REPRODUCTION 399 



mined by analysis. The great variety realized in such a' 

 series is accounted for by assuming that the molecules 

 are differently constructed; they contain the same atoms, 

 but the grouping is not the same. The more complex the 

 molecules with which we have to reckon, the larger the 

 number of ways in which their constituent atoms can be 

 arranged. 



The molecules which have to do with the transmission of 

 hereditary qualities are presumably of the protein order or 

 nearly allied thereto. The number of ways in which the 

 atoms of a protein molecule can be combined is, to all in- 

 tents and purposes, infinite. It may be that we can sat- 

 isfy all the demands of a theory of heredity without intro- 

 ducing any factor but the spatial ordering of the atomic 

 clusters and chains. Supposing this pattern to be modi- 

 fied at the time of fertilization, we can see that the mathe- 

 matic chances are entirely favorable to a unique product 

 and not to a repetition of anything that has ever existed 

 before. 



