248 Mr. H. Saunders on the Geographical 



without a struggle, and goes elsewhere. This has been 

 proved by Dr. Bureau on the north-west coast of France. 

 In 1890 I was surprised to find at Geneva and Lausanne 

 examples which had been obtained on Lake Leman in 

 May ; and I assumed that these were occasional migrants, 

 deflected from a supposed line of migration up the 

 Rhone valley from the Western Mediterranean, where, 

 as already stated, the species was known to occur irregu- 

 larly. No one has yet obtained the Roseate Tern on the 

 coasts of the Iberian Peninsula, the north-west shores of 

 Africa, or in the Canary Islands ; but it occurs in Madeira, 

 as well as in the Azores. Passing westward, we find it in 

 the Bermudas ; the West Indian Islands generally, from the 

 vicinity of Venezuela upwards ; and along the east side of 

 America up to Massachusetts ; not on the Pacific side, even 

 where the continent is narrowest. Returning to the eastern 

 hemisphere, the Roseate Tern has been taken at the Cape of 

 Good Hope and in South-eastern Africa; breeds in the 

 Mascarene Islands, Ceylon, and the Andamans ; can be 

 traced by Tenasserim, Malayasia, and the Moluccas to 

 Australia, and even to New Caledonia — its most eastern 

 breeding-place; while it ranges along the China Seas to the 

 Loo-Choo Islands, wandering to Hitachi, Japan. 



Now it will be seen that there are two very important 

 gaps in its distribution : no authentic specimens being known 

 from West-African waters between Madeira and the Cape of 

 Good Hope, on the one side, or between the Mediterranean and 

 the Indian Seas on the other. But when — as Mr. Whitaker 

 has shown — a colony exists on the coast of Tunisia, it seems 

 not improbable that the line of continuity should be sought 

 eastward, along the coast of Africa, and southward down the 

 Red Sea to the Indian Ocean. It is quite conceivable that 

 the Roseate Tern may not breed on the islands of the Red 

 Sea, because there, as well as at the Laccadive Islands and 

 along the Malabar coast, we find — thrust in like a wedge — 

 ;S^. albigena, an allied species, which may prove inimical to 

 S. dougalli, just as S. fluviatilis is, under certain conditions, 

 further north. But it strikes me now that if a look-out 

 is kept for the Roseate Tern along the Red Sea in April and 



