Total mercury in hair of polar bears (Ursus maritimus) from Greenland and Svalbard ERIK W . BORN, ARISTEO RENZONI and RUNE DIETZ Born. E . W . , Rcnzoni. A . & Dietz. R . 1991: Total mercury in hair of polar bears ( U r s u s moririmlts) from Greenland and Svalbard. Polar Research 9(,7). 113-120. Conccntratinns (ppm = pg/g dry wcight) of total mercury ( H g ) were determincd in hair of polar bean ((!r\uT marirbnus) from northwestern Greenland ( N : 2 2 : pcriod of sampling: 197b1989). castern Grccnland ( N = 44: 19841989) and Svalbard ( N = 31: 1980). For subadults (2-h yearr of lifc). adults (7-10 yeara). and old bears ( > I 0 years). concentrations of total Hg in hair were not found to he dcpcndent on agc or SCX. A decreasing trend in Hg conccntrations was found from west to east. The mean conccntrations of total Hg in hair (cobs o f the year and yearlings cscludcd) were: northwcstern Grccnland. X = 8.38 ppm (min.-max.: 1.71-14.19 ppm. N = 21): castern Greenland: X = 1.28ppm ( m h - m a x . : 2 . 5 C 8 . 8 3 ppni. N = 41): and Svalbard. X = 1.98ppm (niin.-mas.: 1.02-4.SSppm. N = 2Y). Concentrationa found i n north- wcatern Greenland werc similar to those reported by others from the hair of polar bears samplcd within management zone F of thc eastern Canadian Nigh Arctic. Concentrations of total Hg in polar hear hair from castern Greenland were similar to concentrations found by other5 i n contemporary (1Y88) material collected during spring in western Svalbard. However. the mean concentration of total Hg in the 1980 Svalhard material, which was collected during July-September. was significantly lower than concentrations found in samples taken during late winter and spring in eastern Greenland and at Svalbard. respectively. Presumably the relatively low concentrations found in the 1980 Svalbard sample arc attributable to the period of moult and hence a larger proportion of newly grown hair in the individual samples. In a subsample consisting of internal tissues from 19 polar bears from eastern Greenland (19841987). concentrations of total Hg in hair correlated positively with concentrations of total Hg (wet weight) in muscle ( N = 6). liver (N = 19) and kidney (N = 19) tissue. For liver and kidney tissue these relationships were statistically significant. Erik W . Born. Greenland Fisheries Research Institute. Tagensuej 135, DK-2200 Copenhagen N , Denmark; Aristeo Renzoni, Dipartimenro Biologia Amhientale, Universita di Siena. Via delle Cerrhia 3, 53100 Siena. truly: Rune Dietz. Greenland Enuironmental Research Insriture. Tagensoej 135, DK-2200 Copenhagen N . Denmark. Introduction Hair samples of mammals are considered to be good indicators of the body burden of mercury (Johannesson et al. 1981; Hansen 1981; Renzoni 1989). Concentrations in hair reflect blood mer- cury or body burden of mercury during time of hair growth (Sexton et al. 1978; Kershaw et al. 1980). Being a top predator which feeds almost exclusively on seals and in particular ringed seals (Phoca hispida; e.g. Stirling & Archibald 1977; L0n0 1970), the polar bear can be considered a good indicator of the degree of bioaccumulation of persistent contaminants in its ecosystem. Based on analyses of 128 recent (1977-1980) and 18 museum samples (1910-1927), Eaton & Farant (1982) presented concentrations of total mercury (Hg) in polar bear hair from eight different localit- ies in the Canadian High Arctic. Liver samples from 67 polar bears from 1982 were analysed for mercury and 21 other elements by Norstroni et al. (1986). Renzoni & Norstrom (1990) gave con- centrations of total Hg in hair sampled between 1976 and 1988 from a total of 141 polar bears in various localities in the Canadian Arctic, the USSR, and at Svalbard (Norway). Concen- trations of total and organic Hg in muscle, liver and kidney in polar bears from central-eastern Greenland were presented in Dietz (1987), Dietz et al. (1990) and Dietz & Agger (in press). Here we present the results of analyses of total Hg in hair from 66 polar bears from three localities in Greenland with a note on the relationship between concentrations of total Hg found in hair and in muscle, liver and kidney tissue. For com- parative reasons analyses of total Hg in 31 polar bear hair samples obtained at Svalbard during July-September 1980 are included. 