Geographical Distribution (2000) and Temporal ... - ACS Publications

Polybrominated Diphenyl Ethers in Peregrine Falcon (Falco peregrinus) Eggs from ... Da Chen , Mark J. La Guardia , Ellen Harvey , Michael Amaral , Kat...
0 downloads 0 Views 153KB Size
Environ. Sci. Technol. 2002, 36, 4783-4789

Geographical Distribution (2000) and Temporal Trends (1981-2000) of Brominated Diphenyl Ethers in Great Lakes Herring Gull Eggs R O S S J . N O R S T R O M , * ,† M A R Y S I M O N , † JOHN MOISEY,† BRYAN WAKEFORD,† AND D. V. CHIP WESELOH‡ Environment Canada, National Wildlife Research Centre, 100 Gamelin Boulevard, Hull, Quebec K1A 0H3, Canada, and Environment Canada, Canadian Wildlife Service, 4905 Dufferin Street, Downsview, Ontario, M3H 5T4, Canada

Geographical distribution of brominated diphenyl ether (BDE) flame retardants in the North American Great Lakes ecosystem in 2000 was determined by analysis of herring gull eggs (13 egg pools) from a network of 15 monitoring colonies scattered throughout the lakes and connecting channels. ΣBDEs were found at concentrations ranging from 192 to 1400 µg/kg, mean of 662 ( 368 µg/kg (wet weight of egg contents). Highest concentrations were found in northern Lake Michigan and Toronto harbor (1000-1400 µg/ kg) and lowest in Lake Huron and Lake Erie (192-340 µg/ kg). The distribution suggested that input from large urban/ industrial areas through air or water emissions contributes local contamination to the herring gull food web in addition to background levels from regional/global transport. The congener composition was similar among sampling sites. Major congeners were BDE-47 (43%), BDE-99 (26%), BDE-100 (13%) BDE-153 (11%), BDE-154 (4%), BDE183 (2%) and BDE-28 (1%). Temporal trends of BDE contamination, 1981-2000, were established by analysis of archived herring gull eggs (10 egg pools) from colonies in northern Lake Michigan, Saginaw Bay, Lake Huron and eastern Lake Ontario. BDE-47, -99 and -100, and BDE-153, -154 and -183 concentrations were grouped separately for analysis because these two groups had different trends and are primarily associated with the Penta BDE and Octa BDE flame retardant formulations, respectively. ΣBDE47,99,100 concentrations were 5-12 µg/kg (wet weight) in 1981-1983 and then increased exponentially (p < 0.00001) at all three sites to 400-1100 µg/kg over the next 17 years. Doubling times were 2.6 years in Lake Michigan, 3.1 years in Lake Huron and 2.8 years in Lake Ontario. ΣBDE154,153,183 concentrations generally increased but varied in an erratic fashion among sites and decreased as a fraction of ΣBDE over time. Concentrations of ΣBDE154,153,183 were 100-200 µg/kg in eggs from all three colonies in 2000. Therefore, most of the dramatic increases in ΣBDE concentrations observed over the past 20 years in the Great Lakes aquatic ecosystem seem to be connected with the Penta BDE formulation, which is mainly used as a flame retardant in polyurethane foam in North America. If * Corresponding author phone: (819)997-1411; fax (819)953-6612; e-mail: [email protected]. † Environment Canada, National Wildlife Research Centre. ‡ Environment Canada, Canadian Wildlife Service. 10.1021/es025831e CCC: $22.00 Published on Web 10/15/2002

Published 2002 by the Am. Chem. Soc.

present rates of change continue, concentrations of ΣBDEs will equal or surpass those of ΣPCBs in Great Lakes herring gull eggs in 10-15 years.

