I WISH to correct slightly my recent letter in NATURE1 on this subject. The number of neutrons emitted by a radium plus beryllium neutron source has been accurately measured (C. J. Bakker2 and other observers) and found to be (2·1 ± 0·2) × 104 neutrons per sec. per mC. radium. The source of 120 mgm. used in our experiments emitted during nine days about 1·96 × 1012 neutrons. If each neutron transformed an atom of sulphur into radio phosphorus (ideal case certainly not realized) we should obtain 1·96 × 1012 atoms of 3215P weighing about 10-10 gm. Hence the coloration and the increase in weight of 3·7 mgm. of the electrode in our experiment cannot be due to the radio-phosphorus.