114 E . W. Born, A. Renzoni & R . Dietz Material and methods catch of polar bears. Between January 1984 and August 1989 a total of 40 samples (year/sample A total of 66 hair samples (28 F, 37 M , 1 undet.) from Greenland and 31 samples (15 F, 16 M ) from Svalbard have been analysed for contents of total Hg. The majority of the Greenland samples were collected by hunters from their subsistence size: 1984/11; 1985/2; 1986/1, 1987/4; 1988/8: 1989/13; unknown/l) were obtained from the Scoresby Sound region of eastern Greenland (Fig. 1). Four samples (2F, 2M) were obtained in 1989 from the Ammassalik area in southeastern Green- Frans Josef Land 0% % 9 4 Svalb rd A- B % Sound 4 .$ Arnrnassalik I Fig. I . Area5 in uhich polar bcar hair samples anal@ in thts atudy were obtained. T h c borders o f the Canadian polar heal management zone F nrc l h m n with a dotted line (......). Total mercury in hair of polar bears 115 land; a total of 22 samples (8 F, 14 M) were collected in the Avanersuaq (Thule) area in north- western Greenland. Four of these samples were collected by scientists in May 1978; the remainder were collected by local hunters in 1988 (N = 7) and 1989 (N = 11). A total of 31 hair samples were collected by scientists during a Swedish ice- breaker expedition (YMER) operating in the Sval- bard area in 1980. The vast majority of the Green- land samples were from late winter and spring (January-May) while the Svalbard-material was collected during “summer” (July-September; Fig. 2). For 19 of the polar bears from the Scoresby Sound material (1984-1987), additional infor- mation was available on total mercury content in muscle ( N = 6), liver (N = 19) and kidney ( N = 19) tissue. The hair samples were kept frozen in poly- ethylene plastic bags until analysis (with the exception of the 1978 Thule samples, which were stored dry in paper bags). The samples were carefully washed in acetone. rinsed and dried 14 12 10 8 6 4 2 0 NO. L .mu o s c o N= 22 36 1 2 3 before being analysed for mercury residues at the laboratory of Dipartimento Biologia Ambientale (Siena) by atomic absorption spectrophotometry as described in Renzoni et al. (1986). Samples of internal tissues were kept frozen and stored in polyethylene plastic bags before being analysed at the laboratory of Greenland Environmental Research Institute by a method described in Dietz et al. (1990). To detect whether concentrations of total Hg in hair correlated with age, the polar bears were grouped into five age classes: cub of the year = COY, yearling, subadult. adult, and old. The age classes were based either on age obtained from reading of the annual layering in tooth cementum or on field age class estimates and body length. For a sub-sample from Scoresby Sound (N = 19; 1984-1987), individual ages were determined from the counting of the annual layering in the cementum of the lower premolars as described in Stirling et al. (1977). For the 1978 Thule-sample and the 1980 Svalbard-sample, ages were esti- Amm B S v a L 4 5 6 7 4 31 8 9 10 11 12 Month F I ~ . 2 . Seasonal distribution of polar bear sampling from three different regions in Greenland (Thu = Thule: Sco = Scoresby Sound; Amm = Ammassalik) and Svalbard (Sva). Four samples from Scoresby Sound presented in Table 2 had no information on the date of sampling. 116 E . W . Borri. A . Retizoni & R . D i e n Trrhlr I Polar bear age cl.i\\c\ with zoological hod! lengths used in a n a l l z e s o f rclation5hip between t o t a l Hg i n hair and age 1 - I 4 I4 I A 15 I- 111 mated i n the field by scientists. T h e majority of the Greenland material was categorized into age classe\ according t o e 4 r n a t e s of age given b!. the h u n t e n i n the field together w i t h zoological body length measured from tip of snout to tip of tail along the back \ \ h e n the bear was positioned on its belly. L e n g t h data for these age classes are presented in Table 1 where adult and old females are pooled hecause there was no difference in average body Icngth of these age u . 