and water since the time when birds began to bc migratory is indeed probable, but unless crossing the sea means a distinct advantage it implies the retention of a habit which would not only be useless but might be a positive danger to the species.
In the Gulf of Mexico and the Caribbean Sea there is evidence of perhaps the most recent land-bridge in the chain of islands from Florida to Venezuela, collectively known as the West Indies. Although vast numbers of North American birds winter in South America only a few of the species which annually pass from one continent» to the other make use of this comparatively easy passage. One might naturally conclude that the final severance of England from the Continent was in the neighbourhood of the Straits of Dover, yet this short passage is only used by a comparatively small number of our migrants.
Mr Dixon indeed argued that there is no greater barrier to migration than even a narrow arm of the sea (26). He refers to many Continental species which are common breeders in France but are unknown as nesting species in the British Islands, and others which are found in England but not in Ireland. But this is surely but an incident of distribution; the narrow strait or even river may for a time mark the limit of expansion of a species, just as at the present time the westward and northward unseen barrier prevents the range of the nightingale
