THE KAW MATERIALS OF EVOLUTION 117 



biennis and Penstemon wrightii, and we have 

 decisive evidence that environmental stimulus 

 acting directly on the germ-cells may induce 

 striking variations. This is a very important 

 result, for it is evident that the germ-cells of 

 animals may in a similar way be naturally stimu- 

 lated to vary by chemical changes in the vascular 

 fluids. With the flowers, as with the beetles, there 

 were not only losses and augmentations of what 

 was previously present, but there were distinct 

 novelties which maintained their distinctness when 

 crossed with the parental strains. 



After we have worked for years along the lines 

 of these various suggestions that have been offered 

 as to the causes of variations, we shall be better 

 able to say how far they account for what we believe 

 has occurred in the past, and for what we know 

 to occur at present. Perhaps they will prove 

 insufficient, and evolutionists will be forced to 

 recognise that variability is, like growth, a primary 

 quality of living things, and that " breeding 

 true " has arisen secondarily as a restriction. 

 The relation of genetic continuity between succes- 

 sive generations is an economical arrangement 

 which secures relative constancy amid continual 

 flux, but in spite of this the Proteus continually 

 asserts itself. Its essence is creative power. It 

 lives because it changes, it changes because it 

 lives. From generation to generation there is a 

 continuous lineage of germ-cells; but just as we 

 see a youth growing and changing, taking time 

 into himself and making himself, in some ways, 

 new by his experience, so it may be that there is a 

 power of growing and varying inherent in the germ- 

 cells as also in the unicellular organisms in the 



