Two studies of the new alpha-glucosidase inhibitor, miglitol, in patients with non-insulin-dependent diabetes mellitus (NIDDM) are reported. In the first, 13 patients, poorly controlled on sulphonylureas, received miglitol 50mg three times daily for 4 weeks. Post-prandial blood glucose was reduced after breakfast, lunch, and tea compared with placebo (p less than 0.05-0.01) but there was no improvement in fasting blood glucose, serum fructosamine or haemoglobin A1. In a dose-response study the effect of a single dose of miglitol (0,50,100,150 or 200mg) on post-prandial glycaemia after a test breakfast was assessed in 20 patients with mean +/- SEM fasting blood glucose 9.9 +/- 0.4 mmol/l. With 50mg miglitol, there was a significant reduction in blood glucose from 30 to 120 min post-prandially compared with placebo. Increasing doses of miglitol further depressed the post-prandial rise in blood glucose and with 200mg there was no significant change from fasting levels. Side-effects were limited to flatus and loose stools particularly with the higher doses but were not severe. Miglitol effectively reduces post-prandial blood glucose rise in NIDDM with as little as 50mg but there is considerable individual variation. Larger doses may be necessary in patients already poorly controlled on sulphonylureas.