Mohammed, R., Khorasani, R. G., Goonewardene, L. A., Kramer, J. K. G. and Kennelly, J. J. 2011. Persistency of milk trans-18:1 isomers and rumenic acid in Holstein cows over a full lactation. Can. J. Anim. Sci. 91: 147-167. A long-term lactation study was undertaken to determine whether the previously reported short-term persistency in vaccenic acid [VA; trans(t)11-18:1] and rumenic acid (RA) could be maintained. To test this hypothesis, 24 Holstein cows were allotted to two experimental diets (control and test) from 2 wk before calving until they were 270 d in milk (DIM). The test diet was similar to the control diet, but supplemented with sunflower seed (11.2% diet DM), fish oil (0.5%) and monensin (22 mg/kg DM) by replacing an equivalent amount of barley grain. The forage: concentrate ratio was 50:50 (DM basis) with 35% barley silage and 15% alfalfa hay. Milk was sampled every fortnight from the start of lactation until cows were 270 DIM. Data obtained were averaged into three equal periods of 90 d each, representing three stages of lactation (SOL): early-lactation (EL), mid-lactation (ML) and late-lactation (LL). Dry matter intakes were not different between treatments with greater intakes observed during ML than during EL or LL. Milk yield was not different between treatments and decreased with increasing DIM. Milk fat content and yield showed interaction between treatment and SOL with lower values observed for the test diet than control diet during EL and ML. De novo synthesized fatty acids (4:0-15:0), 16:0-16:1 and preformed fatty acids (17:0 and above) showed interaction between treatment and SOL with the former two being greater for control diet than test diet and the latter greater for the test diet than control diet within each SOL. Milk t10-18:1 (% fatty acid methyl esters, FAME) was greater for the test diet compared with control diet (4.38 vs. 1.32) and was greater during ML (3.79) than during EL (2.38) or LL (2.38). Milk VA and RA showed interactions between treatment and SOL with greater values observed for the test diet than the control diet within each SOL. When analyzed by treatment, milk VA was not different across SOL for both diets. Milk RA was not different across SOL for the test diet, but was different for the control diet; it was lower during EL than during ML. Step-wise regression analysis revealed that the variability in milk RA for the control diet (P<0.01; R²=0.97) was determined by VA (70%) and RA/VA (27%); and for the test diet (P<0.01; R²=0.987) by VA (88.7%), RA/VA (5%) and t10-18:1 (3.8%). Desaturase index based on RA/VA showed an interaction between treatment and SOL; it was greater for the control diet than the test diet within each SOL. Overall findings revealed that the differences in milk t10- and VA across SOL reflected possible differences in starch and PUFA intakes, respectively. Differences in milk RA across SOL for the control diet could be attributed to possible differences in mammary desaturase activity based on differences in RA/VA.