The resolving power of genetic markers is determined by the level of polymorphism detected, which is determined by the mutation rate at the genomic sites involved. Variation at allozyme loci is caused by point mutations, which occur at low frequency (<10-<sup>6</sup> per meiosis). Moreover, only mutations modifying the net electric charge and conformation of proteins can be detected, reducing the resolving power of allozymes. In contrast, mutations at minisatellite and microsatellite loci, mainly due to changes in the number of repeat units of the core sequence, have been estimated to occur at the relatively high frequency of 10-3-10-2 and 10-5-10-2 per meiosis, respectively. The other markers presented generally show intermediate levels of polymorphism, resulting from base substitutions, insertions or deletions which may alter primer annealing sites and recognition sites of restriction enzymes, or change the size of restriction fragments and amplified products. In choosing the appropriate technique, the level of polymorphism detected by the marker needs to be considered in relation to the presumed degree of genetic relatedness within the material to be studied. Higher resolving power is required when samples are more closely related. For example, analyses within species or among closely related species may call for fast-evolving markers such as microsatellites. However if the objective is to study genetic relatedness at higher taxonomic levels (such as congeneric species), AFLPs or RFLPs may be a better choice because co-migrating fast-evolving markers will have less chance of being homologous. A primary guiding principle in marker selection is that more conservative markers (those having slower evolutionary rates) are needed with increasing evolutionary distance and vice-versa.