The objective of this study was to investigate the effects of baker's yeast as a protein source instead of soybean meal in dairy cow diets. Four dairy cows were utilized in a 4 x 4 Latin Square experiment. Diets included 0.0% (Control Diet), 6.6% (Diet 1), 13.2% (Diet 2), and 19.8% (Diet 3) baker's yeast containing 3.56 x 10(8) CFU/g. Forty percent of the dry matter intake of dairy cows was supplied from concentrate, and forage was offered ad libitum. The dry matter intake, milk yield, and composition of milk were similar. While the rumen NH3-N level was significantly low (p < 0.05) 3 h post-feeding, acetic acid concentration was significantly high (p < 0.05) 12 h post-feeding in cows that consumed diets containing yeast. Propionic acid concentrations were lower (p < 0.01) 3 h post-feeding in cows that consumed Diets 2 and 3 compared with those that consumed the other diets. While serum triglyceride levels were significantly higher (p < 0.05) in cows that consumed diets containing baker's yeast than in those that consumed the control diet, serum protein, urea, glucose, calcium, phosphorus, gamma glutamyl transferase and aspartate aminotransferase were similar in cows that consumed different diets. It was concluded from this study that there were some positive effects, such as low rumen NH3-N level and high acetic acid concentration, in cows that consumed diets containing yeast, and thus, baker's yeast can be utilized in dairy cow diets as a protein source.