Voluntary intake of silages in dairy cows depending on chemical composition and in vitro gas production characteristics

A study was conducted to investigate the possibilities to develop models for predicting the relative silage dry matter intake (SDMI) in dairy cows utilising information on chemical composition and in vitro gas production (GP) kinetics of silages. In five experiments, each with an average of 38 lactating dairy cows, SDMI was recorded for 15 grass silages made from primary growth and regrowth swards of timothy (Phleum pratense L.). The silages were characterised by chemical analysis and by utilising an automated in vitro GP recording technique with end point measurements of substrate residues. The silage samples were analysed both as dried and wet samples to evaluate the effects of sample preparation techniques on GP kinetics and their relations to SDMI. Relationships between feed variables and SDMI were investigated utilising simple linear and multiple regression. The wet silage samples had higher cumulative GP and different GP curves compared to the dried samples. The linear relationships between, GP variables, harvest number (first or second cut) of the grass, chemical characteristics of the silages and the relative SDMI show that the GP technique is a powerful tool to detect silage quality. By using the parameters from the dried samples the multiple regression analysis resulted in a relationship, relative SDMI (kg per 100 kg live weight (LW)) per day = 0.071 + 0.0029 × NDFD − 0.266 × C (R² = 0.82, S.D. = 0.06). NDFD is the degradability of the neutral detergent fibre (NDF) (g/kg NDF) and C is the variable that regulates the switching characteristics of the GP profiles. By using the wet silage samples the multiple relationship did not include any GP variable; relative SDMI (kg per 100 kg LW per day) = 1.86 − 0.008 × acetic acid (g/kg DM) + 0.024 × ethanol (g/kg DM) (R² = 0.62, S.D. = 0.08). The results from the regression analysis and the experience of the laborious sample preparation technique for wet samples, give the conclusion that dried silage samples are recommended for determining feed characteristics using the GP technique in intake studies.