Before satisfactory conclusions can be drawn from these figures it is necessary to know how many of the 10,036 persons became sick before vaccination, and whether or not the rate of decrease in this group was not similar to that shown by the number of patients who developed influenza during the intervals between their doses of vaccine. The relatively high percentage of cases following the first and second doses are capable of explanation on one, or perhaps on all, of the three following grounds: (a) the general subsidence of the epidemic, which showed a rapid decrease by the time the third dose was given; (b) the increased protection afforded by the three doses of vaccine, and (c) the broken resistance of the patient following sudden sensitization by the vaccine. 3. Rosenow (21) prepared a mixed vaccine by growing the various bacteria in glucose broth, for from 18 hours to 36 hours, centrifuging and suspending the sediment in salt solution and making up the vaccine on a percentage basis. FORMULA OF VACCINE Pneumococci, Types I (10 per cent.), II (14 per cent.) and III (6 per cent.) 30 per cent. Pneumococci Group IV and the allied green-producingdiplostreptococci described 30 per cent. Hemolytic Streptococci 20 per cent. Staphylococcus Aureus 10 per cent. Influenza bacillus 10 per cent. Most of the vaccine was distributed within a radius of 200 miles of Rochester, Minn., but samples were furnished to physicians all over the country, who agreed to return statistics on its use. No evidence was found that this vaccine caused a temporary break in the resistance of the user. Out of a total of 20,972 persons vaccinated, 14.6 cases of influenza, 1.8 cases of pneumonia, with 1.8 mortality, occurred per thousand in the six weeks following vaccination. As controls, he took “such persons in institutions, colleges, factories and communities where vaccine was used, and included only those reports which contained accurate data as to the incidence and mortality among them.” Among 61,753 such controls he found 229 cases of influenza, 15.7 cases of pneumonia and 3.4 deaths per thousand. He concluded from his results that “it appears possible to afford a definite degree of immunity by prophylactic inoculations to persons against the more serious respiratory infections during the present epidemic.” It is quite difficult to agree with Rosenow in his interpretation of the figures as presented by him, inasmuch as he made no allowance for the stage of the epidemic at which vaccination was carried out, either among the vaccinated or the non-vaccinated. Such a comparison would be well nigh impossible where the vaccine was sent in varying quantities to such a large number of places.