NATURAL SELECTION 387 



order to prevent swamping out through intercrossing with the 

 parent-type. 



5. Objection has frequently been made to Darwin's idea of the 

 purely fortuitous or chance character of variations. According to 

 this view variations occur in all structures and in all directions at 

 haphazard, so that there would be the widest possible opportunity for 

 a given adaptive variation to occur just when the circumstances 

 would demand. It now appears that variations do not occur in all 

 directions in random fashion, but that they tend to follow certain 

 definite paths of change; in other words, variations are, to a consider- 

 able extent at least, orthogenetic. If variations really tend to follow 

 certain definite lines, owing to purely internal causes, natural selection 

 would be unnecessary, at least until orthogenesis went too far for the 

 good of the species, or far enough to be of real importance in the 

 struggle for existence. 



6. The difficulty of explaining how natural selection could make 

 use of the initial stages of adaptive structures is obvious. It is incon- 

 ceivable that the first, almost imperceptible variation in a favorable 

 direction could be of selective value, so as to effect the survival of the 

 individual or the relative number of its offspring. What would be the 

 advantage of the first few hairs of a mammal or the first steps 

 toward feathers in a bird when these creatures were beginning to 

 diverge from their reptilian ancestors? This objection is, of course, 

 based on the fluctuating- variation idea. If the mutation idea were 

 substituted, the difficulty would, to a great extent, clear up; for a 

 mutation might be of sufficient importance in one generation to have 

 selective value from the very first. 



7. Natural selection is said to be incapable of explaining the origin 

 of coadaptive and highly complex adaptations whose effectiveness 

 depends upon the perfection of their adjustments to one another. For 

 example, we may refer to some of the perfected adaptations described 

 in chapter xiv. In the case of the electric organs of certain fish, the 

 Darwinian assumption would be that the first step in the direction of 

 an electric organ would be a very small one, and that it was built up 

 little by little by means of natural selection. But, say the critics, the 

 electric organ would be of no value until it became powerful enough 

 to impart an effective shock to the intruder, and this would not be 

 possible if the character began in a small way. The whole phenome- 

 non of protective resemblance is open to the same type of criticism. 

 As a specific example of this we may cite the case of the dried-leaf 



