Health and ruminal variables were intensively measured during adaptation to grain-based diets in 6 beef cattle with fistulated rumens. The cows had been maintained on prairie grass hay-supplemented diets, and were converted to a grain-based finishing ration by feeding each successive diet (diets 1-4, respectively) for a period of 7 days. Each cow was evaluated and samples were obtained 3 times each day for the first 5 days that each diet was fed. Health variables monitored were rectal temperature, pulse, respiratory and rumen motility rates, fecal consistency, demeanor, blood pH, and blood glucose and L(+) lactate concentrations. Ruminal variables monitored were pH and glucose, DL-lactate, and volatile fatty acid concentrations of rumen contents. Data were analyzed by use of a multivariate ANOVA. We determined that most of the health variables were within reference rang limits throughout the adaptation period; however, analysis of pulse and respiratory rates indicated that diets 2 and 4 were stressful. Although blood pH continually decreased during feeding of the 4 diets (7.38 to 7.30), blood L(+) lactate and glucose concentrations had large increases only within diet 4. The pH of ruminal contents decreased progressively from 6.8 to 5.3. Rumen glucose concentration was low (< 1 micromole/ml), except with diet 4 in which values were 8 times higher than for other diets. By the end of the study, the ruminal contents of all animals were acidic (pH < 5.5), and, on the basis of higher than background amounts of ruminal glucose and DL-lactate, it was determined that rumen microbial equilibrium had not yet been achieved. Analysis of results of this study suggested that ruminal imbalance could be evaluated by monitoring pulse and respiratory rates, blood pH, and blood glucose concentrations. Assessment of the rumen alone could be accomplished by monitoring the variables of rumen pH, rumen glucose, and DL-lactate concentrations. Respiratory rate, blood and rumen content pH, and blood L(+) lactate concentrations were significantly (P < 0.001) affected by time after feeding.