ALASKA INDUSTRIES. 117 



does not live who can stampede rookery bulls. No smoke of vessels or 

 presence of sliips ever cause the stampede of an entire breeding ground. 

 Such things have been reported but no one has ever seen it, and it 

 would require persistent effort to accomplish such a result. I have had 

 cause to send natives on several occasions entirely across a rookery, 

 and no stampede ensued. I have thrown eggshells filled with blue 

 paint at female seals, for the purpose of marking them, until rocks and 

 seals were a mass of blue color, but with no disturbing effect. In the 

 prosecution of my investigations I have shot females with a noiseless 

 rifle upon a small detached breeding ground, have crawled in and 

 dragged out the seals killed without causing the other mothers to recede 

 more than 20 feet, and in fifteen minutes thereafter the breeding grounds 

 presented their wonted appearance. 



After two seasons' observation I unhesitatingly state that I do not 

 believe there has ever been breeding grounds stampeded in such a whole- 

 sale manner as to cause the death of pups. If such occurred in 1891 

 and 1892 it is certainly extraordinary that only the starvelings met death. 



The true explanation of the deaths upon Tolstoi this year is not 

 readily found, and must be sought in local causes other than those indi- 

 cated above, and I am confident that to none of those causes can be justly 

 attributed the dead pups of 1891 and 1892. The following explanation, 

 based upon my acquaintance with the facts, is offered in a tentative way : 



A glance at the map will show that the location and topographic 

 character of this rookery have no counterpart elsewhere on the island. 

 The rookeries upon which deaths are infrequent are those which are 

 narrow and upon the rear of which are precipitous bluffs that prevent 

 the wandering of pups backward. The larger part of Tolstoi, as will 

 be seen from the map, extends far back and has great lateral dimen- 

 sions. Much of it is composed of drifting sands and it has rather a 

 steep inclination down to the sea. ' The shore is an open one, and the 

 surf, either gentle or violent, is almost constantly present. As the time 

 for learning to swim approaches the pups find it easy to come down the 

 incline. They congregate in large numbers upon the sandy shore and 

 begin their swimming lessons. This is at a period when they are still 

 immature and not very strong. The buffeting of the waves exhausts 

 them and coming ashore they either wander off, or struggling a certain 

 distance up the incline, made more difficult of ascent by the loose sand 

 of which it is composed, lie down to rest and sleep, and are overlooked 

 by their mothers returning from the sea. I have seen mother seals go 

 up the entire incline seeking their pups. 



I find nothing in the history of dead pups upon the island this year 

 which does not confirm my belief that the great mortality of the season 

 of 1891 was due to pelagic sealing in Bering Sea. Had it not been so, 

 there is no reason why the deaths in 1892 should not have been as 

 widely distributed as they were the previous year. 



During the past summer particular care was taken to have the drives 

 conducted in the same manner as in previous years, in order that the 

 effect of driving upon the young males might be noted. 



From June 10 (the day after my arrival) to the close of the season, on 

 August 9, there were eleven drives made, the longest one being from 

 Middle Hill, about 2 miles from the village killing ground. With two 

 exceptions, no drives were made from the same hauling grounds except 

 at intervals of two weeks. As the killing this year was limited to 

 7,500, there could be but few seals taken each week, and this necessita- 

 ted turning back to the water, about 200 yards distant from the killing 



