The objective of the present study was to investigate the effects of starches with differing rates of hydrolysis on exposure to pancreatin in vitro on postprandial carbohydrate metabolism in healthy subjects and in subjects with type 2 diabetes. Two test starches, prepared from uncooked native granular starch products, and naturally enriched with 13C, were consumed in a randomized crossover design by eight healthy and thirteen type 2 diabetic subjects. One starch was characterized in vitro as being rapidly hydrolysed (R, 94% after 180min), and the other was more slowly hydrolysed (S, 51% after 180min). Each subject consumed 50g of each test starch. In addition, the type 2 diabetic subjects consumed 89·7g of the S starch on a separate occasion. Blood samples were taken at 10min intervals for 3h, and at 20min intervals for a further 3h during a 6h postprandial period. Breath 13CO2 enrichment was measured at the same time points, and indirect calorimetry was performed for seven 20min sessions immediately before and during the 6h postprandial period. With the R starch, plasma glucose concentrations and serum insulin concentrations rose faster and the maximum glucose change was approximately 1·8 times that for the S starch, averaged across both subject groups. The areas under the curves for glucose and insulin were, respectively, 1·7 and 1·8 times higher for the R starch compared with the S starch, averaged across both subject groups. The rate of 13CO2 output and the proportion of 13C recovered in breath after consumption of the R starch was similar for both subject groups. The results provide evidence that starches which have different rates of hydrolysis in vitro result in different patterns of glycaemia and insulinaemia in both healthy adults and in diet-controlled type 2 diabetic subjects. Data from the hydrolysis of novel starch products in vitro, therefore, are useful in predicting glycaemic responses in vivo.