TCP in adipose tissue and livers of seals from the Baltic. There is little ... (1:l v/v). The sample intake was l-5 g for all cod liver, eel and seal ... octane as a keeper.
Environmental
Pollution, Vol. 93, No. 1, pp. 39-47, 1996
Copyright 0 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0269-7491/96 $15.M)+O.OO
Pll:SO269-7491(96)00017-6
DETERMINATION OF TRIS(4-CHLOROPHENYL)METHANOL AND TRIS(4-CHLOROPHENYL)METHANE IN FISH, MARINE MAMMALS AND SEDIMENT
Jacob de Boer,= Peter G. Wester,a Erik H. G. Ever@ & Udo A. Th. Brinkman’ “DLO-Netherlands Institute for Fisheries Research, P.O. Box 68, 1970 AB IJmuiden, The Netherlands hNational Institute for Coastal and Marine Management, P.O. Box 20907, 2500 EX Den Haag, The Netherlank cFree University, Department of Analytical Chemistry, De Boelelaan 1083, 1081 HV Amsterdam The Netherlands
(Received 24 March 1995; accepted 18 January 1996)
Until now data on the presence of TCP and TCPMe in biota and sediment from the North Sea and Dutch rivers were not available. In the present paper a method was developed for the determination of TCP and a screening was carried out in cod liver, mussels and marine mammals from the North Sea and Dutch Wadden Sea, and in eel and sediment from Dutch rivers and the Yssel Lake.
Abstract Tris(I-chlorophenyl)methanol (TCP) and tris(I-chlorophenyl)methane (TCPMe) were determined in aquatic organisms and sediment by a method based on Soxhlet extraction, gel permeation chromatography, fractionation over silica and gas chromatography/mass spectrometry (GCIMS) analysis. TCPMe was identiJied as the 4-substituted isomer after synthesis of this compound. TCP could be determined by GC/MS with negative chemical ionistation (GCINCI-MS) with a detection limit of 0.02 g kg-’ and a recovery of 90%. TCP concentrations in marine mammals,from the North Sea and Dutch Wadden Sea ranged from 0.2 to 2 mg kg-‘, and those in marine andfreshwaterfish samples from 0.005 to 0.4 mg kg-’ on a lipid wt basis. TCP concentrations in two Rhine delta sediment samples were 1.2 and 3.0 pg kg-’ dry wt, respectively. TCPMe concentrations, determined by GC/MS with electron impact (GCIEI-MS), were lo-50% of the TCP concentration in all samples analysed. Copyright (0 1996 Elsevier Science Ltd
MATERIALS AND METHODS The samples used for this study were selected to carry out a screening on the presence of TCP and TCPMe in Dutch coastal waters and freshwater. Cod liver and eel were selected because of their high lipid content, which facilitates the analysis. Mussels were included as an example of a lean biological analysis, lower in the food chain, while some marine mammals were included to assess a possible biomagnification of TCP and TCPMe. Two sediment samples from the river Rhine estuary were included in this screening. Marine sediments were not included, because TCP and TCPMe levels were expected to be below detection limits. Marine mammal samples, common seals, different types of dolphins and a harbour porpoise, were obtained from Dr P. Reijnders, DLO-Institute for Forestry and Nature Research, Den Burg, Texel, The Netherlands and Dr R. A. Kastelein, Marine Mammal Park, Harderwijk, The Netherlands. All samples of these animals were taken within hours after their death. Fish, cod (Gadus morhua), yellow eel (Anguilla anguilla) and mussel (Mytilus edulis) samples consisted of a number of individual specimens from which equal quantities of the required tissue were pooled. Sediment samples consisted of some 10 sub-samples taken at a depth of O-l 5 cm from within an area of 1000 m2. Sample data are given in Table 1 and sample locations are shown in Fig. 2. TCPMe was synthesized and TCP (Lancaster Synthesis GmbH, Miillheim/Main, Germany) purified by Dr Meinema, TNO-TU Delft, Delft, The Netherlands. The purity of TCP was > 99% and that of TCPMe 97%. All
INTRODUCTION The presence of tris(4-chlorophenyl)methanol (TCP) (Fig. 1) in environmental samples from various parts of the world has been reported by several authors. Walker et al. (1989) reported the presence of TCP in seals from Puget Sound, Northwest USA. Jarman et al. (1992) also reported its presence in various birds and marine mammals from the Arctic and Antarctic, Australia and the USA, and Zook et al. (1992) reported the presence of TCP in adipose tissue and livers of seals from the Baltic. There is little information available on sources of TCP-possibly, it is a metabolite of tris(Cchlorophenyl)methane (TCPMe) or tris(4-chlorophenyl)methylchloride (TCPC), which are used in the production of dyes (Michaels & Lewis, 1985, 1986; Jarman et al., 1992). DDT may also be a potential source of TCP and TCPMe (Buser, 1995). Jarman et al. (1992), Zook et al. (1992) and Buser (1995) also reported the presence of TCPMe in the animals studied 39
