A set of 434 Trypanosoma cruzi stocks from a wide ecogeographical range was analysed by Multilocus Enzyme Electrophoresis for 22 genetic loci. Strong linkage disequilibrium, not associated with geographical distance, and 2 main genetic clusters each considerably heterogeneous, was observed. These results support the hypotheses previously proposed that T. cruzi natural populations are composed of highly diversified genetic clones distributed into 2 main phylogenetic lineages: lineage 1, the most ubiquitous in the endemic area, was more frequently observed in sylvatic cycles, whereas lineage 2, predominant in humans and domestic cycles, in the southern part of the area surveyed, was further partitioned into 5 lesser genetic subdivisions. T. cruzi appears therefore subdivided into at least 6 ‘discrete typing units’ or DTUs (Tibayrenc, 1998a–c). We have identified various specific isoenzyme markers (‘tags’; Tibayrenc, op. cit.) suitable for the routine identification of these DTUs for epidemiological tracking purposes. We discuss the correspondence with previous classifications and with the recent recommendations of the 90th anniversary of the discovery of Chagas disease symposium, as well as the impact of T. cruzi genetic variability on this parasite's biomedical diversity.