Lithium has been reported to be goitrogenic when used for the treatment of manic-depressive psychosis. To investigate the effects of lithium on iodine metabolism, male Sprague-Dawley rats were placed on a low iodine (LID) or normal iodine diet (NID) containing enough Li(2)CO(3) to give serum lithium levels of 0.23-0.86 mEq/liter (human therapeutic range is 0.6-1.6 mEq/liter). The following effects were noted with lithium treatment: (a) thyroid weight increased concomitant with a slowing of thyroidal iodine release; (b) the ability to concentrate iodide was increased only after goiters were established; (c) on the LID, (131)I uptake was elevated throughout all phases of treatment, even when the release rate was normal; (d) iodine organification was unaffected but the proportion of (131)I present as iodothyronines was decreased; (e) the thyroidal (127)I content was increased; (f) despite these changes, the serum PBI remained normal as did the thyroxine turnover rate; and (g) thyrotropin (TSH) levels in serum were the same as controls except for a slight elevation early in the course of treatment; TSH levels did not correlate with goitrogenesis. When LiCl was injected in large doses into intact rats (giving serum lithium levels of 3.08-3.89 mEq/liter), the iodide concentrating mechanism, (131)I uptake, and (131)I release rates were depressed. Similar experiments in hypophysectomized rats receiving TSH demonstrated these to be local antithyroid effects not mediated through the pituitary. The discrepancy between acute and chronic responses to lithium, and the dissociation between the inhibition of iodine release and stimulatory effects is discussed.