82 THE MIGRATIOX OF BRITISH BIRDS 



furthest south, we can readily understand \vh}' eastern 

 individuals of the same species visit the Cape in winter, 

 whilst western individuals go no further south than 

 North-west Africa and the Canary Islands — their base 

 of northern extension. 



The species marked B in the above table are birds 

 that go much further to the south in East Africa than 

 in West Africa ; not necessarily because the}- breed any 

 further north in East Europe or West Asia than in West 

 Europe, but because the passage up or down the Nile 

 valley has always been a more continuous one, whilst in 

 West Africa the Sahara Sea or its now sand>- wastes 

 were an insurmountable barrier to emigration. The 

 Terns, being thoroughly oceanic species, do not come 

 within the scope of these remarks. The evidence also 

 seems to suggest that some of these migrants, after 

 going down the Nile valley, spread westwards across 

 the Soudan, even to the Atlantic sea-board, as if they 

 were following the southern coast-lines of the ancient 

 sea ! 



There is another small group of Summer Migrants to 

 the British Islands which appears to me to have unques- 

 tionably dwelt in Refuge Area III. during the Glacial 

 Epoch. The winter quarters of these birds may be 

 either in the Mediterranean basin or in Tropical Africa. 

 They appear to reach those districts by way of Iberia, 

 thence eastwards through Algeria and Tripoli, and 

 southwards down the Nile valley. This class of sum- 

 mer migrants illustrates very vividly the intricate paths 

 followed by birds on passage, and demonstrates how in 

 the remote past the colonists gradually extended their 

 range, first eastwards and westwards across the Soudan, 



