The effect of starvation on adrenal cortical function was studied in 12 obese female subjects. Measurements of plasma and urine 17-hydroxycorticoids (17-OHCS) as Porter-Silber chromogens, urine 17-ketosteroids (17-KS), and increase in plasma 17-OHCS after ACTH infusion, were made before and after a 13-day period of starvation. Before starvation mean levels for plasma 17-OHCS at 8:00 pm, 8:00 am, and after ACTH, and urinary excretion of 17-OHCS and 17-KS were within the normal range. After 13 days' starvation, during which the average weight loss was 23 lb, the mean 24-hr urinary excretion of 17-OHCS and 17-KS was reduced by 27 and 47%, respectively. Both the evening and morning levels of plasma 17-OHCS became elevated after starvation, reaching values as high as 150 μg/100 ml (mean values: 8:00 pm, 52.4 μg/100 ml; 8:00 am, 61.9 μg/100 ml). Comparisons of the differences between 8:00 pm and 8:00 am plasma 17-OHCS and increases after ACTH stimulation, before and after starvation, revealed that no significant change had occurred, although absolute levels of plasma 17-OHCS had increased after starvation. Measurement of plasma cortisol by a chromatographic isotope dilution method after starvation in 2 subjects yielded normal values (6.3 and 14.4 μg/100 ml) despite high values (68.0 and 150 μg/100 ml) measured as Porter-Silber chromogens. Blue tetrazolium staining of the chromatograms ruled out the presence of other alpha-ketolic steroids in the plasmas, giving the Porter-Silber reaction. These observations indicate that, although starvation was associated with a marked increase in plasma 17-OHCS measured as Porter-Silber chromogens, this was not a reflection of the actual level of cortisol in the plasma. The increased chromogens in the plasma are most likely nonsteroidal material produced during starvation.