TABLE II., TO SHOW THE RELATIONSHIP EXISTING BETWEEN THE WASSERMANN REACTION, THE BUTYRIC ACID REACTION, AND LYMPHOCYTOSIS

TH## || [+++HH++H+HH++ | [H+|+|| ++H++ ||| ++HHH++++ THAAT1+++ || THHHH++H+++ | | | |+HH

112 211

21214

112

9 15.0 8 18.3 43 71.7 8 5.0 2 3.3 55 91.7 7 11.7 3 5.0 50 88.80 0 1

249 39 15.7 29 11.6 181 72.7

9.2 202 81.2 25 10.0 18

venereal history or of clinical manifestations, syphilis can still not be excluded, must lose force when large numbers of such cases have accumulated in our experience.

However, there being no data giving us any accurate knowledge of the frequency of syphilis among sane persons of an urban population, we cannot really tell whether or not these percentages are abnormally high.

As to Noguchi's butyric acid reaction, it would seem that the greater frequency of its occurrence in general paresis shows it to be more sensitive than the Wassermann reaction; the lesser frequency of its occurrence in other psychoses-namely in 10.8 per cent of our cases-would point to greater specificity (if we leave out of consideration the fact that it occurs practically in all inflammatory conditions of the central nervous system); and the simplicity of its technique and the ease with which the readings are taken as indicated by the small percentage of doubtful findings would render it more widely utilizable. Unfortunately, however, this reaction also is not specific as is indicated by its occasional occurrence when there is no other indication of syphilis : in 16 out of our series of 333 cases of psychoses other than general pareses and cerebral syphilis, as shown in Table II. In some of these cases perhaps there is latent syphilis. But the argument that syphilis cannot be excluded can be of no greater weight here than in connection with the Wassermann reaction.

It will be observed from Table II that the butyric acid reaction occurs in connection with the Wassermann reaction either in the blood, or in the cerebro-spinal fluid, or in both, much more often than alone. Possibly the occurrence of both reactions together is a more trustworthy indication of syphilis than the occurrence of either reaction alone.

We have, then, in the cytological examination a method which enables us to detect the existence of an acute or chronic inflammatory lesion of the central nervous system, and in the Wassermann reaction and the butyric acid reaction very useful, if not entirely reliable, tests which help us to determine whether or not the inflammation in question is of syphilitic nature or origin.

The certain differentiation of general paresis from cerebral syphilis, when it cannot be made clinically, cannot be materially aided by these methods of investigation; the only measure that may

help to establish the diagnosis is the test of specific treatment, as was well shown by our case of cerebral syphilis in which the original diagnosis was general paresis in the last stage; in that case treatment by means of intragluteal injections of salicylate of mercury was instituted, not with the hope of obtaining a cure, but with the object of observing the effect of such treatment upon the Wassermann reaction and the butyric acid reaction in general paresis. Our case surprised us by making a recovery within eight weeks, and owing to this fact the diagnosis was changed to cerebral syphilis.

Turning now to Table III we are struck by the almost complete lack of correspondence between positive or negative histories of syphilis and positive or negative results, respectively given by the reactions. In the case of general paresis this is not discomfiting as it is now almost universally acceded that syphilitic infection is a constant precedent of that disease; so that when we find a positive Wassermann reaction or a positive butyric acid reaction and get in the clinical history a denial of specific infection, we simply assume the history to be unreliable.

It may be pointed out that a positive history of syphilis is of higher value than a negative one; in other words, a positive history is more to be relied upon in establishing the fact of actual infection than a negative one in excluding it-for reasons that are sufficiently obvious-although it must be borne in mind that even in the case of positive histories there are still possibilities of error due to mistaken diagnosis of venereal sores, various skin lesions, etc. Thus it may appear strange that of those cases of psychoses, other than general paresis or cerebral syphilis, which gave a positive history of syphilis practically no higher percentage gave a positive Wassermann reaction or butyric acid reaction than those in which negative histories were obtained.

An individual study of our cases revealed the fact that, outside of general paresis, positive reactions were obtained chiefly in cases with a history of recent syphilitic infection, that is where there was any history at all, while in many cases of arteriosclerotic brain disease, involutional psychoses, etc., negative reactions were regularly obtained in spite of a positive history of old syphilitic infection.

It is to be regretted that, owing to special difficulties encountered