The addition of vegetable fibres to the diabetic diet has been reported to ameliorate glycaemic and plasma lipid profiles, and Guar flour seems to obtain the best results. At its usual dose, Guar produces several gastro-intestinal side effects. A lower dose (4 + 4 g/day) was therefore employed in 10 non-insulin dependent diabetics (NIDD). The following parameters were measured at the end of treatment and after a control period: HbA1 levels, hepatic glucose production (3H-Glucose infusion), peripheral sensitivity to insulin and insulin secretion (hyperglycaemic clamp), and specific insulin binding to isolated monocytes. The ultracentrifugal plasma lipid pattern was also measured. No significant body weight change was recorded during the study. A significant glycaemic and insulinaemic decrease in the fasting state was observed after Guar, together with a significant decrease of HbA1 levels (from 8.5 +/- 0.4 to 7.9 +/- 0.4%, p less than 0.05) and amelioration of peripheral sensitivity to insulin (M/I = 14.3 +/- 6.6 versus 24.3 +/- 8.8, p less than 0.025; 50% increase of insulin binding to circulating monocytes) without significant variation of the fasting hepatic glucose production. Decreased B-cell stimulation by flattening post-prandial glycaemic peaks may be an explanation of the reduction of insulin resistance via down-regulation mechanism. As far as the lipid profile is concerned, a significant reduction in total and LDL cholesterol (p less than 0.05 and p less than 0.01) and an increase in HDL-phospholipids (p less than 0.05) were recorded after Guar. These results suggest that Guar in low doses is well accepted and can contribute to a better glycaemic and lipaemic control in NIDDM.