THEORIES OF EVOLUTION 25 



directions by acquirements occurring elsewhere, as to believe 

 that the adherent cell is so influenced. It is sometimes 

 argued that, since the different parts of the body influence 

 one another, there is no reason why they should not influence 

 the germ-cells. This is quite true ; but the fact that A in- 

 fluences B is no proof that C influences D. The various 

 co-ordinations of the body, such as the influence which the 

 nervous system exercises over muscles and glands, or the 

 influence which the primary sexual characters exercise over 

 the secondary sexual characters, are limited in number, have 

 been evolved through a slow and difficult process, and have 

 appeared only because they are highly useful. But acquired 

 characters are limitless in number, they are often injurious, 

 and frequently they are new to the race. In the one case 

 there is a reason why the correlations should occur, but in the 

 other very cogent reasons why they should not. 



42. What evidence is there that acquired characters are 

 acquired by offspring ? Though for many years biologists have 

 ransacked the plant and animal kingdoms, no single indubit- 

 able instance of the transmission of an acquirement has yet 

 been proved. It is not necessary to review at length the 

 immense mass of negative testimony. In the course of this 

 work we shall have to deal incidentally with evidence which it 

 is believed the reader will consider conclusive. Meanwhile, 

 before taking leave of this part of our subject, it will be well 

 to take note of one or two fallacies which are current in 

 popular and medical literature, though happily, in the latter 

 case, to a much less extent than formerly. 



43. Mutilations are still cited as evidence of the transmis- 

 sion of acquirements, though the belief in their inheritance 

 has long been abandoned by all, or nearly all biologists. But 

 mutilations are so very many and the instances of their 

 transmission so very few that all such instances may be 

 dismissed as mere coincidences, as examples of confusion 

 between post hoc and propter hoc. Terriers still retain their 

 tails, after many generations of mutilation. The use of ear- 

 and nose-rings during centuries has left no trace on certain 

 sections of the human race. How often do medical men see 

 the mutilations caused by disease or the surgeon's knife 

 reproduced? The parent acquires a mutilation, and the 

 child in very rare cases happens to vary from the parent in 

 such a way that the acquirement is apparently reproduced ; 

 but only apparently. One swallow does not make a summer. 

 Were mutilations transmissible, then, not in one apparent 

 instance out of a million, but in all or most instances, would 



