Forty-eight Holstein cows at Utah State Dairy Farm were blocked statistically according to date of calving, previous milk production, and numbers of lactation at parturition. The cattle were assigned randomly to one of four treatments within blocks. The four treatments included a base ration (control, treatment #1), base ration plus .8% of sodium bicarbonate (treatment #2), base ration plus .4% of magnesium oxide (treatment #3), and base ration plus both .8% of sodium bicarbonate and .4% of magnesium oxide (treatment #4). The research was conducted from February 1983 to November 1984. A formagraph was used to measure milk coagulation parameters and pH was determined. There was no significant difference in milk coagulation parameters or pH between the control and the buffer treatments. Milk parameters were significantly different in individual cow, week, and milk pH. Milk parameters did not appear to be dependent upon season. Curd firmness was significant in interaction of season and treatment. Significant variations in milk pH were observed in relation to week, season, and individual cow. Overall treatments, the clotting time, K20, and pH value increased each week, and A30 decreased each week. The milk parameters and pH in each treatment were significant between weeks except K20 and A30 in treatment #3, and A30 in treatment #2 (p>0.05). The clotting time and K20 were negatively correlated with firmness, and there was positive correlation between Ct and K20 as expected. Somatic cell count was positively correlated with clotting time, K20, and pH and negatively correlated with A30. Milk pH was the most significant and had positive correlation coefficient with clotting time and firming rate and negative correlation coefficient with curd firmness.