The Migration of North American Sparrows 21 



Pa., March 31, 1893; Princeton, N. J., February 6, 1872; New Haven, Conn., 

 December 17, 1878; Providence, R. I., March 14, 1896; Boston, Mass., Decem- 

 ber 26, 1906, April 10, 1907, and November i, 1910; abundant at Revere Beach 

 and Nantasket Beach, February 19-22, 1883; Westbrook, Me., January 26- 

 February 27, 1896, and December 12, 1903; Houghton, Mich., November 20, 

 1904; near Iowa City, la., January 18-25, 1896; Meridian, Wis., January 9, 

 March 26, 1896, and Magnolia, Colo., December 9, 1895. This last individual 

 had traveled twice as many degrees to the west as to the south. 



HOARY REDPOLL 



The Hoary Redpoll breeds in the high Arctic regions of North America, and 

 comes south in the winter as far as the northern United States. The beginning 

 of the fall migration was noted September 19, 1903, when flocks appeared at 

 Fort Franklin, Mackenzie. Some fall or early winter records in the United 

 States are: Cambridge, Mass., November 15, 1880; Swampscot, Mass., 

 November 16, 1878; New Haven, Conn., November 24, 1906; Meridian, Wis., 

 December 13, 1895; Sault Ste Marie, Mich., December 7, 1899, and Fairbault, 

 Minn., December 15, 1883. It was noted in southern Ontario at Guelph 

 December 8 and 26, 1903, and was fairly common at Milton the winter of 

 1882-3. 



It has remained at Cambridge Mass., in the spring until March 20, 1888; 

 Hamilton, Ont., April 6, 1885; Meridian, Wis., March 26, 1896; Miles City, 

 Mont., March 12, 1900; Winnipeg, Monitoba, April 3, 1900; Indian Head, 

 Saskatchewan, April 17, 1892, and Fort Simpson, Mackenzie, April 30, i860 

 and May 10, 1904. 



Another subspecies of this bird — the Greenland Redpoll — has only one 

 record in the United States, that of a single bird taken March 29, 1900. at 

 Sault Ste. Marie, Mich. 



PURPLE FINCH 



The breeding range of the Purple Finch includes southern Canada east of 

 Alberta, and the neighboring portions of the United States south to Minnesota, 

 Michigan, Pennsylvania (mountains), and Long Island. The great bulk of the 

 individuals winter south of the breeding range, but a small percentage remain 

 at this season, farther north in the southern part of the breeding range, and 

 sometimes even to the middle part. There is therefore a broad belt, covering at 

 least a third of the entire range of the species, in which migration dates are 

 unsatisfactory, because the records of real spring migration are so mixed with 

 notes on birds that have wintered. The case is made more involved by the 

 fact that the Purple Finch is normally a late migrant, so that there are, in 

 reality, two sets of notes, one of birds that have wintered unnoticed in the deep 

 woods and are recorded when they spread to the open country during the first 