Introduction Growth in interest in the brominated diphenyl ether (BDE) flame retardants has been as exponential as their apparent increase in the environment over the past 20-25 years. The first indication that they were persistent organic pollutants (POPs) of possible global concern were reports of their presence in fish from Swedish waters in 1981 (1) and harbor seal blubber, ringed seal blubber and guillemot eggs from the Baltic Sea, North Sea and Arctic Ocean in 1987 (2). The major compound in the chromatograms from these samples was a tetrabromodiphenyl ether matching the first peak in the Penta-BDE commercial fire retardant mixture, Bromkal 70-5DE, which we now know to be 2,2′,4,4′-tetrabromodiphenyl ether (BDE-47) (3). Since that time, interest in BDEs as environmental contaminants expanded greatly in Europe, mainly in Sweden and The Netherlands, and in Japan (4, 5). Worldwide interest quickened with a report that concentrations of BDEs in human milk in Sweden had been increasing exponentially (doubling time ) 5 years) from at least the mid-1980s to 1997, concomitantly with a decrease in other halogenated POPs such as PCBs (6). Although there were occasional earlier reports of the presence of BDEs in humans, fish and wildlife in North America (7-9), it was not until 1999 that systematic studies on the presence of BDEs in the North American environment began to be reported (10, 11). There are suggestions that there is also an exponential increase in BDEs in North American breast milk at a faster rate than found in Sweden (12, 13). It seems clear that it is the Penta BDE commercial product that is of most concern, since the majority of the BDE congeners found in both humans and wildlife arise from this mixture (3). The Penta mixture was 24% of total BDE demand in North America in 1999 (14). Although 72% of commercial BDE demand in North America in 1999 was the Deca BDE mixture (14), it is has very low bioavailability, does not bioaccumulate, and is not found in biota (15). The Octa BDE mixture was the least used in North America (4% of total BDE demand) (16). Octa BDE consists mainly of hepta- and octa-BDE congeners which also appear to have low bioaccumulation ability (16). However this mixture may be the major contributor of some of the hexa-BDEs found in the environment (17). The Deca and Octa BDE formulations are used primarily in rigid thermoplastics (computer and TV housings, electronics), while in North America the Penta BDE mixture is almost entirely used in polyurethane foam, much of which ends up in upholstered furniture (16). Both the physical-chemical properties and use patterns favor release of congeners associated with Penta BDE to the environment. To put BDEs into context with other POPs, the ΣBDE concentrations in rural Ontario air in April, 2000 were 881300 pg/m3, while those of ΣPCBs were 96-950 pg/m3 (18). The toxicology of BDEs has not been thoroughly studied (19). Two of the major congeners in the Penta BDE mixture, BDE-47 and BDE-99, have been shown to be developmental neurotoxins in rats (20). The Canadian Wildlife Service has a program to monitor distribution and trends of persistent organic pollutants throughout the Great Lakes by analysis of herring gull (Larus argentatus) eggs on an annual basis since the mid-1970s (21). The herring gull is an ideal biomonitoring species in the VOL. 36, NO. 22, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4783

FIGURE 1. Sites of herring gull colonies in the North American Great Lakes sampled in 2000 for BDE analysis: (1) Granite Island, Lake Superior; (2) Agawa Rock, Lake Superior; (3) Big Sister Island, Green Bay, Lake Michigan; (4) Gull Island, Lake Michigan; (5) Double Island, North Channel, Lake Huron; (6) Chantry Island, Lake Huron; (7) Channel-Shelter Island, Saginaw Bay, Lake Huron; (8) Fighting Island, Detroit River; (9) Middle Island, Lake Erie; (10) Port Colbourne, Lake Erie; (11) Niagara River (above the falls); (12) Hamilton Harbor, Lake Ontario; (13) Leslie St. Spit, Toronto Harbor, Lake Ontario; (14) Snake Island, Lake Ontario; (15) Strachan Island, St. Lawrence River. lower Great Lakes, because the adult population is resident year-round, nests in colonies, and eats primarily alewife (Alosa pseudoharengus) and rainbow smelt (Osmerus mordax), the two most abundant prey fish in the Great Lakes. Because of winter movements from the more ice-bound Lake Superior and northern Lake Huron to the lower lakes, gull eggs from the most northern colonies may be only a partial reflection of contamination of the lake on which the colony is situated in colder years (22). The protocol calls for collection of 1013 randomly selected eggs from each of 15 colonies chosen to represent as wide a geographical range within the Great Lakes as possible. Only southern Lake Michigan, which has no suitable habitat, has no herring gull colonies. Portions of the eggs are archived at -40° as individual homogenates and as equal weight pools (n ) 10-13) for each colony, in the Canadian Wildlife Service Specimen Bank at the National Wildlife Research Centre. To study recent distribution of BDEs in the Great Lakes aquatic ecosystem, pooled homogenates of herring gull eggs collected from fifteen locations on the Great Lakes and connecting channels in 2000 were analyzed. Temporal trends, 1981-2000, were established for colonies in Lake Michigan, Lake Huron and Lake Ontario by analysis of pooled homogenates of eggs retrieved from the Specimen Bank. These colonies were chosen to provide a wide geographical coverage.

