The Optimal Dose of Antigen for the Wassermann Test

The representative protocols herewith submitted serve to illustrate several considerations which bear upon methods of choosing the proper antigenic dose. Selection of an arbitrary fraction such as one–fourth of the anticomplementary dose is obviously faulty because of the marked variations in antigenic potency among different extracts, and because with strong antigens this amount would be large enough to inhibit fixation except with strongly positive sera (table 2). Determination of the antigenic “unit” (with its implication of constancy) and the use of a certain number of units as the dose empirically found to be optimal is a practical method leading to fairly accurate results but is faulty in principle because the positive serum used is a variable, and upon its potency depends the size of the antigenic unit. The attempt to control serum variability by using a large pool of strongly positive specimens is only a fairly practical device, which does not overcome the objection to “standardizing” one unknown by means of another. Kolmer favors 10 antigenic units while Kahn and Johnson's tables show the optimum to be 1.5 to 2 units. In table 1, above, it may readily be seen that if a serum pool had the strength represented by the 0.05 ml. amount the optimal antigenic does would contain about 2 units, whereas were it twice as strong the same optimal dose would contain at least 10 units. Our experience indicates that the optimal dose is that quantity which gives strongest fixation with minimal amounts of reagin, that it can be closely approximated by means of appropriate titrations with any positive serum, and that it is uniform in its reactivity with serum from any stage of syphilis. In accordance with a general immunologic principle a constant may thus be determined by which serum units can more definitely be measured, thereby permitting, when desired, the establishment of a more significant and reproducible “unit” of antigen. For practical routine tests, however, with the potent types of antigen now in common use, such circuitous measurements of the antigenic unit seem unessential.