The strategy of integrating quantitative and multicategorical data was performed to evaluate genetic divergence among 88 accessions of Coffea arabica. Twelve quantitative traits were evaluated: total score of cup quality; weight of 3 L of "farm coffee"; grain yield; weight of 3 L of "farm coffee" with 11% moisture; weight of 3 L of processed coffee with 11% moisture; total "farm coffee" fruit in 1 L; percentage of dry grains, overripe grains, cherry grains, green-cane grains and green grains in 1 L of "farm coffee"; percentage of fruits with empty-locus in 1 L of "farm coffee". Six multicategorical traits related to cup quality (sweetness, acidity, body, taste, after taste, and general aspect) were also evaluated. Low genetic divergence among accessions was found when the analysis of quantitative and multicategorical data was taken individually. However, the integration of the traits favored a better visualization of genetic divergence among the evaluated accessions when the quantitative data were converted into multicategorical ones. This occurred because of the increased number of groups formed when the integration strategy was used.
Coffea arabica L.; multivariate analysis; germplasm bank
