1. A new in vitro method using the intestinal fluid of the pig for predicting the digestibility of diets was proposed.

2. A 0.5 g sample of pig diet was placed in a 100 ml Erlenmeyer flask, 20 mg pepsin (EC 3.4.4.1) in 10 ml 0.75 M-hydrochloric acid was added, and the mixture was incubated for 4 h at 37°. Then, after neutralization with 0.2 M-sodium hydroxide, 10 ml of intestinal fluid was added and incubated for an additional 4 h at 37°. This fluid was obtained from a pig fitted with a simple cannula in the upper jejunum. After the two-stage incubations, the contents of the flask was centrifuged for 10 min at 1250 g and the residue transferred to preweighed filter paper for dry matter (DM) and crude protein (nitrogen × 6.25; CP) determinations. The in vitro DM and CP digestibility was calculated on the basis of the original DM and CP content of the diet respectively.

3. The intestinal fluid could be stored at ∓ 20° for 60 d without losing its activity on DM and CP digestion.

4. A high correlation was found between this in vitro method and the standard in vivo procedures, for seven diets commonly used for growing or adult pigs, when diets were ground to pass through a 1 mm screen for the in vitro experiment, but there was an obvious dissimilarity between the two methods, especially in the case of CP: DM, Y = 1.04X +0.0806 (r 0.98, P < 0.001); CP, Y = 1.70X − 0.6092 (r 0.98, P < 0.001), where Y and X are in vivo and in vitro digestibility respectively.

5. This method is rapid and reproducible and particularly suited for the evaluation of a large number of samples. Since this method assesses only stomach and small intestinal digestion, it therefore may require further modification when used for diets which include large amounts of crude fibre. Also, depending on the needs of the investigator, the addition of a correction factor for endogenous faecal nitrogen loss may be required.