Summary—A 3‐year interlaboratory proficiency testing for lymphocyte subset phenotyping was initiated as part of the Etalonorme national quality control program. Specimens consisted of fresh whole blood and of lyophilised mononuclear cells (Cytotrol Coulter). The number of participating laboratories was 62 in 1990, 99 in 1991 and 129 in 1992. Statistical analysis indicated that results of phenotyping, expressed as percentages of positive cells, are not related to reagents, instruments or differences in methodology (like, eg time and temperature of incubation). The highest dispersion was observed for total lymphocyte counts and was found to correlate with the type of calibration of the instrument. The coefficients of variation for different lymphocyte subsets were similar if phenotyping was performed on whole blood or lyophilised cells and varied inversely with the percentage of positive cells in each specimen. The consistency of the results indicated that they could serve as a basis for clinical decisi