CHILDS and GAFFKE: MERCURY CONTENT OF GROUNDFISH 



Sebastes pinniger; flag rockfish, Sebastes rubri- 

 vinctus; and rougheye rockfish, Sebastes aleu- 

 tianus were significantly lower than 0.30 ppm 

 at the 0.005 level. The means of all species 

 were significantly lower than 0.50 ppm at the 

 0.005 significance level, with the exception of 

 lingcod (the significance level was 0.01) and 

 spiny dogfish (not significantly lower). Spiny 

 dogfish were further investigated to determine 

 if there was any difference in mercury content 

 with regard to sex. The mean of all male samples 

 was 0.58 ppm with a 99% confidence interval 

 of 0.41-0.76 ppm, and the mean for female 

 samples was 0.65 ppm with a 99% confidence 

 interval of 0.53-0.77 ppm. The confidence inter- 

 val ranges overlapped considerably and t- 

 function analysis revealed no significant dif- 

 ference in mercury content based on sex. 



Numerous species of soles and rockfishes 

 were obtained. The mercury content of all 

 soles was similar suggesting that their degree 

 of exposure and metabolism of mercury may be 

 similar. However, the mercury content of rock- 

 fishes varied widely, with yellowtail rockfish 

 having a 99% confidence interval of 0.30-0.45 

 ppm, canary rockfish 0.08-0.32 ppm, and flag 



rockfish 0.08-0.19 ppm. These data make it 

 difficult to predict the mercury content of 

 marine fish on a phylogenetic basis. 



For species of fish which were available along 

 the entire coast, data were analyzed separately 

 for each area (Table 2). There was variation 

 among the coastal areas. Lingcod from the 

 northern coast had a mean mercury content 

 of 0.403 ppm, from the central coast 0.274, and 

 from the southern coast 0.400. Conversely, sand 

 sole from the central coast had more mercury 

 than samples from the southern coast and Dover 

 sole from the northern coast was lower than 

 either of the other coastal areas. Thus, no 

 relationship is immediately discernible between 

 catch area and mercury content. 



In freshwater fish it has been noted that there 

 is a correlation between length and mercury 

 content (Bligh, 1972). All present data were 

 scrutinized by regression analysis to determine 

 if a similar correlation existed in Oregon 

 groundfish (Table 3). No highly discernible 

 linear correlation (i.e., correlation coefficient 

 greater than 0.90) between either length or 

 weight and mercury content was noted. Graphs 

 of mercury concentration vs. length are pre- 



Table 2. — Mercury content of Oregon groundfish according to coastal area. 



Table 3. — Regression 



analysis of mercury content of Oregon groundfish vs. length 

 and weight. 



715 



