40 PROBLEMS OF EVOLUTION 



improbable hypothesis) to confess that he has failed to marshal 

 his determinants, may say, " After all they can marshal them- 

 selves without my drilling them." 



Metaphors are always misleading. When we try to describe 

 the cell, the term architecture is unsatisfactory, because fixity is 

 the character of a building. The architecture of the germ-cell 

 takes a peculiar form only to metamorphose itself into another. 

 The changes suggest nothing so much as crystallisation. But 

 whereas each substance that crystallises water, quartz or what- 

 ever it be forms its proper number of facets and there rests, the 

 nucleus goes through a whole series of changes, even if we take 

 into account only those which can actually be seen. The 

 nucleus is, in fact, a substance which crystallises repeatedly, each 

 form following its predecessor in regular sequence. Equally 

 suggestive of crystallisation is the growth of a complex animal 

 from a single cell ; at each stage the proper pattern is formed as 

 unerringly as water freezes into its hexagons. So long is the 

 series of changes that we might almost say that the organism 

 must have a memory to guide it. But that is to explain a 

 wonderful thing by something still more wonderful. There is 

 in fact nothing in nature by which we can illustrate the ordered 

 series of kaleidoscopic changes through which the organism 

 passes. 



As yet I have been speaking only of the wonderful structures 

 which the single cell unfolds. It might also seem that there 

 arise in it as it grows powers different altogether in kind, not 

 only in degree, from those which its minute self possesses. But 

 there is reason to believe that they are all there in rudimentary 

 form. The amceba, with nothing to help it but the resources of 

 its own little blob of protoplasm, eats, assimilates, breathes, 

 moves, secretes from itself a hard chitinous cyst, has sense of 

 touch and, therefore, potential nerves all our five senses are 

 varieties of one sense and the brain itself is but a very com- 

 plicated combination of nerve ganglia and, as if by volition, 

 selects its food. The Amceba, therefore, is a perfectly 

 generalised cell having all the characteristics of life, so that 

 in the course of evolution we have not to imagine the addition 



