The collection of skeletal material at the University of Maryland School of Medicine now includes a small sample of scapulae and long bones of the upper extremities taken from cadavers of known handedness. This sample has been used to begin studies for determining the accuracy currently used for diagnosing handedness. Measurements have been taken of the deflection angle that results from dorsal inclination of the glenoid fossa, humeral length and head diameter bicondylar width, and radial and ulnar lengths. A radiohumeral index and the total combined length of the long bones for each extremity as well as side difference in total length were calculated. These data plus the presence or absence of an extensor facet onthe dorsal margin of the glenoid fossa have been compared with the known hand preference. Thus far, the data indicate that the extensor facet, greater dorsal inclination of the glenoid fossa, greater total length of long bones, and greater bicondylar with correlate with the dominant side. Since the sample size is small, further work is needed before a valid statistical analysis can be made.