collected in the summer. Body size and AHH activity were usually not significantly 

 correlated. AHH activity and hepatic P-450 content were significantly correlated. Fish 

 that were fed oil and /or PCB showed significant increases in AHH activity relative to 

 controls. Fish caught in areas near sources of contaminants or near natural oil seeps always 

 had higher AHH activities than those caught in Monterey Bay, a relatively 

 uncontaminated embayment. 



In conclusion, it appears that AHH activity was less sensitive than some of the other 

 measures, that within-site variability was relatively moderate, that the measure is 

 responsive to exposure to hydrocarbons, and that it may be influenced by gender, species, 

 stage of gametogenesis, and seasons. Fish exposed to hydrocarbons often demonstrate a 

 distinct difference in activity relative to fish from uncontaminated areas. AHH 

 measurements complement other measures of the cytochrome P-450 system and are often 

 correlated with them. 



Reproductive Success. Since all steps in steroid synthesis and metabolism are regulated by 

 MFO enzymes, the potential exists in any of these steps to interfere with the proper 

 quantities or timing of synthesis or metabolism of steroid hormones. Interference may be 

 expressed as measures of the concentrations of precursor hormones or the concentrations of the 

 hormones themselves or of the percent of reproductive products that survive to various stages 

 in development. In the present evaluation, the concentrations of two steroid hormones were 

 measured and selected fish were spawned to determine fertilization, hatching and 

 embryological success. The measures of steroid hormone content and fertilization/ 

 hatching/embryological success are important indicators of reproductive condition and 

 success. They have been used successfully in other studies. However, in this evaluation the 

 steroid hormone measures were relatively insensitive and the sample size available for the 

 fertilization/hatching/ embryological success tests was too small to provide useful data. 



Summary Comparative Evaluation. 



Each of the biological tests is compared in a subjective rating matrix in Table 43. The 

 biological end-points are compared with five criteria for which data were collected in the 

 present evaluation. These five criteria are a subset of the original eight that were initially 

 used to select the biological measures for evaluation. Sediment profiling photography was 

 omitted, since it was performed to determine geographic patterns in the San Francisco Bay 

 estuary, not to compare its performance with the other tests. The sample sizes for the 

 reproductive success end-points were too small to evaluate their performance and the benthos 

 analyses are incomplete. 



The 'sensitive' criterion reflected the ability of the biological test to determine 

 differences either between a test sample and respective controls (sediment tests) or between 

 two sampling sites (fish tests). The sediment toxicity tests were rated a "yes" for sensitive 

 if the end-point indicated that one or more of the samples was significantly more toxic than 

 respective controls (Table 13). The measures of fish health were rated a "yes" for sensitive 

 if one or more of the sites were indicated as significantly different than other sites (Table 

 32). Both the sediment and fish tests rated a "yes" for the "correlated with toxicants" 

 criterion of the rank correlations with toxic chemicals were equal to or exceeded .500 in 

 Tables 19 and 39, respectively. Similarly, both sediment and fish tests generally rated a 

 "yes" for the "not correlated with nuisance variables" criterion if the ranked correlations 

 were less than 500 in Tables 19 and 35, respectively. The sediment and fish tests rated "yes" 

 for the "low analytical variability" criterion if the average CVs were <30 percent (Table 

 17) and <75 percent (Table 39), respectively. The sediment and fish tests rated a "yes" for 

 the "high discriminatory power" criterion if the quotients of the range over the average SD 

 were >3.0 (Table 17) and >2.0 (Table 39), respectively. All of these yes/no thresholds were 

 arbitrarily selected and the ratings for many of the tests could change if different values 

 had been selected. 



Among the sediment toxicity tests, all but the end-points of avoidance by the two 

 amphipods were sensitive to at least one of the samples tested. R. abronius survival, and 

 M. edulis abnormal development and survival were the most sensitive end-points. The 



93 



