36 U- S. BUREAU OF FISHERIES 



Continuing tlie migration studies, tagging was again carried on 

 during the spring at the Sitka and Craig spawning areas, which are 

 known to contribute the greatest supply of herring to the commercial 

 catch. In this experiment approximately 14.000 individuals were 

 marked at Sitka and 11,000 at Craig. The electronic detector was 

 again operated in Southeastern Alaska for the recovery of tagged 

 individuals and 113 recoveries were made during the season. 



From these recoveries, together with others made in previous years, 

 the migration habits of the commercially important populations have 

 been established. It is evident that: (1) the fishery in the Cape 

 Ommaney area, from which 60 to 95 percent of the total catch is 

 made, draws almost exclusively from the population which spawns in 

 the vicinity of Sitka; (2) the Warren Island fishery draws from an 

 intermixture of the Craig and the Sitka populations, and a small 

 influx of herring from minor spawning grounds south of Craig at 

 least as far as Rose Inlet, approximately 40 miles south of the major 

 spawning area; (3) the summer fishery conducted on the west coast 

 of Kuiu Island, including Malmesbury, Tebenkof, and Pillar Bays, 

 is composed, as is the Warren Island fishery, of herring from the 

 Craig and Sitka spawning areas; (4) the Douglas Island fishery is 

 supported by the Juneau spawning populations; and (5) the Icy 

 Strait and Tenakee fisheries are composed of an intermixture of 

 Juneau fish with some unknown population, most probably that of 

 the Kootznahoo spawning area. 



The herring in the Cape Ommaney fishing area from which the 

 bulk of the commercial catch in southeastern Alaska is made, have 

 shown a marked decline in abundance during the past few years. 

 The measure used in evaluating this abundance is derived by a com- 

 parison of the catch per unit of gear per day's fishing with the aver- 

 age catch per day's fishing established over a 9-year period. In de- 

 riving these indices, the fleet has been divided into two groups (those 

 of over 35 net tonnage capacity and those under this size) to mini- 

 mize the effect of the differencein efficiency of the larger and smaller 

 vessels. 



The index for the 1937 season for the larger vessels was 71 com- 

 pared with 73 for the 1936 season, 140 for the optimum year of 1932, 

 and 62 for the minimum year of 1935. The indices for the vessels 

 not over 35 net tons for 1937 was 75, compared with 60 for the mini- 

 mum vear of 1936 and 164 for the optimum year of 1932. 



This continuing low level of abundance, while doubtless due in 

 large measure to an overly intensive fishery, was brought about in 

 part by the virtual failure of the three successive spawnings of 1932, 

 1933, and 1934, to contribute the normal increment of young fish to 

 the populations. The entrance during the 1937 season of a large per- 

 centage of the new age class resulting from the more successful 

 spawning of 1935 may be expected to result in a rise in the level of 

 abundance. This year class, which appeared in large numbers as 

 2-year-olds in the 1937 catch, may be expected to contribute a large 

 portion of the catch as 3-year-olds during the coming season. 



The low catch per unit'^of gear in the Ommaney area during the 

 1937 season was due in part, also, to the emigration out of that area 

 of the age group which had supported the fishery during the three 

 previous years, the abundant 1931 year-class. Tagged individuals 

 belonging to this age-class were taken in large numbers in the Teb- 