40
J. de Boer et al.
solvents used were of nanograde quality and were obtained from Promochem, Wesel, Germany. Both the biological samples and the sediments were dried with Na2S04, which had been heated for 24 h at 400°C. After drying for 6 h they were Soxhlet extracted for 6 h with n-pentane/dichloromethane (1:l v/v). The sample intake was l-5 g for all cod liver, eel and seal samples, and around 40 g for mussels and sediments. Therefore, the mussels and sediments were transferred to larger Soxhlet thimbles and extracted for 12 h. The sediment extracts were desulphurized by shaking with tetrabutylammonium sulphite (Jensen et al., 1977). Gel permeation chromatography (GPC) was carried out on S-X3 Bio beads (column length 33 cm, internal diameter 2 cm). Dichloromethane/hexane (1: 1, v/v) was used as a solvent. By application of a low nitrogen pressure (0.5 bar) during elution of the TCP fraction, the elution rate was accelerated to 10 ml mini. TCP and TCPMe eluted between 70 and 150 ml, whereas 99% of the lipids eluted before 70 ml. The GPC step
was repeated to remove the lipids which remained in the extract after the first elution. The GPC eluate was concentrated on a rotary evaporator after adding 2 ml isooctane as a keeper. Final concentration to 2 ml took place under a gentle nitrogen stream. A fractionation was subsequently carried out on 1.8 g SiO2.2% HZ0 columns to separate TCP and TCPMe, together with the chlorinated pesticides, from the PCBs. TCP and TCPMe eluted in the second fraction of 10 ml diethylether/iso-octane (1585 v/v), after a first 11 ml iso-octane fraction which contained all PCBs. TCP and TCPMe were quantified using a multiple-level calibration with five different concentrations (range: TCP (NCI) 15690 ng ml-‘, TCPMe (EI) 30-3000 ng ml-‘). A standard solution was submitted to the entire extraction and clean-up procedure. The data calculated by the multiple level calibration method were not corrected for the recoveries of TCP and TCPMe in the standard solution, which were normally in the range of 88-100%.
b
a
Fig. 1. Structures of (a) TCP and (b) TCPMe. Table 1. Sample data Species
Tissue
Location
Year
Number”
Age (year)
Sex
Common seal (1) (Phocu
blubber
Wadden Sea
1990
1
2
F
blubber blubber
Wadden Sea Wadden Sea
1992 1992
I 1
0.01 5
h F
blubber blubber
Central North Sea Wadden Sea
1989 1992
I 1
d e
G
blubber blubber
Dutch coast Dutch coast
1990 1990
1 1
10 > 10
M M
liver liver whole mussel muscle muscle muscle muscle muscle muscle
Dutch coast Northern North Sea Wadden Sea Rhine, Lobith Hollands Diep Roer, Vlodrop Yssel Lake, Urk Haringvliet-east Niers Haringvliet-east Nieuwe Merwede
1993 1992 1994 1994 1994 1994 1994 1994 1994 1992 1992
25 25 200 25 25 25 25 25 25
ndf nd nd nd nd nd nd nd nd
nd nd nd nd nd nd nd nd nd
vitulina)
Common seal (2) Common dolphin (Delphinus delphis)
Dolphin’ Whitebeaked dolphin (Lagenorynchus albirostris)
Whitebeaked dolphin Harbour porpoise (Phocaena phocaena)
Cod (Gadus morhua) Cod Mussels (Mytilus edulis) Yellow eel (Anguillu anguillu) Yellow eel Yellow eel Yellow eel Yellow eel Yellow eel Sediment Sediment
-
a If > 1, then pooled sample. bSex unknown. cPrecise species unknown. dAge unknown, length 220 cm. ‘Age unknown. cm, weight ca 85 kg. fNot determined.
