LIFE HISTORIES OF NORTH AMERICAN PETRELS AND PELICANS. 83 



till there is but little room left within the circle that they almost invariably 

 form. The first gentle breeze will start them on their journey again, and l 

 have learned to have confidence in a breeze that starts them flying, for as far 

 as my observations go, they only rise if the wind is to be continuous, and will 

 pay no attention to a gentle puff that will die out in a few minutes. 



None of our I'acific coast sea birds adhere so closel.v to established fly lines 

 as do the three species of Pufflmis ; even when flying fifty miles or more from 

 land the first flock that passes will, with almost absolute certainty, mark the 

 line which the next Avill follow, even though they be an hour behind. And I 

 have long since discovered that in order to secure specimens of these shy 

 species the boat must be placed in their fly lines. A flock will, on encountering 

 a skiff directly in their path, either divide and pass on either side or all 

 swerve slightly to one side, immediately resuming their line of travel in either 

 case. At times, however, they are easily turned from their course. 



The fli<T:Iit of the l)hick- vented shearwater is more erratic, Avith more 

 constant flapping:, than that of other species. Mr. A, B. Howell 

 writes me that he has seen them "" progress by a series of ellipses 

 when feeding, one side of which is within a foot of the water and 

 the other eight or ten feet above." 



Mr. Anthony (18966) describes tlie notes of this species, as he 

 heard them about Guadalupe Island, as follows: 



It would be impossible to describe accurately these notes. They were a 

 series of gasping, wheezy cries, resembling somewhat the escape of steam 

 through a partly clogged pipe, uttered in a slightly varied key and repeated 

 from four or five to ten times. During calm weather in January, February, and 

 March flocks of a dozen to several hundred of these .shearwaters often collect 

 on the water well offshore, and at such times I have heard the same notes from 

 two or more birds as they chased each other, half running, half flying, over the 

 water. From the notes that came from the cliffs, I thought that the birds were 

 chasing one another, and a little later many of them came down to the water 

 and were occasionally seen as they flashed by within the circle cast by our 

 anchor light. After an hour or so the outcry somewhat subsided, and I think 

 n)ost of the birds went offshore to feed, returning before daylight, for during 

 nearly two Aveeks spent in cruising about the island only one flock of shear- 

 waters was seen in the daytime. 



Winter. — At the close of the breeding season these shearAvaters 

 Avander from their breeding grounds northAvard, and perhaps south- 

 Avard as Avell. They are frequently seen along the California coast in 

 the fall and Avinter, as far north as the Columbia River. These wan- 

 derings could, perhaps, hardly be considered a migration, as they are 

 in the opposite direction to that usually taken by birds that breed in 

 the Northern Hemisphere; but many of them certainly spend their 

 season of leisure far north of their breeding grounds, to which they 

 return during the latter part of Avinter or early spring. 



DISTRIBirnON. 



Breeding range. — Islands off the west coast of LoAver California, 

 Mexico (San Benito, Natividad, and Guadalupe Islands). 



