been undertaken. An ophthalmologist suspected some impaired vision affecting the upper temporal quadrants; otherwise the eyes were normal. Because of recent activity, the tumour was given a dose of 4,300 rads with cobalt y-rays in twelve fractions over twenty-three days; estimated to be equivalent to about 4,500 rads given in twenty fractions over twenty-eight days using orthovoltage X-rays. After satisfactory clinical remission the patient developed defective vision in one eye about fourteen months later. Exploratory craniotomy for suspected recurrence revealed no evidence of tumour, but atrophy of the optic tracts, regarded as secondary to radiation damage. There was no evidence of injury to other related structures. Since this episode Jackson Richmond (1958), Crompton & Layton (1961), Bignami et al. (1963) and Almquist et al. (1964) have reported other cases of radiation damage not only to the optic chiasma, but also severe and fatal damage to the related hypothalamic region, following doses of radiation to the pituitary of similar magnitude. Almquist et al. (1964) comment on the relative radioresistance of the optic chiasma and tracts in contrast to the hypothalamus and white brain tissue and -refer to doses of about 100,000 r n-radiation, from implanted radio-yttrium seeds which were required to cause optic tract damage in monkeys. They consider that estimates of radiation dosage which produce radionecrosis of the optic nerve should remain open pending further investigations. A case of optic nerve damage only, from external irradiation, has been reported by Lawrence et al. (1963) two years after ablation of the pituitary for breast cancer, using a narrow beam of ac-particles from a 900 MeV accelerator. Pituitary doses of about 15,000 rads are given in six treatments over eleven days, which may, by estimate, amount to four or five times the maximum equivalent dose given by Xor y-rays to a pituitary adenoma. These workers have also irradiated the pituitary with a-particles for acromegaly and Cushing's disease, but the -fuil -value *of' this method, involving the use of very costly equipment, has yet to be assessed. Experience with a relatively small number of patients confirms that most cases of pituitary tumour sent for radiotherapy are chromophobe adenomata. Corroboration by Horrax (1958) and Sheline et al. (1964) of the earlier findings of Henderson (1939), that irradiation after surgery significantly increased the control rate, suggests that the combined attack is still to be preferred to surgery alone. There is as yet insufficient evidence of their radiosensitivity to warrant irradiation only of these adenomata, in preference to surgery: this however, may not apply with the same force to eosinophil tumours. Although response to irradiation of pituitary adenomata increases with size of dose, a slight risk of radiation damage to the hypothalamic region and optic tracts seems to exist when doses exceed the equivalent of about 4,500 rads delivered to the gland with orthovoltage X-rays in twenty fractions over twenty-eight days. A maximum dose of 3,750 r (3,650 rads) over a similar period as suggested by Jackson Richmond (1958) appears reasonably safe, but might be raised slightly when employing cobalt y-rays, or equivalent high energy radiation. Full cancericidal doses, risking morbidity, would be justifiable in treating malignant tumours. Pituitary adenomata associated with Cushing's disease have not been treated in recent years by external irradiation.