length 216
Determination of TCP and TCPMe in$sh, marine mammals and sediment TCP and TCPMe were determined by GC/ECD (GC, with electron capture detection), GC/EI-MS and GC/NCI-MS-the conditions are given in Table 2. 1,2,3,4-Tetrachloronaphthalene (TCN) was used as an internal standard. The detection limits obtained by the three different methods are given in Table 3. The recovery of a TCP spike to a seal blubber sample (306 ng TCP in 1 ml iso-octane added to 1 g seal blubber, ground with 15 g Na2S04) which was submitted to the entire method was 90%. The reproducibility of the method was determined by the replicate determination of TCP and TCPMe in a sample. The relative standard deviation for the determination of TCP (NCI-MS) was 12% (n = 5) and of TCPMe (EI-MS) 22% (n = 6). The lipid contents in the marine mammal and yellow eel samples were extractable lipid contents, determined after Soxhlet extraction. The lipid contents in the cod liver and mussels are total lipid contents, determined according to a modified Bligh and Dyer method, based on a chloroform/methanol extraction (Bligh & Dyer, 1959; de Boer, 1988). The organic carbon contents in the sediment samples were determined by a titration after oxidation with dichromate, according to the method of Mebius (1960).
RESULTS
AND DISCUSSION
Analysis
A synthesized TCPMe standard was used for the first time. TCPMe was indeed identified in the samples as the
4’1
501
Fig. 2. Sample locations in The 2. Dutch coast, 3. Yssel Lake Nieuwe Merwede, 6. Hollands Niers, 9. Roer
’ 69
7’1EL
Netherlands: 1. Wadden Sea, (Urk), 4. Rhine (Lobith), 5. Diep, 7. Haringvliet-east, 8. (Vlodrop).
41
with chlorine substitution in the para position, as was expected, based on the similarity with TCP before its synthesis. The use of florisil, applied by several authors (Walker et al., 1989; Jarman et al., 1992; Zook et al., 1992) requires an extensive pretreatment and, in addition, florisil is only stable for a short period of time. Therefore, alumina column elution (de Boer, 1988) and treatment of the extract with sulphuric acid (Wester et al., in press) were tested as alternative methods for the separation of lipids and TCP. Initial experiments showed that it was not possible to elute TCP from an alumina column without co-eluting the lipids present in the extract after Soxhlet extraction. Furthermore, TCP was not stable when mixed with sulphuric acid. Finally, GPC was tested to effect the separation of TCP and lipids. Lipid determinations in the TCP/TCPMe fractions obtained after a repeated GPC elution showed that these fractions did not contain any significant amount of lipids ( < 0.01%). Although the necessary repeated GPC clean-up is somewhat time consuming, this method enables TCP and TCPMe to be separated completely from the lipids. Figure 3 shows NC1 and EI mass spectra of both TCP and TCPMe. Clear differences between the spectra can be observed with much more fragmentation in the EI spectrum than in the NC1 spectrum. The ions preferred for quantification were 139 (EI) and 362 (NCI) for TCP, and 311 (ET) and 346 (NCI) for TCPMe. As is to be expected, the most prominent peaks in the NC1 mass spectra are due to the molecular ions. These also show up in the EI spectra, but the main peaks can now be attributed to [M2PhCl-OH] + in the case of TCP (m/z 139) and to [M-Cl]+ with TCPMe (m/z 311). Generally, compounds with less than four chlorine atoms are rather difficult to determine in environmental samples by means of GC/NCI-MS, because of too low a sensitivity (de Boer, 1995). In the present instance the electronegativity of the oxygen apparently helps to increase sensitivity and, thus, to create good analyte detectability for TCP. GC/NCIMS clearly is not the method of choice for TCPMe because of low sensitivity (Table 3). This confirms the role of the oxygen atom observed with TCP. Both GC/EI-MS and GC/ECD are suitable alternatives, but MS-based detection is preferred because of its higher selectivity. GC/EI-MS can in principle also be used for the determination of TCP, with m/z 139 to be used for quantification. However, quantification on such a low mass is easily disturbed by interferences of mass fragments of other compounds. This is clearly demonstrated by comparing EI and NC1 results for TCP in a variety of samples. Depending on the sample, results up to 10 times too high were obtained by GC/EI-MS (Table 4). Such problems are not normally encountered in the GC/ EI-MS determination of TCPMe for which the relatively high m/z value of 311 is used. The different sensitivity isomer
