30 PROBLEMS OF EVOLUTION 



Evolution it w jii b e we ll now t o consider the evolution of Weismann's 

 mann's views, for they too have had their stages of development. His 

 views essa y s> as they have appeared in succession, the earliest of the 

 series in 1 88 1, supply us with a very interesting history of the 

 evolution of his theory in an exceedingly critical environment. 

 Every fresh development has come in response to a stimulus 

 from without, some criticism, intended to be fatal, but which 

 has produced merely some readjustment, usually slight, some- 

 times considerable. The theory has, in fact, adapted itself to 

 the environment of facts. 



Weismann saw two very remarkable phenomena for which a 

 theoretical basis must be found. First, heredity, the recognised 

 fact that sons closely resemble their parents ; secondly, the fact, 

 as he himself considered it to be, that acquired characteristics are 

 not inherited : that if the speed of a horse is increased by training, 

 the increase of muscular power, so obtained, is not transmitted. 



These two phenomena he accounted for by his theory: (l) 

 the germ-plasm was continuous ; (2) the cells which contained 

 it were localised and isolated. The first and great article of his 

 faith he has made good ; as to the second, he has had to give 

 way to some extent. He started with the localisation of the 

 germ-cells from the very outset in the body. It turned out 

 that in only one case was this literally true ; that in some cases 

 they were never completely localised. If their isolation is in- 

 complete, there is a possibility of the modifications of the soma 

 being transferred to them, and the question of the transmission 

 to the next generation of the characters thus acquired, though 

 the continuity of the germ-plasm makes this transmission im- 

 probable, must be settled by direct evidence. A variety of cases, 

 where acquired powers seem to reappear in the next generation, 

 must be examined. If no reasonable explanation but that of 

 transmission is possible, then we must assume that it has taken 

 place, even though Weismann has shown how difficult it is to 

 find, between the bodily organ that is modified and the germ- 

 plasm, any channel along which the modifying influence can pass. 

 The direct evidence, for and against, I shall consider in the 

 next chapter. 



