Magnetic resonance imaging (MRI) thermometry has been utilized for in vivo evaluation of thermal exposure induced by a focused ultrasound beam. A simulation study of the focused ultrasound beam was conducted to select imaging parameters for reducing the error due to the spatial and temporal averaging of MRI. Temperature imaging based on the proton resonance frequency shift was utilized to obtain the temperature distribution during sonication in the skeletal muscle of eight rabbits. MRI-derived temperature information was then used to calculate the thermal dose distribution induced by the sonication and to estimate the coagulated tissue volume. The tissue changes were also evaluated directly by taking the T2-weighted and the contrast agent enhanced T1-weighted MR images. Errors in the temperature and thermal dose measurements were found to be minimal using the following parameters: slice thickness = 3 mm, voxel dimension = 0.6 mm, and scan time per image = 3.4 s. The estimated dimensions of the coagulated tissue volume were in good agreement with the tissue damages seen on the contrast agent enhanced T1-weighted images. The tissue damage seen on the histology was closely matched to the ones seen on the T2-weighted images. This study showed that MRI thermometry has significant potential for both monitoring the thermal exposure and evaluating the tissue damage. This would allow real-time control of the sonication parameters to optimize clinical treatments.