Following conflicting results from Hammersmith and Edinburgh, the 62.5 MeV (p-->Be+) Douglas cyclotron was installed at Clatterbridge in order to carry out further studies with fast neutrons. Several features were incorporated into the study design to achieve as unbiased as possible a comparison between 62.5 MeV neutrons and conventional 8 MV x-ray therapy. Interim analysis of 151 patients in the pelvic study in the autumn of 1989 revealed a trend towards a worse survival in the neutron therapy group which soon became significant, leading to study termination in February 1990. The reasons for this diminished survival were unclear; with no significant difference in morbidity. Although the incidence of metastases was initially higher in the neutron patients than the photon patients this difference was not sufficient to explain the survival difference. Considerable morbidity would be expected from photon therapy using the same fractionation as was used in the neutron arm of the trial. If further neutron therapy at this energy is planned consideration should be given to the use of smaller fractions.