Results: Age of gymnasts was positively associated with BMD at all measured sites (p <0.001; r=0.62-0.68). Weight was positively associated with BMD at all measured sites (p <0.001; r=0.82-0.90). Lean body mass was positively associated with BMD at all measured sites (p <0.001; r=0.74-0.87). Body fat percentage was positively associated with BMD at all measured sites (p <0.001-p=0.01; r=0.39-0.54). However, calcium intake was not significantly associated with any of the BMD sites. Sunlight exposure and indirect estimates of vitamin D were not significantly associated with any of the BMD sites; all r-values indicated a weak positive association with BMD. Of the gymnasts who had experienced menses (n=15), those with regular menstrual periods (n=8) had significantly higher BMD values at the arm, leg, trunk, rib, and spine, and total body than those who did not have regular menstrual periods (n=7). There was no significant difference in BMD for gymnasts who had regular periods at the pelvis. A regression analysis was performed. The predictors total BMD values from the regression equation were the following: regular menses, height, weight, percent kilocalorie requirement consumed from predicted kilocalorie needs, calcium intake with supplements, lean body mass, hourly deficits >300 kilocalories from predicted kilocalorie needs, and hourly surpluses >300 kilocalories from predicted kilocalorie needs.