Experiments were carried out in vitro and in normal human subjects to evaluate alternative food-grade viscous polysaccharides as agents for reducing postprandial hyperglycemia and to assess the relationship between the in vitro and in vivo performance of the polysaccharides. A 1:1 mixture of xanthan and locust bean gum (X/LBG) had the greatest viscosity at equivalent concentrations and shear rates and was more effective than guar gum, xanthan, or locust-bean gum at inhibiting glucose movement in vitro. It was not, however, more efficient in lowering postprandial blood glucose and plasma insulin in human subjects when incorporated in a drink containing 50 g glucose. When the different gums were acidified and reneutralized to mimic conditions in the gut, there was a better correlation between viscosity and blood glucose and plasma insulin levels. This effect may explain why X/LBG was no more effective than the other gums in reducing postprandial hyperglycemia in man.