The rodent calvarial defect model is commonly used to investigate bone regeneration and wound healing. This study presents a micro-computed tomography (micro-CT) methodology for measuring the bone mineral content (BMC) in a rat calvarial defect and validates it by estimating its precision error. Two defect models were implemented. A single 6 mm diameter defect was created in 20 rats, which were imaged in vivo for longitudinal experiments. Three 5 mm diameter defects were created in three additional rats, which were repeatedly imaged ex vivo to determine precision. Four control rats and four rats treated with bone morphogenetic protein were imaged at 3, 6, 9 and 12 weeks post-surgery. Scan parameters were 80 kVp, 0.45 mA and 180 mAs. Images were reconstructed with an isotropic resolution of 45 microm. At 6 weeks, the BMC in control animals (4.37 +/- 0.66 mg) was significantly lower (p < 0.05) than that in treated rats (11.29 +/- 1.01 mg). Linear regression between the BMC and bone fractional area, from 20 rats, showed a strong correlation (r(2) = 0.70, p < 0.0001), indicating that the BMC can be used, in place of previous destructive analysis techniques, to characterize bone growth. The high precision (2.5%) of the micro-CT methodology indicates its utility in detecting small BMC changes in animals.