Potassium iodide (KI) was fed to male and female rats before and during breeding, to females only during gestation and lactation, and to their offspring after weaning (day 21 after birth) through to day 90, at levels of 0, 0.025, 0.05 or 0.1% (w/w) of the diet. Dams in a fifth group (positive controls) were given 4 mg/kg ip of the anti-mitotic/cytotoxic drug 5-azacytidine on day 17 of gestation. All offspring were reared by their natural dams and were evaluated blind with respect to treatment in a battery of standardized behavioural tests between 3 and 90 days of age. KI produced no significant reductions in parental body weight or food consumption, though it significantly reduced litter size and increased offspring mortality at the highest dose, and decreased weight gain at the two highest doses throughout the first 90 days after birth. Functionally, KI delayed auditory startle at the two highest doses, delayed olfactory orientation to the home-cage scent at the middle dose and decreased female running-wheel activity at all dose levels. In rats killed on day 90 after birth KI reduced brain and body weight at a dose of 0.1% of the diet, and reduced body but not brain weight at a dose of 0.05% of the diet. No significant effect was found on absolute or relative thyroid weight at 90 days of age. Several additional behavioural effects were observed in the low-dose KI group, but because these effects were not dose-dependent, they were not regarded as reliable. 5-Azacytidine produced evidence of substantially greater developmental toxicity than KI. It was concluded that KI produced evidence of developmental toxicity consistent with a picture of impaired thyroid function. The inclusion of tests of functional development added useful evidence to the overall picture of KI developmental toxicity.