Résumé : In Central Africa, all of northern Zaire is very severely deficient in iodine. A peculiar feature of this endemia is that iodine deficiency and the ensuing thyroid gland stimulation not only leads to goitre formation but also to progressive thyroid involution and to myxoedematous cretinism. An iodine supplementation trial based on oral administration of small doses of iodine was made in 81 schoolchildren. All of them received a small dose of iodine (0.1 ml containing 48 mg) per os and the thyroid status was followed during 4 months. Blood and urine samples were collected at the start of the study, then 2 weeks, 2 months and 4 months after iodine administration. Before iodine supplementation the mean urinary iodine level was 0.18 +/- 0.02 micromol/l, and 10% of the subjects had a urinary iodine level below 0.08 micromol/l. Fifty-two percent of the subjects had a serum thyrotropin (TSH) level above 10 mU/l. All the subjects responded to the administration of iodine. and all of them recovered a euthyroid status. Most of them were still euthyroid at the end of the study. However. within 4 or even 2 months, some subjects (15 % of the total) reverted to hypothyroidism. At the entry of the study these subjects were all hypothyroid and had elevated TSH and paradoxically low serum thyroglobulin (TG) values. In myxoedematous cretins living in the same area, even lower serum TG levels were found. Together with the absence of goitre, a paradoxically low serum TG Suggests a low thyroid reserve, and in the present case a reduced amount of functional thyroid tissue. We show that the serum TG/TSH ratio may be used as a predictive index of thyroid reserve and of positive response to iodine administration. These data further suggest that thyroid damage is not confined to myxoedematous cretins. but is widely distributed in the phenotypically normal population. Widely distributed thyroid damage may render iodine prophylaxis based on oral administration unpredictable.