Major fatty acid composition of 19 almond cultivars of different origins. A chemometric approach

Multivariate chemometric techniques have been applied to the fatty acid composition of gas chromatographic data from 19 almond cultivars. Analysis of variance indicates the important contribution of all major fatty acids to the typification of almond cultivars. Principal component analysis applied to all of the individual values of fatty acids of the different cultivars leads to three new variables which accumulate about 90% of the total variance. The projection of different cultivars in the reduced space allows the visualization of some different cultivar groups. Cluster analysis classifies the studied almond cultivars into three groups. Within a large group many cultivars of the Mediterranean area and the American cultivar Non Pareil are found. The second group includes some Italian and American cultivars. A third group covers one American and one Australian cultivar, in association with two Spanish ones. Discriminant analysis applied to the groups previously established by cluster analysis and principal component analysis brings out two discriminant functions, which explain 68.53% and 31.47% of the total variance, respectively. Application of these discriminant functions verified a correct assignment (100%) of each member to its group.