1 92 1.] Bird-Migration Inj the Marlinfj Method. 469 



that migration is ii far too complex and also a far too regular 

 phenomenon to be created anew each season merely under 

 stress of circumstances ; moreover, it is known that migration 

 begins before the need is in the least nressino-. The more or 

 less indefinite wanderings of some sea-fowl, the irregular 

 dispersals of some other lairds, and the late " weather move- 

 ments " rhat occur in severe seasons may be attributed 

 to immediate causes, but a deeper seated origin — not 

 necessarily identical for every species — there must surely be, 

 for the highly develoj)ed habit of some of our more typical 

 migrants. 



A little consideration will show how speculation regarding 

 this origin is rendered futile for lack of a certain kind of 

 fact. For instance, there is the perhaps rather far-fetched 

 theory that the migrational habit was established by some 

 great meteorological change in the distant past — say by a 

 Glacial Epoch, as has been suggested, which drove the birds 

 resident in northern latitudes towards the Equator, and made 

 them form there a second homo : to this they would annually 

 return, it is supposed, after the cessation of the unfavourable 

 conditions had allowed them to ro-colonise thoir original more 

 northerly area as a summer home, the individuals continuing 

 to use the routes followed by the species at the time of the 

 first great movement. Then there is the more recent 

 theory [cf. Pycraft, History of Birds, 1910, p. 100) that the 

 migrational habit arose from the gradual northward spread 

 of a species from a supposed original southern area in search 

 of fresh feeding and breeding grounds, the birds withdrawino- 

 to this original area each winter. Without discussing thes(> 

 theories, it may be noted how their proof or disproof would 

 necessarily rest on a knowledge of the facts concernino- the 

 relation of particular summer-quarters to the correspondino- 

 winter-quarters, and of the routes connecting them. Thus it 

 is often suggested that the members of a species summerino- 

 farthest north winter farthest south, and that those mid- 

 way are more or less stationary {cf. Swallow, Section X.) ; 

 but the observer only sees a general southward movement, 

 and typical individuals must be singled out for study before 



