Efeito do desbaste seletivo sobre a divergência genética em progênies de Pinus caribaea Morelet var. bahamensis
Título alternativoEffect of selective thinning on the genetic divergence in Pinus caribaea Morelet var. bahamensis progenies
Data de publicação2007-03-01
Direito de acesso
MetadadosExibir registro completo
The objective of this work was to study the effect of selective thinning on the genetic divergence in progenies of Pinus caribaea var. bahamensis, aiming to identify the most productive and divergent progenies for the use of improvement program. The test of progenies containing 119 progenies and two commercial controls were planted in March 1990, using 11 × 11 square lattice design, sextuple, partially balanced, disposed in lineal plots with six trees in the spacing of 3,0 × 3,0m. 13 years after planting thinning was realized (selection for DBH), with 50% selection intensity based on Multi-effect index, leaving three trees per plot in all the experiment. The evaluations were done at four situations: A (before the thinning); B (thinned trees); C (remaining trees after thinning) and D (one year after thinning). The analyzed traits were: height, diameter at breast height (DBH), volume, form of stem and wood density. The genetic divergence among the progenies was studied with aid of the canonical variables and of clustering of Tocher method, using the generalized distance matrix of Mahalanobis (D2) as estimate of the genetic similarity. The progenies were grouped in four groups in situation A, fourteen in the situation B, two in the situation C and three in the situation D. The selective thinning of the trees within of the progenies caused a change in the genetic divergence among the progenies, genetically homogenizing the progenies, as demonstrated by the generalized distances of Mahalanobis, clustering of Tocher' and canonical variables methods. The thinning made possible a high uniformity in respect to the relative contribution of the traits for the total genetic divergence. The techniques of clustering were efficient to identify groups of divergent progenies for the use hybridization and little divergent progenies for the use in backcross program.