32 Heredity. 



*^ simply uataral" that a bird's Qgg should g'lA-e rise to a 

 bird, and a horse-ovum to a horse, but no one would 

 accept the statement as an explanation. 



We have in the natural selection of variations a true 

 explanation of the manner in which an unicellular rlii- 

 zopod has been slowly and gradually modified by an almost 

 infinite number of slight changes, extending through 

 countless millions of generations, into a bird. The change 

 is one of the most wonderful of the j^henomena of nature, 

 but it is in no sense a mvsterv, for the skill of the 

 breeder may even now, by the employment of the same 

 means, i:)roduce similar results, only on a much smaller 

 scale ; by the methodical selection of congenital varia- 

 tions an organism may be, in a few generations, slightly 

 modified in any desired direction, and we can fairly and 

 truly affirm that we understand the evolution of birds 

 from their unicellular ancestors ; but the resemblance 

 between the evolution of birds from these remote an- 

 cestors by natural selection, and the development of an 

 individual bird from an unicellular ovum, is simply an 

 analogy. It is true that it is an analogy of the greatest 

 significance, but we must not lose sight of the fact 

 that the means by which the end is accomplished — the 

 natural selection, through a long series of generations, 

 of congenital variations — is absent in the second case. 

 If the epigenesis hypothesis is true, if the ^gg is simply, 

 like the rhizopod, an unspecialized cell ; if the ^gg of 

 a bird does not differ from the Qgg of a star-fish in any 

 essential points, we must acknowledge that the mystery 

 of individual development is not only as yet unsolved, 

 but absolutely insoluble. 



The student at the sea-shore may collect at the sur- 

 face, with his dip-net, three similar transparent siDherical 

 eggs. Each of these is, optically, simply a nucleated 



