Sequences corresponding to some of the polymorphic loci previously reported from Entamoeba histolytica have been detected in Entamoeba dispar. Comparison of nucleotide sequences of two loci between E. dispar strain SAW760 and E. histolytica strain HM-1:IMSS revealed significant differences in both repeat and flanking regions. The tandem repeat units varied not only in sequence but also in number and arrangement between the two species at both the loci. Using the sequences obtained, primer pairs aimed at amplifying species-specific products were designed and tested on a variety of E. histolytica and E. dispar samples. Amplification results were in complete agreement with the original species classification in all cases, and the PCR products displayed discernible size and pattern variations among the isolates.The realization that Entamoeba histolytica and Entamoeba dispar are two distinct but morphologically identical species (8) has had a major impact on all aspects of amebiasis research, most notably epidemiology (2, 3). The 10% prevalence estimate made by Walsh (21), though still widely quoted, was based on data collected before the formal redescription of the species and so does not distinguish between infections with E. histolytica, which can cause invasive intestinal and extraintestinal disease, and those with E. dispar, which do not. It is now known that even in areas where invasive amebiasis is common E. dispar is by far the more prevalent species (11).Tools that allowed accurate differentiation of the two species were clearly needed, and in the past decade differentiation based on DNA amplification has been a research focus of many groups. Species-specific primers that amplify regions of several different genes have been used (1,7,14,17,18,19). Using trophozoites in culture, comparisons showed that PCR is more sensitive and specific than the enzyme-linked immunosorbent assay-based stool antigen detection kits, which employ monoclonal antibodies for the detection and differentiation of E. histolytica and E. dispar (16). A significant percentage of individuals in areas of high endemicity could be simultaneously infected with both E. histolytica and E. dispar (16). However, the major drawback of using culture is that mixed infections are overlooked and were only detected after PCR was used on DNA extracted directly from stool (12). Field studies that compared PCR and antigen detection methods directly on stool samples suggest that these methods perform equally well (13).But is E. dispar really nonpathogenic, and should it on this basis be completely dismissed as a subject for further investigation? It has been shown to be capable of producing variable focal intestinal lesions in animals (4, 9, 20) and of destroying epithelial cell monolayers in vitro (10). There is also some evidence that pathological changes may occur in some humans (15), though invasive lesions and symptomatic infections have to date not been reported. Whether these characteristics are variable among strains is unknown. None of the above s...