We have measured plasma thyroxine (T4), triiodothyronine (T3), and TSH with specific radioimmunoassays in rats during adaptation to severe iodine deficiency after they had previously received regimens supplying various quantities of iodine. Rats were maintained on a high-iodine diet (HID) containing 3 mg iodine/kg or a low-iodine diet (LID) containing 30 mug iodine/kg supplemented with 0.1, 0.2, or 0.4 mug iodine/ml of drinking water before swtiching to KID alone. Frequent serial blood samples were obtained up to 3 months, using 6 or more animals for each time interval. In animals originally fed HID, T4 remained at 4-6 mug/100 ml unitl the tenth day of LID, then rapidly decreased to a value of less than 0.4 mug/100 ml at 1 month. TSH was initially 50 muU/ml and increased linearly to 165 muU/ml on day 16. Thence there was a much more rapid rate of rise to 640 muU/ml at 38 days. The rats changed to LID alone after having been fed LID with iodine supplementation underwent similar qualitative hormonal changes. However, the decrease in plasma T4 and the increase in plasma TSH occurred sooner in the rats which had drunk water containing only 0.1 or 0.2 mug iodine/ml than in the previous experiment. Rats which had received 0.4 mug iodine/ml showed a pattern essentially identical to that of the animals which had been fed HID. plasma T3 did not change significantly in any of the experiments, remaining at 60-90 ng/100 ml, although there was a tendency for the values to be somewhat lower after several weeks of LID. There was a highly significant negative correlation of plasma T4 with plasma TSH. There was no significant correlation of plasma T3 with either plasma T4 or plasma TSH. It is concluded that the combined physiologic effect of plasma T4 and T3 concentration is more important in determining TSH secretion through negative feedback effects on the hypothalamus and/or pituitary than is the concentration of plasma T3.