Four non‐lactating cows, each fitted with a rumen cannula, were given a diet of grass silage, molasses, barley, soya bean meal and a mineral mix (70:14:9:6:1 on a dry matter basis) in eight meals a day. Four experimental treatments, consisting of intraruminal infusions of sodium bicarbonate at 0, 300, 600 and 900 g d −1in a 4×4 Latin square design, were used to study the effects of added bicarbonate on the rumen environment and on the predicted rumen degradability of soya bean meal, rapeseed meal and fishmeal. Mean rumen pH values were 6.26, 6.53, 6.73 and 6.87 for the control and bicarbonate treatments, respectively; corresponding values for the fractional outflow of rumen liquid were 0.129, 0.142, 0.151 and 0.153 h −1 (SED 0.01; P>0.05) and for the fractional outflow rates of Cr‐mordanted protein particles 0.035, 0.040, 0.048 and 0.054 h −1 (SED 0.001; P<0.01). Sodium bicarbonate had no effect on the disappearance of soya bean meal or rapeseed meal nitrogen from Dacron bags incubated in the rumen, but at the 900 g d −1 level there was an increased nitrogen disappearance from fishmeal at all incubation times; disappearances after 24 h were 0.64 and 0.68 (SED 0.025; P<0.05). Using the values for the disappearance of nitrogen and the outflow rates of Cr‐mordanted protein the predicted rumen degradabilities of soya bean meal and rapeseed meal were reduced by about 8% by the 900 g d −1 infusion, but for fishmeal there was no change in degradability since the effects of an increased outflow rate were counter‐balanced by an increased rate of loss of nitrogen in the Dacron bag incubations.