The study was carried out by means of the in vitro gas production technique described byTheodorou et al. (1994). For that the forages Pennisetum purpureum cv. CT-115, PP; Panicummaximum cv. Likoni, PM; Leucaena leucocephala, LL; Gliricidia sepium, GS; Morus alba cv.Tigreada, MA and Trichantera gigantea, TG were used. The mixtures of these were inproportions 60:40 (grass: supplement). Eoch combination = experimental treatment (8): 1)PM-LL; 2) PM-GS; 3) PM-TG; 4) PM-MA; 5) PP-LL; 6) PP-GS; 7) PP-TG; 8) PP-MA. Thesuppliers of the inoculum were two Holstein Friesian cows, with ruminal cannulae, adapted toa 50:50 ration. The in vitro gas production was determined 3, 6, 9, 12, 15, 24, 36, 48, 60, 72and 96 h post-inoculation. The grasses showed a higher gas production potential than the treesor shrubs, although with exception of LL the first stage of fermentation (until 12 hours)showed similar trends. Fermentative potential: PP>PM and non leguminous species (MA andGS) > leguminous trees (LL and GS). When they were combined in a 60:40 relationship(grasses: forage trees), the differences observed during the incubations of the forages alonedecreased considerably, possibly due to complementary effects. In the rations with PM, thecombination with GS was higher, followed by MA, TG and LL. In the treatments with PP, thesupplementation with all the species showed gas production values similar to and lower thanthe grass alone.