The objective of this study was to evaluate the effect of a fermentation by-product on rumen function, microbial yield, and composition and flows of nutrients from the rumen in high-producing lactating dairy cattle. Eight ruminally cannulated multiparous Holstein cows averaging (mean ± standard deviation) 60 ± 10 d in milk and 637 ± 38 kg of body weight were randomly assigned to 1 of 2 treatment sequences in a switchback design. Treatment diets contained (dry matter basis) 44% corn silage, 13% alfalfa silage, 12% ground corn, and 31% protein premix, containing either a control mix of urea and wheat middlings (CON) or a commercial fermentation by-product meal (Fermenten, Arm and Hammer Animal Nutrition, Princeton, NJ) at 3% diet inclusion rate (EXP). The trial consisted of three 28-d experimental periods, where each period consisted of 21 d of diet adaptation and 7 d of data and sample collection. A triple-marker technique and double-labeled 15N15N-urea were used to were used to measure protozoal, bacterial, and nonmicrobial omasal flow of AA. Rumen pool sizes and omasal flows were used to determine digestion parameters, including fractional rates of carbohydrate digestion, microbial growth, and yield of microbial biomass per gram of degraded substrate. Fermentation by-product inclusion in EXP diets increased microbial N and amino acid N content in microbes relative to microbes from CON cows fed the urea control. Microbial AA profile did not differ between diets. Daily omasal flows of AA were increased in EXP cows as a result of decreased degradation of feed protein. The inclusion of the fermentation by-product increased nonmicrobial AA flow in cows fed EXP versus CON. Average protozoal contribution to microbial N flow was 16.8%, yet protozoa accounted for 21% of the microbial AA flow, with a range of 8 to 46% for individual AA. Cows in this study maintained an average rumen pool size of 320 g of microbial N, and bacterial and protozoal pools were estimated at 4 different theoretical levels of selective protozoa retention. Fractional growth rate of all microbes was estimated to be 0.069 h-1, with a yield of 0.44 g of microbial biomass per gram of carbohydrate degraded. Results indicated that fermentation by-product can increase omasal flow of AA while maintaining adequate rumen N available for microbial growth and protein synthesis. Simulations from a developmental version of the Cornell Net Carbohydrate and Protein System indicated strong agreement between predicted and observed values, with some areas key for improvement in AA flow and bacterial versus protozoal N partitioning.