AIMS: To evaluate a commercially available enzyme immunoassay based on a monoclonal antibody to a genus specific Cryptosporidium (IDEIA Cryptosporidium; Dako) antigen for detecting Cryptosporidium oocysts in faecal and environmental samples. METHODS: 435 human faecal samples and post-filtration deposits from 10 reservoir samples, and from six tap water samples seeded with Cryptosporidium oocysts, were examined by EIA according to the manufacturer's instructions, and by microscopic examination of phenolauramine stained smears. Samples giving discrepant results were examined by specific immunofluorescence, before and after concentration of oocysts. RESULTS: Sixteen (3.6%) faecal samples were positive by both microscopy and EIA; five (1.1%) were positive by microscopy of auramine-phenol stained smears (but were not confirmed by specific immunofluorescence) and negative by EIA; one (0.2%) was positive by EIA alone, but confirmed by specific immunofluorescence; and 362 (83.2%) were negative by both microscopy and EIA. Compared with immunofluorescence positive faecal samples, the sensitivity of conventional microscopy and EIA were 94% and 100%, and specificity 76.4% and 100%, respectively. Fifty one (11.7%) were not examined by microscopy due to detection of other pathogens in a previous sample from that patient, but were found to be negative by EIA. Ten reservoir water samples (not suspected of being linked to cases of cryptosporidiosis) were negative by both microscopy and EIA. Of six samples of tap water seeded with varying concentrations of Cryptosporidium oocysts, two (10(2) and 10(3) oocysts/l) were positive by both microscopy and EIA, two (10 and 1/l) by EIA alone, and two (0.1/l and unseeded water) were negative by both microscopy and EIA. CONCLUSIONS: The kit is simple and rapid to use and offers a less subjective method than microscopy for detecting Cryptosporidium in faecal samples submitted to a busy diagnostic laboratory.