196 The Wilson Bulletin — No. 65. 



every hand there were rifled nests, and we found the birds 

 greatly excited, so much so that ahnost no pictures could be se- 

 cured. We indulged in a good deal of speculation as to the 

 probability of a landing by the Indian eggers on Carroll, and its 

 probable results. Fortunately for us they had not dared to 

 make the passage which we were to essay within the next half 

 hour. 



Without going into the details of the two mile row against 

 heavy seas and the equally rough landing om' a southerly spur 

 of Carroll, nor speaking of the tedious carry from the high tide 

 mark to- the top of the island, let us pass at once to the work in 

 hand. 



Seaward Carroll Islet presents a rock precipice some 20U feet 

 in hight. A stone dropped from the top, within two rods of our 

 camp, would fall clear into the ocean below. Landward the is- 

 land slopes at first gently, but finally at an ang)^ of nearly 70° 

 to within thirty feet of the water, ending in another precipice 

 there. It was only along the landward side that ascent was pos" 

 sible, and even there one must clamber up vertically for ten 

 or more feet, finding foothold in the weathered rock. Two sharp 

 rock ridges jut out, one at the north-east corner the other land- 

 ward easterly. The gentler slope of the top is covered with Sitka 

 spruce trees, two of them' old moiiarchs, with a few deciduous 

 trees, growths of elder bushes, a sort of a red raspberry bush, 

 and the ever-present salal bushes. Bordering the woods on the 

 steeper slopes there is a growth of grass clinging to masses of 

 soil which has lodged in the interstices between rock chips. In 

 some places this grass is seen clinging tO' shelves on the face of 

 precipices. Exposed rock faces are pitted and hollowed by the 

 elernients into nesting places for cormorants and gulls. Other 

 rock masses, a good deal worn down, project from the other 

 angles of the island. The waves have worn a hole completely 

 through the island parallel to the landward side and about a 

 hundred feet from it. 



Studies of the breeding birds may very well be made by 

 species separately. A description of the White-crested Cor- 

 morant colonies has already been given. The most abundant 



