LIFE HISTORIES OF NORTH AMERICAN PETRELS AND PELICANS. 139 



colony of birds ; we dug out some 30 burrows, 18 of which contained 

 a single egg each, which were fresh, or nearly so, covered by an 

 incubating bird. We failed to find a mated pair of birds in any 

 of the nests, which is usually the case before the egg is laid, from 

 which we inferred that the courtship period had passed and that all 

 the eggs had been laid. I noticed that the entrances to many of 

 the burrows were partially overgrown with vegetation, indicating 

 that they had been occupied for successive seasons; undoubtedly 

 luch is the case; they are probably repaired, enlarged, or extended 

 a little each year. The burrows varied in length from a little over 

 1 foot to 3 feet or more; the entrance, v,diicli was large enough to 

 admit my hand, sometimes ran straight downward for a few inches, 

 l)ut was more often slanting inward from the side of some little 

 eminence or on sloping ground; the passageway, which was often 

 (luite tortuous, usually ran along horizontally only a few inches be- 

 low the surface, just beneath the roots of the grass. I could easily 

 open the burrows by running my arm in and tearing up the sod. 

 Sometimes the passages intersected or branched in several direc- 

 tions. At the end of the burrow was an enlarged chamber contain- 

 ing the nest, which was loosely made of dry grasses, bits of sticks, 

 and weed stems, mixed with pieces of bark and sods. Sometimes 

 the egg was laid on the bare, soft soil without any attempt at nest 

 building. The nest cavity was so much larger than the nest that 

 the bird, when exposed to view, could crawl away almost out of sight 

 under the overhanging soil. The birds are always very stupid un- 

 der such circumstances; they make no attempt to escape, but try to 

 avoid the light by hiding under any shelter available; if thrown 

 into the air, however, they fly straight out to sea with a wavy and 

 uncertain flight. 



My second visit to this island was made on July 12, 1915, and I was 

 disappointed to find the petrel colony very much reduced in size; 

 there were certainly not over 50 pairs nesting here; this may be 

 accounted for by the fact that some fishermen had been camping on 

 the island and keeping a dog there, which means death to petrels. 

 I had hoped to find some specimens of downy young petrels, but I 

 was too early; there were plenty of heavily incubated eggs, but I 

 did not find any young. I did not, of course, want to dig out all 

 of the burrows and perhaps break up the colony. 



On Seal Island, Nova Scotia, in July, 1914, I saw a large and 

 populous colony scattered over the heavily wooded portion of the 

 island. In a large burnt area their burrows were scattered thickly 

 among the stumps and on the edges of the spruce forest along the 

 shores, where there was plenty of soft soil, the gi'ound was fairly 



