The United Kingdom Cyclosporin Quality Assessment Scheme has now over 6.5 years experience of the measurement of cyclosporin. The scheme has more than 154 member laboratories in 30 countries. Each laboratory is sent three samples a month for analysis and the returned results allow the laboratories to measure their performance relative to their peers. The data generated by the centres makes it possible to assess independently the performance of the methods available for the measurement of cyclosporin. The data for the 12 month period June 1989 to May 1990 were analysed. Over that period, the proportion of laboratories using plasma rather than whole blood fell from 9% to 7%. Eight different methods were available for the measurement of the drug but three techniques accounted for over three quarters of the results returned; the specific radioimmunoassays Cyclo Trac-SP (48%), and Sandimmun-SP (21%), and high performance liquid chromatography (8%). The specific immunoassays showed a positive bias, approximately 10%, relative to HPLC when measuring the drug in patient samples. However, there are no data to suggest these differences are of clinical significance and the specific immunoassays can be recommended in preference to HPLC because of their better precision and ease of use. The poor performance of some laboratories and the increasing number of methods available for the measurement of cyclosporin suggest that there is a continuing need for the external quality assessment of this measurement.