5 ~ ~ . d i = I !IS. P o . 0 5 ) . Concentrations of total H g are given as ppm (1cg;g. d . ~ . ) in h a i r and 21s ppm (kcg'g. w . w . ) in other tiswes. The distrihutiun of Hg in hair i n the differ- ent areas did not differ significantly from nor- mality (Kolmogorov-Smirnov one-sample tests: P > 0.05) but had significantly different variances (Bartlett's test. X' = 19.567. P < 0.01). All data were In-transformed and A N O V A analyses were performed using the factors: area, sex, age. and the respective interactions o f these. Statistics were made using Statview 11 and SuperANOVA for a Macintosh microcomputer. H g itr h i t Concentrations of total H g found in polar hear hair in the four different areas are presented in Table 2 . Values for the two youngest age groups are given separately because young bears (i.e. cubs of the year = COYS and yearlings) are assumed to be dependent o n milk until they are well into their second year of life (Lqmo 1970). Most cubs become independent when they are ii little over 24 months old (Larsen 1986). In none of the samples were concentrations of total Hg in hair related t o either age (age classes: suhadults, adults. olds) or to sex ( t w o factor A N O V A s ; P > 0.05). Therefore. data from these two cat- egories were pooled for further analyses. As con- centrations of total H g in polar bear from Ammassalik in southeastern Greenland did not differ significantly from concentrations found in h a i r from the Scoresby Sound region ( F = 0.446. P < 0.05. df = 1/39). the samples froni these 7;7b!r 2 . C ' o n i c n t r d r i o n \ ( i q g d w . ) of t o t a l rncrcur! in hdir of polar bear\ troni Greenland (l97&-19W) and Svalhard ( I Y K O ) . Thc a v e r . i g c and crror r a n p c s arc calculated f r o m thc aiitihigarithni\ d t h e mean and the \tandard deviation of the In-transformed ~ ~~ data T h u l c I978-198Y Yearling 2 ?cars and older l Y S 1 I Y X Y Yearling?. 2 !car\ .ind vldcr ? !car\ and oldcr 1 "I IYSII C u b 01 the !car Yenrlin$ 2 \ e a r \ a n d d d s r Y 51 8 38 I X I 1 h2 1.11 0 34 I IN 1 4 s - 6 . 3 b l l . 0 1 1.71-13 19 I h.LZ (11 1.65-2.w 3.3%6. 17 3 I b S . S i 3 6%4.X' i . 4 1-6.02 1 21 3 27 4 I I 29 Total mercury in hair of polar bears 117 J - areas were pooled. The geometric mean con- centrations of Hg differed significantly between the three main areas ( F = 122.09, df = 2/88, P < 0.01). The northwestern Greenland Hg-con- centration was 1.8 times and 4 . 3 times higher than mean concentration in eastern Greenland and at Svalbard, respectively (Table 2). Concentrations found in hair from eastern Greenland were significantly higher than con- centrations found in the sample taken at Svalbard in 1980, but similar to concentrations reported by Renzoni & Norstrom (1990) for a sample obtained from Svalbard during the spring of 1987 and 1988 (Tukey's test; P < 0.01). In the East Greenland sample the concen- trations of total Hg in hair collected during July and August (X = 5.23 ppm, rel. SD = 1.26, min.- max.: 3.54-6.62ppm, 5M and 1F) did not differ significantly from concentrations in hair at other times of the year (F = 0.066; df = 3/33; P < 0.05). Insufficient seasonal representation did not allow similar analyses to be performed for the other samples. lE Hg in hair H g in muscle, kidney and liver The relationships of concentrations of total Hg in hair to concentrations of total Hg in muscle. kidney and liver tissue from 19 polar bears sampled in the Scoresby Sound region (eastern Greenland) between 1984 and 1987 in the period December-April are shown in Fig. 3. Con- centrations of mercury in hair were found to be positively correlated with concentrations in tissue from kidney (r = 0.480, N = 19, P < 0.05), liver (r = 0.685, N = 19, P < 0.05), and muscle (r = 0.574, N = 6, n.s.). The slopes of the regression lines were significantly different from 0 at the same probability level for kidney and liver tissue, whereas no significance was detected for muscle tissue, probably due t o the low number of samples. 