RECOUNTS OF ST. GKORGE ROOKERIES. 559 



Mr. Lucas'.s count ol' this rookery on .Inly !• giive 4!t7 cows and 40 harems. 



It is easy to see lliat tliis rookery was uudercstimated last year. The count of 

 cows on July 9 gave ouly 355, whereas 1,3.30 pups were counted in August. Last year 

 when Dr. Jordan and I counted this rookery we saw it ouly from a i)oint midway of 

 the rookery. Seen from the jutting elilf which marks the western end of the rookery 

 it is entirely different. Almost a third of the harems must have been out of sight 

 from the point where the count was made last year, and the number of seals on little 

 east last year must have been more nearly GUO than 355. That the rookery has plainly 

 sliruuk away from the slope above the bowlder beach is very evident. 



ME. LUCAS'S NOTES. 



The large mass of seals on Polovina as compared with the photograph of last 

 year is thinner and narrower on the left, while to the right and up the center the 

 conditions are much the same as last year. An attempt was made to count the cliff 

 portion of the rookery, but without much success. There are 01 harems and 747 

 cows, which is jirobably an underestimate. 



On Zapadni Keef a bull is seen to land and seize the nearest cow, carrying her 

 out and swimming 75 or 100 yards with her, most of the time under water. He lands 

 in a break in the rookery, but fails to hold the cow. 



JULY 16. 



I made a count of harems on Xorth rookery this morning, finding a total of 214. 

 The increase in harems over Mr. Lucas's count of the 9th is doubtless due to the 

 formation of small harems. The scattering out of the cows already noted on St. Paul 

 is plainly marked here. Twenty- five harems of single cows were noted. Most of 

 these with many of the 2 and 3 harems have been formed since the 9th. The 

 ap]iearance of North rookery at this date is in marked contrast to what it was last 

 year when first seen on the 8th. No diminution could be detected under the cliffs, 

 though it may easily be present in a general thinning out. There are now no large 

 harems like the one of 135 cows counted last year. Tlie cows are scattered and the 

 identity of individual harems is not clear except from the presence of the bulls. On 

 the western end, where the harems extend up the slope, a diminution is apparent, if 

 a shrinkage in the outer line of the band of bulls could be taken for a guide. 



Zapadni rookery of St. (ieorge shows the beach portions very much broken and 

 scattered. One ouly extends above the line of the bowlder beach, and there are but 

 .3 small harems on the flat. The northern portion, which was counted last year by 

 Dr. Jordan and Professor Th(mi])son and found to contain about 30 harems, had 

 this year only 2 L The slope jtortion of the rookery at the southern end is densely 

 covered with cows. A number of harems are in the bench at the top and beyond 

 them a great pod of pups, which must have come up from the harems below. I did 

 not see Zapadni rookery in the season last year and can therefore give no comparative 

 estimate. There are apparently 133 harems on the rookery today. 



There are about 700 bachelors in the hauling ground, which runs out in the 

 ravine at the foot of the hill. 



Staraya Artel shows a very different aspect from that seen by us on July 9, 1896. 

 The dense mass of seals is broken up and evenly distributed among the rocks. 

 Undoubtedly a large proportion of the cows are away. A count of the harems from 



