FISHERY BULLETIN: VOL. 71. NO. 3 



Table 1. — Catch of salmon by species, age offish, and unit of gear. 30 September and 2 October 1964. 



of Bristol Bay origin) south of the Alaska 

 Peninsula and in the eastern Bering Sea in June 

 and early July. 



Fishing gear consisted of multifilament gill 

 nets of 133-mm mesh. Four 91.5-m nets, 

 7.3-m deep, were joined together to form a 

 basic unit of gear 366-m long. For some sets in 

 areas where large catches were expected, the 

 basic unit was reduced to 2 shackles (183 m). 



Dropout of salmon from gill nets was esti- 

 mated by regression analysis of catch ratio 

 obtained from the catch for a 6-h period and the 

 catches for the first and second halves of the 

 same period (Figure 5). Assuming Ni and N2 

 are the numbers of fish available to the gill 



nets during the first and second halves of a 

 given 6-h period, the expected catches of the 

 three units of gear are: 



f 1 = .s'o ;/i Ni for the first 3-h unit, 



Co = ^'o id N2 for the second 3-h unit, 



C = Ni No ^'1 iVi + No !(2 No for the 6-h unit. 



Where Hi is the fraction of the fish available 

 during a 3-h period which are captured by the 

 unit of gear, .so is the fraction of those captured 

 which are retained by the gear until the time of 

 hauling, and hi is the fraction of the fish 

 captured and retained by the 6-h unit of gear 

 during the first half of the period, which are 



Table 2. — Catch of salmon in gill nets fished continuously and in gill nets hauled at intervals, and estimated loss of salmon 



from nets fished continuously. 



' Cumulative catch divided by cumulative hours fished. 



•^Adjusted to equal time period of nets fished continuously; catch per hour times number of hours fished of nets fished continuously. 

 STotal in adjusted cumulative catch for nets fished short intervals minus total for nets fished continuously, divided by adjusted cumulative 

 catch. 



850 