7 6 5 4 3. Discussion Concentrations of total Hg in polar bear hair were highest i n northwestern Greenland. However, the concentrations were in the same order of mag- nitude as those reported by Eaton & Farant (1982) and Renzoni & Norstrom (1990) from the eastern Canadian High Arctic (i.e. within the Canadian management zone F; Fig. 1). Recaptures in north- Hg in hair El . 0 0 0 * % 0 . . . I 0 0 a r = ,685 8 7 lz Hg in hair 0 6 5 4 b 0 0 0 r = ,574 3 1 - 2l I Hg in liver llg in muscle 0 0 .02 .04 .Oh .08 . I . I 2 . I 4 I . *. . . . . . 0 . 0 I Hg in k i d n e y II 6 ' b 10 20 30 40 SO 613 118 E . W . Born, A . Renzoni & R . Dierz western Greenland of polar bears which were marked i n zone F (Rosing-Asvid & Born 1990) indicate that polar hears move freely between these areas. Concentrations of total H g in hair from polar bears from eastern Greenland were similar to values reported by Renzoni & Norstrom (1990) for a sample taken during spring in western Sval- bard. Satellite tracking (Larsen e t al. 1983) and movements of marked bears (Larsen 1986) indi- cate that exchange occurs between polar bears in eastern Greenland and in the Svalbard - Frans Josef Land region. However, t o what extent local groups of bears exist within this range still remains undetermined (Born & Rosing-Asvid 1989). Eaton & Farant (1982) and Renzoni & Nor- strom ( 1990) suggested that geographical vari- ation in concentrations of total Hg in the hair of polar bears in the Canadian High Arctic t o some extent reflects different ratios of the bottom-feed- ing bearded seals (Erignurhus burbutus) to the more pelagic ringed seals in the diet of the bears in the various areas. T h e very limited and highly heterogeneous information available on the feed- ing of polar bears i n Greenland (Pedersen 1945: Born upubl. d a t a ) and at Svalbard ( L o n 0 1970) does not allow for an evaluation of geographical variation in polar hear diet in these areas. O n the other hand. Dietz & Agger (in press) state that the only differences found in mercury levels in bearded seals and ringed seals from NW Green- land are found in liver tissue. as also found in Smith & Armstrong (1975) i n Canada. whereas no differences were found between these two species in blubber and muscle and kidney tissue. We suggest that the differences found in our study in concentrations of total H g in hair from Greenland polar bears mainly reflect variations in the overall mercury burden of the environment rather than different proportions of various prey species in their diet. Norstrom et al. (1986) found that mercury con- centrations in polar bears from the western part of the Canadian Arctic tended t o be higher than in the eastern part. Likewise mercury con- centrations in the northern areas at Melville Island were higher than in southern regions. Dietz & Agger (in press). who found that con- centrations of total H g in the liver tissue of polar bears from eastern Greenland were lower than those reported by Norstrom et al. (1986), confirm a decreasing trend in H g concentrations from west to east. Higher concentrations of H g were found in polar bear hair sampled at southwestern Svalbard in 1987 and 1988 (Renzoni & Norstrom 1990) than in a sample from the pack ice areas around the archipelago in 1980. To our knowledge there is n o information which indicates that the total mercury burden of the marine environment of the Svalbard region has increased within the last decade and we suspect that the relatively low concentrations of total H g found in the 1980- sample from Svalbard reflect seasonal variation related to moulting of hair. T h e 1980 Svalbard- samples (this study) were collected from “summer fur“ while the samples analysed by Renzoni & Norstrom (1990) were taken during March-April from “winter fur.” T