Experimental Section The locations of the sampling colonies in 2000 are shown in Figure 1. All samples were pooled (n ) 10-13) on an equal wet weight basis prior to analysis. Homogenates of egg pools (n ) 10-16) from three of the monitoring sites: Gull Island, Lake Michigan; Channel-Shelter Island, Saginaw Bay, Lake Huron and Snake Island, Lake Ontario (Figure 1), which were archived at -40° in the Canadian Wildlife Service Specimen Bank, were retrieved for approximately every other year from 1981 to 1999 and analyzed to study long-term historical trends in BDE contamination at these sites. A 1 g (wet weight) of each egg pool homogenate was ground with 10 g of anhydrous sodium sulfate and poured into a 18 cm long glass column containing 20 mL of pesticide analysis grade 1:1 dichloromethane:hexane (DCM:HEX). For the temporal trend study, the extraction column was spiked with 10 µL of BDE surrogate spiking solution (EO-4981, Cambridge Isotope Laboratories) containing 13C12 labeled 4784

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 22, 2002

BDE-47 (1000 pg/g), and 13C12 labeled BDE-100, -99 and -126 (1500 pg/g each), and eluted with 150 mL of DCM: HEX. The final volume of the extract was reduced with a rotary evaporator to 10 mL. Because the BDE concentrations were high in 2000, for the geographical comparison samples an aliquot of the extract equivalent to 0.2 g (wet weight) of egg was spiked after extraction (rather than on the column) with BDE surrogate spiking solutions, 20 µL of EO-4981 (Cambridge Isotope Laboratories) containing 13C12 labeled BDE-47 (10 ng/g of egg), and 13C12 labeled BDE-100, -99 and -126 (15 ng each/g of egg); and 10 µL of MBDE-MXB (Wellington Laboratory) containing 13C12 labeled BDE-28, -154 and -183 (25 ng each/g of egg). Lipid content was determined gravimetrically from a 0.1 g aliquot of the extract. Lipid was removed by gel permeation chromatography (23). Final cleanup was done with a Florisil column (8 g, 1.2% water deactivated) eluted with 95 mL of DCM:HEX. The eluate was reduced to ca. 0.5 mL under a gentle nitrogen stream, and the solvent was exchanged to toluene and reduced further to ca. 100 µL. The cleaned-up sample was spiked with 10 µL of volumetric standard (EO4151, Cambridge Isotope Laboratories) containing 13C12 BDE-77 (10 ng/g of egg) and reduced with gentle nitrogen flow to a final volume of 10 µL for analysis. Samples were analyzed with a VG AutoSpec double focusing GC-MS operated at 7000 resolution in the electron ionization mode (70 ev) using selected ion monitoring. GC conditions were as follows: 30 m DB-5 fused-silica column, 0.25 mm ID, 0.25 µm film thickness; 45 cm/sec (100°) carrier gas (He) velocity; 1 µL splitless injection; injection port temperature 260 °C; oven temperature program 100 °C, hold 3 min.; 20 °C/min to 180 °C; 5 °C/min to 325 °C; hold 4 min. Preliminary screening monitoring the two strongest ions in the molecular ion cluster for mono- to deca-BDE indicated that there were no detectable nona-BDEs or deca-BDE. Therefore, five chromatographic windows were used to monitor the following isomer groups: mono- + di-BDEs, tri- + tetra-BDEs, tetra- + penta-BDEs, penta- + hexa-BDEs, hexa- + hepta-BDEs. The two strongest ions of the molecular cluster for each native analyte and labeled surrogate were monitored. BDEs were quantitated using the isotope dilution method (12C12/13C12 response). For minor congeners which had no corresponding internal standards, a 13C12-BDE isomer with the same number bromines was used as reference.

