Biodiv Sci ›› 2010, Vol. 18 ›› Issue (1): 37-43.DOI: 10.3724/SP.J.1003.2010.037

• Original Papers • Previous Articles     Next Articles

Phenotypic diversity of sugarcane variety germplasm

Xinlong Liu1,2, Qing Cai1,3, Caiwen Wu1,2, Li Ma1,2, Xiongmei Ying1,2, Xin Lu1,2, Yuanhong Fan1,2,*()   

  1. 1 Yunnan Key Laboratory of Sugarcane Genetic Improvement, Kaiyuan, Yunnan 661600
    2 Sugarcane Research Institute, Yunnan Academy of Agricultural Sciences, Kaiyuan, Yunnan 661600
    3 Biotechnology & Genetic Resources Institute, Yunnan Academy of Agricultural Sciences, Kunming 650223
  • Received:2009-04-09 Accepted:2009-08-23 Online:2010-01-20 Published:2010-01-20
  • Contact: Yuanhong Fan


Sugarcane varieties are important in providing parental resources in sugarcane breeding programs. In order to improve efficiency in selecting sugarcane varieties as parents, to provide guidance in choosing the best parent combinations for crossing, and to assist in choosing core collections of parents, the genetic variation, genetic structure and relationships of a range of traits were studied in a set of sugarcane varieties from different origins. Seventeen qualitative traits and five quantitative traits were studied, and a range of statistics were determined including genetic variance for each quantitative trait, genetic distance, coefficient of genetic divergence, and genotypic relationships based on cluster analysis. The coefficient of variance (CV) of all quantitative traits differed among different groups of varieties based on their origins, that population from Hainan Province exhibited the largest genetic diversity. The analysis of genetic diversity of qualitative traits revealed that variety populations from the USA, Taiwan of China, and Australia had high levels of genetic diversity in trait values, suggesting there had been greater use of diverse parental materials in programs in those places. Results from analysis of coefficients of genetic divergence (Gst = 0.0479 for quantitative traits, Gst = 0.0950 for qualitative traits) indicated that a high proportion of total genetic variation was retained within the groups of origin, with a high level of overlap in variation between most groups, and the high gene flow (Nm = 4.7632 for qualitative traits, Nm = 9.9493 for quantitative traits) showed the active genetic exchange was found between different origins. Based on analysis of genetic distance, relationship between variety populations from different origins was examined. The genetic distance between origins ranged from 0.0261 to 0.2945, with the greatest similarity being evident in variety populations between Fujian and Guangdong provinces (China), Cuba and USA, Guangxi and Yunnan provinces (China), Australia and Philippines, Jiangxi and Sichuan provinces (China), Brazil and France. This may reflect similarities in ancestors of parents used in these pairs of programs, in particular for the variety populations from Fujian and Guangdong which were the most similar. It is suggested these sugarcane variety populations which possess rich phenotypic diversity should be utilized and those crosses between sugarcane variety populations within the same groups from cluster analysis should be avoided. Sugarcane variety populations from Mexico, which are quite dissimilar to most other groups, should be given a higher priority in breeding programs.

Key words: coefficient of variance, quantitative traits, qualitative traits, origins