A two-stage in vitro procedure was used for assessing the activity of parotid saliva to enhance rumen digestion of tanniniferous browse foliage. The procedure consisted of pre-incubation in saliva for 4 h at 39 °C followed by incubation in diluted buffered rumen fluid. Using this procedure, a study was conducted to examine the effects of pre-incubation in sheep (SS), quebracho-supplemented sheep (qSS) and goat (GS) parotid saliva or in McDougall's artificial saliva (AS, used as control) on in vitro rumen fermentation kinetics (estimated using the gas production technique) of browse foliage from six shrub species (Cytisus scoparius, Genista florida, Rosa canina, Quercus pyrenaica, Cistus laurifolius and Erica australis) collected over two seasons (spring and autumn), thus varying the in vitro digestibility (from 0·597 to 0·903) and tannin contents (from 3 to 130 g tannic acid equivalent/kg dry matter (DM)). Saliva was collected from four sheep and four goats fed alfalfa hay, and from four sheep fed the same alfalfa hay but supplemented with quebracho (rich in condensed tannins) for 60 d, through a cannula inserted in the parotid duct, and rumen fluid was always from sheep fed alfalfa hay. The extent of degradation when browse foliage was pre-incubated in qSS was similar to that observed with control AS (0·449 v. 0·452, respectively), and 8% less than the value with pre-incubation in SS (0·490). In vitro fermentation kinetics (gas production parameters) of browse foliage were not significantly enhanced with pre-incubation in qSS compared with SS, whereas in vitro digestibility and extent of degradation in the rumen were significantly reduced with qSS compared with SS. After pre-incubation in sheep and goat saliva, the extent of browse foliage degradation was significantly increased by 4–8% compared with pre-incubation in the control AS. Fermentation efficiency of browse foliage was increased (P<0·05) with pre-incubation in GS compared with SS. Sheep or goat saliva may have some activity to affect in vitro rumen fermentation of the foliage samples incubated, enhancing extent of degradation of tannin-rich browse. However, a relationship between the magnitude of this effect and the tannin content of the browse foliage could not be established, suggesting that sheep and goat saliva may not be particularly important in neutralizing tannins.