A dose intercomparison was carried out by the National Physical Laboratory between the seven radiotherapy centres which contributed the largest number of patients to the British Institute of Radiology fractionation study of three fractions per week versus five fractions per week in clinical cancer treatment. Six of the centres showed remarkable agreement within the acceptable limits of error of the measurements. In one centre there appeared to be a physical dose discrepancy of 2.8% which was materially less than could be detected clinically.