This study compared the recovery of Cryptosporidium oocysts and Giardia cysts ((oo)cysts) from raw waters using 4 different concentration-elution methods: flatbed membranes, FiltaMax foam, Envirochek HV capsules, and Hemoflow ultrafilters. The recovery efficiency of the combined immunomagnetic separation and staining steps was also determined. Analysis of variance of arcsine-transformed data demonstrated that recovery of Cryptosporidium oocysts by 2 of the methods was statistically equivalent (flatbed filtration 26.7% and Hemoflow 28.3%), with FiltaMax and Envirochek HV recoveries significantly lower (18.9% and 18.4%). Recovery of Giardia cysts was significantly higher using flatbed membrane filtration (42.2%) compared with the other 3 methods (Envirochek HV 29.3%, FiltaMax 29.0%, and Hemoflow 20.9%). All methods were generally acceptable and are suitable for laboratory use; 2 of the methods are also suitable for field use (FiltaMax and Envirochek HV). In conclusion, with recoveries generally being statistically equivalent or similar, practical considerations become important in determining which filters to use for particular circumstances. The results indicate that while low-turbidity or "finished" waters can be processed with consistently high recovery efficiencies, recoveries from raw water samples differ significantly with variations in raw water quality. The use of an internal control with each raw water sample is therefore highly recommended.