140 



THE ASIATIC FUR-SEAL ISLANDS. 



Upon my arrival, in 1895, 1 impressed upon Mr. Grebnitski the desirability of 

 having such a census prepared, and suggested that Selivanof, the kossak in charge 

 of the rookery, be ordered to undertake the work. Mr. Grebnitski, fully aware of the 

 great importance of knowiag exactly what classes were represented in each drive, at 

 once took up the suggestion and ordered SeJivanof to make a detailed tally of each 

 drive according to the scheme I furnished. The drive on July 19 I counted myself 

 conjointly with Selivanof, and the tally sheet is here produced to show how the work was 

 done and how much reliability can be placed upon it. The seals killed and those escap- 

 ing from each pod, as it was culled and slaughtered, were separately counted, Feoktist 

 Ivanof Korsakovski counting the dead ones, Selivanof and I those allowed to escape. 



Tally of drive taken, July 19, 1895, north rookery, Bering Island. 



The accuracy of the above tally is attested by the fact that the number of skins 

 taken in this drive was 545. Sometimes the killed ones of the previous pod were 

 lying so close to those being counted that it was difQcult to ascertain the exact number, 

 in which case the smaller figure was noted. And so with the escaping ones. Selivanof 

 and I counted separately; if we differed, and a recount was not practicable, we took 

 the lowest figure. The percentages are, therefore, very nearly correct. If there is 

 any error, it is in understating the number of females, but I am sure that the possible 

 error does not exceed 1 per cent. 



The figures of the 13 drives in the table previously given were ascertained in the 

 same manner, and I have no doubt that they are essentially correct. No tally was 

 kept previous to the drive on July 14, and I failed to obtain the details of the drive 

 on July 24, but there is no reason to believe that the percentage of the classes was 

 different in these drives, except that I was informed that there were no females or pups 

 in the first drive, June 13. In order to complete the record of this rookery for 1895, 



