An oral trimethylamine challenge test has been used to confirm the heterozygous status of patients with 'fish-odour syndrome'. By measuring the percentage of total urinary trimethylamine-related material excreted as the N-oxide, no discrimination could be made between obligate heterozygotes (parents of 'fish-odour syndrome' patients) (n = 15; 96 +/- 2%, range 92-98%) and control individuals (parents of unaffected children) (n = 16; 96 +/- 2%, range 93-99%) on a normal diet. However, after ingesting a trimethylamine load (600 mg base) the obligate heterozygotes were clearly distinguishable (76 +/- 3%, range 71-79%) from controls (95 +/- 2%, range 91-99%) (t-test; p <0.001). One of a hundred apparently normal volunteers who were subsequently challenged with trimethylamine had a N-oxidation capacity which fell within the range found among the obligate heterozygotes.