Counts of Cryptosporidium oocysts in 100-L volume "spot" samples of treatedwater have been simulated for conditions representative of an outbreak. The counts agree well with those reported in outbreaks in the United Kingdom and the United States. Whereas 33 percent of 100-L samples contained no oocysts, a small proportion contained hundreds and even thousands of oocysts. The risk of Cryptosporidium infection is directly related to the total number of oocysts entering the distribution system in the treated water as represented by the arithmetic mean oocysts density. Monitoring programs based on taking single spot 100-L volume samples will tend to miss the "all important" high-count samples and thus underestimate the arithmetic mean and the risk to public health. Spot samples recording 0 oocysts/100 L therefore provide very little reassurance. To detect temporal variation in the form of infrequent high-count spikes, the Cryptosporidium regulations in the United Kingdom require continuous monitoring of treated water as it leaves the treatment facility. Includes 37 references, tables, figures.