The aim was to evaluate under protein-limiting conditions the effect of different supplemental energy sources: fermentable NSP (fNSP), digestible starch (dStarch) and digestible unsaturated fat (dUFA), on marginal efficiency of fat deposition and distribution. A further aim was to determine whether the extra fat deposition from different energy sources, and its distribution in the body, depends on feeding level. A total of fifty-eight individually housed pigs (48 (SD 4) kg) were used in a 3 £ 2 factorial design study, with three energy sources (0·2 MJ digestible energy (DE)/kg 0 ·75 per d of fNSP, dStarch and dUFA added to a control diet) at two feeding levels. Ten pigs were slaughtered at 48 (SD 4) kg body weight and treatment pigs at 106 (SD 3) kg body weight. Bodies were dissected and the chemical composition of each body fraction was determined. The effect of energy sources on fat and protein deposition was expressed relative to the control treatments within both energy intake levels based on a total of thirty-two observations in six treatments, and these marginal differences were subsequently treated as dependent variables. Results showed that preferential deposition of the supplemental energy intake in various fat depots did not depend on the energy source, and the extra fat deposition was similar at each feeding level. The marginal energetic transformation (energy retention; ER) of fNSP, dStarch and dUFA for fat retention (ERfat:DE) was 44, 52 and 49 % (P. 0·05), respectively. Feeding level affected fat distribution, but source of energy did not change the relative partitioning of fat deposition. The present results do not support values of energetic efficiencies currently used in net energy-based systems.Fattening pigs: Energy sources: Fat deposition: Fat partitioning: Energy utilisation Data on the contribution of different energy sources to growth performance, and protein and fat deposition, are limited. Essential information is missing on the mechanism by which various energy sources affect fat deposition and its distribution within the body. A mechanistic growth model for pigs (1,2) clearly indicated the higher potential of dietary fat for body fat deposition compared with carbohydrates. Equal intakes of energy from glucose, long-chain fatty acids and SCFA may result in differences in fat deposition rates (3,4) . It is, however, unknown to what extent distribution of body fat over tissues is affected by the type of nutrient absorbed. Unlike the effect of energy source on ATP-generating potential, these effects should be studied at feeding levels above maintenance energy, and preferably under protein-limiting conditions. Also, in many studies (5 -8) , increased intake of one energy source is balanced by decreasing the content of other energy sources, maintaining isoenergetic intakes, but nonetheless complicating the interpretation of causal relationships. It has been reported that the energetic efficiency of digestible energy (DE) intake for energy retention depends on energy supply (9) . Therefore, ...