238 FROM COMTE TO BENJAMIN KIDD PART iv 



and postulates a separate speck of matter or living 

 vibration for each quality distinguishable from the 

 rest in human thought and speech. 1 The description 

 applies, among other hypotheses, to Darwin's " pro- 

 visional hypothesis of pangenesis." According to 

 Darwin's view, each part of the adult and vigorous 

 organism gives off extraordinarily minute "gem- 

 mules." These work their way to the parts of sex, 

 and pass on as " packets," one paternal "packet" 

 blending with one maternal "packet" in the embryo, 

 and gradually reconstituting a body, each gemmule 

 helping to build up an organ, or limb, or tissue, like 

 that from which it sprang. Facts, however, insist on 

 a serious qualification, the facts known as atavism. 

 Often, or always in some features, the child resem- 

 bles a grandparent or remote ancestor more than it 

 resembles either parent. How is this to be explained ? 

 Again we are forbidden to fall back on mysticism, or 

 to omit the discovery of a physical and mechanical 

 cause. There must be gemmules from far-away an- 

 cestors developing in each child. It follows that in 

 each embryo some gemmules must fail to develop, 

 but, instead of perishing, must pass on as gemmules, 

 with all their latent qualities ; must enter with other 

 gemmules into new packets constituting ova or sper- 

 matozoa, and must find their chance of development 

 in a later generation by a triumph of atavism. Thus 

 it is only partially true in Darwin's opinion that the 

 parent organism and the reproductive material are in 

 full sympathetic reciprocity. Distinct part of the 

 latter, according to Darwin, though in this genera- 



1 This criticism is urged very tellingly by Mr. George Sandeman in 

 his Problems of Biology. 



