FUR-SEAL CENSUS, PRIBILOF ISLANDS, 1919. 107 



^the truth is all that can be expected. The best census is a very careful 

 estimate made by utilizing as many of the known factoi-s in the 

 biology of the species as possible. 



The great difficulty in counting bulls is to find a point back of the 

 large rookeries from which all the animals can be seen. In order to 

 overcome the trouble we climb a high ladder which can be moved 

 from place to place, and whenever the weather and sea will permit 

 boats are utilized, because of the better view thus usually secured. 

 But both of these methods are unsatisfactory and certain construction 

 work has been outlined elsewhere. Possibly the work could be done 

 with absolute accuracy from an airplane with the aid of photography, 

 a means first suggested, it is thought, by Dr. Paul Bartsch, of the 

 f nited States National Museum. 



With the decrease in the number of bulls and the increase in the 

 average harem in 1919 less difficulty was experienced in counting 

 pups than in 1917 and 1918. The rookeries are so large on St. Paul 

 now, however, that a complete pup count could hardly be attempted 

 without a considerable force of men. And even then it is believed 

 the loss whicli the herd would suffer from the count would not be 

 warranted. If the persons intrusted with the census work can give 

 the subject their uiKli\itled attention during the season the count of 

 pups on a few of the smaller rookeries should suffice to establish the 

 average harem, provided, of course, that the harem count be accurate. 



SUGGESTIONS FOR FUTURE CENSUSES. 



In connection with work on the fur-seal rookeries it is impossible 

 to forecast even one year in advance the new problems which may 

 be forthcoming and the difliculties which may arise to interfere with 

 a satisfactory taking of the census. However, several years' ex- 

 perience have led to certain conclusions which may be worth recording 

 tor such use as future workers may care to make of them. 



It is anticipated that the buU counts on St. George Island can be 

 made with comparative ease at the heiglit of the breeding season; no 

 material increase in the idle and harem bull classes is expected for at 

 least two or three yeai-s. Staraya Artil and North Rookeries are 

 much more satisfactorily counted from a boat than from the land side 

 if the weather is suitable for one to get out on the sea. Should it be 

 necessary to count from the land however, the use of a long ladder is 

 recommended in order to permit observation of the entire areas of 

 the breeding masses. East Reef and East (lifTs Rookeries can be 

 seen better from the water, but satisfactory counts can be obtained 

 from the land. A considera})le risk is run in the case of the latter 

 rookery on account of the dense vegetation growing on a ragged 

 slope. Should one slip here he would fall among the bulls anfl be in 

 imminent danger of losing his life. However, should timidity be felt 

 regarding the cliff, it seems entirely feasible for the person doing the 

 counting to be supported from above with a rope. Zapadni Rookery 

 can be counted by exercising the ordinary care one acquires in creeping 

 upon the breeding rookeries; and the same is true in the case of 

 South. 



The much larger areas on St. Paul Island, however, can not be 

 disposed of so easily. There is not likely to be any material diminu- 

 ,tion in the numbers of bulls, which are a constant menace to observers 



