ESTIMATES MADE BY TKUE AND TOWNSEND. 85 



It is not possible for us to snsgest any explanation or justilication for the vasaries 

 wbicli these estimates of Mr. Elliott show, and they need not be further disotissed 

 here. In an ajjpendix to the recent republication,' by order of Coiij^ress, of repoits 

 of agents and otiiers connected with the fur seal islands, they have been considered 

 at length in connection with the subject-matter of the reports of which they are 

 a part. 



THE TRUE AND TOWNSEND ESTIMATE FOR 189.=5. 



The most recent computation of the seals by acreage measurements is tliat made 

 by Messr.s. True and Townsend in 1895. In this a decided improvement was made in 

 securing the space unit occupied by the individual seal. Instead of using an arbitrary 

 estimate, a count of the cows was made on Kitovi and Lagoon rookeries and on [)arts 

 of Lukauin and Tolstoi. The area of the counted districts was then taken from the 

 current maps, and the average space occupied by the individual seals found. For the 

 4,110 cows counted, this average was found to be 4G square feet, ranging from (J5 

 square feet on Lagoon rookery to 29 on Tolstoi. As the spaces counted were all of 

 the scattered or "unmassed" sort, an arbitrary reduction to one half of this space, or 

 23 square feet, was made for the crowded or " massed" breeding grounds. 



Taking these averages and applying them to the acreage extent of the breeding 

 grounds as obtained from tiie current maps of the rookeries, an estimate of the popu- 

 lation of all the rookeries was arrived at. The total number so obtained was about 

 75,000 adult breeding seals. To make it comparable with the former estimates of Mr. 

 Elliott we may add the 70,000 pups, making a total of 145,000 " breeding seals and 

 young." 



THE ESTIMATE MUST BE DOUBLED. 



In this enumeration it was assumed that, at the time the census was made, all, or 

 practically all, the animals were present, including the yearling and 2-year-old females. 

 The effect of this assumption we have already alluded to in coTinection with Mr. 

 Elliott's estimate. The fact is that at no time during the season are more than lialf 

 the cows present. The estimate must therefore be doubled at least to make it 

 represent actual conditions. 



IT ANTICIPATED THE .SEASON. 



But as a inatter of fact, for the estimate of 1895 this will not be sufficient. The 

 counts on which the estimate is based were made before the real maximum of poi)ula- 

 tion on the rookeries was reached. The counting was done between the 8th and 

 loth of 'Tuly, whereas the investigations of 1890 and 1897 show that the maximum of 

 poi)ulation is probably not reached until about tlie 15th of the month. Mr. Townsend 

 himself, in referring to the estimate of 1895, remarks that "the rookeries may not 

 have (as yet) reached their breeding height." 



ARBITRARY REDUCTION FOR MASSED AREAS. 



Another weakness in this estimate lies in the arl)itrary reduction to one-half in 

 obtaining the space for the massed rookeries. Our investigations on this point seem 



' "Seal and Salmon Fisheries, and General Resources of Alaska," vol. 3. 



