We hypothesized that dietary hydrolyzable tannins would not act as digestibility reducing substances but would be excreted in the feces if the tannin were ingested by rats producing salivary proline-rich proteins (PRPs). To test that hypothesis we used two groups of Sprague-Dawley rats: tannin-naïve rats that were secreting basal levels of salivary PRPs and tannin-habituated rats that were secreting elevated levels of PRPs. The animals were fed for 10-18 d on diets containing 3% (w/w) purified hydrolyzable tannin [pentagalloyl glucose (PGG)] that was periodically spiked with chemically synthesized, radiolabeled 1,2,3,4,6-penta-O-galloyl-[U-14C]-D-glucopyranose (1 microCi per gram diet). The PGG-habituated rats excreted three times more of the consumed 14C in their feces than did the PGG-naive rats (11.4% for PGG-habituated rats vs. 3.5% for PGG-naïve rats, P < 0.05). The addition of 3% PGG to the diet of the PGG-naïve rats had no significant effect on apparent dry matter or nitrogen digestibility (P > 0.05). However, dry matter digestibility and nitrogen digestibility were significantly decreased by PGG in the diets of the PGG-habituated rats (7 and 25%, P < 0.001, respectively). Production of PRPs increased the amount of PGG excreted intact in the feces but at the cost of diminishing apparent dry matter and nitrogen digestibility.