REPORT OP BRITISH COMMISSIONERS. 103 



0,357,750 for the total iiuiuber of seals. He explains, however, that the 

 space given to each seal by this hypotliesis was too small, and, con- 

 sequently, reduces his estimate by one-fourth, making- it 4,768,300.* 



362. It will be observed that Elliott's mode of computing the space 

 occupied by the breeding seals has been made the basis for sub- 



65 sequent calculations, though both Maynard and Tingle took the 

 liberty of essentially changing the results as they would have 

 appeared if this method had been strictly followed. Neither wholly 

 believed in it, but neither saw his way to substituting a more accurate 

 basis, and both, therefore, merely modified its results by guessing at 

 additions or subtractions. 



363. Elliott's basis of computation must, however, be taken subject to 

 his own measurements of an adult female, which are as follows : Length, 

 50 inches; girth, 36 or 37 inches. Such an animal, in a recumbent 

 position, would be contained in a rectangle of as nearly as possible 4, 

 instead of 2, square feet, and as it is not the normal habit of seals to 

 lie overlapped one upon another, or to stand upright on their hind flip- 

 pers, it is surely clear that his unit of measurement is an erroneous one. 

 This appears to have occurred to the author himself, for, in stating tlie 

 totals of various rookery areas, he writes, cautiously, "making ground 

 for" so many seals, and it is not till he proceeds to make up his grand 

 totals that this statement is suddenly exchanged (though in the same 

 tables) for one representing actual number of seals. 



364. This fact of measurement is not, however, the most i)alpable 

 source of error in these calculations, for the nature of the ground occu- 

 pied by the breeding seals in itself renders them wholly inapplicable. 

 A first inspection of the territory covered by any one of the Pribyloft" 

 rookeries is sufficient to show this, and the fact becomes more and more 

 obvious as they are examined in detail. The notes already given (§ 256 

 et seq.) on the character of the rookery grounds may indicate the reason 

 of this criticism, but it would be difficult to convey an adequate idea of 

 the rocky and broken character of some of them by any description. 

 Photographs may serve to exhibit better their general nature, and it 

 appears to be reasonably within limits of error to conjecture thai:., in the 

 aggregate of the Pribyloff Rookery grounds, not more than one-half the 

 whole space included by their outer limits can, under any circumstances, 

 be assumed to be a surface so level as to be " ground for the resting- 

 place of seals." 



365. It has been considered necessary to deal with this subject because 

 of its direct bearing upon the question of the fluctuation and general 

 diminution of the seals upon the rookeries, and the evidence that it 

 aifords of the now scarcely-questioned fact, that the estimates made in 

 the earlier years of the control of the islands by the United States were 

 absurdly high. It may be added that no single individual of the many 

 questioned by us who had been familiar witli the Pribyloft' or Com- 

 mander Islands, or both, for longer or shorter i^eriods, was found to be 

 ready to maintain even the approximate accuracy of the statements of 

 number of seals according to the above-discussed method of enume^-a- 

 tion. 



366. By way of further substantiating the conclnsions arrived at, how- 

 ever, it may be well to quote a few published opinions bearing on it, 

 which occur in the Congressional Inquiry into the Fur-seal Fisheries of 

 Alaska, made in 1888: t 



* House of Representatives, Report No. 3883, 50th Congress, 2nd Session, pp. 163 

 and ] 77. 

 t House of Representatives, Report No. 3883, 50tli Congress, 2ud Session. 



