50 



40 



E 



Q. 



00 



o 



o 



30 



20 



10 



70 



SELENIUM mg/kg (ppm) 

 Figure 5. — Relationship between total mercury and selenium in liver of 46 blue marlin from the Hawaiian 



Islands. 



Concentration through the food chain is appar- 

 ently the reason for the elevated levels of mercury 

 found in the marlin. Stomach contents examined 

 in 76 blue marlin captured in 1973 showed that 

 the most common food items were tuna, mostly 

 Katsuivonus pelamis (SS'/r occurrence); mackerel 

 scad, Decapterus pinnulatiis (36'yf ); squid {2Y'/()\ 

 spiny puffer, Diodontidae (19^^); and dolphin, 

 Coryphaenidae ( 12% ) according to Naughton (see 

 footnote 1). 



Mercury levels were reported for yellowfin tuna, 

 Thunnus albacarcs, 0.54 mg/kg; skipjack tuna. A'. 

 pelamis, 0.38 mg/kg; and for dolphin, Coryphaena 

 hippurus, 0.25 mg/kg; by Rivers et al. (1972). 

 These are pelagic species and as such, are exposed 

 to the same amount of mercury in their physical 

 environment; yet all have mercury levels nearly 

 an order of magnitude less than that of the marlin. 

 In addition, their mercury content was essentially 

 all methyl mercury. 



If most of the mercury entered the marlin via 

 the food chain as methyl mercury it would seem 

 that demethylation to inorganic mercury had oc- 

 curred. Conversion of methyl mercury to inor- 

 ganic mercury has been shown to occur in bluegill 



(Burrows and Krenkel 1973) and in rats (Norseth 

 and Clarkson 1970). 



It is indicated in Figure 1 that the mercury is 

 accumulated with size but at different rates for 

 males and females. This observation contrasts 

 with previous year's results in which no sex- 

 related concentration disparities were found 

 (Shultz and Crear 1976; Shultz et al. 1976). The 

 reason for this year's anomaly is unclear and not 

 enough information is available to determine if we 

 are dealing with more than one population. 



Selenium levels in the marlin are presented in 

 Table 1 and show an increase with weight and age 

 (Figure 2). The high correlation of mercury and 

 selenium with weight (Table 2), and the observa- 

 tion that selenium modifies the activity of mer- 

 cury in experimental animals, indicate that they 

 are highly correlated with each other, as is seen in 

 Figures 3-6. The precise nature of the mercury- 

 selenium interaction is not known, although a 

 number of suggestions have been advanced. 

 Pah'zek et al. ( 1971) demonstrated with rats that 

 the protective effect of selenite against mercuric 

 chloride was not related to an increase in mercury 

 excretion but to a decrease, resulting in a change 



876 



