Prediction of nutritive values in grass silages: II. Degradability of nitrogen and dry matter using digestibility, chemical composition, and fermentation data

Abstract

One hundred thirty-six perennial rye-grass silages with wide variations in quality were evaluated for N and DM degradability in three beef steers offered grass silage of medium quality ad libitum. The silages were incubated in the rumen of each animal in triplicate for 0, 6, 12, 24, 48, and 72 h, respectively. The disappearance rates of N or DM were used to calculate the readily soluble fraction ("a" value), potentially degradable fraction ("b" value), and the fractional degradation rate of"b" ("c" value). The effective degradability (P) of N or DM was then estimated assuming a ruminal outflow rate of 0.02, 0.05, or 0.08/h (P0.02, P0.05, or P0.08). The objective was to use these data to develop prediction equations for N and DM degradability in grass silages. There were considerable variations in "a," "b," and "c" values and the P0.02, P0.05, or P0.08 of N and DM (e.g., the P0.02 of N ranged from 75.0 to 93.4% and the P0.02 of DM from 51.5 to 82.5%). The P0.02, P0.05, or P0.08 of N and DM were negatively related (P < 0.001) to ADF, NDF, and lignin concentrations but positively related (P < 0.001) to protein fractions (CP, soluble CP, and true protein concentrations) and digestibility of DM, OM, GE, CP, and NDF and digestible OM in the total DM (measured with sheep). The N and DM degradability data were also positively related to silage lactic acid concentration, but the relationships between DM degradability data and pH, ammonia N/total N, and VFA concentration in silages were negative (P < 0.05). Several sets of prediction equations (linear and multiple) were thus developed for N and DM degradability using CP or NDF concentration, "a" value or digestibility data as primary predictors, together with or without other nutrient concentration and silage fermentation variables. All these relationships were highly significant (P < 0.001), and each predictor had a significant effect on the relationship (P < 0.05). The R2 values in multiple regression for N and DM degradability were generally over 0.70 and higher than in linear regression equations. Four equations were also developed to convert N and DM degradability at a given ruminal outflow rate, predicted using the above-mentioned equations, to their counterparts at any ruminal outflow rate (0.02 to 0.10/h), respectively.