S tudies of the rate of extrathyroidal conversion of thyroxine (T4) to 3,5,3′-triiodo-L-thyronine (T3) were carried out in rats. Total body homogenates were prepared and extracted with ethanol 48, 72, and 96 hr after the intravenous injection of ¹²⁵I-T4. ¹³¹I-T3 was added, and the paper chromatographic purification of T3 was effected by serial elution and rechromatography in three paper and one thin-layer cycles. The ratio of ¹³¹I-T3 and ¹²⁵I-T3 counting rates in the final chromatograms, which was identical in three different paper chromatography systems, was used to calculate the proportion of ¹²⁵I-T3 to ¹²⁵I-T4 in the original homogenates. In order to discount the effects of in vitro monodeiodination of T4 during extraction and chromatography, we killed control animals immediately after injection of ¹²⁵I-T4 and processed them in a similar fashion to the experimental groups. The average ratio of ¹²⁵I-T3 to ¹²⁵I-T4 in carcass extracts of animals killed between 48 and 96 hr after isotopic injection was 0.08 whereas the average ratio of ¹²⁵I-T3 to ¹²⁵I-T4 in chromatograms of control animals was 0.01. On the basis of the proposed model, calculations indicated that about 17% of the secreted T4 was converted to T3. Assuming values cited in the literature for the concentration of nonradioactive T3 in rat plasma, these findings would suggest that about 20% of total body T3 is derived by conversion from T4. Moreover, since previous estimates have suggested that in the rat, T3 has about 3 to 5 times greater biologic activity than T4, these results also raise the possibility that the hormonal activity of T4 may be dependent in large part on its conversion to T3.A necessary assumption in calculating T4 to T3 conversion in this and other studies is that the 3′ and 5′ positions are randomly labeled with radioiodine in phenolic-ring iodine-labeled T4. Evidence supporting this assumption was obtained in the rat by comparing the amount of labeled T3 produced after injection of phenolic and nonphenolic-ring iodine-labeled T4.