21G Canon Tristram on the Polar Origin of Life. 



the origin of which I propose to explain by the heredity of 

 attachment to their place of origin. It is trne that there 

 are many more species of Swallows which breed in the southern 

 than in the northern regions. But these are simply^ like the 

 Ethiopian and Neotropical Thrushes, races which have become 

 sedentary, and thus been specialized into local species. In 

 the countries in which these local species are found, the indi- 

 viduals of the migratory sorts, spending our winter in South 

 Africa, exceed ten-fold or fifty-fold the number of all the 

 sedentary species. Why should this be, but because the 

 continued habit of reversion to the north, and the intercourse 

 with their fellows from other southern regions whom they 

 there meet, have secured permanence of type and checked 

 the tendency to variation by segregation which marks the 

 permanent dwellers in Southern Africa ? If the Hirundinidse 

 had had a southern origin, how could the habit of northward 

 migration have originated ? Not certainly by imitation, 

 for their flight is more rapid than that of most other 

 travellers. Not from absence of food in the south, for where 

 the sedentary species can find support so could the migratory. 

 Besides, as we know, in the case of Hirundo savignii and 

 Cotile rupestris, there are sedentary species in the north as 

 well as in the south. 



If the Swallow tribe had a southern origin, there has been 

 no theory yet advanced on migration which could possibly be 

 reconciled with the facts of its life-history. Genera are but 

 arbitrary divisions, and the fact that of the eleven genera of 

 Hirundinidse, most of which are very unsubstantial and 

 shadowy abstractions, only one is cosmopolitan, and that now 

 confined to the Palsearctie Region, rather goes to illustrate 

 the enormous powers of flight of the family, and the ease 

 with which so great a change as that from a roving to a 

 sedentary life has modified the specific characters of the 

 group. The one fact that none of the migratory Hirun- 

 dinidse breed in the southern hemisphere, though visiting it 

 at the times of the nidification of the local species, seems to 

 be of itself proof enough of a northern origin. 



