allowed to air dry, and fixed in methanol for 15 min. The blood smears were kept cold (on 

 ice in the field or in a refrigerator in the lab) until they were processed. The smears were 

 stained with May-Grunwald-Giemsa procedure (Preece, 1972) and examined on coded slides 

 under a microscope at high power (1,000 x). One of the three slides was examined for each 

 fish; one of the other two was used if the first was not usable. Two replicate counts of the 

 number of micronucleated erythrocytes per 1,000 cells were made on each smear, and the 

 results reported as the average of the two counts. Two types of micronuclei were reported: 

 detached and attached. 



The degree of nuclear pleomorphism (loss of the usual elliptical shape of the nucleus) 

 was determined for each slide and coded: 1 (<5% of erythrocytes), 2 (5-50%), or 3 (>50%). 

 Severely pleomorphic nuclei had indentations and/or projections. If projection was greater 

 than about one-fourth the nuclear diameter and terminated in a chromatin mass, it was 

 counted as an attached micronucleus. 



Fish Hepatic Cytochrome P-450 Analyses 



Analyses of cytochrome P-450 enzyme activity were performed on liver microsome 

 samples of individual fish by WHOI. Hepatic microsomes were prepared from pieces of 

 liver that had been frozen on dry ice at the time of collection. Methods for microsome 

 preparation were those in use at the LLNL. In addition, microsomes were prepared from 

 gonad from a selected number of individuals. Samples of microsomes were shipped at 

 various times from the LLNL to WHOI following preparation at Livermore by both LLNL 

 and WHOI staff. 



Cytochromes P-450 and b$. Cytochrome P-450 (extinction coefficient (e) = 91 mM'^cm"^) was 

 measured by sodium dithionite difference spectra of CO-treated samples and cytochrome b5 



content (e = 185 mM^cm'^) was determined from NADH difference spectra as previously 

 described (Stegeman et al, 1979). Each cuvette contained approximately 1 mg microsomal 

 protein per mL. 



Assays for cytochrome b5 were usually carried out prior to cytochrome P-450 analysis 

 using dilutions or concentrations like those above. After obtaining an NADH difference 

 spectrum, NADH was balanced in the cuvettes by combining material in sample and 

 reference cuvettes, and adding more NADH. The mix was then treated with CO, re-divided, 

 and the sample reduced by Na2S204 to determine the cytochrome P-450 content. This 

 procedure not only permitted quantitation of cytochrome b5, but effectively eliminated 

 interference by b5 or hemoglobin in the analysis, thereby permitting quantitation of 

 cytochrome P-420. NADH and CO were balanced in all analyses to permit measurement of 

 putative cytochrome P-420, whether or not cytochrome b$ was analyzed. The degradation or 

 denaturation product of cytochrome P-450, if present, could limit some interpretations. 



Ethoxyresorufin O-deethylase. EROD activity was measured by the spectrophotometric 

 method described by Klotz et al. (1984). This method directly measures product formation, 

 like the fluorometric analysis described by Burke et al. (1985), except that the resorufin is 

 detected by absorbance. The reaction mixture contained 0.1 M Tris-Cl, pH 8.0 with 0.1 M 

 NaCl, 2 uM 7-ethoxyresorufin added in methanol, and approximately 100 ug of microsomal 

 protein in a final volume of 0.5 mL. The reaction was initiated by the addition of 0.5 mM 

 NADPH and run at 26°C. The formation of resorufin (e = 73 mM^cm" 1 ) was followed at 572 

 nm on a Shimadzu UV-260 or a Cary 118C recording spectrophotometer. 



Estradiol 2-Hydroxylase. Estradiol 2-hydroxylase activity was routinely assayed by 3H20 

 release from (2-3H)E2 (Kupfer et al, 1981). Our modified assay (Snowberger and Stegeman, 

 1987) consisted of microsomes (0.1-0.3 mg/mL), 25u.M (2-3H)E2 (16 uCi/uxnol), 1 mM EDTA, 

 and 0.3 mM NADPH in 100 mM sodium phosphate buffer, pH 7.4 in a final volume of 200 uL. 



26 



