FISHERY BULLETIN: VOL. 74, NO. 4 



Figure l.-Mean mercury levels in 

 Pacific halibut by area of catch. 



'V° • "^'-st^ 



.=a-=* 



BERING SEA 



0.15 



WASHINGTON -OREGON 

 0.45 



nape section were ground in a Hobart grinder^ 

 equipped with a Vg-inch (3.2-mm) hole stainless 

 steel plate. Larger steaks and fillets were ground 

 in a Hobart Silent Food Cutter (Model 84181). The 

 comminuted flesh was mixed thoroughly before 

 subsampling for analysis. Because samples were 

 often collected more rapidly than they could be 

 analyzed, they were stored at -29°C until analysis. 

 No change in mercury content was observed in 

 halibut that were analyzed immediately or that 

 had been held in frozen storage in either glass 

 vials or aluminum containers if dehydration was 

 prevented. A halibut sample stored in the above 

 manner and used as an analytical control showed a 

 mean mercury content of 0.88 ±0.02 ppm over a 

 2-yr period. This control was analyzed routinely to 

 verify both accuracy and precision of the method. 

 Total mercury was determined at the PURC by 

 either the method of Munns and Holland (1971) or 

 Malaiyandi and Barrette (1970) as modified by 

 Munns (1972). The former method uses sulfuric, 

 nitric, and perchloric acids for digestion with 

 sodium molybdate as a catalyst, while the Munns' 

 modification utilizes nitric and sulfuric acids for 

 digestion and vanadium pentoxide as a catalyst. 

 Some samples were analyzed at the SEURC by the 

 method of Hatch and Ott (1968) as modified by 

 Uthe et al. (1970). This method uses sulfuric acid 

 for digestion and potassium permanganate as an 

 oxidizing agent. 



■'Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



Final quantitation was by flameless spectroscopy 

 using a Perkin-Elmer Model 403 Atomic Absorp- 

 tion Spectrophotometer at the PURC and by a 

 Varian Techtron Model AA5 at the SEURC. In a 

 collaborative study, the mean deviation between 

 laboratories and methods did not exceed ± 0.02 ppm 

 Hg. All samples were analyzed in duplicate or 

 triplicate, depending upon the method of analysis 

 used. We consider ± 0.05 ppm a significant devia- 

 tion; therefore, when differences between rep- 

 licates exceeded this level the samples were 

 reanalyzed. Results are stated in parts per million 

 wet weight. 



RESULTS AND DISCUSSION 



A total of 1,227 halibut were analyzed for mer- 

 cury content. Results indicated a relationship 

 between mercury levels and area of catch, age, and 

 size of fish. The results are broken down by the 

 previously described catch areas (Figure 1). The 

 fish taken from each area were separated by 

 weight classes that approximate those used in the 

 halibut industry; the low, high, and mean mercury 

 values for each weight class are given with a 

 frequency distribution of the fish by increasing 

 mercury concentration (Tables 1 through 5). 

 Because we thought that large fish would be more 

 likely to exhibit higher concentrations of mercury, 

 we attempted to obtain as many large fish as was 

 practicable. For this reason our sampling contains 

 a greater percentage of large fish than do the 

 commercial catches from most of the areas dis- 



784 



