Prediction of forage intake using in vitro gas production methods: Comparison of multiphase fermentation kinetics measured in an automated gas test, and combined gas volume and substrate degradability measurements in a manual syringe system

This study investigated two approaches to in vitro analysis of gas production data, being a three phase model with long ([greater-than-or-equal]72 h) incubation times, to obtain kinetics and asymptotic values of gas production, and combination of gas volume measurements with residue determinations after a relatively short incubation of 24 h. Dry matter intake (DMI) of 24 Ethiopian forages were predicted by these approaches. Combined gas volume and true degradability measurements in vitro were used to calculate a partitioning factor (PF), being the ratio of mg substrate degraded to ml of gas produced. In the automated test, kinetics of gas production were described by a three phase model: y=a1/(1+(b1/t)c1)+[cdots, three dots, centered]+a3/(1+(b3/t)c3), where y is the gas volume at time t and a1-a3 are the volumes of gas in the pools, b1-b3, the time when half of the gas volume of a given pool is produced and c1-c3 describe the switching characteristics of the curves by pool. The model fitted gas production profiles, but there was no relationship between curve parameters and DMI in multiple regressions (R2 = 0.24, P=0.06). In contrast, combined gas volume and substrate degradability measurements at 24 h of incubation accounted for 63, 62 and 56% of the variation (P