however, the Wassermann test is negative, and this fact makes it of little value in recognizing early primary lesions.  In about 20 to 30 per cent of syphilitic individuals the test returns to negative after the active secondary stage is passed.  This does not necessarily mean that the person is recovering.  It is even possible to have the roof fall out of the mouth from gummatous changes and the Wassermann test yet be negative.  It is equally possible, though unusual, for a negative Wassermann test to be coincident with contagious sores in the mouth or on the genitals.  So it is apparent that as an infallible test for syphilis it is not an unqualified success.  But infallibility is a rare thing in medicine, and must be replaced in most cases by skilful interpretation of a test based on a knowledge of the sources of error.  We understand pretty clearly now that the Wassermann test is only one of the signs of syphilis and that it has quite well-understood limitations.  It has revealed an immense amount of hidden syphilis, and in its proper field has had a value past all counting.  Experience has shown, however, that it should be checked up by a medical examination to give it its greatest value.  Just as all syphilis does not show a positive blood test, so a single negative test is not sufficient to establish the absence of syphilis without a medical examination.  In a syphilitic, least of all, is a single negative Wassermann test proof that his syphilis has left him.  In spite of these rather important exceptions, the Wassermann test, skilfully done and well interpreted, is one of the most valuable of modern medical discoveries.

+The Blood Test in the Treatment and Cure of Syphilis.+—­In addition to its value in recognizing the disease, the Wassermann test has a second field of usefulness in determining when a person is cured of syphilis, and is an excellent guide to the effect of treatment.  Good treatment early in a case of syphilis usually makes the Wassermann test negative in a comparatively short time, and even a little treatment will do it in some cases.  But will it stay negative if treatment is then stopped?  In the high percentage of cases it will not.  It will become positive again after a variable interval, showing that the disease has been suppressed but not destroyed.  For that reason, if we wish to be sure of cure, we must continue treatment until the blood test has become negative and stays negative.  This usually means repeated tests, over a period of several years, in connection with such a course of treatment as will be described later.  During a large part of this time the blood test will be the only means of finding out how the disease is being affected by the treatment.  To all outward appearance the patient will be well.  He may even have been negative in repeated tests, and yet we know by experience that if treatment is stopped too soon, he will become positive again.  There is no set rule for the number of negative tests necessary to