I intake and I deficiency were investigated in thirty vegans (eleven males and nineteen females) consuming their habitual diet. I intake was estimated using the chemical analysis of 4 d weighed duplicate diet collections. The probability of I-deficiency disorders (IDD) was judged from the measurement of urinary I excretion in 24 h urine specimens during the 4 d. There was wide variation in I intake. Mean I intake in males was lower than the reference nutrient intake (RNI; Department of Health, 1991) and mean intake in females was above the RNI, although 36% males and 63% females had I intakes below the lower RNI. Mean I intake in subjects who consumed seaweed (n 3) was in excess of the RNI, and approached the provisional maximum tolerable daily intake (World Health Organization/Food and Agriculture Organization Joint Expert Committee on Food Additives, 1989). The probability of IDD in the group investigated was moderate to severe: three of five subgroups were classified as moderate and two subgroups were classified as severe IDD possibility. The findings highlight that vegans are an 'at risk' group for I deficiency. The I status of vegans and the subclinical effects of low I intakes and infrequent high I intakes on thyroid function in this group should be further studied. Our work has also raised the question of adequate I intakes in groups where cow's milk is not consumed, and has exposed a need for more research in this area.