Quantitative dual-energy radiographic absorptiometry (DRA) and dual-photon absorptiometry (DPA) were compared to determine the best means of assessing bone density. Both methods were used to evaluate the lumbar spine in 107 women (aged 35-84 years [mean, 64 years]) referred for evaluation of osteoporosis risk. High correlation was documented between measurements derived by the two techniques, with a .95 linear regression coefficient for the total spine density measurement. Age-related regression equations were similar in slope but manifested different intercepts. Bone mineral density values derived with DRA were consistently lower than those obtained with DPA (conversion equation: DPA density = [1.067 X DRA density] + 0.163). Besides the inherent imperfections of each system, it was found that inaccurate identification of intervertebral spaces on the low-resolution DPA images introduced errors in patient data. DRA may replace DPA as the dedicated projectional densitometric procedure of choice for technical reasons, but at present a conversion equation must be used to compare DRA data to DPA data.