110 THE FUR SEALS OF THE PRIBILOF ISLANDS. 
SUMMARY OF PERCENTAGES. 
From these figures we may draw the following summary of percentages: 
Percentages of decline as shown by counts. 
| | | Percent- 
Count of— 1896, | 1897. | age of | 
; decrease. 
| 
699 633 9.5 
10,198 | 7,307 28, 34 
16,241 | 14,318 11.8 | 
Harems --- 
The results in this limited count of harems are not so striking as in the completed 
count of harems for each season on all the rookeries. These were 4,932 in 1896 and 
4,418 in 1897, a decrease of 10.41 per cent. 
DECREASE IN THE AVERAGE SIZE OF HAREMS. 
In connection with this marked decline in the number of breeding families it may 
be noted that on Kitovi rookery, which we have taken as typical of rookery conditions 
in general, there is also a marked decrease in the size of the individual harem. In 
1896 the apparent size of harem, as shown by a count of cows, was 17.3; in 1897 it was 
13.6, a decrease of 21 per cent. As this rookery was counted on exactly the same date 
and under like conditions these figures may properly be compared and are significant. 
THE COUNT OF COWS. 
The count of cows, which shows a decrease of 28.34 per cent, is less certain but 
i8 still significant. Owing to their constant coming and going, the number of females 
on therookeries in the height of the season varies greatly from day to day. This will 
be clearly seen by reference to the daily counts of cows on Lukanin and Kitovi rookeries 
during the season of 1897, which will be found in Appendix I. The count of cows 
and pups, as recorded above in the case of Zapadni Reef and Polovina cliffs, where the 
latter were three times as numerous as the cows, furnishes a good illustration. At 
the same time, while the decrease shown by the comparative counts of cows can not be 
taken at its full value, the fact of large decline thus shown can not be ignored. 
THE COUNT OF PUPS AN ABSOLUTE MEASURE. 
The final and absolute measure of decline, however, is to be found in the counts 
of pups. As we have seen, the number of harems fluctuates. The cows come and go, 
and throughout the breeding season the rookeries are undergoing constant change. 
With the pups this is not the case. They are fixed upon the rookery to which they 
belong at least for the first six weeks of their lives. A count of these animals, living 
and dead, is an exact index to the number of breeding cows which have during the 
season appeared upon the breeding ground in question. 
