This reasoning, sound enough, helps to a more difficult problem--how the Pacific golden plover, _Charadrius fulvus_, found its way to the Hawaiian Islands, where numbers of the birds winter annually. Roughly the islands are 2000 miles from California, 2400 from Alaska, whence the birds fly, and 3700 miles from Japan. Mr Cooke scouts the idea that any bird flies aimlessly out to sea to find a new winter home, and the chance colonisation by a storm-swept party is as improbable; if this did occur it is hardly likely that they would at once depart, in a single season, from ancient habits and carve out an entirely new migration route. Probably the origin of the route is as follows. The bird breeds on the northern shores of eastern Siberia from the Liakof Islands to Behring Strait, and on the Alaskan side south to the northern base of the Alaska peninsula. It winters on the mainland of south-eastern Asia, in eastern Australia, and throughout the Oceanic Islands from Formosa and the Liu Kiu Islands on the north-west to the Low Archipelago in the south-east. [Illustration: Map showing the evolution of the migration route of the Eastern or Pacific Golden Plover. (From _The National Geographic Magazine._)] It is fairly certain that the original route would be roughly north and south, between Siberia and southern Asia. In time the species spread eastward in winter, to Australia and to islands farther east, whilst the breeding area extended to Alaska. If these extensions took place before any cutting off of corners in the route, Alaska birds would travel 11,000 miles to reach the Low Archipelago, only 5000 miles in a direct air-route (No. 1). Probably shortening began early among the Pacific islands, from the northern islands to the Asiatic coast, and finally to Japan (No. 2). From Palmyra the flight to the nearest of the Marshall Islands is 2000 miles; thence a journey, provided with several possible rests, of 3000 miles would bring them to Japan. A thousand-mile drift through strong winds might cause the birds to reach Hawaii, whence they would find a chain of islands which would help them, and render the last flight to Japan no longer than the one they had been accustomed to. Having once reached the Midway Islands the shortening of the route would be carried on again by lengthening the oversea journey northwards until the Aleutian Islands were discovered (No. 4). The present route, now followed in spring and autumn (No. 5), would be the natural climax of this long evolution. The two golden plovers, sub-specifically distinct, nest little more than a hundred miles apart; their migrations and winter homes are as different as they could be in any two widely divergent species. It is one of the most striking of the ascertained facts in the distribution and habits of birds. * * * * * CHAPTER VII MIGRATION AND WEATHER