1. The only instance in the Pittsburgh community in which vaccination was completed before the epidemic appeared is that reported from the Dixmont Hospital, Dixmont, Pa., and furnished me through the courtesy of Dr. Hutchinson (16). The institution had a population of about 1,000 patients and 300 employees. Prophylactic vaccination was begun on October 20, and was completed about November 6. Each c.c. of the vaccine used contained 200,000,000 each of B. Pfeiffer, Micrococcus Catarrhalis, B. Friedlander, Pneumococci, Streptococci and Staphylococci, both Aureus and Albus. Four doses were given of 4 minims, 8 minims, 12 minims and 16 minims, respectively. Inoculations were carried out at four-day intervals. Owing to the isolation of the institution from the general community, the first case did not appear until two weeks later—namely, on November 20. The results are shown by the table. Population. No. of % of No. of % of Cases. Cases. Deaths. Deaths. Vaccinated 600 44 7.3% 0 0% Unvaccinated 700 69 9.8% 9 1.2% None of the vaccinated patients developed pneumonia, though there were 15 cases among the unvaccinated. This experiment shows a slight percentage in favor of vaccination, and indicates that there was some decrease in the severity of the secondary infections. 2. The experiment reported by McCoy, Murray and Teeter (17) showed quite opposite results from the above, and was an excellent example of a small though completely controlled test. In an asylum for the insane in San Francisco all of the patients under 41 years of age were divided into two groups—one group was kept as controls and the other was given a vaccine furnished by F. O. Tonney, of the Chicago Health Department. The vaccine contained 500,000,000 each of B. Influenza, Pneumococcus I, II and III, 1,500,000,000 Pneumococcus IV, 1,000,000,000 Streptococcus Hæmolyticus and 500,000,000 Staphylococci. Doses of 0.5 c.c., 1 c.c. and 1½ c.c., which were given at 48–hour intervals. Inoculation was completed on November 15, and the first case of influenza appeared on November 26. The table shows the result. Vaccinated. Not Vaccinated. Persons in group 390 390 Cases of influenza 119 103 Cases of pneumonia 23 17 Number of deaths 10 7 3. The report of Minaker and Irvine (18) included several groups of men, the first two of which apparently belonged in our first series. They used a vaccine, each c.c. of which contained 5,000,000,000 B. Pfeiffer, 3,000,000,000 each of Pneumococcus I and II, 1,000,000,000 Pneumococcus III, 100,000,000 Streptococcus Hæmolyticus. In all, they vaccinated 11,179 persons.