THE EVOLUTION OF MIGRATIONS 103 



Finally, we have assumed that the constant repe- 

 tition of a north and south oscillation has finally 

 established the tendency as an inherited instinct. 

 In the light of present biological knowledge such 

 an assumption is almost wholly unwarranted. It 

 supposes that the Lamarckian hypothesis — that 

 acquired characteristics can be inherited — is accep- 

 table. Experiments of great variety and ingenuity 

 have been devised to put this conception to the test 

 but none has been a convincing success. Yet there 

 is this to be said. Failure to prove a given hypothesis 

 is an entirely different thing from disproving it and 

 if we may not accept the Lamarckain view as es- 

 tablished we are still fully entitled to consider it an 

 open question. It has admittedly never been 

 proved, but neither has it been disproved. The 

 answer is in the lap of the gods; the solution lies in 

 the future. But migration must inevitably be an 

 acquired characteristic even if it is not inherited. 

 The ancestors of birds — reptiles, sluggish and earth- 

 bound — could hardly have been migratory although 

 there are migrants among modern reptiles. Archae- 

 opteryx and Archaeornis vj^re dXmost certainly sede- 

 tary. Ichthyornis must have been strong on the wing 

 but the environment of the Cretaceous probably did 

 not demand migration, a necessity imposed by local 

 circumstances. With a few exceptions recent birds 



