SELECTION 449 



passage from one position of equilibrium to another; the 

 Proteus leaps as well as creeps. An advance marked from 

 the first by a certain measure of perfectness is made at a 

 stride, not by minute steps generation after generation. A 

 copper-beech, a laciniate celandine, a hornless calf, a calculat- 

 ing boy, or the like, just appears out of the inexhaustible 

 conjurer's box. Now it is plain that as the list of these 

 mutations or saltations grows in length, the lighter will be 

 the burden that has to be laid on the shoulders of Natural 

 Selection. Apart from the palseontological record it is only 

 by analogy from the present that we can argue back to what 

 occurred in the distant past, but it looks as if mutations were 

 much more frequent than has been till recently supposed, 

 and the more frequent mutations were in the past, the less 

 work would there be for Natural Selection to do in the way 

 of fostering small increments in a particular direction. 



It is quite premature, however, to think of abandoning 

 the idea so characteristically Darwinian of the cumulative 

 importance of minute advances. Many paleontologists in- 

 sist on the origin of new characters " by excessively fine 

 gradations which appear to be continuous" (Osborn), and 

 also on the frequent occurrence of orthogenesis, i.e., change 

 in a definite direction without marked divagations. As Prof. 

 H. F. Osborn says (1919), the palreontological record often 

 confirms the prophetic judgment of Aristotle : " Nature pro- 

 duces those things which, being continuously moved by a 

 certain principle contained in themselves, arrive at a certain 

 end." 



We must be on our guard, however, against the possible 

 fallacy of concluding, from the apparent orthogenesis in 

 fossilised and surviving stages along an evolutionary line, 

 that there was no zigzagness and pruning in the process. 



