166 BULLETIN 121, UNITED STATES NATIONAL MUSEUM. 



who supposed it was identical with the stormy petrel {Thalassidroma 

 pelagica), thought it bred in the Bahama and Bermuda Islands, as 

 well as on the coast of Florida and Cuba. Audubon (1840) evi- 

 dently confused it with the Leach petrel, for he says: "Wilson's 

 petrel breeds on some small islands situated off the southern extremity 

 of Nova Scotia." In 1881 Mr. William Brewster (1883) found in 

 specimens he had shot in the (jrulf of St. Lawrence between June 17 

 and July 25 no evidence of breeding. He also secured a 3^oung 

 bird on June 18 which was at least two months old. He surmised, 

 therefore, "that Wilson's petrel breeds in winter or early spring 

 in tropical or subtropical regions and visits the northeastern coast 

 of the United States only in the inteHyn between one hreedhig season 

 and the nexty His conjecture was a logical one and correct to a 

 certain extent, but he did not put the breeding place far enough 

 south. As late as 1884 Baird, Brewer, and Ridgway, in The Water 

 Birds of North America, say of this bird : " Its breeding places have 

 been and to some extent remain in doubt," and they instance rec- 

 ords of its being resident about the Azores and of " eggs pur- 

 porting to belong to this species said to have been taken near 

 Madeira," They quote, however. Dr. J. H. Kidder's belief that 

 these birds nest at Kerguelen Island, in 60° south latitude, and of 

 his report of a nest and eggs of this bird found by Rev. A. E. Eaton 

 on Thumb Mountain of that island. This discovery by the Rev. 

 Mr. Eaton was the first definite knowledge we had of the true nest- 

 ing of the species. The second edition of the American Ornitholo- 

 gist's Union Check List, published in 1895, gives Kerguelen Island 

 as the only breeding place of this bird. Since then the Wilson petrel 

 has also been found breeding on the South Shetland and the South 

 Orkney Islands, as well as on the great Antarctic Continent itself, 

 in South Victoria Land. Capt. Robert F. Scott (1905), of sad but 

 glorious memory, says : " We twice saw it apparently exploring 

 the great ice barrier, in latitude 78°, some 20 miles from the nearest 

 water, where alone it would find its food." The mystery is solved; 

 the Wilson petrel breeds only in the Antarctic regions in the sum- 

 mer months of the antipoles, namely December, January, and Feb- 

 ruary, and migrates north during the antipodal winter and spends 

 it in the northern summer. Its life is therefore one long summer, 

 albeit a stormy and cold one. 



Nesting. — Like most of the petrels this species prefers to nest 

 in colonies. Tliere is considerable variation in the nesting site and 

 nest, dependent undoubtedly on the character of the country and 

 the material at hand. Thus Wm. Eagle Clarke (1906) speaking of 

 the South Orkneys, says : 



There was no attempt at nest making, the egg was simply laid in a hollow 

 in the earth in narrow clefts and fissures in the face of the cliffs, under boulders, 



