344 IlEV. J. T. GULICK ON 



from a diversity of binary combinations coexisting witli segregated 

 varieties ; but it is evident that the number of divergent arrange- 

 ments that may be produced by any given number of varieties 

 exceedino- two will be much larger if to the above arrangements 

 are added ail that may be produced by arranging single with 

 trinary, and binary with trinary ; and if more complex com- 

 binations are introduced, the number may be still further in- 

 creased. 



Of the five suppositions with which we started *, the second 

 and third assume a uniformity in the contingencies relating to 

 the number and character of the individuals never realized in the 

 different sections of a species that is divided by natural barriers ; 

 and the fifth assumes a uniformity in the changes afi'ecting the 

 environment which, though not often realized, is here assumed 

 for the sake of showing that divergence of character is not de- 

 pendent on the organism being exposed to diiferent environments. 

 In connection with the fourth supposition, it would have been 

 in accordance with the usual conditions of nature to have assumed 

 that, besides the many kinds of food of which only a very small 

 fraction of the species could avail themselves, there would be a 

 few kinds on which much larger numbers could feed ; and that 

 when the numbers that could partake of one kind of food were 

 sulScient to ensure the propagavtiou of those thus adapted, that 

 variety would survive in both districts. But such certainty re- 

 lating to the propagation of some of the varieties would not 

 prevent the contingencies and the divergences that would arise 

 in the propagation of the much rarer or less favoured varieties. 

 It is also evident that similar contingencies would arise whenever 

 the pressure of population on the supply of food should render it 

 necessary for large numbers to seek new resources. The divergent 

 tendency of such pressure, from whatever cause the pressure 

 arises, is in no respect an arbitrary supposition; and the arbitrary 

 assumptions which I have introduced in order to simplify the 

 problem remove from consideration some of the contingencies 

 that must produce still greater divergence. 



Fecukdal iNTEisrsroN, 



or Segregation and Divergence produced hy Independent Fecundal 

 Transformation, that is by difterent relative degrees of fertility 



* See page 338 of this paper. 



