The Authenticity Identification And Variation Of Interspecific Hybrids Between Miscanthus Floridulus And M. Sacchariflorus

Miscanthus, which belongs to Poaceae, Subfam. Panicoideae A. Braun, Trib. Andropogoneae Dumortier, Subtrib.Saccharinae Grisebach, is a kind of C4perennial grass. For its high biomass yield and it can grow on non-cultivated land such as desert, Miscanthus can bring both ecological and economic benefits. So it has been an important direction for the study of Miscanthus that cultivating Miscanthus hybrids which have high quality and can be used for industrial production. In this study, Miscanthus floridulus and Miscanthus sacchariflorus were selected as parents to cultivate artificial hybrids and our purpose is to breeding for new species which have strong resistance, high-yield and general adaptability through the analysis of genetic variation of hybrid progeny combining with the excellent characters of the parents. The results were as follows:1.The study of hybrids authenticity identificationSSR markers and morphological markers were used for identifying the authenticity of the hybrids resulted from the interspecific crosses between Miscanthus floridulus and M. sacchariflorus. The results showed that two SSR markers (HAU-170and A3) identified100%of plants from M.floridulus×M. sacchariflorus and77.46%of plants from M. sacchariflorus×M. floridulus to be the true hybrids. On the other hand,100%of plants from M. floridulus×M. sacchariflorus and77.67%of plants from M. sacchariflorus×M. floridulus were considered as the true hybrids by morphological markers. Consistency between the results obtained by SSR markers and morphological markers reached97.3%.2. The analysis of hybrids reproduction biologyThe flowering phenology, pollen sterility, seed setting and the meiotic process of pollen mother cells of artificial hybrids from the interspecific crosses between Miscanthus floridulus and M. sacchariflorus were observed. The results showed that:(1) The flowering duration of F1plants was long, meanwhile, the flowering between individuals had high synchronicity. The flowering phenology of F1was influenced by both parents and it had two peaks of flowering that one was in June which had overlap with parent M. sacchariflorus and the other one was in September which had overlap with parent M.floridulus.(2) The pollen sterility of hybrid plants was lower than the parents, but in barren soil, the seeds setting percentage was higher than the parents. The hybrid plants showed some heterosis.(3) The correlation analysis between flowering phenology and seed setting showed that there was a negative correlation between seed setting and the first flowering date, at the same time, there was a positive correlation between flower number and flowering duration and seed setting.(4) The meiotic process of pollen mother cells of Fl artificial hybrids was generally normal, only a few of pollen mother cells appeared abnormal behavior of chromosome. The incidence of abnormal behavior at different development stages was lower than2%and it illustrated that there was a close genetic relationship between M.floridulus and M. sacchariflorus.3. The analysis of hybrids genetic variationThe total of22phenotypic traits of F1artificial hybrids between M. floridulus and M. sacchariflorus were analysed by statistics. The results showed that the average of coefficient of variation was32.42±36.49and the variation of some traits related to yield was larger such as stem number per plant (30.21%), dry weight per tiller (24.39%) and dry weight per plant (40.40%), etc. It also showed that the dry weight per tiller had a extremely significant positive correlation to Axillary bud number per tiller, node number per tiller, first node short axis, first node long axis, dry weight of stem, dry weight of leaf, dry weight of panicle, panicle main axis length, third class branch number of panicle, largest leaf length, flag leaf length and plant height, but it had a negative correlation to second class branch number of panicle and flag leaf width. There was a significant positive correlation between dry weight per plant and stem number per plant. It showed that there was obviously negative heterosis in F1hybrids and there were not significant difference of the mid-parent heterosis and heterobeltiosis for the reciprocal cross hybrid plants and traits related to yield. F1reciprocal cross individuals was not clustering respectively, but clustering dispersedly in clustering analysis.