Immunomagnetic separation (IMS) has been specified as a standard method for the measurement of Cryptosporidium in some countries. In this study, the IMS method was evaluated on the basis of the recovery efficiencies of Cryptosporidium oocysts at various IMS operation conditions. The average recovery for different Cryptosporidium concentrations in deionized water was 82.6 +/- 18.2% (n = 52), No significant change in recovery was observed by altering the debris ratio of the water samples. The efficiency was increased by prolonging the reaction time, and by increasing the amount of immunomagnetic beads. The recoveries of oocysts seeded in an Eppendorf with a small reaction volume were similar to those seeded in glass tubes with 10 times the reaction volume. The recovery efficiency of oocysts was reduced significantly when the reaction buffer was replaced by PBS. In conclusion, this method has good reproducibility and high recovery.