The nylon bag method and in vitro gas production techniques were used to describe the kinetics of fermentation based on the exponential model P = a + b (1-e –ct). Alfalfa hay (ALF), phragmites australis (PA) or wheat straw (WS) were offered twice a day (8:00 and 16:00) at about maintenance level to 3 ruminaly fistulated Sistani (Bos indicus) and 3 Holstein (Bos taurus) cattle in a completely randomized design of factorial arrangement. Dry matter (DM) and crude protein (CP) degradability and gas volume (ml / 200 mg DM) were measured after 0, 4, 8, 12, 24, 48, 72 and 96 hours of incubation. Taking into consideration the in vitro gas production as related to nylon bag results, there were no significant differences observed in DM and CP degradability of ALF, PA or WS incubated in the rumen of the two cattle breeds, the same pattern being observed for total cummulative in vitro gas production. When feed samples were incubated for 24 hours, gas volume (ml / 200 mg DM) of ALF (41.63) was higher than in PA (29.17) and in WS (26.02) samples (p<0.05). The same pattern was observed for DM and CP degradability of ALF (46.23%, 79.63%) , PA (37.67%, 58.42%) and WS (36.24%, 41.95%) respectively (p<0.05). The high correlation coefficient (r) of in vitro gas production with degradability coefficient for ALF, PA and WS provide validation as
to the use of gas production test as an appropriate in vitro technique for evaluating these roughage feeds.
)