MIGRATION — MEINERTZHAGEN. 341 
ters of migratory birds breeding in northern latitudes coincide with 
the old breeding quarters of the same bird's ancestors in the post- 
Pliocene glacial period. It seems probable that a species with a con- 
fined breeding area and an extensive range in winter had its original 
home in the confined breeding area to which it is most attached, for 
this area is much more exact and local in influencing the bird's life, 
and becomes the focus of its migrations. On the other hand, it may 
be that a species with a wide breeding range and a confined winter 
quarters was originally evolved in its present winter quarters, which 
retains the hereditary attraction due to the love of a bird for its old 
home. In this and in other ways geographical distribution, when 
closely studied, will be found to be most suggestive of a bird's past 
migratory history. 
In this connection it is interesting to note that, though a particular 
form of bird chooses for its winter quarters an infinite variety of cli- 
mate, in most cases the breeding quarters in the breeding season show 
no great variation of climate, though these may cover a vast lati- 
tudinal area. 
The much-debated question of trinomials is outside the scope of 
this paper. The value, however, of subspecies to the student of 
migration is immense, and the more a species can be split into geo- 
graphical forms the easier becomes its migration problem and the 
determination of its correct geographical distribution. Throughout 
the southern part of the Palaearctic region we frequently find more 
than one form of a single species wintering in the same area, and with 
the help of subspecific differences, however small or distasteful to the 
conservative binomial ornithologist, we can at once recognize the 
breeding area of the bird in question and its probable migration route, 
provided we have reliable information regarding its geographical 
distribution. 
Geographical distribution includes, in the case of migratory birds, 
the breeding area, the winter quarters, and the routes of migration 
connecting these areas in spring and autumn. A^ery few species in 
the Palaearctic region can be classified as true residents throughout 
all seasons, though many might appear to fulfill the conditions of a 
resident species until their movements are closely studied. A disre- 
gard of the importance of a species' distribution at all seasons has 
largely discounted the value of many ornithological works and 
papers, for the mere mention of a species occurring at a certain 
locality, without date or further detail, does not really advance our 
knowledge of the geographical distribution of that species, but rather 
confuses it and encourages misleading deductions. 
In writings on the birds collected in a certain area we frequently 
see a great amount of detailed description of the birds collected, their 
12573°— 21 23 
