Variability of quantitative indicators of walnut fruits

Introduction. Obtaining homogeneous products is associated with the use of species that have low variability of economically significant characteristics. For an objective assessment, it is advisable to establish it using several methods. Walnut is a valuable food plant. Significant selection quantitative characteristics of its fruits are the weight of the fruit and kernel, the kernel yield, the total score of selection value. The goal of the research is to study the variability of quantitative indicators of walnut fruits using various methods. The objectives are the following: determination of the mass of fruits and kernels, kernel yield, total score of selection value in individual specimens of the species and establishment of their variability indices. The research methods. The indices of 6 forms, 70 nuts in each, were studied. The variation coefficient, relative entropy, variability indices of Shannon, Sukhorukikh-Biganova, Margalef, Menkhinik, polydominance, Berger-Parker were determined by known methods with the replacement of the number of species by the number of classes of distribution of the trait. The results were processed in the Stadia 8.0 program for Windows. The Results. When divided into 10 classes, the sum of the variability ranks has made: kernel yield - 25, fruit weight - 24.5, kernel weight - 22, total selection value score - 8. Nonparametric analysis of ranks has not revealed a reliable difference between the variability of nut weight, kernel, and kernel yield (Friedman ρ = 0.2-0.5, significance - 0.4795-0.6547). The variability of the total selection value score is significantly different from the others (Friedman ρ = 8, significance 0.00468). Consequently, significant efforts will be required for walnut breeding for uniformity according to the first three indicators, and the least efforts according to the last one. Low statistical correlation has been observed between the Berger-Parker and Margalef, Menkhinik indices (r = 0.4336), the variation coefficient and other indices (r = 0.3115-0.4339). Strong statistical correlation has been found between other indices (r = 0.7052-1).