V0l "i?i? XI1 ] Willett, Birds of Forrester Island. 301 



until 2 a. m. the air swarmed with petrels of both species. There is a 

 considerable difference in their notes while in the air, and the notes of the 

 white-rumped bird were in a preponderance of about three to one. Many 

 of this latter species were not in the air, however, but were in burrows and 

 in crevices in the rocks in pairs, this being the height of their courting 

 season. Their cooing love notes could be heard emanating from the ground 

 during the entire night. 



Oceanodroma leucorhoa kaedingi. K.eding's Petrel. — The 

 white-rumped petrel of Forrester Island is exactly the same as the bird 

 that nests on St. Lazaria Island, Sitka Bay. In previous articles on the 

 birds of that reservation (Bird-Lore, XIV, 1912, pp. 419-426: Condor, 

 XVI, 1914, pp. 71-91), I referred this petrel to the form 0. beali de- 

 scribed by Emerson (Condor, VIII, 1906, p. 54). 



Through the kindness of the authorities of the United States National 

 Museum, I secured for comparison with St. Lazaria and Forrester Island 

 birds a series of nine adult specimens of 0. leucorhoa leucorhoa from the 

 north Atlantic, six specimens from the Aleutian Islands and Bering Sea 

 and four specimens from near Midway Island, Ter. Hawaii. Also through 

 the courtesy of the Oregon State Game Commission, I obtained twelve 

 breeding specimens of 0. leucorhoa kcedingi from Three Arch Rocks, off the 

 Oregon coast. The following conclusions were arrived at by a careful 

 study of the above mentioned material in comparison with series from St. 

 Lazaria Island and Forrester Island. 



Average measurements. Wing. Tail 



Nine specimens, north Atlantic 6 . 24 3 . 45 



Six specimens, Aleutians & Bering Sea 6.22 3.14 



Twenty specimens, Sitka Bay 6 . 05 3 . 04 



Twenty specimens, Forrester Island 6 . 03 3 . 05 



Twelve specimens, Three Arch Rocks 5.98 3.01 



From the above measurements it will be seen that the southeastern 

 Alaska birds are much nearer kcedingi than leucorhoa. The birds from 

 Bering Sea and the Aleutian Islands are nearer leucorhoa but with a tend- 

 ency toward koedingi. There are exceptional specimens from both St. 

 Lazaria and Forrester Islands that measure nearly as large as the average 

 of leucorhoa. For these two latter reasons it would seem that kcedingi must 

 be regarded as only subspecifically distinct from leucorhoa; therefore I 

 have used the trinomial. The measurement of the forking of the tail which 

 has been extensively used by some writers is very variable. The two races 

 O. beali from Sitka Bay, and O. beldingi from Netarts Bay, Oregon, de- 

 scribed by Emerson (1. c, p. 54^ seem to be founded on characters too 

 minute to be worthy of recognition. The birds from Sitka and Forrester 

 Island possibly average slightly lighter on the back and darker on the 

 under parts than specimens from the Oregon coast but in several specimens 

 at hand these differences cannot be detected. 



