A nested-PCR assay, incorporating an internal positive control, was developed for Cryptosporidium monitoring in finished water. This assay was capable of reproducibly detecting 8 oocysts in spiked-filtered water samples collected from 5 South Australian water treatment plants. The RT-PCR assay of Kaucner and Stinear (Appl. Environ. Microbiol. 64(5) (1998) 1743) was also evaluated for the detection of Cryptosporidium parvum. Initially, under our experimental conditions, a detection level of 27 oocysts was achieved for spiked reagent water samples. This level was improved to 5 oocysts by modification of the method. Untreated South Australian source waters concentrated by calcium carbonate flocculation were found to be highly inhibitory to the RT-PCR assay. Concentration of similar samples using Envirochek filters appeared to eliminate PCR inhibition. While both methods possessed similar sensitivities the nested-PCR assay was more reproducible, more cost effective, simpler to perform and could detect both viable and non-viable intact Cryptosporidium parvum oocysts, which is an important consideration for plant operators. These factors make the nested-PCR assay the method of choice for screening large numbers of potable water samples, where a reliable low level of detection is essential.