STOUT: ORGANOCHLORINE RESIDUES IN FISHES 



the DDE. All of the TDE and DDT eluted into the 

 benzene fraction. For dieldrin and endrin 

 analysis, tissues were saponified with KOH in 

 aqueous ethanol, extracted with hexane, and 

 cleaned up on Florisil. Since the specimens, origi- 

 nally collected for trace-metal analysis, were 

 stored in polypropylene containers (Falcon No. 

 4014) with polyethylene lids (Falcon No. 4017), 

 the containers and lids were also analyzed to as- 

 sure freedom from interfering substances. The de- 

 tails of our procedures were published previously 

 (Stout and Beezhold 1979). Blanks carried 

 through the whole procedure for either PCB and 

 SDDT or dieldrin and endrin showed no 

 chromatographic peaks which interfered with 

 quantitation of the chlorinated hydrocarbons. 



The extracts of the fishes were quantitated by 

 electron-capture gas chromatography. A Varian 

 600 C gas chromatograph with a titanium tritide 

 detector was fitted with a 1.5 m (5 ft) x 0.32 cm 

 (0.125 in) o.d. glass column containing a mixture 

 of equal parts of 15% QF-1 on 80-100 mesh Gas 

 Chrom Q and 10% DC-200 on the same support 

 (Burke and Holswade 1966). Reference standards 

 of p,p'-DDE, p,p'-DDMU, p,p'-TDE, p,p'-DDT, 

 dieldrin, and endrin were obtained from the U.S. 

 Environmental Protection Agency, Health Effects 

 Research Laboratory, Research Triangle Park, 

 N.C. Aroclor 1254 obtained from the Monsanto 

 Company, St. Louis, Mo., was the standard for the 

 PCB because the residues in the fishes matched 

 this Aroclor most closely. Standard curves of peak 

 height versus concentration were used to deter- 

 mine the concentrations of components in the ex- 

 tracts. The sensitivity throughout each run was 

 assured by frequent injections of standard solu- 

 tions. The quantifiable limit was about 0.003 ppm 

 for DDE, TDE, DDT, dieldrin, and endrin, and 

 about 0.04 ppm for PCB depending on daily sen- 

 sitivity. The mean relative standard deviation for 

 samples analyzed in duplicate was 11%. The aver- 

 age recovery of samples spiked with standards was 

 85%. The values reported were not corrected for 

 recovery. Residue values were calculated on the 

 basis of micrograms chlorinated hydrocarbon per 

 gram wet tissue or parts per million (ppm). 



The PCB were quantitated by summing the 

 peak heights corresponding to the five major peaks 

 in Aroclor 1254 after omitting the twin peak with 

 a retention time similar to that of p,p'-DDE. As 

 Veith ( 1975) also concluded, use of five peaks in- 

 creased the accuracy of PCB measurement by 

 reducing the effect of minor variations in concen- 



tration of individual components in the samples 

 (Figure 1). 



Since part of the DDE eluted from silica gel with 

 the PCB fraction and one of the major peaks in 

 Aroclor 1254 overlapped the DDE peak in the 

 gas-chromatographic traces, a special technique 

 was needed to quantitate the DDE in the PCB 

 fraction. First the gross DDE concentration was 

 determined in the usual way from the peak height 

 versus DDE concentration curve. Next, the con- 

 tribution of PCB to that overlapping peak was 

 calculated based on the assumption that the 

 height of Aroclor peak 3, the peak which over- 

 lapped DDE, was proportional to the heights of the 

 five PCB peaks used to calculate the concentration 

 of PCB. To make this calculation, the sum of the 

 peak heights of the five major PCB peaks exclud- 

 ing the "DDE" peak was plotted against the peak 

 height of the "DDE" peak in PCB standards of 

 increasing concentrations. From the sum of the 

 five PCB peaks in the sample, the peak height of 

 the PCB portion of peak 3 in that sample was 

 determined. This peak height was converted to 

 concentration of DDE via the peak height versus 

 concentration curve for DDE. The apparent con- 

 centration of DDE from PCB peak 3 was sub- 

 tracted from the gross DDE concentration in the 

 "DDE" peak to obtain the net concentration of 

 DDE in the PCB fraction. The elctron-capture de- 

 tector is so much more sensitive to DDE than PCB 

 that this method of calculation affected the accu- 

 racy of DDE determination only to a small extent 

 (Figure 1). Use of a minicomputer expedited these 

 calculations. 



Confirmation of DDT and its metabolites and 

 PCB was effected by saponifying separate portions 

 of tissue (Reinert 1970). DDT is dehydrochlori- 

 nated by base to DDE, and TDE similarly to 

 DDMU. PCB are stable to base. The levels of diel- 

 drin and endrin were too low to warrant confirma- 

 tion studies. 



RESULTS AND DISCUSSION 



The marine fishes from the northwestern Atlan- 

 tic Ocean and northern Gulf of Mexico analyzed in 

 this study contained relatively low levels of SDDT 

 and PCB. Of the 70 composite samples, only 29 

 contained more than 0.05 ppm SDDT and 0. 1 ppm 

 PCB in the edible portion (skinless fillets), and 

 only one as much as 1 ppm SDDT and PCB. Red 

 grouper contained the lowest levels of both chlori- 

 nated hydrocarbons. The mean IDDT content for 



53 



