Agrupamento multivariado de curvas na seleção de cultivares de alface quanto à conservação pós-colheita

One of the main problems for the commercialization of lettuce is the short shelf life due to loss of quality. One way to study the decline of plant quality in function of time after harvest is the graphical method, however this strategy may become difficult if a great number of treatments is evaluated. So, the grouping of curves using multivariate analysis can be a viable alternative. The objective of this work was to select lettuce cultivars presenting better post-harvest conservation through multivariate clustering curves. Fifteen lettuce cultivars were evaluated in terms of cumulative loss of fresh matter (PAMF) and discoloration degree (GD) during 72 hours after harvest. Thirteen regression models were adjusted, checking the best fit of the first-degree polynomial model for the PAMF and the logistic model for the GD. For both traits, the cultivars were grouped into six groups. The cultivars 'Romana Balao' and 'Vitoria de Santo Antao' are the most resistant to dehydration and discoloration in storage at environmental temperature, being recommended to production, marketing and as parents in breeding programs.