Erythromycins often cause gastrointestinal side-effects due to an increase in motility or to change in the intestinal bacterial flora. In order to evaluate the effect of erythromycin on gastrointestinal motility. 11 healthy volunteers were given placebo, erythromycin stearate (ES) 1000 mg or a therapeutically equivalent single dose of erythromycin acistrate (EA.2'-acetyl erythromycin stearate) 800 mg in a double-blind trial. The oro-caecal transit time was measured using the hydrogen breath test with lactulose as the substrate. The transit time was estimated from the H2-peak (ppm) in end-expiratory breath by two methods, t1 representing the "front" and t2 the "bulk" of lactulose reaching the colon. t1 was 51 min in the placebo group, 38 min in the EA and 31 min in the ES group (p less than 0.05, ES vs placebo). t2 was 74 min, 64 min, and 46 min, respectively (p less than 0.05, ES vs placebo). The difference between EA and ES was also significant. Six subjects in the ES group but none in the EA group recorded adverse gastrointestinal effects attributable to medication. It was concluded that erythromycin shortens the oro-caecal transit time in man and that EA effects the transit time slightly less than ES.