xii BIRDS OF AMERICA 



the Atlantic coast to Florida and continued southward; others went in the opposite direc- 

 tion, traveling westward to southern Alaska and southward along the Pacific coast to 

 Guatemala. Thus birds which were neighbors in summer became separated nearly 3,000 

 miles during migration, to settle finally in close prox'mity for the winter. 



The Connecticut Warbler, choosing another eccentric course, adopts different routes 

 for its southward and northward journeys. All the individuals of this species winter in 

 South America, and so far as known all go and come by the same direct route between Florida 

 and vSouth America across the West Indies; but north of Florida the spring and fall routes 

 diverge. The spring route leads the birds up the Mississippi Valley to their summer home 

 in southern Canada; but fall migration begins with a 1,000-mile trip almost due east to New 

 England, whence the coast is followed southwest to Florida. The Connecticut Warbler is 

 considered rare, but the multitudes that have struck Long Island lighthouses during October 

 storms show that the species is at least more common than would be judged from spring 

 observations, and also show how closely it follows the coast line during fall migration. The 

 breeding of the Connecticut Warbler offers a fruitful field of investigation for some bird 

 lover during a summer vacation, for there undoubtedly is a large and as yet undiscovered 

 breeding area in Ontario north of Lakes Huron and Superior. Incidentally this route of the 

 Connecticut Warbler is a conclusive argument against the theory that migration routes 

 always indicate the original pioneer path by which the birds invaded the region of their 

 present summer homes. 



Another species having an elliptical migration route is the White-winged Scoter. This 

 Duck breeds near fresh water in the interior of Canada and winters entirely on the ocean 

 along the Atlantic and Pacific coasts of the United States. From its summer home west 

 of Hudson Bay individuals that are to winter on the Atlantic travel 1,500 miles almost due 

 east to the coast of the most eastern part of Labrador; thence they cross the Gulf of St. 

 Lawrence and follow the New England coast to their winter home, which extends from 

 southwestern Maine to Chesapeake Bay, with the center of abundance off Long Island 

 and Massachusetts. In spring the birds return to their breeding grounds by an inland route 

 traversing the valleys of the Connecticut, Hudson, and Ottawa rivers. Individuals that 

 winter along the Pacific coast from Washington to southern California are known to pass 

 by thousands up and down the coast as far north as that coast has a generally north and 

 south trend; but as soon as the coast line turns westward near the northwestern part of 

 British Columbia the birds disappear and are not known anywhere in the 500-mile strip 

 between the Pacific coast and the Mackenzie Valley. Apparently this region is crossed at 

 a single flight from the salt water of the coast to the fresh-water summer home on the great 

 lakes of the Mackenzie Valley. 



A migration route entirely different from any thus far mentioned is that of the Western 

 Tanager, or Louisiana Tanager, as it was formerly called. From its winter home in Guate- 

 mala it enters the United States about April 20; another 10 days and the van is in central 

 New Mexico, Arizona, and southern California, marking an approximately east and west 

 line. The next 10 days the easternmost birds advance only to southern Colorado, while 

 the western have reached northern Washington. May 10 finds the line of the van extending 

 in a great curve from Vancouver Island northeast to central Alberta and thence southeast 

 to northern Colorado. It is evident that the Alberta birds have not reached their breeding 

 grounds by way of the eastern slope of the Rocky Mountains, a route which would naturally 

 be taken for granted by anyone examining a map of the winter and summer homes. On 

 the contrary, these Alberta breeders must have come by way of the Pacific coast to southern 

 British Columbia and then crossed over the main range of the Rocky Mountains, which at 

 this season (May 20) are still cold and parth^ covered with snow. 



The shape of North America tends to a converging of the lines of migration toward the 

 Gulf of Mexico, and consequently the east and west breadth of the migration route just 

 south of the United States is usually less than the corresponding breadth of the breeding 



