22 PROBLEMS OF EVOLUTION 



Natural Selection the variations which it requires as its raw 

 material : the production of variations is their sole raison d'etre. 



" In the diptera (among them our common flies) certain cells, 

 containing unaltered germ-plasm, are from the first set apart. In 

 other cases the separation takes place later in the course of de- 

 velopment, in the hydra even after the formation of a fresh 

 person or individual (neither term is satisfactory) by budding. 

 Nevertheless, it is a part of the original germ-plasm. A 

 sharp distinction must be drawn between the germ-cells and the 

 soma : between the cells whose function is reproductive, and all 

 the others of which the individual is composed. From a fertilised 

 germ-cell a new individual springs: in this individual, germ-plasm 

 is set apart unaltered, and so the race is continued. From the 

 isolation of the germ-cells, it follows that only by a process very 

 difficult to imagine can characteristics acquired by the soma 

 be transferred to them. All variations originate in the germ- 

 plasm. 



"The idea that death is an attribute of life, that it is the 

 nature of living things to die, has led to much error. Conjuga- 

 tion, owing to this assumption,- has been considered to be a 

 vitalising process, and the experiments of M. Maupas on In- 

 fusorians might seem to prove it. 1 They only prove that in 

 certain species reproduction follows a cyclic method, taking 

 place now by fission, now by conjugation, the latter process in- 

 tervening occasionally to produce variations for Natural Selection 

 to work upon. In Cypris, the water flea, parthenogenesis may, 

 as experiment has shown, continue practically for ever. 2 Death 

 is the result of Natural Selection; the species that were cumbered 

 with aged or out-of-date individuals were at a disadvantage and 

 disappeared. Sexual reproduction is not a means of restoring 

 lost vigour but of producing variations. Favourable variations 

 would obviously be of advantage to the species. Those in- 

 dividuals that multiplied by fission only, though potentially 

 immortal, disappeared in the struggle for existence, because the 



1 See p. 10. 



2 Forty parthogenetic generations of Cypris reptans were observed by Weismann. 

 See Euayi, vo.l ii. p. 198. 



