Variations in palliative response of breast cancer to adrenalectomy can be explained either by variations in the glands removed, by variations in the biology of the neoplasm, or by variations in the host herself. Most of the reported studies of this subject have been devoted to the latter two concepts; seeking to discern either in the tumor (the anatomy of its metastases, its growth rate), or in some aspect of the host (the free interval, age, or menopausal status) some indicator of the anticipated response. Recently, attention has been devoted to the urinary hormone excretion as a measure of adrenal function, and subsequent derivation from that information of a discriminant or predictor of clinical response. A previous report compared the extent of adrenal atrophy observed after prolonged corticosteroid treatment with that observed following hypophysectomy in patients with advanced carcinoma of the breast.1Hypophysectomy produced a far more profound adrenal