Amebiasis is a major health problem in developing countries. Entamoeba histolytica infects 5 x 10 8 people worldwide, more than 90% of them living in West and South East Africa, China, SE Asia, México, the western portion of South America and India (Smyth 1994). Only 10% of all infected individuals eventually develop invasive amebiasis ranging from mild (i.e. self-limited) or serious intestinal amebiasis, to life-threatening hepatic invasion (Walsh 1986).It has been suggested that there are in fact two different species of Entamoeba. One, the classic pathogenic E. histolytica discovered by Schaudinn in 1903. The other, the non-pathogenic E. dispar described by Brumpt in 1925, based mainly on epidemiological considerations. The concept of two Entamoeba species began to be objectively established by functional in vitro studies (Martínez-Palomo 1982, Mirelman & Chayen 1990) and zymodeme patterns (Sargeaunt 1988), culminating now with molecular genetic data (Tannich & Burchard 1991, Tannich 1996.Thus, the majority of ameba infected individuals may actually harbour E. dispar, leaving the E. histolytica as the agent of the relatively rare invasive intestinal and extraintestinal disease. Amebic abscess of the liver (AAL) on the other hand, occurs in less than 1% of individuals with invasive intestinal amebiasis (Martínez-Palomo et al. 1993).Why the vast majority of individuals (≥99%) appear to be resistant to extraintestinal invasion remains unknown but some inroads have now been made: a significant increase in the SCO1 complotype and in HLA-DR3 was found in 160 Mexican mestizos (adults and children) with AAL when compared to healthy controls, to adult patients with amebic rectocolitis or to asymptomatic amebic carriers (Arellano et al. 1996). DNA subtyping is now in process to confirm these findings.Studies of human amebiasis and of experimental animal models of amebic disease have suggested that cell-mediated rather than humoral immune mechanisms, are responsible for acquired immunity to AAL (De León 1970, Salata & Radvin 1986, Kretschmer & López-Osuna 1990. Moreover, axenically grown trophozoites of E. histolytica kill in vitro normal human neutrophils and eosinophils (not-withstanding the aid by antibody and complement), normal monocytes and lymphocytes, without suffering changes in their own viability (Jarumilinta & Kradolfer 1964, Artigas et al. 1966, Guerrant et al. 1981, Salata et al. 1985, 1987, López-Osuna et al. 1986, López-Osuna & Kretschmer 1989 fectively destroying virulent amebas, even though finally succumbing in the process as well (Salata et al. 1985). In our laboratory we found that the same occurs with activated human peripheral blood eosinophils (López-Osuna et al. 1992).There is no circulating eosinophilia in most protozoan infections, yet this is not necessarily an argument against their possible participation in AAL, specially because eosinophils are regularly found in the early inflammatory stages of experimental amebic liver disease (Tsutsumi et al. 1984), and since the bulk of eosinophils in ...