23 



ivie .. 

 Xapuiliiii' (rrct) 



:. 



Little) 

 I'olii\ inn (Little) 



. 



North-c;i>t 



,. 



(ioi-Kitch . . 



I'olavinn .. 



'..>i 



( 



1-8 



2-7 

 3- 



1 



6'2 



7-7 



9-5 



13 1 



16 9 



The- iirrat dilYercnc.es here tabulated go hand in hand in a way that is clear on the 

 whole, it' not i 't t r.uvable in every single instance, with broad differences in the natuie of 

 the ground. Tin rockv rookeries show the least mortality; the stony beaches come i 

 in order; the latx'c rooUeries of the Iteet'anJ North-cast Point stand midway; Polavina, 

 with its flat, level expanse, stands higher, and Tolstoi and Zapadnie owe their pre-eminence 



>ndy inter-pace^ among the rocks, so fatal to the pups that we came to speak of them 

 as " death-traps." The project of removing these last sources of danger by filling up the 

 sandy hollows with rocks ; n<l stones has been put forward by the American Commissioners. 

 The scheme i- a big one, and I am for myself inclined to think that the labour involved 

 would lie very meai indeeu, and beyond the power of the islanders to accomplish. 



After tiie middle of August my journey to the Commander Islands and other 

 matters occupied my time, and I made no more autopsies. Accordingly, I cannot speak 

 from personal observation of the cause of death indicated by the bodies of the dead pups 

 later in the season. 



The M r ,rtiiliti/ xub.ifi/nciit to August 15. 



The second count of the dead pups was postponed until the end of September, in 

 order that the pelagic fishery might produce its full effect. This count was made for 

 St. I'aul Island by Messrs. Macoun and Clark, with the assistance of Judge Crowley, 

 Colonel Murray, and Mr. Barrett Hamilton, between the 28th September and the 1st 

 October. 



The count on St. George Island \\as made on the 6th October by Mr. Barrett 

 Hamilton and Mr. James Judge on Kast and Little Kast Rookeries, and by Mr. Macoun 

 and Mr. Clark on /apadnie, Staraya Atil, and North Rookeries. 



The following is the complete result of the October count, the number found in August 

 being deducted from those found in October, to give the number that had died in the 

 interval, and an addition of 20 per cent, being made in the case of St. Paul Island to fully 

 cover the po-sible loss by putrefaction and other causes between August and the 

 1st October : the addition is a liberal one : 



