In vivo neutron activation analysis (IVNA) allows in vivo measurement of total body calcium and is considered the gold standard technique. Dual-energy X-ray absorbtiometry (DXA) estimates total body calcium. Although there are a few comparisons of the two techniques, there are none with obese women. The present study compared the estimation of total body calcium by DXA with the measurement of Ca by IVNA in 69 obese women. Sixty-nine obese women (age 44 +/- 10 years, body weight [BW] 93 +/- 10.4 kg, body mass index [BMI] 35 +/- 4, fat mass 42 +/- 7 kg, 48 pre- and 21 postmenopausal, 30 Caucasian and 39 African-American) were recruited for a six-month body composition study without weight loss intervention. Total body (TB) Ca was measured by IVNA (CaIVNA) and by DXA (CaDXA) at baseline (n = 69) and again at six months (n = 40). At baseline, CaDXA was significantly higher than CaIVNA (1018 +/- 130 g vs. 706 +/- 92 g, p = 0.0001). The difference remained after six months (1028 +/- 153 g vs. 715 +/- 96 g). There was a small but significant increase in TBCa at six months by both DXA (16.5 +/- 40, p < 0.02) and IVNA (8 +/- 22 g, p < 0.03). However, this change was not significantly different between the two methods (p = 0.2365). The correlation between the two methods was very high (r = 0.937) and was independent of either race or menopausal status (p = 0.5163). We generated a regression equation: CaIVNA = 0.659 x CaDXA + 34.77; SEE = 32.2, R = 0.937. We present a model integrating the effect of weight and height on the relationship between CaIVNA and CaDXA. The higher the BMI, the bigger the difference between CaIVNA and CaDXA. In our population with a mean BMI of 35.3, CaIVNA = 69% CaDXA.