1. The cell-wall material of white lupin (Lupinus albus L.) cotyledon is characterized by low contents of cellulose (47 g/kg) and lignin (17 g/kg) and a high content of pectic substances (710 g/kg). The digestion of lupin cell-wall material by adult cockerels was estimated using gas-liquid chromatographic analyses of alditol acetates derived from polysaccharide sugars. The analyses were performed in the destarched water-insoluble fractions of feed and excreta. Digestibility measurements were carried out using a 3 d balance period including a 2 d feeding period and a 24 h final starvation period.
2. In the first experiment, six animals were given a diet containing 510 g white lupin cotyledon flour/kg which was the only source of protein and cell walls in the diet. The apparent digestibility of cell-wall components was near zero.
3. In the second experiment, three diets were prepared by diluting a fibre-free basal diet (diet A) by a semi-purified cell-wall preparation introduced at two different levels: 100 g/kg (diet B) and 200 g/kg (diet C). The semi-purified cell walls were prepared from the white lupin cotyledon flour used in the first experiment. The true digestibilities of polysaccharides measured in birds given diets B and C were near zero. It is suggested that the measurement of the neutral-detergent fibre (NDF) content according to Van Soest & Wine (1967) is not a suitable procedure for estimating the undigestible fibre content in poultry nutrition as the cell-wall pectic substances are not included in the NDF measurement.
4. Addition of the semi-purified cell-wall preparation (Expt 2) resulted in a slight decrease in the apparent protein digestibility. This decrease might be explained by the addition of undigestible cell-wall protein.
5. Addition of the semi-purified cell-wall preparation (Expt 2) had no effect on the apparent lipid digestibility.
6. The metabolizable energy values of the basal diet fraction of diets B and C were calculated assuming that the added plant cell-wall fraction was of no energy value. These calculated values were similar to the measured metabolizable energy value of diet A (basal). Thus, the tested pectic plant cell walls seemed to act as a diluent. It is suggested that, on the point of the digestion yields, all types of plant cell walls would act as diluents, in poultry.