Spiked recovery of p,p‘-DDD in eight species averaged 79.5% (85.2% with 
the exclusion of one questionable recovery result). Replicate analysis 
of seven fish samples having detectable p,p'-DDD showed good agreement in 
two cases and less than detectable results in replicate samples in five 
cases (all were within 3 ppb of the method detection limit for this 
compound). In actual samples, p,p'-DDD was detected in 35 of 67 cases, 
with a range and average amount of 1.7-7.8 and 2.8 ppb, respectively. 
These levels are consistent with previous results and with the levels of 
p,p'-DDE reported. 
Spiked recovery of p,p'-DDT in eight species of fish averaged 112%. Of 
the eight fish samples run in replicate, this compound was detected in 
only one (non-replicated) instance. In actual samples, p,p'-DDT was 
found in 17 of 67 examples, with a range and average amount of 1.8-7.5 and 
2.9 ppb, respectively. These levels are close to the method detection 
limit, but are consistent with previous reports and with the levels of 
p,p'-DDE and p,p'-DDD seen in these samples. 
Spiked recovery of RGBs was evaluated using a mixture of seven isomerids 
(dichloro- through octachlorobiphenyl). This task was complicated by the 
significant background of environmental RGB compounds in the samples. 
Gorrection for unspiked background yielded an average recovery for eight 
species of fish of 115%. Replicate fish analysis showed agreement that 
averaged 4.4% RSD. Analysis of actual samples gave detectable RGB 
compounds in 67 of 67 cases, with a range and average sum of 13-456 and 
84.3 ppb, respectively. Estimation of total Aroclor level, based on 
these results, gave a range and average of 19-684 and 125 ppb, 
respectively. These results are in agreement with previous reports. 
Level 2 trace organics analysis . GG/MS analysis of Level 2 fish confirmed 
the presence of RGBs and chloronaphthalene (spiked QG compound), but failed 
to confirm the lower level analytes (e.g. hexachlorobenzene, DDT) seen in 
Level 1 analysis. No other chlorinated xenobiotic agents were identified 
from these samples, with an estimated detection threshold of 1-10 ppb. 
Re-analysis of the Level 2 samples by GG/EGD failed to detedt any of the 
following pesticides (above an estimated detection limit of 1 ppb wet 
weight): (alpha, beta, gamma, delta)-BHG, aldrin, heptachlor epoxide, 
gamma-chlordane, dieldrin, endrin, beta-endosulfan, endrin aldehyde, 
endosulfan sulfate, methoxychlor, mirex. A chromatographic peak at the 
correct retention time for heptachlor was observed in several samples in 
amounts equivalent to 1.6-11.5 ppb. None of these results were confirmable 
by mass spectrometry, although the highest samples were above the nominal 
instrument detection limit. It is currently believed that this peak was 
an interferent. The results of these analyses are shown in Table 4. 
Rolynuclear aromatic hydrocarbons were detected in two assays: HRLG/UV 
absorbance/fluorescence and GG/MS. The results for both are presented in 
Table 4. In the present study, these methods should be viewed more as 
complementary than comparable, since fluorescence and absorbance methods 
provided more sensitive detection of the key 5-ring RAH compounds than did 
126 
