Adrenocorticotrophin levels, measured by a cytochemical bioassay, were determined in the plasma and cerebrospinal fluid (CSF) of adult female rhesus monkeys which were ovariectomized and receiving oestrogen replacement therapy. In control monkeys, ACTH bioactivity was found in both CSF (10.2 +/- 1.8 ng/l) and plasma (186 +/- 51 ng/l) in samples taken at 14.00 h (lights on: 07.00-19.00 h). Dexamethasone treatment (0.2 mg/kg) twice daily for 4 days suppressed plasma ACTH levels (52.8 +/- 25.2 ng/l) but had no effect on CSF levels (7.6 +/- 2.7 ng/l). Raising plasma ACTH, either by daily injections of a long-acting preparation of ACTH (1-24) for 6 days or by bilateral adrenalectomy (and subsequently with-drawing cortisol replacement therapy) also resulted in no detectable changes in ACTH levels in the CSF. A regression analysis between ACTH in the plasma and CSF from samples taken throughout the experiments showed no correlation. In contrast, measurement of ACTH by radioimmunoassay, whilst satisfactory for determination of this peptide in plasma, could not identify authentic ACTH in the CSF. It is concluded that bioactive ACTH does not enter the CSF in detectable quantities from either the peripheral vascular compartment or from the animal's own pituitary gland, and that reducing ACTH secretion from the pituitary also has no effect on levels of ACTH in the CSF. This is in marked contrast to other pituitary peptide hormones, including prolactin, which is secreted together with ACTH during 'stress' but which, unlike ACTH, enters the CSF relatively easily.