170 THE FUR SEALS OF THE PRIBILOF ISLANDS. 
THE REMOVAL OF THE EARLY DEAD. 
In the work of the present season a distinct step in. advance was made in the 
enumeration of the starved pups. One element of confusion in last year’s count 
resulted from the difficulty in distinguishing between the early and the later dead. 
To obviate this the pup carcasses on Kitovi and Lukanin rookeries were carefully 
removed on August 12, before the starvation from pelagic sealing began. Frequent 
counts of the dead, as they accumulated from and after the 12th, were kept up until 
the 10th of September, and the accessions, all plainly due to starvation, were found to 
be fairly constant from day to day. These counts will be found in the daily journal, 
under date of September 6 and 10. 
On October 15 a count by Colonel Murray of the dead on. these two breeding 
grounds gave a total of 1,057. These had died after August 12, and may be taken as 
a very exact measure of the contribution of these two rookeries to the general quota 
of death caused by pelagic sealing. 
THE ESTIMATE OF STARVED PUPS, 1897. 
There were in round numbers about 9,500 pups born on Kitovi and Lukanin 
rookeries in 1897. The number starved was therefore about 11 per cent of the 
birth rate. Applying this percentage to the total birth rate of the islands, the total 
death rate from starvation in 1897 must have been approximately 14,000. In 1896 
the percentage of females taken in Bering Sea was 84. From the greater scarcity of 
males this year and the closer killing practiced on the islands, we are justified in 
assuming that the percentage for this year could not have been below 90 per cent. 
This would give a total of about 15,000 females killed. The difference of 1,000 
will account for the small proportion of virgin two-year olds, and the adult cows in 
the pelagic catch, which had already lost their pups through natural causes. 
THE IMPORTANCE OF THE FIGURES. 
These data regarding the results of starvation are very important, and from them 
we can review our findings of last year. Applying a proportionate relation of starved 
pups to the pelagic catch of 1896, we find that the estimate for last year should have 
been about 24,000 instead of 16,000. 
It is not desired, however, to press either of these calculations too closely. The 
percentage of the two rookeries counted may be slightly too high for the other 
rookeries. It certainly is too high for those rookeries on which the early mortality is 
great. 
THE DESTRUCTIVE EFFECTS OF PELAGIC SEALING ESTABLISHED. 
But the mere matter of the number of pups which starve is not important. The 
essential thing is that a very large number of pups do starve. This is settled beyond 
cavil. As we know the pups are wholly dependent upon their mothers’ milk for 
nourishment until fully a month after pelagic sealing ceases, it follows necessarily 
that the pup dies as a result of the mother’s death, if it has not already died from 
other causes. That more than 16,000 pups, which had otherwise escaped accidents on 
the rookeries in 1896, and about 14,000 in 1897 died of starvation is sufficient proof of 
