112 ON VARIABILITY AND ADAPTATION 



temporary loss of pathogenic power — of virulence — the case is 

 rather different. It would seem to be most difficult to bring 

 about a loss of virulence that does not tend to be permanent. 

 That is to say, very special methods have to be employed whose 

 action is exerted upon more than one " generation " of the 

 microbes, such as passage through a series of animals less and 

 less susceptible to the invasion of the given microbe in order to 

 exalt organisms, like the pneumococcus of Talamon-Frankel, 

 whose virulence can only be retained outside the body by constant 

 removal to fresh media ; others, like the typhoid bacillus, which, 

 according to Chantemesse and Widal, 1 are so susceptible that 

 a loss of virulence is almost immediate. This, after all, is only 

 what is to be expected if, as seems most probable, the pathogenic 

 property is that which has been latest acquired. A late acquisi- 

 tion is most easily altered, and the tendency to reacquirement 

 of the same is least evident. Intensification of pathogenic 

 properties can, however, be induced by other means than by 

 passage through animals. Hueppe's observations would seem 

 to show that the Bacillus cholerae asiaticae grown anaerobic- 

 ally gains in virulence. 



This principle — that it is difficult to stamp a modification 

 firmly upon a species of bacterium, as, indeed, upon any animal 

 or vegetable species — is exemplified throughout the long series 

 of attempts that have been made to produce new races of micro- 

 organisms, and now, when I proceed to recount some of the more 

 successful of these attempts, the principle will be seen to be 

 constantly in evidence. 



Modifications of Longer Duration 



From the transient modifications which I have noted above 

 it is easy to pass to a series of instances in which the return to 

 the normal only follows after an increasing number of genera- 

 tions, in which numerous generations retain the acquired char- 

 acteristics, and the original properties only gradually reassert 

 themselves. Take, for example, almost any of the chromo- 

 genic bacteria. If a minute quantity of a high-coloured typical 

 growth be removed and spread over the cut surface of a sterilized 

 potato it will be found that the individual colonies which result 



1 Chantemesse et Widal, Annates de VInst. Pasteur, ii., 1888, p. 54. 