TABLE 1. Concentrations (µg/kg, Wet Weight) of the Seven Major BDE Congeners and ΣBDEs in Herring Gull Eggs (Pools of 10-16) from Routine Monitoring Colonies throughout the Great Lakes, Connecting Channels and the St. Lawrence River, 2000 (site, Figure 1)

BDE-28 BDE-47 BDE-100 BDE-99 BDE-154 BDE-153 BDE-183 Σ7BDEs ΣBDEs Σ7/ΣBDEs lipid % moisture %

Granite Isl. L. Superior (1)

Agawa Rock L. Superior (2)

Big Sister Isl. Green Bay (3)

Gull Island L. Michigan (4)

Double Island L. Huron (5)

Chantry Island L. Huron (6)

Chan. Shel. Isl. Saginaw Bay (7)

Fighting Isl. Detroit R. (8)

3.8 253 83.6 202 25.4 71.6 7.1 646 664 0.973 9.7 76.5

3.1 323 113 284 28.8 106 8.0 866 887 0.976 10.0 75.7

5.1 522 167 459 59.5 143 7.1 1362 1400 0.973 8.5 76.6

8.2 602 203 323 55.6 118 22.0 1332 1366 0.975 10.2 76.2

2.5 146 45.2 74.6 15.0 22.8 2.6 309 320 0.966 11.3 73.9

2.5 127 37.3 77.7 11.7 36.4 7.0 299 308 0.971 9.7 76.7

6.9 291 89.5 161 29.1 50.6 7.9 635 652 0.975 9.5 75.5

6.3 322 92.6 130 17.6 53.5 4.5 627 639 0.981 9.5 76.6

(site, Figure 1)

BDE-28 BDE-47 BDE-100 BDE-99 BDE-154 BDE-153 BDE-183 Σ7BDEs ΣBDEs Σ7/ΣBDEs lipid % moisture %

Middle Isl. Lake Erie (9)

P. Colbourne Lake Erie (10)

Niagara River (11)

Hamilton Har. Lake Ontario (12)

Toronto Harbor Lake Ontario (13)

Snake Island Lake Ontario (14)

Strachan Island St. Lawrence R. (15)

4.8 163 51.8 52.0 10.3 37.5 10.2 329 340 0.969 10.4 76.3

1.0 70.0 24.6 55.9 7.2 25.6 3.3 188 192 0.975 10.5 76.1

1.7 168 53.0 111 17.9 57.6 13.8 423 432 0.979 9.0 76.2

7.0 361 102 167.0 28.8 67.0 5.1 738 755 0.978 9.1 77.1

2.7 401 108 322 38.9 89.9 12.5 975 1003 0.972 9.5 76.5

3.9 220 66.5 113.0 31.8 65.2 15.1 515 530 0.971 9.7 76.4

2.8 220 56.6 89.8 20.0 47.7 7.3 444 453 0.980 10.0 76.1

Blank samples were carried through the whole procedure every 10 analyses. Recoveries of the surrogate standards were calculated by comparison to an external standard solution in order to ensure that the method was functioning properly. To minimize carry-over (ca. 1%) between samples in the GC analysis, an injection of solvent was made after every standard injection. For the temporal trend study, the samples were analyzed in sequence from the earliest to latest (lowest to highest concentrations). Whenever there was a high followed by a low concentration sample, the latter was reinjected. The laboratory participates in the certification program of the Canadian Association of Environmental Analytical Laboratories. BDE concentrations are given on the basis of wet weight of egg homogenate.

Results Geographical Distribution, 2000. A total of 25 di- to heptaBDE congeners were identified in herring gull eggs throughout the Great Lakes system. No mono-, octa-, nona- or decaBDEs were found at the detection limit of the analysis, which was 0.01-0.05 µg/kg wet weight of egg. Seven congeners, BDE-28, 47, -99, -100, 153, 154 and -183, constituted 97.5 ( 0.5% of ΣBDEs in the 15 herring gull colonies that were sampled in 2000 (Table 1). BDE-47 was the dominant congener, followed closely by BDE-99. Of the remaining 18 congeners, the structures of 8 could be determined. These were BDE-15, -17, 49, -66, -119, -85, -155 and -140. Only 10 minor congeners were not identified. Recoveries of the 13C12-BDE internal standards spiked to the samples prior to sample cleanup averaged 75 ( 11%. Concentrations were corrected in each analysis using the recovery of the internal standard with same number of bromines. Blank samples

