There is considerable evidence indicating an association between aflatoxin ingestion and liver cancer in humans. The development of methods that would permit the monitoring of aflatoxin exposure in individuals would provide useful information in assessing human risk from this toxin. In this study, an ELISA technique, using a polyclonal antibody raised against aflatoxin B1 in the rabbit, was developed for monitoring the levels of aflatoxin excreted in human urine samples. Urine could not be used directly in the assay because high blanks were observed from the presence of some aflatoxin-like substances. A 'clean-up' procedure using Sep-Pak C[13] cartridges and immunoaffinity columns was developed. The methods were validated using [3]H-AFB[1] in both buffer and presumed uncontaminated human urine samples from Europeans. As AFB[1] itself is unlikely to be found in human samples, urine from marmoset monkeys treated with [14]C-AFB1 were also used in the validation processes as they are likely to contain a spectrum of aflatoxin metabolites similar to those in human urine. The levels of AFB[1]. equivalents were monitored by both radioactive counting and ELISA. The radioactive measurements demonstrated that the overall recovery of the ELISA method was approximately 50%. The failure to detect all of the aflatoxin contamination in the urine may be due to the inability of the antibody to detect most of the polar aflatoxin metabolites. When the ELISA method was used in monitoring aflatoxin excreted in patients with or without liver disease from Thailand, a range of aflatoxin levels in the urine samples were obtained. Since in Thailand, liver fluke infection, caused by Opisthorchis viverrini is a serious health problem in the area where there is high contamination with aflatoxin in food and a high incidence of liver cancer, the interaction between liver fluke infection and aflatoxin ingestion was also studied using hamster as an animal model. The results indicated that liver fluke infection may alter the pattern of aflatoxin metabolites excreted in hamster urine, which in turn could affect the results obtained from the ELISA. Measurements by the ELISA method of aflatoxin excreted in urine samples from Thai vegetarians; who are a population at high risk of exposure to aflatoxin suggested that this group of people excreted higher levels of aflatoxin, but a further study with a greater number of subjects would be necessary to confirm this finding.