Serum concentrations of 3,3',5'-triiodothyronine (reverse T3, rT3) were measured in adult patients with several systemic illnesses whose serum total and/or free T3 were low, serum total T4 was low or normal, and free T4 was either normal or elevated. The mean serum rT3 was 76, 46, and 77 ng per 100 ml in patients with hepatic cirrhosis, chronic renal failure, and acute febrile illnesses, respectively; the values in patients with hepatic cirrhosis and acute febrile illness were significantly higher than, and values in patients with renal failure did not differ significantly from, the mean serum rT3 (41 ng per 100 ml) in normal subjects. The mean serum rT3 in another group of patients from Calcutta, India, who had severe protein calorie malnutrition (PCM), was 53 ng per 100 ml; it was significantly higher than the corresponding value, 22 ng per 100 ml, in the same patients after feeding treatment. Mean serum rT3 in patients with systemic illnesses was not so high as that (151 ng per 100 ml) in the normal newborn, who also has low serum T3 and normal or high T4. High serum rT3 in patients with systemic illness could not be attributed to increased serum protein binding of rT3; whenever studied, the dialyzable fraction of rT3 was not decreased but actually increased. The mean serum-free rT3 was 450,207, and 366 pg per 100 per 100 ml in patients with hepatic cirrhosis, chronic renal failure, and acute febrile illnesses, respectively; each of these values was significantly higher than the corresponding value, 98 pg per 100 ml, in normal subjects. The mean serum free rT3, 516 pg per 100 ml, in newborn cord sera was similar to that in patients with hepatic cirrhosis but was higher than that observed in patients with chronic renal failure and acute febrile illnesses. High serum rT3 and low serum T3 in patients with PCM improved to normal or towards normal after feeding treatment. Since the peripheral metabolism of T4 is normally the predominant source of T3 as well as rT3 in man, our data, demonstrating reciprocal changes in serum rT3 and T3 and no consistent change in serum T4, suggest that body metabolism of T4 may be so altered in systemic illness that the conversion of T4 to rT3 may be increased while that to T3 is decreased. The mechanism or the biological significance of such a diversion of T4, from the normally occurring conversion to highly potent T3, to the generation of poorly calorigenic rT3 in systemic illness, is not clear at this time. The data in patients with PCM demonstrate, however, that such a change in the metabolism of T4 can be reversible.