Tamanho ideal de amostra em diferentes espaçamentos entre plantas para avaliação de famílias de cana-de-açúcar

Abstract

Selection in the initial phase of the sugarcane breeding program, during which are assessed the families with high genetic variability between individuals, is a complex process and, most of the time, inefficient. Selection between families has been used in an attempt at increasing this efficiency, first identifying the best families and, later, selecting within the families with greater potential. This process is normally done during trials in which are collected phenotypical data from the families and, from them, are estimated the values of the traits with agronomic interest. Generally, the assessment of these traits is done by sampling. The size of the samples may vary according to the studied trait and the experimental design adopted. Thus, the present work has the objective of identifying the ideal sample size to represent a families’ potential in order to identify the superior families and determine with which spacing a better phenotypical expression can be obtained for the progenies. For such, five sugarcane families of full-sib brothers, with 360 individuals each, planted in three experiments, in a randomized blocks design, with three replicates and three spacings between plants in the planting line (50 cm, 75 cm and 100 cm), and 150 cm between planting lines. The resampling method was used to determine the ideal sample size, as well as the best spacing for assessment, followed by analyses of variance for the variables: number of stalks, average diameter of the stalks, average height of the stalks and weight of the stalks. Based on the evaluated data, the 100 cm spacing presented the highest means for number of stalks, average diameter of the stalks and estimated weight of the stalks, while the spacing of 50 cm presented the highest values for average height of the stalks. The spacing of 75 cm between plants presented higher discrimination power between families for all evaluated traits. With this spacing and a sample of a minimum of 30 plants per replicate, considering three replicates, it was possible to identify superior families.