AMERICAN WIDGEON 
193 
Chiapam, Lake Atitlan (where it is common). Laguna, Acapam and San Geronimo (Sahun and 
Godman, 1897-1904). It has been traced even farther to Nicaragua at El Boquete on the lake, where 
it has been seen in flocks (Rendahl, 1919). De Armas (1893) includes the species in his list of the 
birds of Colon. 
The Baldpate occasionally visits some of the Pacific islands. Beck (1907) saj^s he saw many on 
Clipperton Island, and it is occasionally found on Oahu, Maui and Laysan in the Hawaiian Islands 
(R. C. L. Perkins, 1903). In recent years it has been taken even in Japan, near Haneda Pacific 
(between Tokio and Yokohama), on December 4, 1908, and January 16, 1918 (Kuroda, Islands 
1920). 
In the Atlantic the species wanders south as far as Trinidad (F. M. Chapman, 1894) and occurs 
occasionally on St. Thomas (Corj% 1889), Porto Rico (Wetmore, 1916) and Jamaica Trinidad 
(P. L. Sclater, 1910). According to Gundlach (1875) it is common in Cuba from West 
September to April or May, but this must have been under exceptional conditions. It Indies 
has been taken on Andros in the Bahamas (Riley, 1905) and is known to have straggled to the 
Bermudas, notably in 1854 (J. M. Jones, 1859; Reid, 1884). In the Azores it has been ^Qj.gg 
taken on San Miguel, and is said to be not uncommon (Hartert and Ogilvie-Grant, 
1905). There are a number of references for Europe, many of which are undoubtedly for escaped 
birds. The only valid ones seem to be three for Great Britain (Brit. Ornith. Union British 
Checklist, 1915) and one for Le Crotoy, France, April 13, 1875 (Marmottan and Isles 
Vian, 1879). Recently the Baldpate has been twice taken in Scotland: in Fife and Stirling (J. A. 
Anderson, 1920; Rintoul and Baxter, 1920). 
Migration 
Throughout the west-central parts of the United States the Baldpate occurs almost exclusively 
on passage. This area includes chiefly Kansas, Nebraska, Missouri, Iowa, Illinois, Wisconsin, 
Indiana, Michigan and Ohio. The migration appears to be primarily from southeast to northwest 
and return. Representative dates of arrival as given by Cooke (1906) are: western New York, March 
23; Erie, Pennsylvania, March 24; Oberlin, Ohio, March 17; southern Michigan, March 25; Keokuk, 
Iowa, March 15; central Nebraska, March 17; Loveland, Colorado, March 10. After reaching this 
latitude the birds proceed more slowly, arriving at Heron Lake, Minnesota, on March 29; in southern 
Manitoba, April 20; at Terry, Montana, April 8; at Indian Head, Saskatchewan, April 24; and at 
Osier, Saskatchewan, May 2. Cooke (1906) makes some interesting observations as to the rate of 
progression. He reckons an average speed of seventeen miles per day from central Nebraska to 
Heron Lake, Minnesota, and of eighteen miles per day from there to southern Manitoba. The 
average rate from Colorado to Montana he figures at sixteen miles per day and thinks this rate is 
maintained northward to Saskatchewan. In his opinion the MacKenzie River breeding birds do not 
come from the Atlantic nor up the Mississippi, but from the Pacific coast, as the known rate of 
advance for the Mississippi River migrants would bring them to the far-northern breeding grounds 
at a date much later than is actually the case. Banded specimens may some day settle questions like 
this. Widgeon reach the mouth of the Yukon early in May, and the earliest record for the lower 
MacKenzie is late April, which must be regarded as verj’ exceptional. 
In the autumn the first birds appear in the northern States in mid-September, and on this southern 
migration the species is more common in the northeastern districts, Ontario, Quebec and the Mari- 
time Provinces, than in spring (J. and J. M. Macoun, 1909) and the same is true in the New Eng- 
land and Middle States. At Squibnockett Pond, Martha’s Vineyard, Massachusetts, in 1919, the 
first arrivals were six or eight, noted on August 31. By September 15 there were thirty there, and on 
the 21st of September about seventy-five. I saw a single specimen on the Kennebec River, Maine, 
on September 5, 1919. This shows that there is a flight probably of young birds very early in the 