(m/z)
(m/z)
Ions used for quantification
3 min 90°C 30°C min-215°C 40 min 215°C 5°C min’ -270°C 35 min 270°C
139, 141, 251, 362, 364
1.10-6 3 min 90°C 30°C min-215°C 40 min 2 15°C 5°C min-‘-270°C 35 min 270°C 111, 139, 141, 251, 362, 364 (TCP) (200 ms, 4&44 min) 119, 165, 199, 235, 311, 313, 346 (TCPMe) (250 ms, 3640 min) 139, 141, 251, 362, 364 (TCP) 3 11, 313, 346, 348 (TCPMe)
1.10-6 3 min 9O”C, 30°C min-210°C 10 min 210°C 5°C min-290°C 30 min 290°C 111, 139, 141, 251, 362, 364 (200 ms, 4&44 min)
Ions used for identification
200 280 120175 -
200 280 120175 -
CP Sil 8 50 0.15 0.30 1 44 280 360
H2
splitless 3
HP 5988A 70 splitless 3 He CP Sil 8 50 0.21 0.20 3 20 285
HP 5988A 70 splitless 3 He CP Sil 12 45 0.25 0.20 3 28 285
Elmer 8500
Perkin
Instrument Ionization voltage (eV) Injection method Splitless time (min) Carrier gas Column Length (m) Internal diameter (mm) Film thickness (pm) Injection volume (~1) Linear gas velocity (cm s-i) Injector temperature (“C) Detector temperature (“C) Source temperature (“C) Transferline temperature (“C) Analyser temperature (“C) Methane pressure plasma (torr) Source pressure (torr) Temperature programme oven
+ TCPMe)
GC/EI-MS(TCP
GC/EI-MS(TCP)
of TCP
GC/ECD(TCP)
and GC/MS
Method
Table 2. Conditions for GC/ECD
362, 346, 348
100 280 120175 1 2.10-5 3 min 9O”C, 30°C min-‘-210°C 10 min 210°C 5°C min-‘-290°C 30 min 290°C 362, 364, 366 (250 ms, 4W4 min)
HP 5988A 200 splitless 3 He CP Sil 12 45 0.25 0.20 3 28 285 -
GC/NCI-MS(TCP)
R
43
Determination of TCP and TCPMe in fish, marine mammals and sediment
mammals and cod liver and mussels, therefore, suggest a strong biomagnification of TCP. The number of samples analysed in this study was relatively small. Therefore, it is difficult to say something about the natural variation of TCP concentrations in marine mammals. However, data of other halogenated micro-contaminants, such as PCBs, in marine mammals show that the natural variation can be relatively large (Duinker et al., 1989). The range of TCP concentrations found in the marine mammals (Table 5) therefore indicates that significant differences in biomagnification between the different species are unlikely. The concentrations of TCP and TCPMe in fish, which are often in the ratio of (22lO):l, seem to be related to industrial activity and human population density, with higher concentrations in river Rhine fish and cod liver from the southern North Sea, and lower concentrations in northern North Sea cod liver. This is confirmed by the only dolphin sample from the central North Sea, which shows lower TCP concentrations than the other marine mammal samples. The TCP and TCPMe concentrations in eel from the river Roer are relatively low, indicating that there is presumably no relationship between mining activities-the river Roer flows through an area with mining industries-and TCP. The low TCP and TCPMe concentrations in eel from the river Niers, which originates in Germany, confirm the hypothesis about the relationship between high TCP/TCPMe concentrations and densely populated, industrialized areas. All other eel samples were taken at locations directly influenced by the river Rhine. The lowest TCP and TCPMe concentrations were found in the mussels. This is explained by the low fat content of the mussels and by the lower position of mussels in the food-chain. Contamination of biota and sediment in the Rhine estuary with, for example, PCBs and organochlorine pesticides has remained essentially constant over the last three years (de Boer, 1995). Although this may be different with regard to TCP and TCPMe, a comparison of the sediment samples from 1992 with the eel samples from 1994 from the same area may at least give an indication of the distribution of TCP and TCPMe in the
