It is generally assumed that lactating dairy cows offered grass silage consume less dry matter than cows offered fresh pasture. The aim of this study was to identify the effect of specific components of the ensilage process on food intake and animal performance.

Thirty-six lactating dairy cows were offered the following diets in a changeover design experiment:

Gl = primary growth perennial ryegrass zero grazed between 11/5/94 and 31/5/94 (high digestibility).

Sl,3 = silage made from Gl and allowed to ferment for 3 weeks prior to feeding.

Sl,9 = silage made from Gl and allowed to ferment for 9 weeks prior to feeding.

G2 = primary growth perennial ryegrass zero grazed between 1/6/94 and 21/6/94 (low digestibility).

S2,3 = silage made from G2 and allowed to ferment for 3 weeks prior to feeding.

S2,9 = silage made from G2 and allowed to ferment for 9 weeks prior to feeding.

Each treatment forage was offered for a period of 3 weeks, with food intake and animal production data from week 3 used in the statistical analysis. Forage was obtained from the primary growth of a perennial ryegrass sward, which had received 135 kg N, 19 kg P2O5 and 38 K2O/ha. All grass was pre-mown with a rotary mower, then lifted using a round baler. Bales were offered directly to the cows or wrapped in black plastic, stored and allowed to ferment. No additive was applied at ensiling.