An experiment was conducted to evaluate condensed molasses solubles(CMS, a by-product from monosodium glutamate production) as a source of nitrogen for ruminant with particular reference to its effects on microbial protein synthesis. Four non-lactating dairy cows fitted with rumen cannulas were used in a 4 ${ imes}$ 4 Latin square with 14-day periods. The four treatments were (1) basal diet consisting of barley straw ad libitum and 3 kg/d of rolled barley, (2) basal diet plus 200 gld molasses and 300 g/d water, (3) basal diet plus 200 g/d molasses, 100 g/d CMS and 200 g/d water, (4) basal diet plus 200 g/d molasses, 200 g/d CMS and 100 g/d water. Ruminal pH remained at high levels and showed little variation during the day between treatments. The concentration of total and individual VFA in the rumen was similar between treatments. There was no difference in the concentration of ammonia in the rumen between treatments, although the intake of nitrogen in molassesl CMS mixture treatments was higher than that of control and molasses treatment. But there was a suggestion of an increased synthesis of microbial protein with the higher level of inclusion of CMS when the allantoin/creatinine ratio was used as an index of microbial protein production(P <0.10).