NOTE Brill and Holts. Natural mortality of North Pacific Thunnus alalunga from drift nets 



799 



55 



45 



35° 



25- 



15' 



-^2 



DRIFT NET FISHERY 



180° 



170° 



Hawaiian Islands 

 1 . I 



140° 



"l30° 



120 



160 150° 

 Figure 1 



Area of operation of the U.S. troll fishery (light cross-hatching) catching 

 albacore. Track lines of U.S. vessels carrying National Marine Fisheries 

 Service observers in 1991 are shown by the solid line. Area of operation of 

 foreign drift net fleets (heavy cross-hatching) extends west to approximately 

 140°W. 



infection and long term stress (Blaxhall, 

 1972; Anderson, 1990; Goede and Barton, 

 1990). Although muscle lipid content can 

 be used as a measure of recent feeding 

 history (Dotson, 1978), it was not used in 

 this study. The lipid content of tuna muscle 

 is dependent on sampling site (Dotson, 

 1978; Vileg et al, 1983; Boggs and Kitchell, 

 1991) and this factor could not be rigor- 

 ously controlled. 



Materials and methods 



ENTANGLED 



/x 



LANDED ESCAPE 



DROP OUT DEAD 



ALIVE 



BADLY INJURED NOT BADLY INJURED, 



DO NOT BEGIN FEEDING BEGIN FEEDING 



Figure 2 



The possible fates of albacore that become entangled in high- 

 seas drift nets. Note that the condition of fish which escape 

 drift nets alive probably form a continuum from those that 

 are so badly injured that they succumb quickly, to those that 

 are only slightly injured and resume feeding immediately. 



Connolly, 1989; Cone, 1989, 1990; Murphy et al, 1991). 

 Relative otolith weight (i.e., the ratio of otolith weight 

 to body weight) was used as an indicator of long-term 

 growth rate (Boehlert, 1985; Pawson, 1990; Fletcher, 

 1991). (Because otolith weight is proportional to fish 

 age rather than to body size, slower-growing fish have 

 larger otoliths at a given body size because they have 

 taken longer to reach that body size. ) RNA:DNA ratio 

 of white muscle was measured because it is an estab- 

 lished indicator of short-term (days to weeks) growth 

 rates (Bulow, 1970; Bulow et al, 1981; Ferguson and 

 Danzmann, 1990; Mustafa et al., 1991). Relative leu- 

 kocyte abundance (i.e., the ratio of leukocytes to red 

 blood cells) was assessed as an indicator of bacterial 



During 1990 and 1991, observers were 

 placed aboard ten U.S. trollers by the 

 Southwest Fisheries Science Center 

 (NMFS, NOAA), in cooperation with the 

 Western Fishboat Owners Association. Ob- 

 servers documented the number of net- 

 marked albacore landed and collected data 

 and samples for our study. Boats operated 

 in an area of the North Pacific shown in 

 Figure 1. Albacore were classified according to the se- 

 verity of net damage with a scheme (Table li devel- 

 oped by Bartoo et al. 2 . Fork length, maximum girth 

 (both to the nearest centimeter), and body weight (to 

 the nearest kilogram) were measured. Dried blood 

 smears, sagittae (otoliths), and white-muscle samples 

 (frozen in sealed tubes in the boats' commercial fish 

 freezers) were also obtained from a subsample of landed 

 fish selected by the observers. Muscle samples (taken 

 from the small amount of muscle remaining near the 

 skull following decapitation) and otoliths could be ob- 

 tained only on vessels that processed (i.e., "gilled and 

 gutted") their catches. Complete sets of samples were 

 not, therefore, obtained from every fish. 



In the laboratory, otoliths were dried, cleaned, and 

 weighed to the nearest 0.1 mg on a Sartorius 1207 

 MP2 electronic balance. RNA and DNA were extracted 

 from the frozen muscle samples by standard proce- 

 dures (Hutchison and Munro, 1961; Munro and Fleck, 

 1966; Wilder and Stanley, 1983). The only modifica- 

 tions were that the RNA was extracted with 1.0 N 

 NaOH, and the DNA with 1.2 N perchloric acid, both 

 at 37°C for 60 minutes. RNA. DNA, and protein con- 

 centrations were measured with a Beckman model 35 

 UV spectrophotometer using the dual wave length 

 method (Tsanev and Markov. 1960) and standard ex- 

 tinction coefficients (Wilder and Stanley, 1983). Red 

 cells and leukocytes were counted in Geimsa-stained 

 blood smears with the aid of light microscopy (oil im- 

 mersion lens) in a minimum of three microscope fields, 

 or until at least 200 red cells were counted. Eosino- 

 phils could be easily differentiated from the remainder 



