Start Date: Feb 26, 2010
End Date: Aug 31, 2011
This will be a 3x3 Latin square feeding trial using 24 high-producing lactating dairy cows, including 6 with ruminal cannulas, from the U.S. Dairy Forage Research Center herd. Three diets will be fed: 1) all supplemental protein fed as solvent-extracted soybean meal, 2) all supplemental protein fed as conventional cottonseed meal originating from up-land cottonseed, and 3) all supplemental protein fed as Pima cottonseed meal prepared using the experimental expeller process. All diets will be based on alfalfa and corn silages plus high-moisture corn, formulated to contain 16.5% crude protein and will be fed as total mixed rations. Cows will be blocked into groups of 3 by days-in-milk and, within blocks, randomly assigned to treatment sequences; 2 blocks will be of ruminally cannulated cows. Periods will be 4 weeks long (total 12 weeks); the first 2 weeks will be for diet adaptation; data on milk yield, milk composition, milk component yield, milk urea concentration, body weight change, and feed intake will be collected over the last 2 weeks of each period. Feeding rates will be adjusted daily for ad libitum intake. Blood will be collected on day 28 of each period and analyzed for urea and glucose concentration. Spot urine and fecal samples will be taken on days 27-28. Urine will be analyzed for creatinine, urea, and total N. Data will be used to predict excretion of urea and total N using creatinine as marker. Feces will be analyzed for total N, dry matter, and indigestible acid detergent fiber. Apparent digestibility and fecal output of dry matter and N will be computed using indigestible acid detergent fiber as marker. On days 27-28 of each period, cannulated cows will be used for sampling to determine ruminal pH and concentrations of ammonia, total amino acids, and volatile fatty acids. Data will be analyzed as a replicated 3x3 Latin square using Proc mixed in SAS.