The aim of this population-based case-control study was to examine the association between bone mass and upper limb fractures in children aged 9-16 yr. Areal bone mineral density and bone mineral apparent density (BMAD) were measured by both dual energy absorptiometry (DXA) and metacarpal index (MI) by hand radiograph. A total of 321 fracture cases and 321 randomly selected individually matched controls were studied. For all fractures, cases had lower DXA measures at all sites (1.1-3.3%; all P < 0.05). A larger reduction was observed for those with wrist and forearm fractures (1.2-4.5%; all P < 0.05, except total body BMAD) but not other upper limb fractures (hand, -1.6 to +1.2%; upper arm: 0.9-4.8%; all P > 0.05). For metacarpal measures, cases had a thinner cortical width and lower MI for wrist and forearm fractures only. In multivariate modeling, both spine BMAD (odds ratio, 1.4/SD reduction) and MI (odds ratio, 1.5/SD reduction) remained statistically significant predictors of wrist and forearm fractures. In conclusion, both DXA measures and MI are independently associated with wrist and forearm but not other upper limb fractures. The magnitude of this association is somewhat weaker than in adults but suggests that optimizing age-appropriate bone mass will lessen the risk of fracture in children.