218 U- S. BUREAU OF FISHERIES 



the temperature-iinit sj^stem, it has been found that the rate of de- 

 velopment is accelerated with an increase of temperature, at least 

 during the range of temperature covered by the available observa- 

 tions. In using the temperature-unit system for calculating the 

 time of hatching, the error thus introduced is sufficient to thro\Y the 

 calculation off by approximately 10 per cent at moderate extremes 

 of temperature. 



ALASKA HERRING 



The use of herring for oil and meal or fertilizer in southeastern 

 Alaska and Prince William Sound has developed so rapidly in recent 

 years that it is assuming alarming proportions. In 1926 these two 

 districts produced over six times as much herring oil as in 1922, but 

 the yield of pickled herring, in spite of good prices for the product, 

 amounted to onlj^ one-eighth that in the earlier jesiT. The manu- 

 facture of these bj^-products, at first a side line to utilize the wastage 

 of the salteries, has now become the chief object of the fishery in 

 these districts. These facts, together with a decline in abundance of 

 herring in certain regions, have aroused the fear of depletion in the 

 minds of the herring fishermen themselves as well as in the salmon 

 and halibut fishermen, who are indirectly dependent upon the herring 

 supply. 



Investigation^ of the herring fishery, begun in 1925 by George A. 

 Rounsefell, have been continued. Most of the work has been done in 

 central Alaska, where conditions are most favorable for studying the 

 life history of the herring. Of direct bearing upon the problem of 

 depletion is the discovery of natural variations in abundance due to 

 varying succes,s of spawning, and the existence or nonexistence of 

 dominant age groups, such as are conspicuously present in the 

 Atlantic herring, and which cause enormous fluctuations in abun- 

 dance, have been studied. Although more information on this point 

 is required and i,s being collected, present studies show conclusively 

 that this phenomenon does occur. To discover the effect of such 

 dominant age groups on the commercial catch, detailed statistics 

 showing the time and place of each catch and the amount of gear 

 required are being collected for later analysis. 



Previous studies have demonstrated that the herring population 

 in Alaska is composed of segregated units or local races inhabiting 

 well-defined areas and subject to depletion from overfishing in each 

 locality. It has been shown that local races occur in Prince Wil- 

 liam Sound, Dogfi,sh Bay, Kachemak Bay, Shuyak Strait, Chignik, 

 the Shumagin Islands, Golovin Bay, and the Bering Sea. Growth 

 rates have been calculated for a few localities. Herring in Halibut 

 Cove, Kachemak Bay, and Shuj^ak Strait make the same rate of 

 growth, but the Prince William Sound fish grow much more slowly, 

 reaching a length of but 23 centimeters when six years old, compared 

 with 25 centimeters for fish in the other localities. 



A tagging experiment to define local migrations wa,s begun in the 

 spring of 1927, when 3,000 herring were tagged at Halibut Cove ; but 

 no returns have been received. It is likely that a new type of tag 

 will be required for a fish as small and delicate as the herring. 



