Epidemics of mouse typhoid set up among the Rockefeller Institute strain of mice were studied over a period of 6 months. During this time the relationship of cage number of mouse typhoid bacilli to mortality, total population, and survival time was determined. A single virulence titration of the epidemic strain was made, and at the end of the experiment all survivors were examined for evidence of infection. The following conclusions may be drawn for the data here presented.
1. The available dosage of mouse typhoid bacilli varied directly with the mortality (plus a time constant of 6 to 8 days) and inversely with the survival time.
2. The virulence of the epidemic strain appeared to be practically the same as that of the original stock culture.
3. About 53.5 per cent of the survivors of one epidemic and 68 per cent of those in the other showed, at the end of the experiment, no signs of infection; the others had either specific blood agglutinins, or living bacteria in their heart's blood, spleen, feces, or gall bladder.
4. During the course of the epidemic, the original infecting strain (mouse typhoid Type II—Bacillus pestis caviæ) was almost entirely replaced by an antigenically dissimilar strain (mouse typhoid Type I—Bacillus enteritidis), probably introduced through the inadvertent inclusion of fecal carriers among the normal mice added as contacts.