Two experiments were conducted, the first to determine whether numbers of recovered bacteria differed due to sampling method used or due to location on carcass sampled (breast or leg quarters) and the second to determine if numbers of bacteria differed between the front (ventral) and back (dorsal) side of the carcass. In both experiments, eviscerated broiler carcasses were obtained from a commercial processing plant just before the final inside-outside bird washer. In experiment 1, carcasses (3 in each of 4 replicate trials) were separated into leg quarters and breast quarters (n = 48) and either rinsed or ground and stomached for microbiological sampling. In experiment 2, for 3 replicate trials of 4 carcasses each, necks, wings, and legs were manually removed; the remaining trunks were cut through the sides to produce front (ventral) and back (dorsal) halves (n = 24); and then rinsed. For both experiments, coliforms and Escherichia coli were enumerated. In experiment 1, significantly higher numbers (P < 0.05) of coliforms and E. coli were recovered by rinsing than by grinding from both breast and leg quarters. Leg quarters were found to have higher bacterial numbers than breasts from grind samples, but no quarter differences were found for rinse samples. In experiment 2, higher (P < 0.05) numbers of coliforms and E. coli were recovered from the dorsal carcass half compared with the ventral half. Bacterial counts of broiler carcasses are affected by both the sampling method used and by carcass location sampled.