186 NYROCA COLLARIS 



most record of its occurrence is for the Coban River, Vera Paz, Guatemala, where considerable 

 numbers were seen in November (Salvin and Sclater, 1860). 



The Ring-neck has been known to occur only once on Bermuda, November 13, 1850 (J. M. Jones, 

 1859). It was originally described by Donovan (British Birds, vol. 6, pi. 147, 1809) from a specimen 

 found in Leadenhall Market and said to have been shot in Lincolnshire, England. 



Migration 



The main migration route of this species undoubtedly lies in the Mississippi Valley. In the States of 

 this basin it is a fairly common migrant, passing over in March and again in September. In the west- 

 central States, however, it is so rare as a migrant that in many it can hardly be considered as more 

 than accidental. The small numbers of birds which winter on the Pacific seaboard are perhaps those 

 that breed in British Columbia. A few must reach the coast of southeastern Alaska for A. M. Bailey 

 (MS.) saw about twenty-five in one flock at Kootznahoo Inlet in late October, 1920. The migration 

 by way of the Great Lakes to the Atlantic coast is a small affair and it is later, for most of our New 

 England records are in October and November. Two spring specimens were taken by Comeau on 

 the north shore of the St. Lawrence besides a late September one (Comeau, 1909) and this seems to be 

 as far east as they ever straggle on their flights. On the Carolina coast they hardly appear before late 

 October, and in the spring are not uncommon on the Atlantic coast as late as April. There is no evi- 

 dence, so far as I know, of any circular migration although they are commoner in autumn than in 

 spring. Those that come to the Atlantic coast seem to return in spring over the same route by which 

 they came, while those wintering on the Gulf Coast also appear to return by the Mississippi Valley. 



Of the two specimens banded at Lake Scugog, Ontario, on October 29, 1920, one was shot at Back 

 Bay, Virginia, on November 23, 1920, and the other near Georgetown, South Carolina, on January 

 5, 1921 (U.S. Biological Survey). 



There are eleven returns of birds banded at Belle Isle Lake in southern Louisiana during February, 

 1922 (F. C. Lincoln, 1924), and these show the trend of Gulf Coast birds up the Mississippi Valley 

 into Canada. They were taken at the following stations: eastern part of Galveston County, Texas; 

 Gueydan, Louisiana; near Asbury, Missouri; Solomon River, Kansas; Franklin, Nebraska; Quincy, 

 Illinois; Vilas County, Wisconsin; Meeker County and at Benson, Minnesota; Shellmouth in Mani- 

 toba; and two places in Saskatchewan: Isle a la Crosse Lake (near it) and The Barrier (40 miles west 

 of The Pas, Manitoba). One of the Saskatchewan records is probably a breeding one but most of the 

 others are for the autumn of the same year they were marked. 



GENERAL HABITS 



It is a curious fact that this duck was not described until 1809, and then by Dono- 

 van, an Englishman, who procured a specimen in the Leadenhall Market, London, 

 supposed to have been taken in the Lincolnshire fens. It was recognized, described 

 and figured by Alexander Wilson in 1814 but he confused it with the Tufted Duck 

 of Europe, a species which he evidently had never seen. Prince Bonaparte (1824) 

 pointed out Wilson's error and gave a very accurate description of both sexes, some 

 account of its habits, and notes on its trachea. Audubon gives a full description and 

 some interesting notes on habits. 



This duck has usually been wrongly assigned to a position next the Scaups, but it 

 probably has little affinity with them. It is far more closely related to the true 

 Pochards and stands between them and the Tufted Duck as Hollister (1919a) has 

 pointed out. This is apparent not only from the plumage, both adult, immatures 



