SIVUrClI ROCK AND TOLSTOI SANDS. 407 



Mr. Liicas fiuds tliat when tlu' liver is tliiii and dark, the hings have very 

 l)liu-ki.sh congestion and the intestines contain more or less blackish fecal niattei-. 

 it is a case of starving to death. The congestion of the lungs is probably prodnced 

 by injuries that would not att'ect well pups. Drowned ]>ups have usually a little 

 water in the lungs. The lungs are jtale, the outer organs in good condition ; sometimes 

 there is water in the stomach. 



Kotik. the experimental pup, weighed 1- pounds on the 4th of August. His 

 weight to-day, August 12, is !tj; pounds. He is beginning to show loss of tiesli. 

 The little starving i)nps noticed within the past few days on the rookeries show the 

 same characteristics — hollow places over the shoulder blades, the ribs, and the hips. 



SIVUTCH RUCK. 



Dr. Jordan counted the harems on this rookery and Mr. Clark counted the dead 

 pups. Of the latter there were 50, all told. The number of harems proves to be 

 larger than was supposed, counting with a glass from the Keef, only a part of the 

 breeding ground being visible from that point. There are 03 harems, represented 

 by bulls at this time. This is doubtless an unsatisfactory count, but it is better 

 than the original. This island therefore represents quite a respectable rookery of 

 !,0'.I0 cows — but little smaller than the Lagoon. The ground is very favorable for 

 rookery purposes, as the low death rate shows. There are two hauling grounds 

 occupied by the bachelors — one at the north end and the other at the south end. 



A bull was seen at some distance in the water trying to keep a cow from swimming 

 oti'. He finally picked her up in his mouth as a dog might a duck and carried her in 

 to the shore, depositing her on the rocks and standing guard over her. 



A good many pups obviously starving are still strong and active. That they are 

 starving is shown by the absence of fat over their ribs, scapula, and rumi) and by the 

 disproportionately large size of the head. The plumpness of the normal pup has given 

 way to an unusual slenderness. The starving puj) at the village gives a standard of 

 comparison. While he is still very active and pugnacious, he is plainly starving and 

 the end is not far ofl'. 



TOLSTOI DEAD PTJPS. 



Tlie count of the dead pups on Tolstoi rookery was made by Dr. Jordan, Mr. 

 Macoun, and Mr. Clark in the afternoon. Mr. Lucas, assisted by Professor Thompson, 

 dissected as many dead pups as were fresh enough for examination. Mr. Townsend 

 photographed the dead pups on the sand Hat. 



The great area of hard, sloping sand, with the beach below, was found to contain 

 1,49.5 dead pups, the vast majority of them apparently having been killed at the 

 beginning of the breeding season, when this region was covered so thickly with seals 

 that they looked like a great swarm of bees. This portion of Tolstoi was the densest 

 of all the rookeries in the breeding season. 



The dead pups were es))eciall3' numerous in the center of the large wedge shaped 

 mass as it appears in the earlj^ part of the season, and also underneath the northernmost 

 green clift'. The rocky slope over this cliff contains many dead pups, the rookery floor 

 being here made up of rock in place, with occasional large bowlders. There are many 

 concave depressions, and the few bowlders are too far apart to be of service. 



