side and 2 gill rakers on the upper limb of the sec- 

 ond gill arch on the blind side, it is referred to A. 

 stomias. The other two anomalous specimens also 

 had 2 gill rakers on the upper limbs of the second 

 gill arch of the blind side and were also recorded 

 as A. stomias. 



Discussion 



From this study, it is evident that the two species 

 of Atheresthes can most easily be distinguished by 

 eye position. The number of gill rakers on first and 

 second gill arches can be used to assist and verify 

 identification. 



When identifying specimens, eye position should 

 be examined first. If the upper eye interrupts the 

 profile of the head, this specimen is A. stomias; if 

 the upper eye does not interrupt the profile of the 

 head, the specimen is A. evermanni. If the head is 

 in bad shape (e.g., damaged during the trawl opera- 

 tion) or if the examiner has difficulty using eye posi- 

 tion and head profile to identify a specimen, the gill 

 arches must be examined. Two or more gill rakers 

 on the upper limb of the second gill arch indicates 

 that the specimen is A. stomias; if there is only 1 

 gill raker, the specimen is A. everm,anni. 



The number of gill rakers on the first gill arch has 

 generally been used to distinguish the two species 

 of Atheresthes. However, this study demonstrated 

 a greater overlap between the two species in number 

 of rakers on the first gill arch than the second gill 

 arch (Tables 1, 2), indicating that the second gill arch 

 is a better character for assigning individuals to the 

 species. 



The study also suggests that the number of gill 

 rakers on the upper limb of the first gill arch is 

 species specific. If there are 4 or more gill rakers, 

 the specimen is A. stomias; 2 or fewer gill rakers 

 indicate the specimen is A. evermanni. 



The uncertainty in examining the first gill arch 

 is when there are 3 gill rakers on the upper limb. 

 Approximately 25% of A. stomias and 50% of A. 

 evermanni samples had 3 gill rakers on the upper 

 limb of the first gill arch. Thus, when 3 gill rakers 

 are present on the upper limb of the first gill, the 

 second gill arch must also be examined to distinguish 

 the two species. 



Acknowledgments 



I want to thank Jean Dunn and James Allen for 

 their comments and suggestions. Richard Bakkala 

 and Patricia Livingston reviewed my earlier manu- 

 script; their help is also appreciated. I also want 



to thank the anonymous reviewers for their com- 

 ments. 



Literature Cited 



Norman, J. R. 



1934. A systematic monograph of the flatfishes (Hetero- 

 somata). Vol. I. Psettodidae, Bothidae, Pleuronectidae. 

 Br. Mus. (Nat. Hist.), Lond., 459 p. [Reprinted, 1966, by 

 Johnson Reprint, N.Y.] 

 Ranck, C, F. Utter, G. Milner, and G. B. Smith. 



1986. Genetic confirmation of specific distinction of arrow- 

 tooth flounder, Atheresthes stomias, and Kamchatka 

 flounder, A. evermanni. Fish. Bull., U.S. 84:222-226. 

 WiLiMOVSKY, N. J., A. Peden, and J. Peppar. 



1967. Systematics of six demersal fishes of the North Pacific 

 Ocean. Fish. Res. Board Can., Tech. Rep. 34, 95 p. 



Mei-Sun Yang 



Northwest and Alaska Fisheries Center 

 National Marine Fisheries Service, NOAA 

 7600 Sand Point Way NE 

 Seattle, WA 98115-0070 



PREDATION OF KARLUK RIVER 



SOCKEYE SALMON BY COHO SALMON 



AND CHAR 



The number of sockeye salmon, OncorhynchiLS 

 nerka, in Alaska's Karluk River (Fig. 1) declined 

 from millions to thousands during the early part of 

 the present century. Rounsefell (1958) discussed 

 alternative explanations for the decline including a 

 general loss of fertility of the system as the number 

 of salmon carcasses declined, competition, over- 

 fishing, subtle changes in climate, and predation; he 

 concluded that the combined effect of predation and 

 fishing was the most probable explanation. Later, 

 Van Cleave and Bevan (1973) suggested that the 

 weir constructed in the river each year to facilitate 

 counting the fish as they entered the system was 

 the most probable cause of the decline. It prevented 

 free movement of both adults and juveniles in the 

 river. All of these hypotheses remain as potential 

 explanations for the decline. 



Fredin et al. (1974) described a relation that 

 showed two equilibrium regions between the spawn- 

 ing stock and the resultant run for sockeye salmon 

 in the Kodiak area. We developed a stock-recruit- 

 ment curve (Fig. 2) for sockeye salmon in the Karluk 

 River basin that also showed two equilibrium 

 regions, and suggested that the population had "col- 



FISHERY BULLETIN VOL. 86, NO. 3, 1988 



611 



