The pathophysiological mechanism of sickle cell anemia has been thoroughly studied and is now well understood, in contrast to the extreme clinical heterogeneity of the disease. A possible genetic explanation for this diversity arose from the discovery of an HpaI restriction polymorphism 3' to the beta globin gene, in linkage disequilibrium with the Hb S mutation. This linkage is unequally distributed among ethnic groups in Africa and predominantly found in Central West Africa. A multipolymorphic analysis spanning 60 Kb of the beta globin gene cluster demonstrated that the sickle mutation arose at least 3 times in 3 different geographical areas (Atlantic West Africa, Central West Africa and Equatorial Central Africa) and expanded by malaria selection. Two genetic factors seem to have epistatic effects which differ when comparing the two first groups. The alpha thalassemia gene (-alpha) is distributed equally among African Black control populations (0.10). The frequency is significantly higher in the SS patients of the Benin area (Central West Africa), whereas it is unmodified in the patients of Senegal (Atlantic West Africa). Alpha thalassemia does not seem therefore to have exercised the same selective effect in this latter group. Secondly, fetal hemoglobin is quantitatively and qualitatively different in both groups. A high G gamma phenotype (greater than 60%) is found in Senegal, whereas a low G gamma phenotype is constant in Benin, without overlap between the two series. The total production of fetal hemoglobin is statistically higher, although only moderately, so in the first group.(ABSTRACT TRUNCATED AT 250 WORDS)