Many studies have shown high carbohydrate, high fibre diets to benefit diabetic control, the improvement being attributed mainly to an effect of fibre. This study investigated the possible beneficial effects of the digestible carbohydrate component. A diet rich in carbohydrate was compared with a traditional low carbohydrate diet in 10 Type 2 (non-insulin-dependent) diabetic patients, using a crossover design; both diets contained less than 20 g dietary fibre/day. During 24-h metabolic profiles carried out after 4 weeks on each diet, the mean basal plasma glucose (mean of 03.00, 05.00 and 07.00 h values) was 5.3 mmol/l on the high carbohydrate diet and 5.9 mmol/l on the low carbohydrate diet (p less than 0.05), despite the 2-h post-prandial glucose (mean of three main meals) being higher on the high carbohydrate diet than on the low carbohydrate diet (8.7 versus 7.3 mmol/l, p less than 0.01). Overall diabetic control was the same throughout the study, as judged by a mean 24-h plasma glucose of 6.7 mmol/l on the high carbohydrate and 6.6 mmol/l on the low carbohydrate diet, and haemoglobin A1c percentage of 8.3 on both diets. Mean cholesterol was 4.55 mmol/l on both diets and fasting plasma triglyceride was 2.83 mmol/l on the high carbohydrate and 2.55 mmol/l on the low carbohydrate diet (p = NS). These results indicate that a diet rich in carbohydrate, but restricted in fibre, does not cause overall deterioration of diabetic control or lipid metabolism in stable Type 2 diabetic patients, and suggest that digestible carbohydrate has an effect on basal blood glucose independent of fibre.