One hundred and thirteen patients diagnosed as suffering from one of four common psychosomatic syndromes (psychogenic headache, cardiac neurosis, functional disturbance of the colon, or pruritus) were treated by two groups of general practitioners with flupenthixol or diazepam or sulpiride. Flupenthixol was compared with diazepam in 58 patients in one group of Belgian practices, and with sulpiride in 55 patients in another group of practices and assessed over a 4-week period for therapeutic response and adverse effects. Flupenthixol was given in a dosage of 0.5 to 2 mg a day, diazepam in a dosage of 2.5 to 10 mg a day and sulpiride 100 to 200 mg a day, in identical capsules. In the flupenthixol/diazepam comparison, there were 7 drop-outs (3 flupenthixol, 4 diazepam). Marked or moderate reduction in global assessment of illness occurred in both treatment groups, but there was no significant difference in the therapeutic effect of the two drugs. The incidence of side-effects was low and similar in the two groups. In the flupenthixol/sulpiride comparison, there were 10 drop-outs (7 flupenthixol, 3 sulpiride). A significant reduction in symptoms occurred in both treatment groups, but this was more rapid in the flupenthixol group. Side-effects were infrequent and mild in both groups.