YELLOW FEVER. 419 



merous outbreaks of limited extent, and to the 

 extended epidemic in the United States in 1878, 

 followed by a reappearance of the disease in Mem- 

 phis in 1879, strongly support these propositions, 

 and the inference drawn from them as to the na- 

 ture of the yellow fever poison. It will be seen, 

 however, that our propositions, if accepted as 

 proven, do not necessarily lead us to the conclu- 

 sion that the yellow fever germ multiplies within 

 the bodies of those sick with the disease. On the 

 other hand, if the first proposition is true, it seems 

 altogether probable that it does not multiply with- 

 in the bodies of the sick, but that the poison is 

 evolved as a result of its vital activity during the 

 decomposition of the dead organic material which 

 serves as pabulum for its growth. The observed 

 facts relating to the epidemic prevalence of the 

 disease indicate that decomposing animal matter 

 furnishes a suitable nidus for the germ, and conse- 

 quently the dead body of a yellow fever patient 

 should constitute such a nidus, even if the living 

 body does not. As a matter of fact, infection 

 has very frequently been traced to dead bodies, 

 whereas there is abundant evidence to show that 

 persons contract yellow fever by exposure in in- 

 fected localities, and not by contact with those sick 

 with the disease. Bedding charged with organic 

 emanations from the body of a sick person is also 

 a suitable nidus for the germ. But the infectious 

 character of infected bedding seems to be acquired 

 in infected localities rather than to be due to infec- 



