1 84 EXPERIMENTAL EVOLUTION LECT. 



glycerine, or of clams, or of monkey, hen, or goose, 

 while in rat broth it becomes very feeble. Each 

 micro-organism has its very marked preferences as 

 regards temperature and chemical conditions, and the 

 susceptibility to variation in these conditions is so very 

 marked, that while the fowl is too warm-blooded to 

 allow anthrax to develop, it is enough to cool the 

 animal artificially to render it inoculable. 



These facts show that the correspondence between 

 the environment and the organism is very close, and that 

 very slight alterations are enough to cause death, and 

 the result is that in the experimental investigation of 

 the influence of environment upon variability, we 

 must be careful to handle our methods with great 

 prudence. 



But, on the other hand, while in all cases very 

 slight modifications in environment are apt to 

 destroy life especially as concerns chemical environ- 

 ment we often justly wonder at the extent of the 

 modifications which may be introduced into the latter 

 without impairing the vital functions. Micro-organ- 

 isms afford numerous instances of this fact, but I 

 prefer giving some examples from the higher organ- 

 isms. It is a familiar fact that while most aquatic 

 animals or plants die very quickly when transferred 

 from sea to fresh water or from fresh to sea water, 

 a number of them withstand the change perfectly 



