The effect of variations in kind and amount of dietary carbohydrate on plasma glucose and insulin responses was studied in normal subjects and in patients with chemical diabetes. Plasma glucose and insulin responses fell when the proportion of total calories given as carbohydrate were decreased by 15% (coincidental with a comparable increase in dietary fat). Associated with the lower plasma glucose and insulin concentrations was a fall in fasting and postprandial triglyceride levels. Plasma glucose and insulin responses were also lower when equivalent carbohydrate challenges were given as part of a mixed meal, as compared to a drink. Furthermore, carbohydrate given as starch also led to an attenuated glucose and insulin response when compared to an equivalent amount of glucose administered as either dextrose or sucrose. Finally, significant differences were also seen after the ingestion of different kinds of starch, with potato being the most like dextrose, and rice the least. These results indicate that differences in amount and kind of ingested carbohydrate can modify ensuing plasma glucose and insulin responses, and raise the possibility that such dietary manipulation may have some therapeutic utility in patients with abnormal carbohydrate and lipid metabolism.