ALASKA FISHERY AND FUR-SEAL INDUSTRIES, 1920, £15 
these classes in 1920 than there were in 1917. Actual numbers, how- 
ever, mean nothing in this connection. The important features are 
the ratio of idle bulls to harem bulls and the average harem. Condi- 
tions for the series of years 1912 to 1920 are best shown in tabular 
form as follows: 
VARIATIONS IN ADULT BULL CLASSES AND RATIO OF IpLE TO HAREM Butts, 1912 
To 1920. 
St. Paul Island. St. George Island. Both Islands. 
Year. Aver- Aver- Aver- 
Harem.} Idle. | Ratio.| age j|Harem.!| Idle.| Ratio.| age /Harem.| Idle. | Ratio. age 
harem harem. harem 
Per ct. Per ct Per ct. 
1912..| 1,077 93 8.63 | 65.0 281 20 7.11 42.5 1,358 113 8.32 60. 4 
1913..| 1,142 77 6.74 | 69.6 261 28 10.72 | 49.1 1, 403 105 7.48 65.8 
1914..| 1,316 159 12.0 60.3 243 13 5.3 57.1 1,559 172 11.0 59. 8 
1915..| 1,789 546 30. 52 49, 27 362 127 35. O8 42.51 2,151 673 31. 28 48.13 
1916..| 2,94 2,278 ttoal 33. 53 552 354 64.13 32. 82 3,500 2,632 75. 20 33. 42 
1917..| 4,166 2,341 56. 19 26. 08 684 365 53. 36 28. 26 4,850 2,706 55. 79 26. 39 
1918..| 4,610 2,245 48. 69 26. 59 734 199 27.11 27.65 5, 344 2,444 45. 73 26. 74 
1919... 5073 2,158 47.19 29. 28 585 81 13. 84 39. 76 5, 158 2,239 43. 40 30. 47 
| 3, 542 1,078 30. 43 40. 45 524 83 15. 83 46. 28 4,066 1,161 28. 55 41.20 
Thus the average harem has increased from 26 to 41, whereas there 
are 28 per cent as many idle bulls as harem bulls. When this is 
compared with average harems of 60 and 65 and with percentages of 
idle bulls from 8 to 11 found in 1912 to 1914, it is at once seen there 
was still a small oversupply of males in 1920 if conditions in the 
former years be considered ideal. It has not been contended that 
there were too few bulls in those years for breeding requirements, 
although it is generally believed that the ideal average harem is 
between 40 and 50. This was the condition in 1920, and if it can be 
maintained with about the same averages and ratios throughout 
commercial operations, those having charge of the business may well 
feel gratified at their success. 
If we pass without comment on the good or bad judgment shown 
in creating the enormous surplus of male life with the closed season 
of 1912 to 1917, still those who have had the administration of the 
herd in hand deserve great commendation for bringing back an ideal 
condition in the short space of three years. 
AVERAGE HAREM. 
The average harem was determined from actual pup counts in 
1920 on the same rookeries as in 1919. The gains shown on these 
rookeries were used as the basis in the computation of the average 
harem for all rookeries. Due consideration was given in every case 
to the topography of the various breeding areas, since it is well 
known that harems are smallest on rough, rocky places. Also it is 
known that large massed areas grow more rapidly than small scat- 
tered ones. Thus, while the harem increased but 6 on the small 
rookeries counted, it was computed to have increased by 10 for the herd 
as a whole. A good illustration of this point may be seen in the 
figures for 1913, when there were over 80 cows to each bull on the 
average for four rookeries, while many of the smaller ones ran below 
40. For purposes of comparison the average harems in 1919 are 
given, showing graphically the increase in size of harems. 
