The milk composition responses in five feeding trials were as follows.

1. A study of silage quality showed that the fat content was lower (39.9 v. 43.1 g/kg, s.e. of difference 1.00) and the protein content was higher (35.0 v. 33.7 g/kg, s.e. of difference 0.48) with silage cut three times per season compared with silage cut twice per season.

2. Silage offered as a buffer food overnight in the early part of the grazing season reduced milk yield (18.9 v. 19.9 kg/day, s.e. of difference 0.26) and increased fat content (39.4 v. 35.6 g/kg, s.e. of difference 0.50) when compared with grazing only. Silage offered in the late grazing season did not affect milk yield but increased fat content(44.1 v. 41.3 g/kg, s.e. of difference 0.94). There was no significant effect on protein content.

3. A concentrate containing 18 g crude protein/kg was offered at rates of 0.30, 0.375 and 0.45 kg/kg milk plus silage ad libitum. The fat contents were 38.0, 38.6 and 36.1 g/kg and the protein contents 30.3, 31.2 and 32.1 g/kg, respectively.

4. In two experiments comparing flat-rate concentrate feeding with feeding according to yield, no significant differences were found in milk yield, fat and protein contents.

5. In a study of body condition at calving, cows calving at score 2 had milk fat contents similar to cows calving at score 3 but there were differences in the fatty acid composition of the milks. The cows calving at score 2 produced milk with a higher protein content than those at score 3 (32.8 v. 30.7 g/kg, s.e. of difference 0.85).