contained small concentrations ( Lake Superior > Lake Huron > Lake Erie. In herring gull eggs, the order for these four lakes was Lake Superior > Lake Ontario > Lake Huron > Lake Erie. Some of the differences in order between species are likely to be explained by winter movements of the gulls in the upper lakes. Depending on ice cover and food availability, gulls from Lake Superior frequently spend part of the winter in Lake Michigan (22), where eggs of resident gulls had a higher concentration of BDEs compared to eggs of gulls from the other Great Lakes. The half-life of BDEs in herring gulls appears to be in the order of 100 days (see above) which means that the BDE concentration in Lake Superior eggs may reflect some (but annually variable) carry-over of accumulation from Lake Michigan in winter. The ratio of ΣBDE concentrations in Lake Ontario lake trout to those in VOL. 36, NO. 22, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4787

FIGURE 3. Trends, 1981-2000, in ΣBDE47,100,99 and ΣBDE154,153,183 concentrations in herring gull eggs from sites in Lake Michigan, Lake Huron and Lake Ontario. Lake Erie was similar to that that found in herring gull eggs from the two lakes, 3.5. Considering individual colonies, it is noteworthy that eggs from colonies near large urban/heavy industrial centers (Toronto Harbor, Hamilton Harbor, Detroit River) and a concentration of chemical industry (Saginaw Bay, Lake Huron) had higher BDE concentrations than colonies more remote from population and industry in the same Lakes. This suggests that current emissions of BDEs from large urban areas, whether from air or other routes, contribute part of the contamination found in local herring gull prey fish in the Great Lakes. The high concentration in northern Lake Michigan herring eggs is undoubtedly because this lake has the largest urban/industrial complex of any of the Great Lakes. Strandberg et al. (29) showed that Chicago urban air had 3-10-fold higher levels of BDEs than at other sites in northern Lake Michigan, southern Lake Superior and eastern Lake Erie. Concentrations of BDE-47 in Chicago air (29) and rural Ontario air (18) were as high as those of the most abundant individual PCBs. The reason Penta-BDE mixture congeners are most prevalent in the Great Lakes environment, when the Decaand Octa-BDE mixtures had much higher volume of production, probably lies in the nature and use of the main product in which Penta-BDE is employed as a flame retardant in North America - polyurethane foam (PUF) employed in 4788

