Faced with large variations in iodine, T4 or T3 supply, the fetal brain is able to maintain T3 homeostasis to a greater degree than apparent from changes in plasma and other tissues, such as the liver. Changes in the activity of 5'D-II play an important role in this homeostasis, although this does not exclude other regulatory mechanism (s), such as changes in type III (5D) activity, in uptake of the iodothyronines by the brain, or in cerebral iodothyronine turnover rates. T3 generated locally from T4 is more important than plasma derived T3 as determinant of the total T3 available to the brain throughout the life cycle of the rat. It is especially important during the fetal and neonatal phases of brain development, when the brain depends almost exclusively on the supply of T4. Fetal brain T3 homeostasis is maintained despite large fluctuations in the supply of T4. An excess of T3 also affects brain T3 to a lesser degree than it does other tissues. If present results are relevant to man, they suggest that overtreatment of mother of a congenital hypothyroid fetus with T4 is not likely to be harmful for the fetal brain. Treatment with T3 should be avoided, as it deprives the fetal brain of the main regulatory mechanism involved in homeostasis of brain T3, namely generation of T3 from T4.