h e majority of samples from Greenland were collected between January and May from “winter fur” (Fig. 2) which presumably is also the case with samples in other studies (Eaton & Farant 1982; Renzoni & Norstrom 1990). Although information on t h e timing of the process of moulting in polar bears is somewhat contradictory (Freuchen 1935; Pedersen 1945; Uspenski 1979 citing Rutilevski 1939; Vibe 1981), i t appears that generally the moult begins in late May and that the new winter coat has grown in autumn (September-October?). O u r assumption that hairs of “summer fur” have lower concentrations of total H g than hairs of winter fur was. however, not supported by the finding that in only six samples, taken in July and August in eastern Greenland, the mean con- centration was not significantly different from the mean concentration in hairs sampled in t h e same area during other seasons. We suggest that t h e low H g levels found in “summer” hair are caused by accelerated hair growth during the moult. However, it remains undetermined whether the different concentra- tions found at Svalbard (this study and Renzoni & Norstrom 1990) also reflect sampling of different segments of the polar bear population at different times of the year and/or seasonal variations in the food of polar bears. In samples from 55 polar bears taken between January and May in 1982 i n seven areas within the central Canadian High Arctic. Renzoni & Norstrom (1990) did not tind a statistically signifi- cant correlation between total Hg in hair and live tissue. In contrast. in a smaller sample taken during spring in eastern Greenland we found that concentrations of total H g in hair correlated posi- tively with concentrations of total H g in internal Total mercury in hair of polar bears 119 Bulletin of Environmental Contamination & Toxicology 35. 49g495. Born, E. W. & Rosing-Asvid, A . 1989: I s b j ~ r n e n (Ursus mar- itimus) i G r ~ n l a n d : en oversigt (The polar bear in Greenland: a review). Teknisk rapport - Grgnlands Hjemmestyre, Milj@- og Naturforualtning. 3900 Nuuk. Rapport Nr. 8 - oktober 1989. 126 pp. (in Danish with English summary). Dietz, R . 1987: Tungmetaller i isbj~irn og andre arktiske dyr/ Nannuni issittullu uumasuini allani saffiugassat oqimaatsut (Heavy metals in polar bears and other Arctic animals). Forskning i Gr@nland ltusaat 1987 ( 2 ) . 2-13 (in Danish and Greenlandic). Dietz. R.. Nielsen. C. 0.. Hansen. M. M . . & Hansen, C . T . 1990: Organic mercury in Greenland birds and mammals. The Science of the Total Environment 95 (1990), 41-51. Dietz. R . & Agger. C. T . (in press): Recent studies on heavy metals in polar bears from Greenland with reference to other marine mammals. Abstract "Contaminants in the Arctic." 18th Annual Aquatic Toxicity Workshop, 30 September-3 October 1991, Ottawa, Canada. Eaton. R. D. P. & Farant J . P. 1982: The polar bear as a biological indicator of the environmental mercury burden. Arcfic 35 ( 3 ) , 422-425. Freeman. H . C. & Horne. D . A . 1973: Mercury in Canadian seals. Bullrtin of Enuironmental Contamination & Toxicology 10. 172-180. Freuchen. P . 1935: Polar bears (Ursus maritimw). Pp. 102-1 19 in Mammals, Part 11. Field notes and biological observations. Rep. Fifth Thule Expedition 1921-24. 21455), 68-278. Hansen, J. 1981: A survey of human exposure to mercury, cadmium and lead in Greenland. Meddr. Gr@nland Man. & Society 3 . 36 pp. Itano, K . , Kawai, S . , Miyazaki. N.. Tatsukawa, R. & Fujiyama. T . 1984: Mercury and selenium levels in striped dolphins caught off the Pacific coast of Japan. Agric. Biol. Chen. 48(5), 1 10%1116. Johannesson. T.. Lundc. G. 61 Steinnes. E . 1981: Mercury. arsenic. cadmium. selcnium and zinc i n human hair and salmon fries in Iceland. Acra pharmacol. er towicol. 48. 1x5- 18Y. Kershaw. T. G . . Dhahir, P. J . & Clarkson. T . W . 19x0: The relationship between blood levels and dose of methyl mercury in man. Arch. Enoiron. Healrh 3.5. 28-35. Larsen, T. 1986: Population biology of the polar bear (Ursus maritimus) in the Svalbard Area. Norsk Polarimtitutts Skrifter 184. 