ALASKA FISHERY AND FUR-SEAL INDUSTRIES 179 
be 11 and 10 years old, respectively, this year, and should have been 
in their prime. 
It should also be noted that from 1930 to 1934 there was an average 
harem ranging from 38.39 to 41.95, the lowest average harems known 
for any period since special additional reserves of seals were made in 
1923. In spite of these low average harems there was no marked 
increase in the number of 3-year-olds available for killing in succeed- 
ing years. Theoretically, when the average harems were smallest, as 
in 1931 and 1932, a greater percentage of females should have been 
bred than when there was a larger average harem. However, when 
the progeny of the low-average harems became of killable age, no 
proportionately greater increase was noticeable in the number of 
3-year-olds available. There has been a gradual increase in the aver- 
age harem from 38.39 in 1933 to 46.69 in 1938 and 60.44 in 1939. 
Nevertheless, it is still believed that there should be a considerable 
surplus of breeding males and that under normal conditions the larger 
the surplus of breeding males the faster the herd will increase. 
In 1938 there was a conspicuous shortage of idle bulls on the rook- 
eries and hauling grounds. In 1939 this condition was reversed. ‘There 
was an abundance of large, vigorous bulls on the hauling grounds. 
Why they remained there throughout the season when the harem 
areas contained considerably less than the usual number of bulls is 
not known. 
Because of this sharp reduction in the number of harem bulls, it 
appeared advisable to start creating a much larger reserve of 3-year- 
old males. Commercial sealing, therefore, was discontinued on July 
26. This provision for an additional reserve shouid be continued for 
a number of years and, if conditions warrant, it would be desirable 
to make an even larger reserve than was made this year. 
The desirability of having the harem and idle bull count made by 
the same persons in the same way each year was clearly shown this 
season. Counts of the larger rookeries cannot be made accurately, 
but if the same enumerators use the same methods each season the 
possibility of any wide variation in estimates will be reduced to a 
minimum. With plenty of idle bulls on the hauling grounds this 
season, it would normally have been expected that killings could 
safely be continued until the end of July, but the sharp increase in 
the size of the average harem appeared as a warning to leave a larger 
male reserve. 
Too close killing for a year or two should not result in any con- 
siderable harm, as mature males should be capable of doing harem 
duty from 6 to 8 years. 
No harem count was made on Sivutch rookery, but on account of 
the reduced number of harems on most of the other rookeries a reduc- 
tion was applied also to this rookery. 
Owing to lack of suitable space, some of the rookeries are incapable 
of continuous enlargement. As the 8 percent increase of cows must 
still be applied in order to determine the total increase of the herd, 
the resulting average harem for these rookeries is shown as consider- 
ably larger than it should be. Only the average harem for the entire 
herd, therefore, should receive consideration. 
Except when the Polovina count was made, unusually good weather 
for counting prevailed during the entire period in which harem counts 
and estimates were made. Presumably it is generally understood that 
