Classical breeding methods that were developed to enable indirect selection have rarely been applied to grape breeding. These methods involve evaluating genetic diversity for a broad array of traits to determine whether there are strong associations that might allow indirect selection for traits that are difficult or expensive to measure. In order to evaluate these methods and determine which traits were strongly associated, 113 progeny from a cross between D8909-15 (Vitis rupestris x V. arizonica/girdiana) x B90-116 (V. vinifera) were assayed for 13 traits: number of clusters, leaf morphology, cluster length, peduncle length, number of berries per cluster, weight of 10 berries, seed number, nature of seeds, berry color, Brix, pH, titratable acidity, and anthocyanin content. D8909-15 is a wild source of resistance to Pierce’s disease and the dagger nematode, Xiphinema index, and B90-116 is a large-berried seedless table-grape selection. Multivariate procedures were applied to estimate the genetic divergence among the genotypes. Genetic variability was observed for all traits. The Ward grouping method was able to divide the progeny into 10 separate clusters, some with good yield and high-quality characteristics as well as resistance to dagger nematode and Pierce’s disease. The estimated correlations between the characteristics suggest that selection of genotypes with good productivity (number of clusters), high Brix values, moderate pH and acidity, and few seeds is possible. Principal component analysis found that the nature of the seeds proved to be invariant in the population, and the leaf scores had the lowest relative importance for the characterization and discrimination of the genotypes tested.
- © 2011 by the American Society for Enology and Viticulture