The effects of three fat supplements on milk yield and composition were measured using 12 mid-lactation in-calf Hoistein-Friesian cows in a balanced incomplete change-over design over three periods each of 3 weeks. All cows received a basal diet consisting of 36 kg/day grass silage (dry matter (DM) 270 g/kg, metabolizable energy (ME) 11·6 MJ/kg DM) and 7 kg/day o f a concentrate mixture containing (g/kg) rolled barley (501), molassed sugar-beet pulp shreds (277), soya-bean meal (208) and a standard cow mineral supplement (14). Treatments were CON (control-no supplement); LIN and FISH (250 gl day of either linseed oil or marine oil, providing approximately 0·046 of ME intake) or TOA (95 glday of tuna orbital oil, providing 0·018 of total ME intake).
There were no significant effects on silage DM intake or milk yield (means 9·25 and 17·2 kg/day respectively). The FISH and TOA treatments depressed (F < 0·05) milk fat concentration (45·4, 44·6, 34·5 and 41·6 (s.e.d. 1·08) g/kg for CON, LIN, FISH and TOA respectively; note — the same treatment order is used for all results quoted). Compared with values for CON, yield of f at (glday) was significantly (F < 0·05) greater for LIN and significantly lower for FISH (739, 808, 572 and 732, s.e.d. 28·7). All three oil supplements reduced (F < 0·05) milk protein content (33·6, 32·5, 30·6 and 32·4 (s.e.d. 0·43) g/kg) but, apart from a small increase for LIN, protein yield (glday) was unaffected (545, 586, 510 and 574, s.e.d. 20·2).
The concentrations (g/100 g) of short-chain fatty acids (< C14) and C16 : 0 in milk f at were lower (F < 0·05) for LIN than for the other treatments. All supplements increased the concentrations ofC18:1 (F < 0·05), the value for LIN being greater (F < 0·05) than for the other treatments (21·0, 27·2, 25·3 and 23·7, s.e.d. 0·74). The FISH and TOA treatments increased (F < 0·05) the concentrations of long chain (< C2O) (n-3) poly-unsaturated fatty acids (PUFA), (0·19, 0·17, 0·49 and 0·27, s.e.d. 0·026) but less than proportionately 0·03 of dietary intake of these acids was transferred to milk, probably because they were found to be mostly in the phospholipid and cholesterol ester fractions of plasma. The FISH and TOA treatments increased (F < 0·05) the percentages of total trans fatty acids in milk fat (1·13, 2·19, 10·26 and 3·62, s.e.d. 0·728) whilst a significant (F < 0·05) increase in conjugated linoleic acid (CLA) was observed only for FISH (0·16, 0·28, 1·55, and 0·52, s.e.d. 0·154). Concentrations of CLA and total trans acids in milk were highly correlated (r = 0·91, no. =36, F < 0·001) whilst trans acids in milk were inversely correlated with milk fat content (r = -0·63, no. = 36, F < 0·001) supporting the theory that milk fat depression may be caused by increased supply of trans fatty acids to the mammary gland. The health implications of these changes in milk fat composition are discussed.