BY HEBER A. LONGMAX. ' 



taking as evidence a mere transverse section of the tree of 

 life. At the most, only a few generations of the more 

 complex organisms can be studied. Let me risk re-stating 

 an old argument by outlining the development of an 

 organism in a special and novel environment. It can be 

 demonstrated b}- experiment, and it will be generally 

 admitted, that most organisms tend to var}^ when subjected 

 to unusual conditions, i.e., when placed in a novel environ- 

 ment which is not so abnormal sa to bring about extinction. 

 In the lifetime of the first individual any variation would 

 be classed as "' transient " (Weismann). and the proof of 

 its heritabilit}' is not accepted. But if the novel environ- 

 ment be maintained, as in the substitution of fertile 

 conditions for eremian plants, or lower altitudes for alpine 

 species, each generation would be re-subjected to the same 

 stimuli. Such changes in environment frequently occur in 

 nature owing to geological and climatic factors. The 

 cumulative effects of such responses on the race is to give 

 specific and even generic distinctions from allies in an 

 adjacent primitive environment. Ultimately we find in 

 special environments, which have apparently existed for 

 long periods, remarkable modifications to the needs of the 

 organism in that environment. Natural selection would 

 of course, weed out unsuita})le individuals. It is surely 

 more logical to assume that these once acquired 

 characteristics are heritable to a verj- large extent and 

 that each individual does not go through, as an individual, 

 the whole process of transformation. To assume that 

 natural selection has to wait for the occurrence of favoiu'able 

 fortuitous germinal variations is to blind one"s eyes to the 

 many experiments demonstrating that a multitude of 

 organisms are plastic, and that great modifications in 

 a parficidar direction may be induced h\ an artificial change 

 in environment. 



8ir Ray Lankester has criticised the two laws of 

 Lamarck which have been summarised by Prof. Poulton, 

 as follows : — Lamarck's " first law assumes that a past 

 history of indefinite duration is powerless to create a bias 

 by which the present can be controlled ; whilst the second 

 assumes that the brief history of the present can readily 

 raise a bias to control the f\iture."' Lankester points out 



