|Barton Ii, Franklin|
Submitted to: Near Infrared Technologies and It's Application. 2nd Edition
Publication Type: Book / Chapter
Publication Acceptance Date: December 4, 2000
Publication Date: N/A
Technical Abstract: The analysis of highly fibrous feeds with near-infrared spectroscopy (NIRS) by diffuse reflectance is different from that of grains in several respects. The components of the plant matrix are more complex and involve numerous discrete interactions. The work of Hruschka and Norris (1982) showed that for ground wheat, the summation of spectra of the chemical components (protein, starch, cellulose, moisture and simple sugars) did no adequately reflect the total spectral composition when curve-fitting techniques were applied to the spectra. When the complexities of a forage sample are considered, it becomes obvious that the interaction of protein with lignin and carbohydrate along with minor constituents would make analyses by "pure components" impossible. Alternatively, it is possible to consider analyses based solely on functionality present in the spectrum if their relationship to some measure of quality is known. This requires a much better understanding of both the spectra of forages and what constitutes quality than we currently have. Therefore, the analyses must be made on the basis of the spectra correlated to empirical results. Again, the amount of fiber is much greater, i.e. the fiber is the matrix as opposed to being a component, as in most foods and feeds.