Masters Theses

Date of Award

6-1972

Degree Type

Thesis

Degree Name

Master of Science

Major

Animal Husbandry

Major Professor

M. J. Montgomery

Committee Members

J. T. Miles, D. O. Richardson

Abstract

Thirty-seven forages, used in previous performance trials, were analyzed for in vitro dry matter disappearance (IDMD), nylon bag dry matter disappearance (NDMD), and chemical constituents. The forages were grouped into the following classifications for comparisons: 1) grain silages, 2) legumes, 3) orchardgrass hay, 4) summer and annual hays, and 5) wheat silage. In some analyses legume silages and hays were separated. The IDMD and NDMD techniques were compared as methods of predicting nutritive value„ Correlation and regression analyses were performed both across species and within classifications to evaluate chemical components as predictors of IDMD and NDMD. A correlation coefficient of 0.85 (P < 0.01) was obtained when IDMD and NDMD were correlated across species. The NDMD method tended to yield higher values than the IDMD procedure. The mean dry matter disappearance values for the two methods were 74.8 and 64.9. The higher values obtained using the NDMD method along with the inaccuracies involved in handling the nylon bags, would suggest the in vitro proce-dure to be the method of choice. Correlation coefficients were utilized to express the relation-ship between chemical components and digestibility estimates. Although the correlations of IDMD and NDMD with crude protein, CWC, hemicellulose, estimated cell wall digestibility and calculated digestible dry matter were highly significant, the magnitudes of these correlations were low and of little predictive value. Lignin gave the larger correlations (r = -0.57 to -0.85) when compared to IDMD within classifications. Several combinations of the predictors (lignin, ADF, CWC, and crude protein) were evaluated for their accuracy in predicting IDMD and NDMD. Multiple correlation coefficients were calculated, and the combination of CWC, ADF, and lignin appeared to be the most accurate in predicting IDMD or NDMD. The multiple regression equation developed from all 37 samples using a stepwise regression analysis was as follows: IDMD = 52.38 + 0.70 crude protein - 0.71 lignin + 0.15 cell contents. The coefficient of multiple determination for this equation was 0.58.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS