ACQUIRED CHARACTERS liii 



we may perhaps assume that that influence has likewise 

 to be maintained in operation for an appreciable proportion 

 of the life of the germ : and no proportion would be appreci- 

 able, at all events among the higher animals, unless it extended 

 over many thousand somatic generations. 



I am fully aware of the hazardous character of such a 

 speculation. Indeed, it would scarcely be worth mentioning 

 at all were it not for the suggestion which it involves, that 

 inheritance of acquired characters, if not entirely fictitious, 

 is likely to be true only to this extremely mitigated degree. 

 It suggests, moreover, that the search for evidence would be 

 more fruitful among protozoa and the lower types of metazoa 

 than among more complex forms : for in these primitive 

 animals the soma (where there is one) is less remote from 

 the germ : it has travelled a far shorter way on its develop- 

 mental career. It is indeed among such animals and among 

 plants that the most plausible cases have been cited. But 

 until the process can be actually observed to occur in at 

 least one undoubted instance, use-inheritance must continue 

 to be regarded as an altogether illegitimate hypothesis, and 

 to be rigidly excluded from our account of the factors of 

 organic evolution. 



§ 6. Classification. 



In his classification of animals, Lamarck had recourse to 

 two fundamentally different methods. If animals in nature 

 fall into a linear series, it is obviously a matter of first-rate 

 importance to ascertain the true order in which the species 

 are arranged in this series. In so doing, no divisions or 

 groups of any kind are needed. All that has to be done is 

 to determine the position of each species with reference to 

 other species, and to ascertain the exact point of the series 

 which must be allotted to it. This is what Lamarck means by 

 a distribution générale of animals. 



A second important function of his classification is to 

 draw the lines in the series, which mark off genera, 



