containing 1.2 mg Aroclor 1248/kg fresh weight produced progressive 

 degenerative changes in the liver of lake trout after 9 months (Roberts et al. 

 1978). The current recommended level for PCBs in fish diets of less than 0.5 

 mg/kg fresh weight is based on investigations with striped bass by O'Connor 

 and Pizza (1984). They found that food items containing less than 0.5 mg/kg 

 fresh weight will not, in the course of one growing season, cause body burdens 

 in striped bass to exceed 2 mg/kg fresh weight, a proposed Food and Drug 

 Administration guideline for PCB burdens in fish. Striped bass are mobile, 

 pelagic, and highly migratory; accordingly, dietary levels may have to be 

 revised downwards for benthic, nonmigratory species that frequent localized 

 areas of high PCB contamination (O'Connor and Pizza 1984). 



Whole body residues of 0.4 mg PCBs/kg fresh weight are associated with 

 reproductive toxicity in rainbow trout (EPA 1980). The large discrepancy 

 between this value and the current recommended level of 5.0 mg/kg fresh weight 

 in fish and shellfish (Table 7) will be discussed later. Eggs of rainbow 

 trout containing 0.33 mg PCBs/kg fresh weight showed reduced hatch, and a 

 significant increase in larval deformities (Niimi 1983). PCB residues of 0.12 

 mg/kg fresh weight in gonads of field-collected Baltic flounders may be 

 associated with population declines of that species (Ernst 1984), but this 

 needs to be verified experimentally. 



Birds seem relatively resistant to PCBs. Among sensitive species, female 

 screech owls fed 3.0 mg of PCBs/kg fresh weight diet laid eggs containing up 

 to 17.8 mg/kg fresh weight; however, no other adverse effects were observed in 

 either parents or progeny (McLane and Hughes 1980). Higher dietary exposures 

 of 5 mg/kg in chickens, and 10 mg/kg in mourning doves resulted in 

 reproductive impairment (Tori and Peterle 1983; as quoted in Heinz et al. 

 1984). Fertilized eggs of ringed turtle-doves containing 16.0 mg PCBs/kg 

 fresh weight showed delays in growth and development (Peakall et al . 1972), 

 and residues of this magnitude should be considered as presumptive evidence of 

 significant PCB contamination. Residues in brain appear to be good indicators 

 of PCB exposure in birds. Concentrations in excess of 301 mg PCBs/kg brain 

 fresh weight is strong evidence of PCB poisoning, while concentrations in 

 excess of 54 mg/kg fresh weight were common in brain of various avian species 

 that survived high PCB dosages (Stickel et al . 1984). 



Rats and dogs ( Canis sp.) fed various Aroclor PCBs for 2 years showed no 

 measurable effects at dietary levels equivalent to 0.255 (dogs) and 0.5 (rats) 

 mg/kg body weight daily. Using a safety factor of 100, tolerable exposure 

 limits of 2.5 (dog) and 5.0 (rat) ug PCBs/kg body weight daily were derived 

 (Table 7). For the rhesus monkey, a comparatively sensitive species, the new 

 temporary tolerance exposure is 1.0 ug/kg body weight daily (Grant 1983). The 

 mink is the most sensitive animal tested to PCBs, with death and reproductive 

 toxicity documented at 100 to 640 ug PCBs/kg fresh weight of diet (Table 7). 

 The feeding level at which no measurable effects occur is not known with 

 certainty. However, the calculated maximum tolerance level for mink is less 

 than 1.54 ug PCBs/kg body weight daily. This value was derived by known 

 growth rates of female mink between ages 7 and 31 weeks (NAS 1968)--when their 

 body weight increased from 560 g to 1,130 g--by the percent of body weight 



60 



