246 SIBERIA IN EUROPE. CHAP. xx. 



origin of migration, some ornithologists holding that the 

 original home of a species was its winter quarters ; others 

 supposing that before the instinct of migration was formed 

 the species was a resident in the district where it now breeds. 

 Both views have their difficulties ; but the preponderance of 

 evidence seems to me to be largely on the side of the latter 

 theory. In turning over a box of Transvaal skins, shot 

 during the breeding season between September and March, 

 it seems impossible to come to any other conclusion. 

 Throwing aside the brilliant birds of the district, we shall 

 find, especially if the box comes from Potchefstrom, a 

 variety of reed - warblers and allied birds, which speak of 

 swamps abounding with insects, where birds of this kind 

 delight to breed and find unlimited food. As a matter of 

 fact most of these birds do breed there, and, because the 

 winter is so mild, remain there all the year round. But 

 mixed up with these African types we shall find a fair 

 sprinkling of our own reed-warblers, who have gone down 

 there to avoid our cold winters. These birds are not breed- 

 ing ; they have migrated to the Transvaal to enjoy the 

 mosquitoes of the Potchefstrom swamps, and when the 

 Potchefstrom birds have finished breeding and begin their 

 six months' rest from family cares, they will some of them 

 migrate to the Arctic regions of North Europe to breed 

 amongst the mosquitoes which swarm on the river -banks 

 on the outskirts of the tundra. We can scarcely conceive 

 it possible that these species were ever resident birds in the 

 Transvaal. It seems much more rational to conclude that 

 they were once resident birds in the subarctic regions of 



