Identification of new descriptors for differentiation of soybean genotypes by Gower algorithm

The descriptors used to differentiate new soybean cultivars which ensure their protection, has become insufficient. This makes evident the need to expand the list of that already used. The morphological traits that may be used as descriptors can be quantitative or qualitative; however, in the evaluation of their efficiency, they are not always analyzed simultaneously. Thus, this study aimed to identify new morphological characteristics for purposes of discrimination of soybean cultivars with the aid of Gower algorithm, which allows evaluation of quantitative and qualitative characters simultaneously. Ten soybean cultivars were used in two experiments, one in the summer and another in the winter. They were conducted in a greenhouse. The experiments were carried out in an experimental randomized complete block design with four replications. Each experimental unit consisted of one pot with two plants. Five quantitative and seventeen qualitative characteristics were evaluated and the discrimination analysis was performed by Gower algorithm. It was concluded that there are differences among cultivars for all traits analyzed in both sowing dates, except for the opening of the petioles of the unifoliolate leaves. The use of Gower algorithm was efficient in the evaluation of discriminatory capacity of quantitative and qualitative characteristics simultaneously, demonstrating that they are useful as soybean descriptors.   Key words: Glycine max (L.), law on plant variety protection, productivity, qualitative variables.