Forty-seven market-weight pigs were slaughtered in order to determine percentage of chemical fat and in an attempt to determine an easily obtainable and inexpensive method to predict this value. The hams and 8-9-10 rib loin sections were removed from the left side of each carcass and dissected into subcutaneous and seam fat, individual muscles, skin and bone. Weights and chemical analysis were determined for each component. Numerous weights, measurements and specific gravity were determined on the carcass, ham and loin section of each pig. Percentage of chemical fat of each ham, loin section and carcasses was determined and correlated with the various weights and measurements taken. Stepwise regression was used to develop prediction equations using carcass data, specific gravity, ham or loin measurements or various combinations of these as dependent variables. The single best indicator of the decimal fraction of chemical fat in the pork carcass was determined to be specific gravity of the carcass half, the prediction equation using this variable had an R-square of .64. By adding 10th rib fat thickness to this equation, the R-square increased to .72. The best equation using carcass variables included 10th rib fat and marbling (R-square = .67). The loin section proved to be an accurate indicator of composition; ham measures were not as accurate as specific gravity and carcass measurements for predicting percentage of carcass fat. This research suggests that the percentage of chemical fat in the pork carcass can be predicted by an easy and inexpensive means.