Adaptabilidade e estabilidade de variedades de cana-de-açúcar em Alagoas e Pernambuco

Abstract

The objective of this study was to assess the magnitude of the interaction between genotypes of cane sugar and environments and the adaptability and stability to the character of total sugar recovered (TSR) using the methodology proposed by Wricke (1965), Eberhart and Russell (1966), Cruz, Torres and Vencovsky (1989) and Annic-chiarico (1992). It was evaluated eight genotypes of sugar cane cultivated in Usina Trapiche in the State of Pernambuco and Usina Serra Grande in the State of Ala-goas. The experimental design was a randomized block design with four blocks during the period 2004 to 2008, referring to cuts, sugarcane, ratoon, second ratoon cane and fourth leaf, respectively. It was observed differences among genotypes within each environment. The evaluation of genotypes, has highlighted the G8 genotype to obtain the average productivity of TSR above the overall average, adaptable and suitable for all types of environments, according to Eberhart and Russell (1966), Cruz et al. (1989) and Annichiarico (1992), and high stability by the methods listed above and Wricke (1965). The G5 was the most productive, adaptable to all types of envi-ronments, but as it presents low stability, its behavior is unpredictable. At unfavorable environments, the genotype G4 was the best, with high productivity and good rre-sponse to environmental changes. It was noted a strong association between the results of the methods Eberhart and Russell (1966) and Cruz et al. (1989), which are based on linear regression analysis and simple linear regression bissegmented re-spectively. The use of methods of adaptability and stability in combination can be interesting and add information about the performance of genotypes evaluated, how-ever, the use of these methods, based on the same statistic is not recomendable.