Comparison of two in vitro fermentation gas production methods using both rumen fluid and faecal inoculum from sheep

Two in vitro gas production apparatuses, glass syringes (GS) and pressure transducers (PT), used for measurement of gas production (GP), were compared using rumen fluid (RF) and fresh faeces (FF) as sources of inocula in order to study fermentation patterns of feeds. The RF and FF from ruminally fistulated Merino sheep were mixed with McDougall's buffer and added to fermentation vessels (either fermentation bottles or syringes) containing substrate, being meadow hay (MH), wheat straw (WS), amorphous cellulose (AC) and barley grain (BG), which were evaluated in separate incubations at 39 +/- 0.5 degrees C for 96 h. Total GP was recorded after 6, 12, 24, 36, 48, 60, 72, 84 and 96 h of incubation. Volatile fatty acids (VFA) and in vitro dry matter degradability (IVDMD) were determined at the end of the incubation. Initial gas production was higher with PT versus GS, but rates and extent of gas production in the two apparatuses were well correlated (r=0.998 for total gas production and r=0.765, 0.919, 0.984 for rates of gas production at 3, 18 and 30 h, respectively). The IVDMD and total VIA production did not differ between PT and GS. Total gas production by PT and GS were correlated to both IVDMD (R-2=0.685, 0.630 for PT and GS, respectively) and stoichiometrically predicted gas volumes (R-2=0.938, 0.934 for PT and GS, respectively). However, results obtained by RF and FF inocula differed, with rate and extent of gas production being much lower with FF versus R-F, and data from the two inocula were generally poorly related (r=0.581, 0.809, 0.355, 0.271, 0.109, 0.558, 0.887, 0.909, -0.210) for total gas production, rates of gas production at 3, 18 and 30 h, IVDMD, total VFA produced and molar proportions of acetate, propionate and n-butyrate, respectively). Both PT and GS are suitable to estimate gas production and results from one apparatus may be extrapolated to those from the other. However, this is not possible for data obtained using faecal inocula versus rumen fluid. (c) 2005 Elsevier B.V. All rights reserved.