The objectives of this study were to select lettuce genotypes with potential for cultivation in protected environments; select genetically divergent genotypes to integrate breeding programs; compare multivariate analysis to have a more precise interpretation of the results; verify the relevance of evaluated traits to genetic divergence; and determine the most important characters in the evaluation of lettuce genotypes. Fifteen genotypes were studied in a randomized block design with four replications and 15 plants per plot. The experiment was carried out in protected environments and the following traits were evaluated: plant height, head diameter; head circumference; fresh matter of heads; marketable fresh matter of heads; root fresh matter; dry matter of commercial heads; dry matter of roots and number of commercial leaves. The genotypes were classified in four groups by Tocher's method. The genotypes that should preferably be used in protected environment and in breeding programs were 'Regina 500' and 'Vitória de S. Antão', from group I; 'Black Seed Simpson' and 'Lívia' from group II; 'Branca Boston' from group III and 'Romana Balão' from group IV. Although the relative contribution of the head circumference is only 0.50%, all traits were important in the study of genetic diversity. According to the genotypic correlation study, the assessment of the fresh mass characteristic of the aerial part can substitute the evaluation of commercial fresh and dry mass of the aerial part in the selection process, reducing the time and cost in a breeding program.
Lactuca sativa; multivariate analysis; protected environment; relative contribution; genotypic correlation
