14 



south extremity of the bay a long, high, bare, cindery acclivity, which rises towards the 

 "parade ground " or plateau. 



All along the bay from Zoltoi Sands westwards are first scattered harems under the 

 cliff, then more numerous harems on the broad beach below the stony ground, and lastly, 

 on the lower portion of the szreat slope, a more numerous colony, running up here and 

 there in long lines to nearly half the height of the hill. 



Beyond Gnrbotch, near, but to the west of the extreme point of the peninsula, is a 

 rocky beach with an ascending slope, commanded from above by a parapet of rocks. This 

 spot is known as the " Slide," and Dr. Jordan has accepted for it the Aleut name of 

 Ardiguen (photograph No. 62 ; Macoun's photograph No. 26, 25th July, 1892). This 

 spot was kept under close personal observation by Dr. Jordan, whose account of its daily 

 economy will be found on pp. 54-61 of his preliminary Report. 



The east side of the peninsula constitutes Reef Rookery in the stricter sense. It 

 consists of a broad rocky beach, on which a nearly continuous band of harems runs from 

 the point to the isthmus. Towards the middle of the rookery are two shallow land-locked 

 pools of foul water, through which the bachelor seals flounder, or pass between them to 

 and from the extensive hauling-ground behind this portion of the rookery. The east 

 portion of the rookery does not extend so far to the north as the west, stopping short 

 at the isthmus, the eastern side of which is high and precipitous. Near the north end of 

 the isthmus on the east, opposite Zoltoi Sands, and behind the dunes, is a small bay in 

 which the bachelors haul out, and from the cliffs above which a close view of them may 

 be enjoyed unobserved (photograph No. 50). 



The greater part of the peninsula proper is occupied by a smooth plateau, sloping 

 gently to the east (photograph No. 71), known as the parade ground. It is now for the 

 most part grassy, except near the western edge, where the seals ascend the slope of 

 Garbotch to it in small numbers. Two main and three smaller " pinnacles " rise above 

 the parade ground, and command the best views of the Eastern Rookery. Near the 

 southern end of the isthmus the ground is very rough and stony; near the southern end, 

 by the dunes, it consists of loose-blown sand, a short stretch of which is by far the most 

 arduous part of the journey to the seals driven to the village, 



Between the dunes and the parade ground, on the route of the drives, is an old 

 killing-ground, whose use is not recollected. On this ground seals of all sizes appear to 

 have been slaughtered. 



The smooth slope of Garbotch is the part of the rookery where we might expect 

 the photographic evidence to be clearest, and where we might hope to see most easilv 

 changes in the superficial extent of the herd. As a matter of fact, however, it is in just 

 such a place that the seasonal changes in area are so clearly perceptible and so striking 

 that they hopelessly confuse one's estimate of the changes that may have taken place from 

 year to year. 



In Mr. Macoun's photograph No. 30 (20th July) and in my ]STo. GO and 61 

 (24th July), the seals only fringe the lower portion of the slope, except at the far end, 

 where a wedge-shaped mass runs up to about the middle. The appearance is practically 

 identical with that shown in Mr. Townsend's photograph No. 17 (20th July, 1895). But 

 the older photographs, such as Mr. Macoun's No. 74 (15th August, 1892), show the seals 

 spreading over the face of the slope and reaching its summit at both ends. Coming back 

 to the photographs of this year, we see the seals spreading far up the hill in Mr. Aiacoun's 

 photograph No. 65 (5th August) and reaching the top of it and invading the parade 

 ground in JSTos. 10.% 106 (31st August). It is here, in my opinion, certainly true, as has 

 been already said in so many other cases, that the photographs of 1895 show absolutely 

 no superiority in numbers over 1896, but that already quoted of 1892 does appear to show 

 somewhat more than those of the present year. 



For the whole of Reef Rookery Colonel Murray estimated the number of harems at 

 1,000 for 1895 and 900 for 1S96; Dr. Jordan in the latter year placed the number at 

 831. On Ardiguen " or the Slide " Dr. Jordan counted 27 bulls, 550 cows (an average 

 20-4), and 652 pups (an average of 23 to a harem). 



Besides the figures quoted and compared in the preceding account, we possess yet 

 another estimate of the breeding seals for 1895, that of Judge Crowley, Resident Agent of 

 the United states' Treasury on the islands. Mr. Crowley says in his Report (Sen. 

 Doc. 137, Part I, p. 35, 1896) "The breeding herd has been reduced to such 

 proportions that they can now be counted with comparative accuracy. I made the count 

 as follows : 



" St. Paul Island. Breeding cows, 73,696 ; bulls, 4,372. 



"St. George Island. Breeding cows, 21,240; bulls, 1,180." 



