This study was undertaken to validate the relevance of Chloroquine (CQ) resistance markers pfcrt(76) and pfmdr1(86) in an endemic area in Tanzania. After treatment with CQ, recrudescence was distinguished from new infection by msp2 genotyping, and the number of concurrent infections was also determined. The rate of children with recrudescent parasites at day 7 and/or day 14 amounted to a parasitological failure rate of 22.4% using PCR. The mean multiplicity of infection at day 0 was 3.2 (n=71). The allelic frequencies of the mutated pfcrt(76) and pfmdr1(86) were estimated to be 92 and 77%, respectively. Both values exceeded by far the observed frequency of 14% of recrudescent parasites as calculated on the whole analysed parasite population taking multiple infections into account. Although neither mutant allele is of predictive value for parasitological resistance, there is evidence for a role of pfcrt(76) in CQ resistance in the natural parasite population. All wild-type pfcrt(76) alleles were eliminated before day 3, after the onset of CQ treatment and no recrudescent parasite with the wild-type allele was observed at later time points. The discrepancy between the rate of resistant parasites (14%) and the frequency of the mutant pfcrt(76) allele (92%), however, indicates that other polymorphisms and other factors must be involved in CQ resistance. No selective elimination of the pfmdr1(86) wild-type allele was observed.