The term Cushing's disease is applied to those cases of Cushing's syndrome in which hypercortisolism is secondary to inappropriate secretion of ACTH by the pituitary. Studies on control of ACTH secretion in these patients reveal: (a) that the episodic secretion of ACTH is similar to the normal; however, frequency and amplitude of the secretory episodes lack the normal circadian rhythm; (b) that ACTH release can be stimulated by vasopressin and metyrapone in a normal or above-normal manner; and (c) that it can be suppressed by large doses of corticosteroids. When the dynamic aspects of the ACTH response to corticosteroid administration are studied, it appears that the normally negative differential feedback mechanism is converted into a positive one, whereas the delayed, integral mechanism is undisturbed. Patients with Cushing's disease in the presence of obvious pituitary tumors cannot be distinguished from those without pituitary tumors by studying only the pituitary function. All these and other well-known facts would favor the concept that ACTH secretion in Cushing's disease is under hypothalamic control whether or not a pituitary tumor is present. Moreover, there are observations that suggest that brain centers superior to the hypophysiotropic area of the hypothalamus are involved in the pathophysiology of Cushing's disease. This concept has led to the discovery of neurotropic drugs that are able to induce complete remission of Cushing's syndrome in a cerain percentage of patients. In some patients with severe psychiatric diseases, neuroendocrine abnormalities are present that resemble closely those characteristic for Cushing's disease. With the most refined neuroradiological methods, pituitary microadenomas are demonstrable in approximately 70% of patients with Cushing's disease, and this number compares well with those of earlier autopsy findings (70 to 80%). In a small number of patients (4 to 10%), these tumors are large and can easily be detected by standard roentgenograms of the head. Recent studies on the frequency of these large tumors do not support the hypothesis that adrenalectomy accelerates the progression of these tumors. In this case the term "Nelson's syndrome" would be uncessary. It is established that complete cure of Cushing's disease can be obtained in most patients with selective removal of a microadenoma from the pituitary gland. The current experience with this microsurgical procedure caused a renewed interest in Cushing's original suggestion that the disease is primarily a pituitary disorder. However, there are already a number of enigmatic observations. Possibly, the recent ultrastructural studies using immunocytochemical methods will resolve some of these problems. At this moment it is impossible to decide whether Cushing's disease is primarily a CNS or a pituitary disorder, when all arguments for one or the other hypothesis are taken into account...