509 could increase effect for the total seeds per panicles, seeds setting rate, panicles length and grain weight per plant, cultivar Minghui 417 could increase effect for setting percentage, 1000-grain weight, panicles length and grain weight per plant, cultivar Minghui 118 could increase effect for setting percentage and grain weight per plant, cultivar Minghui 502 could increase effect for panicles length, total seeds per panicles and seeds setting rate, cultivar Minghui 503 and 512 could increase effect for total seeds per panicles, Minghui 398 could decrease effect for plant height and increase effect for 1000- grain weight. |