THEORIES OF MIGRATION 23 



may withdraw once more to their ancient home. 

 According to this conception migrant birds retire in 

 autumn to the original home, and there remain for 

 the winter. With the advent of spring the impulse 

 for reproduction drives them irresistibly out to the 

 area that forms the summer home, where they settle 

 to rear their young. With this accomplished they 

 are again actuated by an overpowering impulse to 

 return to the area from which they originally came, 

 and so return in the autumn migration to the winter 

 home. This has been alleged especially to explain 

 early autumn migration before there is any practical 

 necessity for a shift in base. It is also alleged to ex- 

 plain those instances in which birds tend to follow 

 migration routes that do not carry them directly 

 north and south. The theory has in its favor the 

 fact that many birds breed in areas where they 

 cannot winter, a sufficient basis for migration. To 

 suppose that all have originated in the south is, 

 however, going to a considerable length, as many, if 

 not most of them, in their evolution must antedate 

 the Pleistocene, when physical conditions regulating 

 their distribution were far different from those en- 

 countered at present. 



Mr. J. T. Nichols,' has summarized certain mi- 

 gration theories akin to this in consideration of the 

 irregular southward flights of the red-breasted nut- 



I Science^ Aug. 16, 1918, pp. 168-170. 



