THE ERROR IN PREDICTING PASTURE DRY–MATTER DIGESTIBILITY FROM FOUR DIFFERENT METHODS OF ANALYSIS FOR LIGNIN

The lignin content of 50 samples of five grasses of known in vivo digestibility were determined by the methods of Armitage, van Soest and two modifications of the van Soest technique. The error in predicting DM digestibility varied from ±3.1 for tbe Armitage metbod to ±5.0 for the van Soest method. This error compares nnfavourably witb ±2.1 previonsly obtained on tbe same samples with the in vitro technique of Tilley and Terry.