
calculated for each of these rookeries a square-meters-per-pup value, 
and arranged the photos of the 6 rookeries in order of decreasing pup 
density. (4) They sorted the photos of the remaining 15 rookeries where 
no pup counts had been made and matched them with the photos of counted 
sample rookeries on the basis of topography. For example, Gorbatch was 
matched with Polovina rookery, since each is a smooth, open area lying 
above a narrow, rocky beach line. (5) Finally, they estimated the 
entire pup population on the basis of the square-meters~per—pup value 
for each rookery (table 10). 
In computing ground area from aerial photographs, distortion 
due to parallax as well as differences in land elevation introduced 
errors which could be eliminated in part only. Although the aerial 
photographs were made in 1948 and the sample pup counts in 1949, the 
Similarity of the rookery patterns from one season to the next is such 
that the error from this source is probably less than that introduced 
by some of the other assumptions involved. Secondary importance is 
assigned to population figures derived from the photogrammetric studies. 
The total number of pups born in 1949 as computed by this method was 
found to be, in round numbers, 580,000. 
Rapid field estimate of living pups, 1951.--After a preliminary 
trial in 1950, the results of which were questionable, the experiment 
was repeated in 1951, as follows: 
Living pups by rapid field estimate, St. Paul only . . . 226,700 
Correction factor obtained by carefully counting 
Zapadni Reef as a sample, add 254..... -__50,'700 
Total living pups on St. Paul ... 2 6 «© © 6 © @ © « 6 280,400 
Dead pups counted on St. Paul (including those 
calculated for Sivutch rookery) .. 2... 2 «e+. 72,700 
Correction of 5% for those overlooked .....-e ees 6 
Total pups dead ..... Seay acts, Waites renteuren SLO g300 
Total pups born on St. Paul Telands, calbe: x-Selsi souks ~e « © 359,400 
Total pups born on Pribilofs (utilizing the fact that 
the St. Paul harem bull count equals 82.9 percent 
of all’ Pribilof count) 206. 2. 3 om 2 6 ea See 1 8B 534 
Rounded =to%,« cacnkan-ckahel's Me. want dtet st) A actetee st a HELO S000 
(This is rounded upward a the wonwest cen thousand! 
Experience shows that field counting is likely to 
be an underestimate.) 
Counts of dead pups.--Mortality on land, both natural and artificial, 
is quite easily evaluated, while mortality at sea is largely unknown 
(fig. 11). In the pelagic- sealing days before 1910, when nursing mothers 
were killed at sea, the mortality rate among pups was 7 to & percent 
(table 11). During the years of rapid herd growth, from 1914 to 1925, 
the mortality ranged from 1.7 to 3.1 percent. Now that the herd has 
reached its ceiling, the mortality rate has increased to about 14.6 per- 
cent (average, 1949-51), and the mortality to a total of 60,000 to 80,000 
pups each summer. The mortality rate may be as high as 39 percent on 
individual rookeries (table 12). The majority of deaths are from hookworm 
meolor 
