Nutritional feed values are currently based on aggregate criteria such as the ileal or faecal digestibility. Digestibility is the result of several processes including hydrolysis, absorption, secretion and passage. In order to develop mechanistic models of digestion to be used for feed evaluation, these processes have to be quantified. The aim of the current study was to determine the enzymatic hydrolysis (or: in vitro digestion) kinetics of main constituents (organic matter, nitrogen and starch) in wheat, barley, wheat bran and soybean meal, using a three-step, enzymatic in vitro method that mimics digestion in the stomach, small and large intestine of pigs. The in vitro results were compared with in vivo results. Hydrolysis kinetics (i.e., solubilisation of feed constituents) was modelled using an exponential segmented model estimating the extent and rate of digestion for each enzymatic digestion step. In vitro digestion of organic matter of soybean meal occurred mainly through the action of pepsin (0.67 versus <0.30 for other feeds, P<0.05), which was caused by the enzymatic digestion of protein at this site. Organic matter of cereals was mainly hydrolysed by enzymes mimicking digestion in the small intestine, and was mainly caused by starch digestion. Fractional in vitro digestion rates of organic matter were higher in the stomach than in the small intestine for cereals (0.20-0.34 min(-1) versus 0.02-0.15 min(-1), P<0.05). The potential in vitro digestibility of organic matter was nearly 0.88 for wheat and soybean meal, 0.79 for barley and 0.61 for wheat bran, which correspond to typical in vivo digestibility values. The in vitro digestibility corresponded reasonably well to in vivo results for enzyme systems mimicking ileal and total tract digestion. The results of this study indicate that it is possible to quantify dynamic aspects of digestion of feedstuffs fed to non-ruminant animals.