152 THE DOCTRINE OF DESCENT. 



where, has opposed to the Divergence of Character, of 

 the inchnation of varieties and species to deviate from 

 one another, a Convergence of Character. It is con- 

 ceivable, he thinks, that species derived from different 

 genera, might sometimes approach each other so 

 closely that they would be classed in the same" 

 genus. The author of the theory of Selection has been 

 content to point out the great improbability of such an 

 event, which, moreover, in this simplicity would scarcely 

 impugn the origin and truth of the theoiy. "If two 

 species of two allied genera both produced a number of 

 new and divergent species, I can believe that they m.ight 

 sometimes approach each other so closely that they 

 would, for convenience' sake, be classed in the same new 

 genus, and thus two genera would converge into one ; 

 but from the strength of the principle of inheritance, 

 and from the two parent species already differing, and 

 consequently tending to vary in a somewhat different 

 manner, it seems hardly credible that the two new 

 groups would not at least form different sections in the 

 same genus."'** 



We here see a theoretical objection theoretically 

 refuted. But although the probability of a convergence 

 carried to absolute similarity is extremely slight, and 

 it receives no support from the palaeontological record, its 

 utter a priori impossibility must not be rashly asserted ; 

 and in my researches on the Sponges of the Atlantic, I 

 have pointed out groups of species approximating so 

 closely as to be scarcely distinguishable. Chalina and 

 Reniera are two distinct genera, actually belonging to 

 different families. It is highly probable that the genus 

 Chalinula, with its extremely variable species, are 



