Three experiments were conducted to develop and test equations for predicting carcass composition. In the first study using 52 d-old Cobb ${\times}$ Cobb male broilers, twenty four carcasses were selected from 325 processed birds based upon visual appraisal for abdominal fat (low, medium, high) and assayed for specific gravity (SG), dry matter (DM), fat, protein, and ash. In experiment 2, 120 birds were fed rations containing 2 caloric densities (2,880 and $3,200kcal\;ME_n/kg$ diet) and assayed as described above on weeks 2,3,4,5, and 6. Carcass fat was elevated (p < 0.05) with increased caloric density. In both studies predictive variables were significantly correlated with chemically determined carcass fat, protein, and ash contents. Pooled across the 2 studies, data were used to form SG, DM, and or age based equations for predicting carcass composition. Results were tested in experiment 3, where 576 birds reared to 49-d consumed either 2,880, 3,200, or $3,574kcal\;ME_n/kg$ diet while exposed to constant $24^{\circ}C$ or cycling 24 to $35^{\circ}C$ ambient temperatures. Both dietary and environmental effects impacted (p < 0.05) carcass composition. The fat content analyzed chemically was enhanced from 12.4 to 15.7%, and predicted fat was also elevated from 13.4 to 14.8% with increasing caloric density. Heat distress reduced (p < 0.05) analyzed carcass protein (18.9 vs 18.3%) and predicted protein (18.2 vs 17.5%). Predicted equation values for carcass fat, protein, ash, and energy were correlated with the chemically analyzed values at r=0.96, 0.77, 0.86, and 0.79, respectively. Results suggest that prediction equations based on DM and SG may be used to estimate carcass fat, protein, ash, and energy contents of broilers consuming diets that differ in caloric density (2,800 to $3,574kcal\;ME_n/kg$) and for broilers exposed to either constant ($24^{\circ}C$) or cycling high (24 to $35^{\circ}C$) ambient temperatures during 49-d rearing period tested in the present study.