210 Canon Tristram on the Polar Origin of Life. 



plains nor the increasing cold of their northern cradle. But 

 the southern plains were then much hotter than at present. 

 Some therefore daringly pushed towards the extremity of the 

 continent. Those which struggled as far as Juan Fernandez 

 remained isolated, having reached it after the instinct of 

 migration had been weakened by desuetude, and so became 

 distinct and sedentary species, accommodating themselves to 

 the reduction of temperature. Those which kept to the line 

 of the continent and reached Tierra del Fuego, like Euste- 

 phanus galeritus, retreated again as the temperature fell in 

 winter, and became one of the few exceptions to the rule of 

 northward migration for nidification. A similar modification 

 was induced in the sedentary species of the Andes as the 

 climate cooled ; till noAV some may be found near the snow- 

 line, while others are almost confined to a single crater. But 

 while species multiplied by segregation, whether in islands or 

 on mountain-sides, there were those which never lost their 

 hereditary attachment to the north ; and so we find on the 

 one side Trochilus colubris pushing its adventurous journey 

 as far as Labrador, and on the other Sekisphorus rufus up to 

 Nootka Sound. We can scarcely account for the prodigious 

 annual expeditions by these tiniest specimens of bird-kind on 

 any other principle than that of heredity. It cannot be 

 climate nor the search for suitable food which impels them ; 

 for both suitable food and temperature are to be found 

 thousands of miles nearer their winter-quarters. I think, 

 ' too, that the case of these two Humming-birds disposes of 

 the hypothesis of an acquired habit. No such habit could be 

 acquired unless it were for the benefit of the species, and the 

 loss of life from storm and enemies during this expedition of 

 over 3000 miles must be prodigious. The fact of these two 

 species taking such divergent routes on either side of the 

 continent seems to indicate a very early dispersal of the family 

 over the neotropical area. 



The Thrush tribe is, perhaps, the most universally dis- 

 tributed of the Passerine family. Taking Avhat we may call 

 the true Thrushes — i. e. Mr. Seebohm^s genera of Geocichla, 

 Tardus, Merula, Mimocichla, Catharus, and Monticola — we 



