A Perkin-Elmer Model 5000 flame atomic absorption 

 spectrometer was used to analyze for copper, cobalt, and 

 manganese. The parameters are described in the instrument's 

 operations manual (Perkin-Elmer Corp., 1976). 



A Perkin-Elmer Zeeman Model 3030 graphite furnace 

 atomic absorption spectrometer (GFA AS ) was used to analyze 

 for arsenic, cadmium, and lead. The GFAAS conditions for 

 arsenic are described by Krynitsky (1987). The GFAAS 

 conditions for cadmium and lead are described by Hinderberger 

 (1981). 



Separate 5.0 g (wet weight) aliquots of sediment were 

 digested for mercury analysis under reflux in sulfuric and nitric 

 acids (Monk, 1961). The determinations were performed by 

 cold vapor atomic absorption spectroscopy (CVAAS) using a 

 Coleman MAS-50B-Mercury analyzer. 



The detection limit for arsenic, lead, and cadmium was 

 0.05 ppm based on a 2.0 g sediment sample. The detection 

 limit for mercury was 0.01 ppm based on a 5.0 g sediment 

 sample. The detection limit for copper and cobalt was 0.2 ppm 

 and that for manganese was 1 .0 ppm, based on a 5.0 g sample. 

 All sediment analysis was based on wet weight. 



Biota 



After each station, samples were combined by family 

 name for analysis. The digestion procedure for cadmium, lead, 

 and arsenic was described by Krynitsky (1987). The entire 

 sample (including shell in the benthos) was homogenized in a 

 Virtis Model 45 blender. A 0.5 g wet weight aliquot of tissue 

 was digested in 5 ml of nitric acid and 0.5 ml of 30% hydrogen 

 peroxide as described above. Determinations for lead and 

 cadmium were performed using the instrument parameters 

 described by Hinderberger (1981) and arsenic as described by 

 Krynitsky ( 1987). The detection limit for lead and cadmium 

 was 0.05 ppm and that for arsenic was 0. 1 ppm, based on a 

 0.5 g sample wet weight. For mercury 1 .25 g were digested and 

 analyzed according to the procedure described by Monk ( 1 96 1 ). 

 "The mercury determinations were performed using a Spectro 

 Products mercury analyzer equipped with a Varian VGA-76 

 vapor generation accessory. The detection limit for mercury 

 was 0.05 ppm based on a 1.25 g sample wet weight. 



Moisture Determimitions 



Moisture determinations were performed by weighing out 

 a separate 1.0 g of aliquot of biota or sediment into a tared 

 aluminum pan. The sample was allowed to dry for 24 h at 

 1 10°C and the percent of moisture was calculated. 



Quality Assurance/Quality Control 



The samples were processed in batches of 1 to 20 samples 

 with one matrix spike, one procedural blank, one duplicate 

 sample, and one standard reference material (SRM) for each 

 batch. The SRM used for sediments was provided by the 

 National Institute of Standards and Technology (NIST), 

 fomierly National Bureau of Standards. The SRM used was 

 NIST 1 646 Estuarine Sediment, which contained the following 

 certified values (ppm dry weight) for the metals of interest: 

 arsenic 11.6±1.3, cadmium 0.36 ± 0.07, and cobalt 

 10.5 ± 1.3 . The SRM used for biota was provided by the 



National Research Council of Canada, Ottawa, Ontario, Canada. 

 The SRM used was TORT- 1 , which is a freeze-dried sample of 

 a partially defatted lobster. TORT- 1 contained the following 

 certified values (ppm dry weight) for the metals of interest: 

 arsenic 24.6 ± 2.2, cadmium 26.3 ±2.1, lead 10.4 ± 2.0, and 

 mercury 0.33 ± 0.06. The matrix spike for the sediment 

 consisted of : 300 ^g manganese; 10 |ig each of lead, copper, 

 cobalt, and cadmium; 20 fig arsenic; 0.5 |ig mercury. The 

 matrix spike for the biota consisted of (6.0 |ig arsenic and 

 5.0 |ig each of cadmium, lead, and mercury. 



Recoveries were monitored in both the standard reference 

 materials and matrix spikes. The average recoveries ranged 

 from 87-108% for the metals we investigated. The relative 

 percent differences between duplicate results averaged less 

 than 10%. 



Results and Discussion 



Trace metals, with the exception of mercury at a few 

 stations, were present in all sediment samples collected in the 

 Bering and Chukchi Seas (Table 1). The one deep-water 

 station (Station 3) differed from the shallow- water stations in 

 that mai]ganese and copper were much higher. This was 

 probably due to the remobilization of these metals within the 

 sediment core as a primary mechanism. Other investigators 

 during the Second Joint US-USSR Expedition to the Bering 

 Sea (Summer 1 984 ), who sampled more deep-water stations in 

 the Bering Sea than were sampled during this cruise, observed 

 the same phenomenon with these two metals (Iricanin & 

 Trefry, 1991). 



When comparing the concentrations of trace metals 

 obtained in sediments during the 1 988 cruise to those obtained 

 during the 1 984 cruise (Table 2 ), the concentrations of lead and 

 cadmium are generally higher for the 1988 cruise. During the 

 1988 cruise the stations sampled were primarily shallow-water 

 stations (<200 m) where the primary productivity was high 

 (Zeeman, Subchapter 6.2, this volume), and during the 1984 

 cruise the stations sampled were the deep-water stations 

 (>600 m). These higher concentrations were most likely due 

 to a downward vertical flux of planktonic organisms and other 

 biogenic debris (Iricanin & Trefry, 1991). In general, the 

 values for trace metal residues obtained from both the 1984 and 

 1988 cruises appear to be less than value for trace metals in 

 sediments from some other parts of the world (Table 2), which 

 suggests that the Bering and Chukchi Seas are relatively 

 pristine. 



Only arsenic, cadmium, and lead had reportable values in 

 the biota (Tables 3,4). The values of mercury in the biota are 

 less than the detection limit. The benthos samples might be 

 expected to have residue patterns similar to the sediments they 

 reside in except in cases where bioavailability plays an important 

 role. Even though the lead values in the sediment were 30 to 

 40 times higher than those for cadmium, the cadmium values 

 in the benthos were in most cases higher than the corresponding 

 lead values. This indicates that the benthos investigated 

 bioaccumulated cadmium better than lead. Lead values were 

 also comparable to the arsenic values in the sediment. As with 

 cadmium, we found that the arsenic values in the benthos were 



320 



