Avaliação e repetibilidade de caracteres agroindustriais de genótipos RB de cana-de-açúcar no litoral norte de Pernambuco

Abstract

The sugarcane (Saccharum spp.) presents itself as a culture of great social and economic importance for Brazil. The breeding programs have contributed to increasing productivity in sugarcane areas of the country, due to the replacement and renewal of sugarcane, this increase amounted to thirty percent increases in the last three decades. However, the efficiency of the breeding programs of sugarcane depends on the spatial and temporal repeatability of traits under selection. Thus, assessment surveys genotype x environment interaction and repeatability of characters is crucial to the selection result in success. The objective of this study was to evaluate the behavior of agroindustrial genotypes sugarcane in all three cutting times and reliably determine the number of evaluations needed to select new genotypes based on their production components. The experiments were conducted: in Santa Teresa Mill (Goiana) in the years 2006/2007, 2007/2008, 2008/2009, evaluating 25 genotypes; in São José Mill (Igarassú) in the years 2005/2006, 2006/2007, 2007 / 2008 and 2008/2009, to evaluate the repeatability of characters in 16 genotypes, located on the northern coast of Pernambuco, both in a randomized block design, with plots consisting of five eight-meter grooves and spaced one meter. In each section were measured variables: ton of pol per hectare (TPH), ton of cane per hectare (TCH) pol in cane corrected (PCC), fiber (FIB), purity (PZA), soluble solids (° BRIX ) and total recoverable sugar (TRS). Data were subjected to analysis of variance, the group averages by testing Skott & Knott and repeatability of the experiment than those we used: ANAVA, principal component analysis (variance and covariance) and structural analysis. Based on the results of gross income, the best genotypes for beginning and middle of harvest were RB86 7515, RB92 579 and SP81 3250, to end of season highlight for genotypes RB86 7515, SP78 4764 and SP81 3250. To select new genotypes based on the coefficient of repeatability, methods of structural analysis and principal components were equivalent, and depending on the variable in the analysis and reliability stipulated, the number of assessments varied: for TPH and TCH were needed for reliability of 2 reviews 90% for FIB, PCC, PZA, BRIX and ATR were 3-4 consecutive assessments.