Aims: To characterize the changes in the relative population size (RPS) of select ruminal bacteria and rumen fermentation variables in beef heifers supplemented with a strain of Saccharomyces cerevisiae as viable active dried (ADY) or killed dried (KDY) yeast following an induced episode of ruminal acidosis. Methods and Results: Six ruminally cannulated beef heifers fed a diet consisting of 50% forage and 50% grain (dry matter basis) were used in a replicated 3 x 3 Latin square design with three 28-day periods. Treatments were: (i) control (CTRL; no yeast); (ii) ADY (4 g day(-1) providing 10 10 CFU per g; AB Vista, UK); and (iii) KDY (4 g day(-1) autoclaved ADY). The acidosis challenge was induced on day 22 and rumen samples were collected on day 15 (baseline; BASE), day 22 (challenge day; CHAL), and on day 29 (168th hour post acid challenge or recovery, REC) of each period. Over the study, duration of pH < 5.8 (indicative of subacute ruminal acidosis) was less for ADY and KDY than CTRL, with ADY less than KDY. No treatment effects were observed on relative abundance of ruminal bacteria, but the day effect was significant. The RPS of lactate producers and utilizers was greater while RPS of fibrolytic bacteria was lower during CHAL than BASE and REC. Yeast supplementation, irrespective of its viability, showed beneficial effects on ruminal pH variables in animals more susceptible to acidosis. Conclusions: Rumen microbial population was altered with the induction of severe acidosis. Most of the changes reverted back to baseline values during the recovery phase. Yeast supplementation reduced subacute rumen acidosis in the most susceptible cattle, but failed to attenuate severe acidosis induced by a grain challenge. Significance and Impact of the Study: The study provided valuable insight into the mechanism by which acidosis affects cattle performance. Individual animal variation in ruminal fermentation partly explained the variability in response to yeast supplementation in the study.