Biometrics, the discipline of establishing an individual's identity based upon physical or behavioral characteristics, has become a major research area due to the numerous applications for reliable personal identification. The performance of a biometric system is highly dependent on the chosen biometric identifier. A novel approach for personal identification which utilizes 3D finger surface features as a biometric identifier is presented. Using 3D range images of the hand, a surface representation for the index, middle, and ring finger is calculated and used for comparison to determine subject similarity. The curvature based shape index is used to represent the finger surfaces. A large unique database of hand images supports the research. Data sets obtained over time are used to examine the performance of each individual finger surface as a biometric identifier as well as the recognition performance obtained when combining them. The probe and gallery sets sizes are varied to determine their affect on overall system performance. Performance results for both authentication and identification tasks are presented, which suggest that 3D finger surface is a viable choice as a biometric identifier.