FISHERY BULLETIN: VOL. 71, NO. 3 



rates as determined from the dropout studies, 

 therefore, indicated substantial numbers of 

 salmon lost during high-seas gillnetting. 



To summarize the loss rates obtained, for 

 large mesh nets of monofilament construction, 

 the percentage loss of immature and mature 

 salmon was approximately 33% (Table 10). For 

 mature fish and for multifilament and mono- 

 filament nets the loss rate was about 20% (Table 

 11). These estimates of loss due to gillnetting 

 are similar to those reported by Semko (1964) 

 and by Miyazaki and Taketomi (1963, as cited 

 by Konda, 1966), are smaller than those 

 reported by Doi (1962), and are somewhat 

 larger than the losses reported by Ishida et al. 

 (1969). 



Applying a loss rate of 33% to the average 

 catch of 2.5 million sockeye salmon of Bristol 

 Bay origin taken by the Japanese mothership 

 fleet results in an estimated average of about 

 1.2 million sockeye salmon lost from the gill 

 net fishery annually, with a range of 0.15 to 3.5 

 million fish yearly, depending upon the size of 

 the catch. A loss i-ate of 20% would indicate that 

 0.6 million Bristol Bay sockeye salmon are lost 

 annually on the average (with a range of 0.08 

 to 1.8 million fish). 



Calculation of the total loss of salmon (of 

 North American and Asian origin) from the 

 mothership operation, based on annual catches, 

 resulted in an estimated loss of about 10.8 mil- 

 lion salmon at a 33% loss rate and 5.5 million 

 fish at a 20% loss rate. The range of losses, 

 depending upon the annual catch by the mother- 

 ship fleet, was from 9.4 to 13.1 million fish and 

 4.8 to 6.6 million fish for the two respective loss 

 rates. 



In experiments in which no decoys were lost 

 and in which losses from gill nets were attribut- 

 able only to dropouts (Table 14), the disen- 

 tanglement rate of about 12% also indicated 

 substantial losses of salmon. This loss rate 

 applied to the average annual catches by the 

 mothership fishery, resulted in estimated annual 

 losses of 0.3 million sockeye salmon of Bristol 

 Bay origin (with a range of 0.04 to 1.0 million 

 fish, depending upon the catch) and an average 

 total loss of 3.0 million salmon of all origins 

 (range of 2.6 to 3.6 million fish depending upon 

 catch). 



We have no way of determining the actual 

 mortality of salmon lost from gill nets. We 

 know that some fish sui^vive as evidenced by 

 net-marked salmon observed in spawning 

 streams. In the experiments on viability of 

 adult sockeye salmon that disentangle from gill 

 nets, which were carried out in an enclosed 

 area in relatively calm waters of Puget Sound, 

 it was shown that approximately 73% of fish 

 that disentangled from gill nets died within 6 

 days compared with 10% for control fish. These 

 particular fish were held in a protected en- 

 closure throughout the experiments; mortali- 

 ties of gill net dropouts could be even greater 

 in the open ocean. Therefore, at any likely mor- 

 tality level that we might assign to salmon 

 dropping out of gill nets on the high seas, 

 the resultant waste of the resource is substan- 

 tial. 



In addition to losses due to dropouts and 

 predation, there is the attendant loss of salmon 

 during the haul of gear aboard vessels. It was 

 estimated for U.S. research vessels that about 

 1.4% of the number offish landed fell out of the 

 gill nets at haul. If we consider a 1% fallout rate 

 (the approximate fallout rate of the smallest of 

 U.S. research vessels), the resultant loss of sal- 

 mon during the Japanese mothership fishery 

 would average over 200,000 fish annually. The 

 estimated loss of sockeye salmon of Bristol Bay 

 origin by the mothership fishery would amount 

 to an average annual loss of 25,000 fish. 



From the indirect methods of determining 

 rates of losses, we obtained rates of approxi- 

 mately 32% for immature fish and 27% for matur- 

 ing fish. Either of these rates indicated losses 

 of salmon on the order of or even greater than 

 that estimated from direct observation. As 

 with the majority of direct observations, how- 

 ever, we were unable to differentiate the effect 

 of predators on dropout rates and losses were 

 considered as due to a combination of dropouts 

 and predation. 



The estimated large numbers of salmon lost 

 from gill nets in a high-seas fishery and the 

 probable large mortality of these salmon indi- 

 cate the relatively large waste of part of the 

 resource. In contrast, inshore experiments in 

 Puget Sound indicated that there is not the loss 

 of salmon from gill nets or potential waste of 



872 