of TCP and TCPMe for the two MS techniques necessitates replicate injections of the extract containing TCP and TCPMe. Even though it is time consuming, it results in a sensitive and selective analysis of both compounds. The use of ECD detection only causes a small loss in sensitivity (Table 3) but a considerable loss in selectivity. Therefore, ECD can be used for screening of TCP and TCPMe (Fig. 4), however, confirmation by GC/NCIMS (Fig. 5) or GC/EI-MS (Fig. 6), respectively, is essential. The differences in selectivity between GC/ ECD and GC/EI-MS are obvious when comparing Figs 4 and 6. Concentrations of TCP and TCPMe The results of the determination of TCP and TCPMe in biota and sediment samples are given in Table 5. The TCP concentrations in seal blubber are similar to those reported by Zook et al. (1992) in Baltic seal livers (up to 3 mg kg-’ lipid wt). TCP concentrations in blubber of seals from Puget Sound, US, were distinctly lower (233750 pg kg-’ lipid wt) (Walker et al., 1989). TCP was also determined in cod liver oil by Rahman et al. (1993). In that sample, also originating from the southern North Sea, a comparable TCP concentration (51 pg kg-’ on a lipid wt basis) was reported. The same authors also reported the presence of TCP and TCPMe in human milk at a level of 2.5 and 1.6 pg kg-’ on a lipid wt basis, respectively. A considerable bioconcentration of TCP and TCPMe is expected on the basis of the octanol-water partition coefficients calculated by the C log P method (Hansch & Leo, 1979). Log K,, for TCP is 6.0 and for TCPMe is 6.5. The TCP concentrations in cod liver are comparable to those reported for HCB and p,p’-DDT in the same matrix (de Boer, 1989). Cod liver and mussels are not the best representatives of average food of seals and dolphin. The TCP and TCPMe concentrations in cod liver and mussels on a lipid wt basis will, however, not differ essentially from those in herring or mackerel, as is for example observed for PCB concentrations in these organisms (de Boer et al., 1993). The orders of magnitude difference in the TCP concentrations on a lipid wt basis between marine
Table 3. Detection limits and retention times of TCP and TCPME
Detection limit/ retention time Detection limit (Pg) Detection limit (pg kg-‘) Retention time (min) Retention time relative to TCN
Method: GC/ECD Column: CP-Sil 8 TCP TCPMe 10
10
Method: GC/EI-MS Column: CP-Sil 12/8” TCP TCPMe 30
0.1
0.1
0.2
74.46
67.01
3.22
2.89
a TCP on CP-Sil 12 column, TCPMe on CP-Sil 8 column.
10
Method: GC/NCI-MS Column: CP-Sil 12/V
TCP 3
TCPMe 100
0.07
0.02
0.7
33.46
37.35
33.46
37.35
2.26
1.69
2.26
1.69
44
J. de Boer et al. [M-2PhCCOH]+
a
139 I
[M2PhCI-H]+ [MPhCI]+
111
\
60
2T1
120
160
200
240
260
320
360
b
[M-O] 7 i' l'yi" ~,...,.,,,,,.,,,.,,,,,' 60 120
l/51
_f\,l;" 190
160
, 200
,T'.
,T;, 240
276 2/93 3/16 T"',"','.',"',"',A.,',"' 260 320
346 \ 360
C
[M-PhCI-2CI]+ 166
00
120
160
200
240
260
320
360
Fig. 3. Mass spectra of TCP in (a) the EI and (b) the NC1 mode, and of TCPMe in (c) the EI and (d) the NC1 mode.
Determination of TCP and TCPMe in,fish, marine mammals and sediment Table 4. Differencesin results of TCP determinatiow in various samples by CC/ELMS (m/z 139) and GC/NCI-MS (m/z 362)
and sediment. When comparing the concentrations in fish expressed on a lipid wt basis with those in sediment expressed on a organic carbon basis, the fish/ sediment ratios are in the range 1.1-14 for TCP and 2.8-13 for TCPMe (eel from the Hollands Diep and Haringvliet-east compared to sediment from the Haringvliet-east and the Nieuwe Merwede, see Fig. 2). These ratios roughly correspond with those found for PCBs in the same area (ratio l-5, de Boer, 1995). They are much higher than those for polychlorinated terphenyls (PCTs), ratios 0.2-1, in the same area (Wester et al., 1996). These data suggest that the adsorption of TCP and TCPMe to sediments is comparable to, or somewhat lower than. that of PCBs.
