The objectives of this research were to: 1) evaluate the accuracy of the Molly cow model predictions of ruminal metabolism and nutrient digestion when simulating dairy and beef cattle diets, 2) advance representations of N recycling between blood and the gut and urinary N excretion in the model, 3) improve the representation of pH and to refit parameters related to ruminal metabolism and nutrient digestion in the model, 4) investigate how ruminal pH affects the microbial community, expression of carbohydrate-active enzyme transcripts (CAZymes), fiber degradation, and short chain fatty acid (SCFA) concentrations. To achieve the first objective, a total of 229 studies (n = 938 treatments) including dairy and beef cattle data, published from 1972 through 2016, were collected from the literature and used to assess the model accuracy and precision based on root mean squared errors (RMSE) and concordance correlation coefficients (CCC). Only slight mean and slope bias were exhibited for ruminal outflow of NDF, starch, lipid, total N, and non-ammonia N, and for fecal output of protein, NDF, lipid, and starch. However, ruminal pH was poorly simulated and contributed to problems in ruminal nutrient degradation and VFA production predictions. To achieve the second objective, representations including ruminal ammonia outflow, intestinal urea entry, microbial protein synthesis in the hindgut, and fecal urea N excretion, were added in the model. Total urea entry, gut urea entry, and urinary urea elimination rates collected from 15 published urea kinetics studies were used to derive related parameters. Significant improvements in predictions of variables describing ruminal N metabolism, blood urea metabolism and urinary N secretion were exhibited after the modifications. To achieve the third objective, a dataset assembled from the literature containing 284 peer reviewed studies with 1223 treatment means was used to derive parameter estimates for ruminal metabolism and nutrient digestions. After refitting the parameters, the model is even more robust in representing ruminal nutrient degradation compared to the initial model. Adding ammonia concentration as a driver to the pH equation increased the precision of predicted ruminal pH, and thereby, the precision of predicted VFA concentrations due to an improved representation of pH regulation of VFA production rates. To achieve the fourth objective, six cannulated Holstein heifers with an initial BW of 362 ± 22 kg (mean ± SD) were subjected to 2 treatments in a cross-over design. The treatments were 10 days of intraruminal infusions of both 1) distilled water (Control), and 2) a dilute blend of hydrochloric and phosphoric acids to achieve a pH reduction of 0.5 units (LpH). Statistical analyses indicated 19 bacterial genera and 4 protozoal genera were affected by low ruminal pH. We observed significant correlations between 54 microbes (43 bacterial and 11 protozoal genera) and 25 enzymes, of which 8 key enzymes participated in reactions leading to SCFA production, suggesting that the ruminal microbial community alters fiber catalysis and fermentation in response to altered pH through a shift in carbohydrate-active enzyme transcripts (CAZymes) expression. Overall, after the modifications and reparameterizations, 19.7 to 37.5% of RMSE with essentially no slope bias and minor mean bias were exhibited for of ruminal and fecal outflow of ADF, NDF, fat, and protein, suggesting the model is properly to represent nutrient degradation and digestion in the bovine. Considering ruminal microbes and CAZymes in predicting ruminal volatile fatty acid concentrations could explain more variance of observations. / Ph. D. / The purpose of this research was to improve ruminal nutrient metabolism and nutrient digestion representations in the Molly cow model. First, the model accuracy and precision were assessed using a dataset including 229 studies (n = 938 treatments) conducted with dairy and beef cattle. The model evaluation results indicated the mechanisms encoded in the model relative to ruminal and total tract nutrient digestion are properly represented. However, ruminal pH was very poorly represented in the model with a RMSE of 4.6% and a concordance correlation coefficient (CCC) of 0.0. Although VFA concentrations had negligible mean (2.5% of MSE) and slope (6.8% of MSE) bias, the CCC was 0.28 implying that further modifications with respect to VFA production and absorption are required to improve model precision. As identified by the residual analyses, the representations of N recycling between blood and the gut were improved by considering ruminal ammonia outflow, intestinal urea entry, microbial protein synthesis in the hindgut, and fecal urea N excretion in the model. Observations of total urea entry, gut urea entry, and urinary urea elimination rates were collected from 15 published urea kinetics studies were used to derive related parameters. After the modifications, prediction errors for ruminal outflows of total N, microbial N, and non-ammonia non-microbial N were 39.5, 27.8 and 35.9% of the respective observed mean values. Prediction errors of each were approximately 10% units less than the corresponding values before model modifications and fitting due primarily to decreased slope bias. The revised model predicted ruminal ammonia and blood urea concentrations with substantially decreased overall error and reductions in slope and mean bias. After that, ammonia concentration as a driver was added to the pH equation, and a dataset assembled from the literature containing 284 peer reviewed studies with 1223 treatment means was used to derive parameter estimates for ruminal metabolism and nutrient digestions. Refitting the parameters significantly improved the accuracy and precision of the model predictions for ruminal nutrient outflow (ADF, NDF, total N, microbial N, non-ammonia N, and non-ammonia, non-microbial N), ammonia concentrations, and fecal nutrient outflow (protein, ADF, and NDF). Therefore, the improved model can be used to simulate nutrient degradation and digestion in the bovine. Although minor mean and slope bias were observed for ruminal pH and VFA concentrations, the small values for concordance correlations indicated much of the observed variation in these variables remains unexplained. To further explain variance in ruminal metabolism and understand how ruminal pH affects the microbial community, expression of carbohydrate-active enzyme transcripts (CAZymes), fiber degradation, and short chain fatty acid (SCFA) concentrations, six cannulated Holstein heifers with an initial BW of 362 ± 22 kg (mean ± SD) were subjected to 2 treatments in a cross-over design. We observed 19 bacterial genera and 4 protozoal genera were affected by low ruminal pH, and significant correlations between 54 microbes (43 bacterial and 11 protozoal genera) and 25 enzymes, of which 8 key enzymes participated in reactions leading to SCFA production. In summary, after the modifications and reparameterizations, the model is even more robust to represent nutrient degradation and digestion in bovine compared to the initial model. More variance of observations of ruminal volatile fatty acid concentrations could be explained by considering ruminal microbes and CAZymes expressions in further study.