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 22, 2002

upholstered furniture (16). North America consumes 98% of recent Penta-BDE production (14), most of which ends up in PUF. Hale et al. (16) determined BDEs in frogs living on PUF ‘islands’ in tanks and crickets which were fed to them. The frogs and crickets accumulated 8-13 mg/kg of BDEs in an unspecified time. The PUF was 32 wt % BDEs. Therefore, BDEs in PUF are bioavailable. Hale et al. (16) also showed that within four weeks exposure to sunlight, polyurethane foam began to disintegrate into small particles which could be easily dispersed and release their BDE content to the environment in the course of degradation. The temporal trend results are in general agreement with the findings of Luross et al. (11) for ΣBDEs in lake trout from Lake Ontario. They found concentrations of ΣBDEs to be increasing exponentially with a doubling time of ca. 5 years compared to 2.8 years found for herring gulls in Lake Ontario. The difference in rate of increase is probably because the lake trout data were based on only 4 time points between 1983 and 1998. In both herring gulls and lake trout, the onset of significant contamination by BDEs in Lake Ontario was the early to mid-1980s. In fact, the first notable increase in BDE concentrations in herring gull eggs from Lake Michigan, Lake Huron and Lake Ontario and lake trout from Lake Ontario occurred in 1987. Temporal trends in concentrations of organochlorine compounds in Great Lakes herring gulls are either stable or decreasing slowly (21). It is interesting to note that total neutral organic bromine was 0.6-1% of organic chlorine in Snake Island, Lake Ontario herring gull eggs in 1977 as determined by instrumental neutron activation analysis (30). However, in two other colonies in Lake Ontario, neutral organic bromine was ca. 5% of organic chlorine. In 2000, ΣBDEs concentrations were 2-5% of ΣPCBs concentrations in eggs from the three temporal trend monitoring colonies. Assuming that ΣPCBs remain constant at present levels, and the ΣBDEs continue to increase at current rates, it will only take 10-15 years for BDEs to become the most abundant organohalogen contaminant in the Great Lakes. There are now several published studies of temporal trends of BDEs in wildlife and humans for comparison with the present results. The chronology of BDE contamination of the Great Lakes is quite different from that in Europe. For example, concentrations of BDE-47 and -99 in guillemot (Uria aalga) eggs from the Baltic Sea began increasing in the mid-1970s, peaked ca. 1985, and dropped back to mid-1970s concentrations (ca. 10-fold decrease) by the late 1990s (3133). The guillemot is at a similar trophic level to the herring gull. Both feed primarily on small fish. Therefore, relative concentrations are an indication of the relative contamination of the Baltic Sea and Great Lakes ecosystems. Peak concentrations of BDE-47 in Baltic Sea guillemot eggs were ca. two times higher, and peak concentrations of BDE-99 were similar, to present day concentrations in herring gull eggs from the Great Lakes. She et al. (34) reported on BDE trends, 1989-1998, in stranded, dead harbor seals from San Franscisco Bay. Concentrations increased exponentially during this period with a doubling time of 1.8 years to reach 3-8 mg/kg in lipid in 1998. Ikonomou et al. (35) found that the onset of BDE contamination in ringed seal (Phoca hispida) blubber from the Canadian Arctic was sometime in the 1980s. Levels of tetra-BDEs, penta-BDEs and hexa-BDEs increased exponentially with doubling times of 8.6, 4.7 and 4.3 years, respectively. By contrast, tri-BDE concentrations peaked around 1985 and began decreasing. This suggests that the tri-BDE input, possibly some tetra-BDE input, was from Europe, while penta-BDEs and hexa-BDEs were following the North American trends. Trends in ΣBDEs concentrations in humans seem somewhat disconnected from fish and wildlife trends, at least in

Europe. Concentrations of ΣBDEs in human milk in Sweden were doubling every 5 years from the 1970s to 1997 (6). Concentrations of ΣBDEs in human serum in Norway doubled ca. every 8 years from 1977 to 1999 (36). Median concentrations of ΣBDEs in human blood in Germany increased between 1985 and 1999 but only by a factor of 1.5 (37). Thus, there is no sign of concentrations peaking and decreasing like was found in Baltic Sea guillemot eggs or fish in some Swedish lakes (4). However, there is also no consistency in the rate of increase among countries. deWit (4) suggested that the different trends in humans and wildlife may be due to exposure of humans to BDEs from contact with, or proximity to, textiles and electronic equipment, in addition to exposure from food.

Acknowledgments This study was funded in part by the Toxic Substances Research Initiative, Health Canada. Jason Duffe is thanked for data analysis, Abde Idrissi for sample extraction and cleanup, and CWS specimen bank staff for sample preservation and preparation.

Literature Cited (1) Andersson, O ¨ .; Blomkvist, G. Chemosphere 1981, 10, 1051-1060. (2) Jansson, B.; Asplund, L.; Olsson, M. Chemosphere 1987, 16, 23432349. (3) Sjo¨din, A.; Jakobsson, E.; Kierkegaard, A.; Marsh, G.; Sellstro¨m, U. J. Chromatogr. 1998, 822, 83-89. (4) deWit, C. Chemosphere 2002, 46, 583-624. (5) Watanabe, L.; Kashimoto, T.; Tatsukawa, R. Chemosphere 1987, 16, 525-532. (6) Nore´n, K.; Meironyte´, D. Chemosphere 2000, 40, 1111-1123. (7) Stanley, J. S.; Cramer, P. H.; Thornburg, K. R.; Remmers, J. C.; Breen, J. J.; Schwemberger, J. Chemosphere 1991, 23, 11851195. (8) Kuehl, D. W.; Haebler, R. Arch. Environ. Contam. Toxicol. 1995, 28, 494-499. (9) Loganathan, B. G.; Kannan, K.; Watanabe, I.; Kawano, M.; Irvine, K.; Kumar, S.; Sikka, H. Environ. Sci. Tecnol. 1995, 29, 18321838. (10) Alaee, M.; Luross, J.; Sergeant D. B.; Muir, D. C. G.; Whittle, D.; Solomon, K. Organohalogen Compd. 1999, 40, 347-350. (11) Luross, J. M.; Alaee, M.; Sergeant, D. B.; Whittle, D. M.; Solomon, K. R. Organohalogen Compd. 2000, 47, 73-76. (12) Betts, K. S. Environ. Sci. Technol. 2002, 36, 50A-52A. (13) Ryan, J. J.; Patry, B. Organohalogen Compd. 2000, 47, 57-60.

