Ninety-two swine averaging 104 +/- 4.5 kg and 99 cold carcasses averaging 75 +/- 3.1 kg were measured with a four-terminal plethysmograph. Pigs were transported to the abattoir, fasted 4 h, weighted, and measured for body resistance (Rs, omega), body reactance (Xc, omega), and distance (L, cm) between detector terminals that were located along the dorsal axis of the animal. Pigs were slaughtered 12 h later, carcasses were chilled for 24 h, then weighted (whole carcasses and side carcasses), and cold carcass Rs, Xc, and L measurements were obtained. The right side of the carcass was ground twice, and a 1-kg sample was frozen for later analyses of fat, ash, N, and moisture. Fat-free mass (FFM, kg) was calculated from weight and percentage of fat. Regression analyses were used to develop equations for estimating FFM on a live, adjusted live, whole carcass, and half-carcass basis. Live BW, Rs, and L accounted for the majority of the variation in FFM. Adjusting live BW for head, viscera, and blood weight increased the explained variation for live BW and decreased the variation accounted by Rs. Multiple regression models involving Rs, L, Xc, and weight accounted for 82, 84, and 84% of the variation for FFM expressed on a live, adjusted live, and cold carcass basis, respectively. Results from this study indicate that bioelectrical impedance has excellent potential as a rapid, nondestructive method for estimation of FFM for market swine and pork carcasses.