HERITAGE OF THE INDIVIDUAL 299 



synapsis. That is, one member of a pair of synaptic mates 

 leaves with the other member certain genes for which it gets 

 none in return: only half of the crossing-over process occurs. 

 Such a phenomenon would probably profoundly modify the 

 constitution of both chromosomes involved and accordingly 

 the organisms to which they contribute. And all such types 

 of mutations must be important raw materials for evolution. 



G. SELECTION 



For more than half a century selection has been something 

 to conjure with a sort of creative principle to explain the 

 progressive changes in plants and animals,. It was assumed 

 that the SELECTION of a certain type of individual for breeding 

 would result in a gradual and continuous transformation of 

 the race or species in the direction of the selection. But 

 Darwin recognized that selection in itself can produce nothing 

 - its efficacy depends on the materials afforded by variation. 

 He did not and, in fact, could not make the modern sharp 

 distinction between modifications, combinations, and muta- 

 tions, but accepted all variations as at the disposal of selec- 

 tion. But recent work indicates that selection of certain 

 types of variations effects only an apparent and not a real 

 change. An example will make this clear. (Fig. 153.) 



Take, say, a quart of beans and sort them into groups ac- 

 cording to the weight of each bean. Then put each group 

 into a separate cylinder and arrange the cylinders in a series 

 according to the weight of the enclosed beans. Now if we 

 imagine a line connecting the tops of the bean piles in each 

 cylinder, it takes the form of a typical curve of probability, or 

 frequency polygon. A similar figure would be obtained by 

 the statistical treatment of nearly all fluctuating characters 

 among the members of any large group of organisms, or of the 

 size of the grains in a handful of sand, or the deviations of 



