the fillets and fish weight for the 127 females ( Fig- 

 ure 2). The weight of individual fish was evenly 

 distributed in each of the area samples with the 

 exception of the small sample of eight fish from 

 Seattle. Although these were the largest fish col- 

 lected, they contained less mercury than smaller 

 fish from other areas. The Seattle sample does not 

 appear to be adequate in number and may not be 

 representative of the population. In all areas, ex- 

 cept Seattle, the correlation coefficients were sig- 

 nificant for the relationship of mercury content to 

 weight (Table 2). The correlations between mer- 

 cury content and fish length were significant but 

 slightly lower in four of the five groups showing 



FIGURE 2. — Relationship between weight and mercury concen- 

 tration in female dogfish fillets. 



TABLE 2. — Correlation coefficients (r) and significance level (a) 

 of mercury content to the weight and length of female spiny 

 dogfish fillets from the State of Washington. 



1 Not significant 



positive coefficients. We expected a more sig- 

 nificant correlation with length, since the weight 

 of the females varied as to whether or not they 

 were pregnant and the length of gestation. Childs 

 et al. (1973) stated that mercury is not concen- 

 trated in the fetuses in situ; therefore, the mercury 

 level in the flesh of the female is presumably un- 

 affected by pregnancy. 



The bellyflaps of 107 female and 5 male dogfish 

 were analyzed (Table 1 ). Bellyflaps of the fish from 

 Seattle and 10 small females from Port Susan 

 were not analyzed. The bellyflaps contained 

 slightly less mercury than the corresponding 

 fillets; however, the percentage exceeding the ac- 

 tion level (739?-) was not significantly different 

 from that for fillets. 



The limited data on mercury levels in male 

 dogfish (Table 1) indicated that essentially all 

 male dogfish over the minimum commercial size 

 (81.3 cm) would exceed the FDA action level. Of 

 the 14 males analyzed, 13 (939r ) exceeded the ac- 

 tion level. The mean weight of the males (2,052 g) 

 was less than the mean weight of the females 

 (3,608 g), yet the mean mercury level was higher 

 (0.96 ppm for males and 0.92 ppm for females). 

 This difference may be attributed to the fact that 

 males are smaller than females of the same age 

 (Jensen 1966). Our findings agree with those of 

 Forrester et al. (1972) on the mercury levels in 

 male and female spiny dogfish from inland waters 

 of British Columbia. 



A study by Childs and Gaffke (1973) included 88 

 dogfish taken off the Oregon coast and showed a 

 similar correlation of mercury level to weight and 

 length but a lower mean level of 0.602 ppm mer- 

 cury in all muscle samples. This suggests that 

 dogfish taken from the Pacific Ocean off the Ore- 

 gon coast may contain less mercury than the popu- 

 lation sampled in this study of the inland waters of 

 Washington. Tagging studies by Kauffman ( 1955) 

 and Holland ( 1957) indicated that offshore dogfish 

 populations may be highly migratory. Jensen 

 (1966) noted that the nature of the dogfish's sea- 

 sonal migration in offshore coastal waters was not 

 clearly understood. Alverson and Stansby (1963) 

 stated that the dogfish within Puget Sound show 

 less tendency to migrate and that Puget Sound 

 stocks are apparently somewhat independent 

 from the coastal and offshore stocks. They further 

 stated that some movement of dogfish may occur 

 between ocean areas and Puget Sound. The mer- 

 cury levels found in our study are most probably 

 those of a population indigenous to Puget Sound. 



Acknowledgments 



We thank Raymond Buckley, James Beam, and 

 Mark Pederson of the Marine Fish Program of the 

 State of Washington Department of Fisheries for 

 obtaining most of the specimens used in this 

 study. 



644 



