2. Cases of leptospiral jaundice (Weil's disease) were present among those diagnosed as yellow fever in the recent epidemic in Rio de Janeiro. This is shown by the isolation of cultures of leptospirae from the blood of two patients by H. R. Muller and E. B. Tilden of The Rockefeller Institute, and by the demonstration by us of protective power against leptospirae and absence of protective power against yellow fever virus in the sera from two persons after recovery. The isolation of leptospirae by Noguchi and other investigators from the blood of occasional patients in past epidemics of yellow fever in a number of American countries indicates that leptospiral jaundice was present then as well and was diagnosed clinically as yellow fever.

3. The absence of protective power against leptospirae shown by the Brazilian sera which protected against yellow fever virus and the absence of protective power against yellow fever virus in the sera that protected against leptospirae point to the probability that American yellow fever is not the combined effect of leptospirae and yellow fever virus. The position of L. icteroides, isolated by Noguchi during yellow fever epidemics, now appears to be not that of a secondary invading microorganism in cases of virus yellow fever, but that of the incitant of a form of infectious jaundice, sometimes fatal, often coincident in its appearance with typical yellow fever and apparently indistinguishable from it clinically. This leptospiral disease has not hitherto been separated from true yellow fever. Noguchi's discoveries become; therefore, of the greatest significance in respect to the epidemiology and causation of yellow fever and of infectious jaundice, previously confused one with the other. In all outbreaks of supposed yellow fever hereafter the existence of the two kinds of jaundice, one due to yellow fever virus and the other to leptospirae will have to be taken into account. Only the former probably is spread by mosquitoes and requires anti-mosquito measures for its control.

4. The only difference observed by us between the American and African strains of yellow fever virus was a pronounced difference in virulence for monkeys. The virulence of the two African strains studied was very high while that of the one American strain was highly variable and usually low.