biota
TCP concentration @g kg-‘) EI-MS NCI-MS
Samnle
Cod liver, northern North Sea Yellow eel, Roer Yellow eel, Niers Sediment, Haringvliet-east Sediment, Nieuwe Merwede
32.6
3.0
5.2 2.0 0.6 1.3
1.2 2.2 0.7 1.9
-
45
CONCLUSIONS
In the present screening programme, TCP and TCPMe have been detected and quantified in marine mammals, mussels and cod liver from the North Sea and the Dutch Wadden Sea, and in yellow eel and sediment from Dutch rivers and the Yssel Lake. The analytes were determined by a procedure consisting of Soxhlet extraction, desulphurizing (sediments), repeated GPC, silica fractionation and GC/NCI-MS (TCP, detection limit 0.02 ,ug kg-‘) or GC/EI-MS (TCPMe, detection limit 0.07 pg kg-‘).
Fig. 4. GC/ECD chromatogram of a whitebeaked dolphin extract (Dutch coast) on a CP Sil 8 column (conditions: cf. Table 2).
CB 112
TCP
+ Fig. 5. GC/NCI-MS
t
R
(min.)
trace (SIM) of TCP in a whitebeaked dolphin extract (Dutch coast) (TCPMe below detection limit).
TCP
TCN
TCPMe LL
22
30 -
Fig. 6. GC/EI-MS
40 t
R
(min.)
trace (SIM) of TCP and TCPMe in a whitebeaked dolphin extract (Dutch coast).
46
J. de Boer et al. Table 5. TCP (GC/NCI-MS) and TCPMe (GC/EI-MS) concentrations in biota and sediment samples
Sample
Concentration TCP
Common seal blubber (l), Wadden Sea Common seal blubber (2), Wadden Sea Common dolphin blubber, Wadden Sea Dolphin blubber, central North Sea Whitebeaked dolphin blubber, Wadden Sea Whitebeaked dolphin blubber, Dutch coast
760 420 180 160 350 1000
Harbour porpoise blubber, Dutch coast Cod liver, Dutch coast Cod liver, northern North Sea Mussels, Wadden Sea Yellow eel, Rhine
680 18 3.0 0.2 29
Yellow eel, Hollands Diep Yellow eel, Roer Yellow eel, Yssel Lake Yellow eel, Haringvliet-east Yellow eel, Niers Sediment, Haringvliet-east (dry wt 61%) Sediment, Nieuwe Merwede (dry wt 63%)
39 1.2 23 25 2.2 0.7 1.9
,LL~ kg-’ wet wtn TCPMe
Concentration TCP
nd’ nd nd nd nd
nd nd nd 0.9 co.1 nd nd 0.2 2.2 5.1 0.8 0.1 0.4
2000 750 220 190 570 1400 1000 40 5.7 13 310 360 IO 82 180 12 26 160
pg kg-’ lipid wth TCPMe nd nd nd nd nd nd nd nd 1.7 16 nd nd 1.6 7.9 37 5.8 3.7 13
“In sediment: /lg kg-’ dry weight. Qr sediment: pg kg-’ total organic carbon. ‘Not determined.
A good separation between lipids and TCP and was obtained. Further improvements may be possible by replacing the repeated GPC clean-up by a single stage GPC elution in the future, using higher pressures and different packing materials. Automation of the clean-up would then also become possible. TCPMe was synthesized and identified here for the first time as the isomer with chlorine substitution in the three para positions. TCP concentrations in marine mammals varied between 0.2 and 2 mg kg-’ on a lipid wt basis. TCP concentrations in fish varied from 6 to 360 pg kg-’ on a lipid wt basis-TCPMe concentrations were 2-10 times lower. The sediment data suggest that the adsorption of TCP and TCPMe to sediments is comparable to, or somewhat lower than, that of PCBs. The several orders of magnitude differences between TCP concentrations in cod liver and mussels and those in marine mammals suggest a strong biomagnification of TCP. No data are available on the toxicity of TCP. Its structure rather closely resembles that of malachite green, a well-known carcinogenic dye which easily forms radicals. The presence of a compound resembling such a carcinogenic substance in fish could have serious consequences, therefore, more knowledge on the toxicity and especially on the carcinogenic potency of TCP should become available. It should also be evident that there is a need for further research concerning TCP and TCPMe levels in the aquatic environment.
