The intestinal permeability to markers was assessed in preruminant calves fed different milk substitutes containing skimmed milk powder or whey and soyabean products of differing antigenic activity as the protein sources. In Experiment 1, the 6 h urinary excretion of lactulose transiently decreased with the antigenic soyabean product but not that of sucrose of D-mannitol. In Experiment 2, the 6 h urinary excretion of sucrose and D-mannitol averaged 1-3%, regardless of age and dietary treatment. Cr-EDTA was recovered at rates of 2 and 4% after 6 and 24 h of urinary collection, respectively. The 24 h excretion of Cr-EDTA was lower in the calves fed the antigenic flour than in the controls after 2 weeks of experimental feeding (2.9 vs 6.0%, P < 0.05) but not thereafter. This transient decrease was also observed using D-xylose with the antigenic flour and with the non-antigenic concentrate (17 and 22% respectively vs 37%, P < 0.05). The present markers, including sucrose, may be suitable for assessing intestinal permeability in the calf even though excretion rates differed from one marker to another. Changes in the intestinal permeability to antigenic soya were of low magnitude and were only transiently detected when measured using lactulose, Cr-EDTA or D-xylose, probes which were therefore more sensitive than sucrose or D-mannitol.