PATH ANALYSIS, CANONICAL CORRELATIONS AND GENETIC DIVERGENCE: II. AMONG PRIMARY AND DISEASE CHARACTERS IN BLACK BEAN (Phaseolus vulgaris L.) GENOTYPES

Abstract

Twenty-five bean genotypes were evaluated in Lages, Santa Catarina, studying the influente of seven characters of agronomic importance on thc grain production per unit arca. The experiment, in randomized blocks with three replications per treatment, was conducted in the growing season of 1995/96 (November — January). The objectives of this work were: I ) estimate the direct and indirect effects among the primary yield components and plant diseases, through path analysis; 2) esti mate'the association intensity among the primary yield components and the main bean diseases; 3) estimate the genetic divergente of twenty-five black bean genotypes, using Mahalanobis statistics. The first group of variables (primary components) was composed by the number of pods per plant, number of grains per pod and 1000 grains weight; the second group (secondary components), by the characters of plant diseases such as leaf antrocnosis, pod antrocnosis, leaf bacterias and leaf rust. The canonical correlation analysis showed the importance of the component plant disease over the primar) , ones. The results of the path analysis showed the importance of the number of pods per plant and 1000 grains weight, which presented the largest direct effects and a high phenotypic correlation with grain yield. The Mahalanobis analysis made possible to discriminate the black bean genotypes regarding their degree of genetic similarity. It was also evidenced the wide genetic variability for all the evaluated characters, especially for the number of pods per plant, number of grains per pod, and leaf rust. Dueto the wide genetic variability observed, the use of these genotypes (genetically distant) in programs of bean genetic improvement using artificial hybridization, certainly would not cause any narrowing of the genetic basis (genetic erosion).