The objective of this study was to explore whether the C and N content can be used to estimate the fat content of animal carcasses. Considering the mean C and N contents of body fat and body protein, the fat content (EE) [%] can be predicted from C and N values [%] according to the generally valid equation EE = 1.3038 x C - 4.237 N. The application of this equation to estimate the total fat content of all animal carcasses results in significant differences in fat content between predicted and measured values. Therefore, we derived specific equations for rats, pigs, cattle, sheep, broilers and mice to predict the fat content by dual linear regression analysis (y = EE [% DM], x1 = C [% DM], x2 = N [% DM]) based on measured fat, C, and N contents of animal body samples. The specific equations for different animals showed residual standard deviations of 1.55, 1.63, 1.12, 1.35, 1.85 and 0.92% fat for rats, pigs, cattle, sheep, broilers and mice, respectively.