366 U. S. BUREAU OF PISHEEIES 



It is only three-fourths mile in length, and the varied character of 

 its stream bed provides for the study of practically all the types of 

 spawning grounds found in the streams in southeastern Alaska. Fur- 

 thermore, all of the commercial fishing which affects its salmon popu- 

 lation is carried on just outside of the port and can readily be ac- 

 counted for. Plans are being made for the construction of a perma- 

 nent counting weir in the stream wherein both the adults that migrate 

 into it to spawn and the resulting fry which migrate to the sea may 

 be counted accurately. In this way iDoth the mortality of the brood 

 while in the streams and in the sea may be determined. It is hoped 

 that funds will be available for a permanent observer at the stream 

 for the purpose of making continuous observations of the environ- 

 mental changes in the stream so as to determine the natural factors 

 which influence the mortality of the brood. Once equipped with 

 knowledge of the factors which influence the mortalitj'^ of the broods, 

 the Bureau will be able to develop further its conservation policy so 

 as to provide for the most efficient utilization of the pink-salmon 

 populations. 



HERRING INVESTIGATION 



With tlie exception of a few minor localities, the herring fisheries 

 of Alaska may be grouped into four districts; namely, southeastern 

 Alaska, Prince William Sound, Kodiak-Cook Inlet, and the Aleutian 

 Islands. The fishery of southeastern Alaska is by far the most 

 productive, and at the present time is supporting the majority of the 

 fishing effort. Hence the activities of the herring investigation during 

 the past year were confined mainly to a study of the fishery in this 

 region. Dr. G. A. Rounsefell, who has been conducting the inves- 

 tigation since 1925, was transfeiTcd to the Puget Sound sockeye- 

 salmon investigation in July owing to the need of a trained inves- 

 tigator for this work. E. H. Dahlgren, who has assisted Dr. 

 Rounsefell for a number of years, was jDlaced in charge of the 

 investigation. 



The stock of Alaska herring, being separated into a series of local 

 races, presents a more complex problem than would be the case if 

 it were a homogeneous population. The delineation of these races 

 and a measure of the intermingling which takes place between stocks 

 is of utmost importance. Estimates of changes in abundance and 

 causes for such changes are dependent on this separation. Observa- 

 tions on the size and age compositions of the catch have clearly 

 demonstrated the existence of dominant year classes, but the deter- 

 mination of the exact causes for such varying degrees of success of 

 different spawning is yet to be solved. 



A racial report now in press, by Dr. Rounsefell and Mr. Dahlgren, 

 entitled " Races of Herring, Clupca pallasil, in southeastern Alaska ", 

 presents the progress already attained in the problem of the segrega- 

 tion of populations. The results of an analysis of vertebral counts 

 and of growth rates, the comparison of proportional representations 

 of year classes, and the recovery of tagged herring are presented in 

 this paper. The separation of the herring populations in the vicinity 

 of Petersburg and Noyes Island, and the localities east of Clarence 

 Strait and south of Sumner Strait, including Wrangell, was made 

 evident by comparisons of the means of the vertebral counts. Size 



