Theor Appl Genet (1984) 69:129-131 9 Springer-Verlag 1984 Genetic divergence among thirty-one genotypes of toria (Brassica campestris L.) grown in twelve environments D. Singh and P. K. Gupta Department of Agricultural Botany, Institute of Advanced Studies, Meerut University, Meerut-250 005, India Received January 12, 1984 Communicated by G. S. Khush Summary. Thirty-one toria genotypes were grown in 12 artificially created environments in order to study genetic divergence. D 2 estimates based on 12 characters were used in obtaining the clustering pattern and inter- and intracluster distances. Selection of divergent parents has been made in three ways, i.e., on the basis of (a) genetic divergence exhibited when grown in the richest and most productive environment; (b) stable and consistent values of divergence over all 12 environ- ments and (c) pooled analysis. Out of 31 genotypes, on the basis of stability, high yield and divergence six genotypes were found to be suitable for use in a breeding programme. Key words: Toria - Consistent genetic divergence - D 2 - Tocher method - Stable genotypes - Heterosis Introduction An analysis of genetic divergence among the available germplasm appears to be important for selecting genetically diverse and stable parents for breeding cultivars. However, since genetic divergence is based on metric traits which are greatly influenced by environ- mental fluctuations, the selected genotypes may not carry the genes for stability. For selecting high yielding and stable genotypes on the basis of consistent diver- gence, it would be desirable to conduct divergence analysis in varied environments in order to overcome environmental influences to some extent. In the present study, divergence analysis of 31 genotypes grown in 12 artificially manipulated environ- ments is presented. Based on this analysis, genotypes have been selected for a further breeding programme. Materials and methods Thirty-one toria genotypes were grown in a randomised block design at Meerut for two years (1977-78, 1978-79) and at Kanpur for one year (1978-79) under four micro-environ- ments (narrow and wide spacing; with and without fertiliser), resulting in a total of 12 environments (Singh 1982). Data were recorded on plot basis in each of the three replications. D 2 estimates were made following Mahalanobis' (1936) generalized distance as described by Rao (1952). The 12 characters involved in the analysis were: days to flowering, days to maturity, plant height, branches per plant, siliquae per plant, grains per siliqua, length of siliqua, weight of siliqua, seed index, seed yield, dry weight and harvest index. Clusters were prepared following Tocher's method (Rao 1952) in all twelve environments and by pooled analysis (Singh 1982). For the sake of brevity, only clusters based on the richest and the most productive environment and those based on pooled analysis are presented in this article. Clustering pattern in the 12 environments and the pooled analysis were also used to select genotypes which showed consistent divergence. Results Clustering patterns of 31 toria genotypes estimated from the most productive environment and from data pooled over all environments are given in Table 1. The inter- and intracluster distances estimated from the most productive environment and from data pooled over all environments are given in Tables 2 and 3, respectively. Clustering pattern of 11 high yielding and stable genotypes exhibiting consistent divergence in 12 environments and in the pooled analysis is presented in Table 4. Discussion Toria is a cross-pollinated crop, but the development of hybrid varieties is not practiced due to lack of prerequisites for