FUR-SEAL: CENSUS, PRIBILOF ISLANDS, 1918, 19 
Some, which in 1917 promised to develop, were abandoned. For 
instance, the lone bull failed to return to Otter Island to try to estab- 
lish a harem, and no cows returned to Suthetunga, otherwise known 
as Zapadni Point. There seemed to be a tendency everywhere 
toward crowding in the larger areas, but the records do not bear this 
out. The phenomenal growth of 154 per cent on South Rookery 
is a feature of interest. Especially is this true in the light of condi- 
tions on Zapadni (St. George) about half a mile distant. For some 
unaccountable reason this rookery fails to develop. 
CHARTS OF BREEDING AREAS. 
Unfortunately it was not possible to test the area method of com- 
putation of seals on the breeding grounds of St. Paul Island again 
this year. Charts were carefully made on St. George, however, by 
Mr. A. C. Reynolds, and his areas have been determined as follows 
by means of a polar planimeter: 
AREAS OF St. GEORGE ROOK#RIES. 
Rookery. er Rookery. eames 
Northteeee 2. -eyne nod..-[e.% 3.02 b.1 102° 000" ||Hiast Reok sats. tecza ees ae- aed 35, 600 
BiarayarAvtllvenn. 2s. sce cet aa zeee ews 05; 200) rbast Clifiscs =. anes sac secure een ramaiae 35, 200 
10) ONCE SR OCB SOne = An ap 2 he an Sa ae 17, 200 
No chart of South Rookery was made, because it has been estab- 
lished since the Coast and Geodetic Survey completed its work. In 
other cases their charts were taken in the field and the plotting care- 
fully done there. 
LOSS OF COWS ON ISLANDS. 
On four rookeries of St. Paul Island 29 dead cows were found 
during the counting of 19,398 pups. This proportion applied to the 
entire herd would give 213 as the total dead. This number is to be 
compared with 39 for 1916 and 129 for 1917. The increase in the 
percentage of loss of cows on the rookeries is larger than the nor- 
mal percentage of increase of this class, due solely to the great sur- 
plus of males which has developed in recent years. If the progeny 
of these lost _cqws is ascertained for 20 or more years aed it will 
be found that the herd is suffering a very perceptible injury. It 
can only be avoided by reducing the number of breeding and fighting 
bulls, which will result in an increase in the average harem. 
In 1918 the agents have kept very accurate records of the loss of 
cows during commercial killing. It has always been known that an 
occasional cow will go on the bachelor’s hauling grounds and may even 
give birth to her pup there. Should she be young, it is almost 
impossible to distinguish her from the 3-year-old males without 
capturing her and making an anatomical examination. This is 
impracticable industrially. Occasionally a cow in the drive will be 
hit with a club unavoidably, even though she be recognized. The 
total number killed in the most extensive work since 1889 was St. 
