Therapeutic results in 50 cases of neurosyphilis Cerebrospinal fluid changes were studied in 50 cases of neurosyphilis, which were treated as follows: 1) malaria and arsenotherapy, 10 cases; 2) penicillin alone, 13 cases; 3) penicillin, fever and arsenotherapy, 21 cases; 4) penicillin and fever therapy, 3 cases; 5) arsenotherapy, 3 cases. Results of treatment were evaluated by the study of cell count, total protein contents, colloidal benzoin test and complement fixation tests for syphilis with heart antigen (Wassermann test) and human brain antigen (Steinfeld test). Return to normality was achieved in 34 per cent of the cases; satisfactory results, in 42 per cent; indefinite response, in 22 per cent; failure of treatment, in 2 per cent. The use of t statistics for the evaluation of results evidenced: a) significance of differences between inicial and final means of the cell count, total protein contents and colloidal benzoin test in the whole group of patients, in the 23 cases with 6 months of follow up and in the three main therapeutic groups; b) non significant differences between the final means of the cell count in these therapeutic groups; c) in what regards the total protein contents and the colloidal benzoin test, the statistical study of final means revealed no significant differences between groups 1 and 3, whether group 2 differed significantly from the others; d) there was not significant differences in the percentages of negativation of complement fixation tests in the three therapeutic groups; e) the differences in the percentages of negativation of complement fixation test with cardiac antigen (Wassermann test) and human cerebral antigen (Steinfeld test) were significant. These results lead to the conclusion that penicillin therapy alone, though equivalent to the other methods of treatment in what refers to the cell count, is more effective concerning the total protein contents and the colloidal benzoin test. This is surprising because method 3, which is the sum of methods 1 and 2, was supposed to be more effective. Considering the dangers involved in malariatherapy, penicillin alone is thus the treatment of choice for neurosyphilis, in total doses of ten to fifteen million units. Replacement of crystalline by procaine penicillin (which is on study in the Department of Neurology) seems to be possible according to the experience of American writers.