(14) Renner, R. Environ. Sci. Technol. 2000, 34, 452A-453A. (15) Hardy, M. L. Chemosphere 2002, 46, 757-777. (16) Hale, R. C.; La Guardia, M. J.; Harvey, E.; Mainor, M. Chemosphere 2002, 46, 729-735. (17) Ikonomou, M. G.; Rayne, S.; Fischer, M.; Fernandez, M. P.; Cretney, W. Chemosphere 2002, 46, 649-663. (18) Gouin, T.; Thomas, G. O.; Cousins, I.; Barber, J.; Mackay, D.; Jones, K. C. Environ. Sci. Technol. 2002, 36, 1426-1434. (19) Darnerud, P. O.; Erikson, G. S.; Johannesson, T.; Larsen, P. B.; Viluksela, M. Environ. Health Perspect. 2001, 109, 49-68. (20) Eriksson, P.; Jakobsson, E.; Fredriksson, A. Environ. Health Perspect. 2001, 109, 903-908. (21) Hebert C.; Norstrom R, J.; Weseloh D. V. Environ. Rev. 1999, 7, 147-166. (22) Hebert, C. E. Ecol. Applications 1998, 8, 669-679. (23) Norstrom, R. J.; Simon, M. Intl. J. Environ. Anal. Chem. 1986 23, 267-287. (24) Braune, B. M.; Norstrom, R. J. Environ. Toxicol. Chem. 1989, 8, 957-968. (25) Dodder, N.; Strandberg, B.; Hites, R. A. Environ. Sci. Technol. 2002, 36, 146-151. (26) Clark, T. P.; Norstrom, R. J.; Fox, G.; Won, H. T. Environ. Toxicol. Chem. 1987, 6, 547-559. (27) von Meyerinck, L.; Hufnagel, B.; Schmoldt, A.; Benthe, H. F. Toxicol. 1990, 61, 259-274. (28) Luross, J. M.; Alaee, M.; Sergeant, D. B.; Cannon, C. M.; Whittle, D. M.; Solomon, K. R.; Muir, D. C. G. Chemosphere 2002, 46, 665-672. (29) Strandberg, B.; Dodder, N. G.; Basu, I.; Hites, R. A. Environ. Sci. Technol. 2001, 35, 1078-1083. (30) Norstrom, R. J.; Gilman, A. P.; Hallett, D. J. Sci. Total Environ. 1981, 20, 217-230. (31) Sellstro¨m, U.; Jansson, B.; Kierkegaard, A.; deWit, C.; Odsjo¨, T.; Olsson, M. Chemosphere 1993, 26, 1703-1718. (32) Sellstro¨m, U.; Jansson, B.; Kierkegaard, A.; deWit, C.; Olsson, M. Organohalogen Compd. 1993, 14, 211-213. (33) Sellstro¨m, U. Ph.D. Dissertation, Stockholm University, Stockholm, Sweden, 1999. (34) She, J.; Petreas, M.; Winkler, J.; Visita, P.; McKinney, M.; Kopec, D. Chemosphere 2002, 46, 697-707. (35) Ikonomou, M. G.; Rayne, S.; Addison, R. F. Environ. Sci. Technol. 2002, 36, 1886-1992. (36) Thomsen, C.; Lundanes, E.; Becher, G. Environ. Sci. Technol. 2002, 36, 1414-1418. (37) Schro¨ter-Kermani, C.; Helm, D.; Herrmann, T.; Pa¨pke, O. Organohalogen Compd. 2000, 47, 49-52.

Received for review May 29, 2002. Revised manuscript received August 30, 2002. Accepted September 4, 2002. ES025831E

VOL. 36, NO. 22, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4789