The genetic diversity of most livestock species is being reduced, and it is not possible to preserve all of these livestock breeds. In order to preserve as much of the genetic diversity as possible, we must first have a robust method of measuring genetic diversity among breeds. Three different methods of study that graphically represent relationships among breeds are presented; Weitzman's method, principal component analysis (PCA) and a neighbour-joining tree with allele sharing. Diversity was evaluated on the basis of 15 microsatellite markers typed over a total of 513 DNA samples collected from five Spanish donkey breeds. Breed differentiation was confirmed by the clustering based on the genetic distances between individuals, which essentially grouped all individuals in discrete clusters. The genetic distance among breeds was used to measure the global diversity of the set in breeds considered, and to evaluate the marginal loss of diversity attached to each breed. The Catalana breed appeared to be the most 'unique' in the set considered. In addition to this, the usefulness of global evaluations of diversity using molecular markers to choose breeds is worthy of conservation.