Bone mass accretion during puberty appears to be critical in the development of peak bone mass. Although bone density of females in Korea has been studied, only a few studies have related bone mass with anthropometric patterns or puberty in the pubescent girls. This study was conducted as part of a study of major determinants of bone development during puberty. Subjects were aged 14∼16 yr(mean 14.97), and had no history of disorders or dedication use likely to influence bone or calcium metabolism. Bone mineral density and content were measured by dual energy X-ray absorptiometry using a Lunar DPX+Scanner (Lunar Madison, WI). Also, total body fat, and total lean body mass were assessed using a Lunar DPX dual-energy X-ray absorptiometer, Pubertal status was assessed according to the Marshall and Tanner guidelines. Serum levels of osteocalcin was measured by RIA using a commercial kit assay. Skinfold measurements were taken with a skinfold caliper(Lange Caliper, USA). Data were analyzed using the regression and GLM procedure of the statistical package SAS. The results indicated that the observed means for lumbar spine BMD and femoral BMD correspond to approximately 91% and 96% of the means for young adult females, respec tively. All subjects were menarchal, with the majority being in the middle to end stages of pubertal development. Total body BMD was positively related to fat mass(P<0.001), lean body mass and time since menarche, and negatively related to urine pyridinoline, serum alkaline phosphatase and osteocalcin. The data indicate that girls who reported lower age for menarche had significantly higher bone densities than girls who reported higher age for menarche. Attaining peak skeletal bone mass during puberty may reduce the incidence of osteoporosis in later life. this finding suggests that early menarche may augment peak bone mass, influencing the extent of bone loss later in adulthood. The results suggest that good nutrition in childhood appears to be needed not for growth and development, but possibly also to assure an optimal peak of bone mass and thus greater latitude for the maintenance or skeletal integrity in the face of bone losses. Troeps skinfold thickness was a better predictor of total BMD and total BMC than was any other skinfold thickness. The study did not find a relationship between total BMD and body fat %, but total fat was significantly positively related to total BMD(r=0.49) and total BMC(r=0.60). It supports earlier report that there was a significant correlation between TBMD and body weight. Conclusively, total fat, lean body mass and pubertal development could influence BMD in pubescent girls. Clearly, longitudinal studies are required to assess the effect of puberty on peak bone mass, and to define further the potential determinants of peak bone mass.