Validation of the Rostock fermentation test as an in vitro method to estimate ensilability of forages using glass jar model silages as a basis for comparison

摘要

This study compared the predictive power of an in vitro fermentation test [Rostock fermentation test (RFT)] with common glass jar silages (GLASS) using maize, fresh and wilted perennial ryegrass and fresh and wilted lucerne. Treatments were made in quadruplicate and included a control variant without additives, sucrose (20 g kg(-1)), a homolactic inoculant (Lactobacillus plantarum, 3 x 10(5) cfu g(-1)), a heterolactic inoculant (Lactobacillus buchneri, 1 x 10(5) cfu g(-1)) and combinations of each inoculant with sucrose. The pH was measured in GLASS at days 1.5, 3, 10 and 90, and in RFT at 14, 18, 22, 26, 38 and 46 h, whereas fermentation products were determined at days 3 and 90 in GLASS, and at 46 h in RFT. Linear regressions revealed the closest relation for pH between 10-d silages of GLASS and 38-h measurement of RFT (adjusted R-2 = 0.808) and coefficients of determination for fermentation products were always higher when 3-d instead of 90-d silages were compared with 46-h measurement of RFT. A pH increase in GLASS as indicator of insufficient anaerobic stability was not reproduced by RFT, suggesting that the in vitro test reflects the initial phase of fermentation, which was also indicated by the absence of 1,2-propanediol in RFT.