PREDICTION OF THE IN-VIVO DIGESTIBILITY OF WHOLE CROP WHEAT FROM IN-VITRO DIGESTIBILITY, CHEMICAL-COMPOSITION, IN-SITU RUMEN DEGRADABILITY,IN-VITRO GAS-PRODUCTION AND NEAR-INFRARED REFLECTANCE SPECTROSCOPY

Twenty-six winter wheat forages (cultivars Slepjner, Hussar and Cadenz a) harvested at three stages of maturity in each of two years, were co nserved with or without Maxgrass additive or with urea (20 or 40 g kg( -1) dry matter; DM) in 2001 barrels. The forages were analysed for in vivo digestibility in wethers, chemical composition, in vitro rumen fl uid-pepsin digestibility, in vitro neutral detergent-cellulase plus ga mannase digestibility (NCGD), in vitro fermentation gas production and in situ rumen degradability. Forages were also scanned using near-inf rared reflectance spectroscopy (NIRS) and calibration equations develo ped for predicting in vivo digestibility, In vivo digestible organic m atter content (DOMD) was poorly predicted by cell wall content (r(2) l ess than or equal to 0.19), NCGD (r(2) less than or equal to 0.41), ru men fluid DOMD (r(2) less than or equal to 0.41), rumen degradability (r(2) less than or equal to 0.44) and in vitro gas production (r(2) le ss than or equal to 0.26). Although crude protein content was a better predictor (r(2) less than or equal to 0.48), the relationship differe d (P<0.05) with the year of harvest of the forages. In contrast, NIRS was a more accurate and consistent predictor of DOMD in vivo (r(2) = 0 .87). This study indicates that traditional laboratory-based feed eval uation techniques are unsuitable for predicting the DOMD of WCW, but t hat NIRS holds promise. However, as only 26 forages were used to deriv e the calibration equation, further research is required using large ( 150) data sets to validate the promise shown by NIRS and enable its ad option by the advisory services. (C) 1998 Elsevier Science B.V.