"Mistakes of diagnosis are most apt to be made in those cases having in the early stages pronounced psychotic symptoms and relatively slight defect symptoms, or in cases with few definite physical signs. Lumbar puncture should always be made if there is any doubt about the diagnosis. A Wassermann examination of the blood alone is not sufficient as this does not tell us whether or not the central nervous system is involved." A study of the statistics of the thirteen New York state hospitals in the "pre-Wassermann" days and before we had acquired our present accurate knowledge of the pathology of general paresis shows that there were 84,152 admissions during the fourteen years ending on October 1, 1888. Of this number 5,697, or 6.76 per cent, were diagnosed as general paresis. In the same hospitals, from 1912 to 1919 inclusive, 6,374 cases of general paresis were reported,—12.71 per cent of the 49,640 first admissions. During the years 1918 and 1919 that disease constituted 13.19 per cent of all admissions. This apparent increase undoubtedly is due to the fact that modern methods have materially improved facilities for accuracy of diagnosis. It is not at all probable that the admission rate has doubled during the period in question for any other reason. In the Massachusetts hospitals during the year 1919, only 7.90 per cent of the first admissions were diagnosed as general paresis. There was, however, an unusually high rate of cerebral syphilis. In twenty-one hospitals in fourteen other states, reports based on the present classification show a total of 18,336 admissions, mostly in 1917, 1918 and 1919. Of this number 1,233, or 6.72 per cent, were cases of general paresis. Thus, in a total of 70,987 admissions based on the present classification of psychoses as used by the American Psychiatric Association there were 7,845 cases of general paresis in all,—a percentage of 11.05. It is, of course, a well-known fact that general paresis is largely a psychosis of densely populated communities. This is readily shown by the New York statistics. During the year 1919, 9.6 per cent of the admissions at Binghamton were cases of general paresis. The percentage at Buffalo was 15.5; at Gowanda, 17.3; Hudson River (Poughkeepsie), 9.0; at Middletown, 3.7; Rochester, 8.6; St. Lawrence (Ogdensburg), 9.2; Utica, 10.1; and Willard, 13. In the institutions caring for the insane of New York City 16.3 per cent were reported at the Manhattan State Hospital, 13.5 per cent at Kings Park, and 14.7 per cent at Central Islip. The percentage at the other institutions, except at Buffalo and Gowanda, which care almost entirely for residents of the city of Buffalo, is determined very largely by the transfer of patients from the hospitals of New York City and the metropolitan district. General paresis constitutes approximately ten per cent of the commitments in the city of Boston. On the other hand, we find an admission rate of