Moe, Aaron J.2017-11-092017-11-091982http://hdl.handle.net/10919/80227Corn silages were selected from samples analyzed in the Virginia Tech Forage Testing Laboratory. A sample selection scheme employing observation of silage quality indicators (color, mold, aroma, fermentation type) was utilized to obtain a sample set encompassing a wide range in deterioration. In vitro digestible organic matter (IVDOM) was the criteria used to estimate feeding value. The full data set contained 142 corn silages which averaged 59.8 ± 6.6% and ranged from 38.5 to 74.7% IVDOM. Simple and multiple regression of IVDOM on chemical constituents were performed for the full data set and within fermentation groups. Coefficients of determination (R² ) revealed that acid detergent fiber (ADF) and dry matter (DM) explained the largest proportion of variation in IVDOM. The simple regression on the full data set resulted in R² of .352 and .208 with standard errors of the prediction S_y.x of 5.29 and 5.86 for ADF and DM, respectively. Multiple regression analyses resulted in minor improvements in S_y.x. These data suggest the inclusion of other variables besides ADF and DM in the prediction equation is unwarranted. Corn silage IVDOM was predicted employing Near-infrared reflectance techniques. A calibration resulted in R² of .77, S_y.x of 5.38. The calibration equation was tested for the prediction of IVDOM for an additional 90 corn silages. The regression of NIR predicted IVDOM resulted in an R² of .37 and S_y.x of 4.04. The prediction of corn silage IVDOM by NIR or ADF were similar.vii, 91, [2] leavesapplication/pdfen-USIn CopyrightLD5655.V855 1982.M63Corn -- SilageThesis