For the past few years, I have been somewhat dubious of the true value of the Schick test. This doubt was crystalized during the epidemic of diphtheria in 1923, when several children and nurses who had responded with negative reactions to carefully controlled Schick tests developed the disease.
This was the impetus which started me actually to titrate the blood diphtheria antitoxin in subjects who were known to have either positive or negative Schick reactions and, later, to extend the work to measure the blood antitoxin response of various groups following injections of diphtheria toxin-antitoxin.
In order to accomplish this work, it was necessary to have a technic that was accurate, simple and economical. Two methods were open: that of Zingher,1 which was found to be incorrect in its dilutions, and that of Roemer.2 The latter method was practical, but the end-points were not sharp enough to