In order to determine the effects of a varied level of dietary energy intake during pregnancy and lactation on milk yield and composition, first, second and fourth parity sows (Large White x German Landrace) were provided with energy at a level of either: (i) 100% of ME requirement (MEreq) during pregnancy and lactation, (ii) 120% MEreq during pregnancy and 80% during lactation, and (iii) 80% MEreq during pregnancy and 120% during lactation. In spite of equal target levels feed analysis revealed that gestating first parity sows with 120/80 treatment combination and lactating sows of 80/120 treatment combination received 25, and 11-17% more digestible N than in the respective 100/100 treatment combination. Irrespective of this 120/80 sows responded with the highest milk DM, fat, and energy contents, and the lowest lactose concentrations whereas protein levels where not affected, irrespective of parity (p < 0.05). Milk yield of sows in 1st and 4th lactation was 85 and 106% of that in 2nd lactation, respectively. Average milk composition was 18.1% DM, 4.9% protein, 6.8% fat, 5.6% lactose, and 0.8% ash. Milk composition changes ceased at day 7 of lactation with a reduction of milk GE and protein, and an increase of lactose content. Concentrations of threonine, arginine, valine, leucine, tyrosine, phenylalanine, cystine, and tryptophan, as well as stearic, oleic, and linoleic acid were higher in colostrum than in milk at later lactation stages. In contrast, laurine, myristic, palmitic, and palmitoleic acids were lower concentrated in colostrum. In conclusion, these results illustrate the importance of body reserve mobilization for milk production in sows and indicate that low energy supply during gestation cannot be compensated by higher energy supply during lactation.