334 THE EVOLUTION THEORY 



to the continued selection of the fittest. We require no further principle 

 of explanation for the establishment of a specific type. 



This 'type' is thus not reached by any indefinite varying of 

 the parent form in all directions, but in general it is reached by the 

 most direct and shortest way. The parent form must indeed have 

 become to some extent fluctuating, since not only the variational 

 tendencies ' aiming at the goal,' but others as well, must have emerged 

 in the germ-plasm ; but gradually these others would occur less and less 

 frequently, being alwaj^s weeded out afresh hy selection, until the 

 great majority of the individuals would follow the same path of 

 evolution, under the guidance of germinal selection, which continues 

 to work in the direction that has once been taken. After a short 

 period of variation, which need not, of course, involve the whole 

 organism, but may refer only to certain parts of it, a steady direct pro- 

 gress in the direction of the 'goal,' that is, of perfect adaptation, will set 

 in, as we have seen in the case of the Planorhis snails of Steinheim. 



We must not forget, however, that natural selection works 

 essentially upon a basis of sexual reproduction, which with its reduc- 

 tion of the ids and its continually repeated mingling of germ-plasms, 

 combines the existing variational tendencies, and thus diffuses them 

 more and more uniformly among the individuals of a whole area 

 of occupation. Sexual reproduction, continual intermingling of the 

 individuals selected for breeding, is thus a very effective and important, 

 if not an indispensable, factor in the evolution of the specific type. 



But it is not only in the case of species transformations due 

 to new adaptations that sexual mingling operates; it does so also 

 in the case of variations ' due to purely intra-germinal causes. We 

 have already seen in discussing Isolation that isolated colonies may 

 come to have a peculiar character somewhat different from that of the 

 parent form, because they were dominated by some germinal variational 

 tendency which occurred only rarely in the home of the parent form, 

 and therefore never found expression there. On the isolated area 

 this would indeed be mingled with the rest of tlie existing germinal 

 variational tendencies, but the result of this minoflinp; would be 

 different, and the further development of the tendency in question 

 would probably not be suppressed. 



We need not wonder, therefore, that specific types occur in such 

 varying degrees of definiteness. If a species is distributed over 

 a wide connected territory, sporadically, not uniformly, it will depend 

 partly upon the mutual degree of isolation of the sporadic areas 

 whether the individual colonies will exhibit the same specific type 

 or will diverge from one another. If the animal in question is a slow- 



