RETURN TO VIRULENCE 309 



preserve an action on a living species. Let us take for 

 example the most attenuated, that which is barely able 

 to kill a guinea-pig a day old: if we inoculate its blood 

 into a guinea-pig of the same age, that of the second 

 animal into a third, and so on, we shall shortly see the 

 virulence of the bacteridium return little by little. 

 Soon we shall be able to kill with it guinea-pigs three or 

 four days old, a week, a month old, and finally sheep. 

 By successive cultures in living media, the bacteridium 

 has been restored to its original virulence. 



It is justifiable to form out of these facts a general rule, 

 in accordance with our theory. A microbe introduced 

 into the body of an animal is not living under the same 

 conditions as one sown in an inert vessel; it is subjected 

 to the pressing alternative of living or dying, of being 

 victorious or vanquished. Vanquished, its history is 

 soon written; victorious, it will come out of the struggle 

 strengthened, that is to say, having complied with the 

 conditions of its new medium, it is better prepared to 

 accommodate itself therein anew. If it is transferred 

 several times from individual to individual of the same 

 race, without having been influenced by external con- 

 ditions in the interim between two passages, we may 

 expect to see its virulence augmented and in some degree 

 fixed for the race and for the customary mode of trans- 

 mission in this race. Thus the bacteridium of sheep an- 

 thrax, for example, living for a long time on our soil, is 

 acchmated to some degree in the race which shelters it, 

 and its virulence varies little from one subject to another, 

 and from one year to another for the same country. 

 The same thing is true, to a certain extent, for Jenner's 

 vaccine, if it is transferred directly from arm to arm on 

 unvaccinated healthy individuals, and if it is carefully 

 preserved between the two operations. The same thing 

 is also true for the virus of rabies administered by tre- 



