78 THE FUR SEALS OF THE PKIBILOF ISLANDS. 



Bryant, liudiiij^ the shore exteut and width of the rookeries and allotting a certain 

 space to each individual animal. He, however, worked out the plan in much greater 

 detail. 



IMPdUTANT ASSUMPTIONS. 



In Mr. Elliott's census two important assumptions are made at the outset. The 

 first is that the time when the rookery population has reached its "exact margin of 

 expansion, at the week of its greatest volume, or when the rookeries are as lull as 

 tliey are to be during the season, is between the loth and 2()th of July every year; not 

 a day earlier and not many days later." ' Mr. Elliott assumes as a result of this obser- 

 vation that at the period in question all, or i)ractically all, of the animals were present 

 and would be included in an enumeration made at that time. 



TUE LAW OF DISTRIBUTION. 



He then assumed- "an imperative and instinctive law of distribution, recognized 

 by each and every seal,'' in obedience to which "the breeding grounds occupied by 

 them were invariably covered with seals in exact ratio, greater or less, as the area 

 upon which they rested was larger or smaller;" that the seals "always covered the 

 ground evenly, never crowding in at one place here to scatter out there;" that "on a 

 rod of ground under the face of blutts, which hem it in from the sea, there are just as 

 many seals, no more nor less, as will be found on any other rod of rookery ground 

 throughout the whole list, great or small." 



BOTH ASSUMPTIONS INCOKEECT. 



One who is fivmiliar with the nature of the breeding grounds can not hel^) feeling 

 that in the formulation of this law Mr. Elliott did not have the picture of the 

 rookeries before him. Had he traveled over the length and breadth of the rookeries, 

 as was done in 1890 and 1897, he never would have proposed such a law. That there 

 should be as many seals to the square rod on the jagged and broken lava blocks of 

 Kitovi, or on the broken slopes of Gorbatch, where the animals are now and must have 

 then been separated by bowlders weighing tons, should be the same as on the smooth 

 sand tiat of Tolstoi or the level slope of Ilutchinsou Hill is on the face of it impossible. 



THE TRUE LAW OF DISTRIBUTION. 



The law of distribution which the fur seal obeys is very simple. The gregarious 

 instinct of the animals leads them to crowd together as closely as possible. They are, 

 therefore, even now to be found in as close proximity as the nature of the ground will 

 permit. Where the ground is broken and interspersed with angular bowlders they 

 are necessarily farther ai)art than where the ground is free from obstructions. It is 

 probable that in Mr. Elliott's time the seals, because more numerous, were more evenly 

 distributed, but the nature of the ground would never i>ermit the same distribution 

 everywhere. 



STABILITY OF ROOKERY CONDITIONS ONLY' APPARENT. 



For the first assumption Mr. Elliott has some justification. During the period in 

 (juestion rookery conditions are to the eye of the observer apparently stable and fixed. 



' Elliott, Monograph Fur-Seal Islands, 1881, p. 50. ^' Ibid., p. 49. 



