We investigated the relationship between resting metabolic rate (RMR) and various morphological parameters in non-breeding mice, selected for high and low food intake corrected for body mass. RMR was measured at 30 degrees C, and mice were subsequently killed and dissected into 19 body components. High-food-intake mice had significantly greater body masses and a significantly elevated RMR compared with the low-intake mice. Data pooled across strains indicated that body mass, sex and strain together explained over 56 % of the observed variation in RMR. The effects of strain and sex on RMR and tissue morphology were removed, and three separate statistical analyses to investigate the relationship between RMR and organ morphology were performed: (i) employing individual regression analysis with each tissue component as a separate predictor against RMR; (ii) individual regression analysis with residual organ mass against residual RMR (i.e. with strain, sex and body mass effects removed); and (iii) pooling of some organ masses into functional groupings to reduce the number of predictors. Liver mass was the most significant morphological trait linked to differences in RMR. Small intestine length was significantly greater in the high-intake line; however, no difference was observed between strains in the dry mass of this organ, and there was no evidence to associate variability in the mass of the alimentary tract with variability in RMR. The effects of strain on RMR independent of the effect on body mass were consistent with the anticipated effect from the strain differences in the size of the liver.