We describe a method for measuring the lean body mass (LBM) and total body fat (FAT) by dual photon absorptiometry (DPA). A total body rectilinear scan was employed with a radiation source of 1 Ci 153Gd. The reliability of estimating the lean percent was assessed in vitro using limb phantoms consisting of ox muscle, lard, and human bone. The precision and accuracy in vitro of the lean percent determination were 1.5% and 1.9%, respectively. The accuracy error in vivo of measuring the total mass of soft tissues (TMST) was approximately 1.4%, thus yielding an overall accuracy error of the LBM of about 2.5%. The precision in vivo of the lean percent and the LBM in kg of duplicate measurements on five healthy subjects was 2.5% and 2.2%, respectively. Other estimates of the LBM and FAT, ie, the calculation according to Boddy et al6 and the skinfold thickness measurement (triceps and subscapular), were compared to the DPA measurement in 100 healthy subjects. High correlations were found between the FAT or FAT% by DPA versus (1) the FAT or FAT% calculated according to the formulae of Boddy et al, and (2) the skinfold thickness. The correlations between the FAT and FAT% by Boddy et al and the skinfold thickness were, however, moderate. The correlation between LBM by DPA and LBM by Boddy et al was highly significant (r = 0.96, SEE = 4.4%). We conclude that LBM and FAT measurements using DPA have precision and accuracy errors that are commensurate with a reliable estimation of the gross body composition.(ABSTRACT TRUNCATED AT 250 WORDS)