Keywords
variety-rootstock combinations;
genotype;
regression analysis;
correlation dependence;
yield of standard planting material.
Abstract
In order to determine the effect of varietal characteristics of rootstocks and graſts, used in the production of seedlings, on the yield of graſted standard planting material, a multidimensional cross-correlation analysis was carried out. The experiments were carried out on four varieties of graſts and five varieties of rootstocks in the conditions of graſting complex and an open-earth nursery. The following functions were used: in the first case, the yield of standard graſted cuttings aſter stratification; in the second case, the yield of standard graſted grape planting material. Digital coding of genoplasm presence in the studied varieties or in their
known parental forms in each element of graſted combination was used as regression parameters. Regression models of influence of the genotype of rootstock and graſt varieties on survival rate of graſted stratified grape cuttings (with an accuracy of 69.69%) and on the yield of standard graſted seedlings from an open-earth nursery (accuracy of 64.96%) were developed. The negative effect of presence of Vitis rupestris grape species and ‘Malbec’ graſt variety genoplasm on the affinity and yield of both standard stratified
graſted cuttings and seedlings was established. The positive effect of Vitis riparia grape species and ‘Viognier’ graſt variety on the yield of standard graſted grape seedlings was established. Due to the increase in the influence of uncontrolled environmental factors in an open-earth nursery, as well as previously removed solute graſted cuttings, the effect of genetic factors on the yield of standard seedlings decreases in comparison with production of stratified graſted cuttings. The presented approaches to the development of regression models can be used in long-term forecasting of compatibility levels of variety-rootstock combinations in scientific and production processes.