Four experiments were performed to determine if residual feed intake (RFI) was related to mitochondrial complex I (CI) protein. For Exp. 1, crossbred Angus steers (initial BW 270 ± 2.0 kg) were fed for a total of 170 d (n = 72). For Exp. 2, crossbred Braunvieh steers (initial BW 280 ± 3.0 kg) were fed for a total of 150 d (n = 50). For Exp. 3, crossbred Braunvieh heifers (initial BW 260 ± 3.0 kg) were fed for a total of 160 d (n = 40). For Exp. 4, crossbred Angus steers (initial BW 290 ± 3.0 kg) were fed for a total of 160 d (n = 40). All cattle in all experiments were fed the same diet. The variable RFI was calculated as the difference between predicted and actual DMI. Predicted DMI was calculated from regressing intake on ADG and metabolic body weight. Blood was collected, lymphocytes were isolated, and antibody used to capture CI. For Exp. 1, 2, and 3, CI quantity was measured using an ELISA commercial kit (Mitosciences, OR). For Exp. 4, CI subunits were separated by gel electrophoresis and bands were analyzed for differences in concentration (absorbance) among RFI phenotypes. For all 4 experiments, there was a significant difference (P < 0.05) between RFI and DMI but no difference (P > 0.05) was reported for ADG and metabolic midweight. For Exp. 1, 2, and 3, CI concentration in mitochondria was greater (P < 0.05) for low RFI compared with other treatments. For Exp. 4, animals with low RFI had a trend (P = 0.07) for greater concentration of Band I (protein S1) than high RFI. Correlation between RFI and Band I was -0.72 (P = 0.04). We concluded that mitochondrial function was at least in part responsible for differences among animals in metabolic efficiency.