Abstract: Arithmetic mean method is commonly used to evaluate the yield stability and adaptability of sugarcane varieties, and variance analysis is applied to estimate the errors in regional trials. However, it is difficult to accurately evaluate the differences of the varieties due to the discrepancies across test sites and years. In this paper, GGE-biplot method was adopted to analyze the data from the regional trials with seven sugarcane varieties at five sites from 2008 to 2009, aimed to objectively evaluate the yield stability and adaptability of sugarcane varieties in China. Among the test sugarcane varieties, Funong No.30 had higher cane yield and better yield stability, Yuegan No.18 had higher sugar content and better trait stability, Funong No.28 and Yunzhe 9991 had high sucrose content and trait stability, while Yuegan No.16 had the highest cane yield and sugar content but ordinary stability. In the test sites, Zhangzhou City in Fujian Province and Suixi City in Guangdong Province had the best representativeness and discrimination. This study showed that GGE-biplot analysis provided a simple and effective method to analyze the high yield and stability of sugarcane varieties in regional trials, and supplied the basis for the approval and extension of new sugarcane varieties.

Key words: GGE biplot, sugarcane, cane yield, sugar content.