TCPMe
ACKNOWLEDGEMENTS
The authors kindly acknowledge Dr W. Jarrnan, Joseph M. Long Marine Laboratory, Santa Cruz, California, USA, who made the TCP standard available. Dr P.
Reijnders, DLO-Institute for Forestry and Nature Management, Den Burg Texel, The Netherlands, and Dr R. A. Kastelein, Marine Mammal Park, Harderwijk, The Netherlands are kindly acknowledged for their gift of several marine mammal samples. Dr J. C. Klamer, National Institute for Coastal and Marine Management, The Hague, The Netherlands, is acknowledged for the calculation of the C log P values.
REFERENCES Bligh, E. G. & Dyer, W. J. (1959). A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol., 37, 91 I-917. de Boer, J. (1988). Chlorobiphenyls in bound and non-bound lipids of fishes: comparison of different extraction methods. Chemosphere, 17, 1803-I 8 10.
de Boer, J. (1989). Organochlorine compounds and bromodiphenylethers in livers of Atlantic cod (Gadus morhua) from the North Sea, 1977-1987. Chemosphere, 18,2131-2140. de Boer, J. (1995). Analysis and biomonitoring of complex mixtures of persistent halogenated micro-contaminants. PhD Thesis, Free University, Amsterdam, The Netherlands. de Boer, J., Stronck, C. J. N., Traag, W. A. & van der Meer, J. (1993). Non-ortho and mono-ortho substituted chlorobiphenyls and chlorinated dibenzo-p-dioxins and dibenzofurans in marine and frehwater fish and shellfish from The Netherlands. Chemosphere, 26, 1823-l 842. Buser, H. R. (1995). DDT, a potential source of environtris(4-chlorophenyl)methane and tris(4-chloromental phenyl)methanol. Environ. Sci. Technol., 29, 2133-2139. Duinker, J. C., Hillebrand, M. T., Zeinstra, T. & Boon, J. P. (1989). Individual chlorobiphenyls and pesticides in tissues of some cetacean species from the North Sea and the Atlantic Ocean: tissue distribution and biotransformation. Aquat. Mamm., 15,95-124.
Hansch, C. & Leo, A. J. (1979). Substituent Constants for Correlation Analysis in Chemistry and Biology. Wiley, New York, USA.
Determination of TCP and TCPMe injsh, Jarman, W. M., Simon, M., Norstrom, R. J., Bruns, S. A., Bacon, C. A., Simaret, B. R. T. & R&borough, R. W. (1992). Global distribution of tris(4-chlorophenyl)methanol in high trophic level birds and mammals. Environ. Sci. Technol., 26, 1770-1774. Jensen, S., Renberg, L. & Reutergird, L. (1977). Residue analysis of sediment and sewage sludge for organochlorines in the presence of elemental sulfur. Anal. Chem., 49, 316318. Mebius, L. J. (1960). A rapid determination of organic carbon in soil. Anal. Chim. Acta, 22, 12&124. Michaels, G. B. & Lewis, D. L. (1985). Sorption and toxicity of azo and triphenylmethane dyes to aquatic microbial populations. Environ. Toxicol. Chem., 4, 45-50. Michaels, G. B. & Lewis, D. L. (1986). Microbial transformation rates of azo and triphenylmethane dyes. Environ. Toxicol. Chem., 5, 161-166.
marine mammals and sediment
47
Rahman, M. S., Montanarella, L., Johansson, B. & Larsen, B. (1993). Trace levels of tris(4-chlorophenyl)methanol and methane in human milk. Chemosphere, 27, 1487-1497. Walker, W., Riseborough, R. W., Jarman, W. M., Lappe, B. W., Lappe, J. A., Tefft, J. A. & de Long, R. L. (1989). Identification of tris(chlorophenylmethano1) in blubber of harbour seals from Puget Sound. Chemosphere, 18, 1799-1804. Wester, P. G., de Boer, J. & Brinkman, U. A. Th. (1996). Determination of polychlorinated terphenyls in aquatic biota and sediment with gas chromatography/mass spectrometry using negative chemical ionisation. Environ. Sci. Technol., 30,473480. Zook, D. R., Buser, H. R., Bergqvist, P. A., Rappe, C. & Olsson, M. (1992). Detection of tris(chlorophenyl)methane and tris(4-ch1orophenyl)methanol in ringed seal (Phoca hispi&) from the Baltic Sea. Ambio, 21, 557-560.
