THROTTGH CUMTJLATIVE SEGREaATION". 



195 



(5) " Separation and Variation," that is, Variation not over- 

 whelmed bj crossing, " is all that is necessary to secure a di- 

 vergence of types in the descendants of one stock," though 

 external conditions remain the same, and though the Separation 

 is other than geographical. 



(6) The Separation of which I speak is anything, in the 

 species or in the environment, that divides the species into two 

 or more sections that do not freely intercross, whether the dif- 

 ferent sections remain in the original home or enter new and 

 dissimilar environments. 



Though these propositions were very briefly and imperfectly 

 presented, I am not aware that any better statement of the facta 

 of Segregation had been previously published. 



The present paper is the result of a long continued endeavour 

 to understand the relations in which this factor stands to Natural 

 Selection and the other causes that co-operate in producing 

 divergent evolution ; and though my work has been done under 

 the great disadvantage of entire separation from libraries, and 

 from other workers in similar lines, I trust it may contribute 

 something towards the elucidation of the subject. In expanding 

 my theory I have been unable to make any use of the positions 

 taken in Moritz Wagner's paper, as they seem to me very 

 extreme and far removed from the facts of nature. The two 

 theories correspond chiefly in that they discuss the relation of 

 Separation to the transformation of species ; while the explana- 

 tions given of the nature, causes, and eff'ects of Separation widely 

 differ. I am informed that my paper on " Diversity of Evolution 

 under One Set of External Conditions " was translated and circu- 

 lated in G-ermany ; but whether it had any eff'ect in modifying 

 "Wagner's theory, I have not the means of knowing. 



I have recently discovered that the principle of 'Segregate 

 Breeding, which I have found to be of such importance in the 

 evolution of species, is allied to the law of Segregation pro- 

 pounded by Spencer in his ' First Principles.' By direct con- 

 sideration of the conditions that have been found necessary for 

 the development of divergent races of domestic plants and animals, 

 I have discovered Segregate Breeding as a necessary condition for 

 divergent evolution ; and by direct observation on the propaga- 

 tion of plants and animals under natural conditions, I have 

 discovered Cumulative Segregation as a constant result fron- 

 certain forms of activity in the organism when dealing with a 