5 5 pp. Larsen. T . . Jonkel. C. & Vibe, C. 1983: Satellite radio-tracking of polar bears between Svalbard and Greenland. lnt. Conf. Bear Res. Manage. 5, 23S237. L0n0, 0. 1970: The polar bear ( U r s w maritimus Phipps) in the Svalbard area. Norsk Polarirrrtitutts Skrijter 149. 103 pp. Norstrom, R . I . , Schweinsburg. R . E. & Collins, B. T. 19x6: Heavy metals and essential elements in livers of the polar bear (Ursus maritimus) in the Canadian Arctic. The Science of the Total Environment 48, 191-212. Pedersen, A . 1945: Der Eisbar (Thalarctos maritimus Phipps) - Verbreitung und Lebensweise. Bruun & Co.. K ~ b e n h a v n . 16 pp. (in German). Renzoni, A . 1989: Mercury in scalp hair of Maldivians. Marine Pollution Bulletin 20(2). 9>94. Renzoni, A . & Norstrom, R . 1990: Mercury in thc hairs of polar bears Ursirs maritimus. Polar Record 26(159). 326-328. Renzoni. A , . Focardi, S.. Fossi, C.. Lconzio, C. & Mayol. J . 1986: Comparison between concentrations of mercury and tissues with a capacity for long term accumulation of mercury. A n explanation for this apparent discrepancy found in the two studies could be that samples from various areas in Canada were pooled in the Renzoni & Norstrom (1990) study whereas the East Greenland material is more homogenenous since it was sampled in only one area. The mercury concentrations in the hair as well as in the internal organs reflect the contamination levels in the food of the polar bears. Therefore, it appears less surprising that we found a correlation between H g in hair and in internal tissues. This is in agreement with previous findings by for example Freeman & Horne (1973). who stated that claws and fur from the Gray seal (Halichoerus gr.ypus) and the harp seal (Pagophilus groen- landicus) were good indicators of the degree of mercury contamination in muscle, liver, kidney and heart tissue. Similarly, Bacher (1985) found a significant correlation between mercury in hair and in internal tissues (muscle, liver, spleen, brain) of Australian fur seals (Arctocephalus p u - sillus). Furthermore, if o n e regards hair as o n e of the excretion routes of H g (see Itano e t al. 1984). it seems reasonable to assume that H g con- centrations in hair in a general way reflect the H g level in the internal tissues of an individual. We conclude that concentrations of total mer- cury in polar bear hair from late winter and spring reflect the overall mercury burden in the eco- system within the range of the various polar bear populations. W e suggest that the relatively low concentrations of total mercury found in samples obtained during summer can be related to the process of moulting. Based on a small sample from eastern Greenland we have found indi- cations that concentrations of total H g in hair taken during spring reflect levels of total mercury in liver and kidney tissue. A c k n o d e d g e m e n t ~ . - Wc wish to thank the Grccnland Inuit and our colleagues who provided the samples. Thanks arc also duc to the Canadian Wildlife Scrvicc (Edmonton) for making age determinations of somc of the tceth, and to our laboratory technicians who made the metal analyses. We are grateful for the constructive criticism offered to the draft by J . Hanscn (Institute of Hygienc, University of Aarhus). R. Norstrom (Canadian Wildlife Service), and 0 . Wiig (Norwegian Polar Research Institute) and two anonymous referees. References Bachcr, G . J. 198.5: Mercury concentrations in the Australian fur seal (Arctocephalus pusillus) from SE Australian waters. 120 E . W . Born, A . Rerizoni & R . Dietz other contaminants i n eggs and tissues of Cory's shearwater. Culoriecfrrs diomedeu collected on Atlantic and Med- iterranean islands Eni~iron~neitfal Poilurion. Ser. A 40. 17- 3 5 . Kosing-Asvid. A . & Born. E. W . 1990: Fangst af ishj0rn (Ursus maritimur) I Avanersuaq og Upernavik kommuner: en inler- viewunders~gelse (Polar bear hunting in the Avanersuaq and Upernavik areas: a n interview survey). Teknrsk rupporf - Grdnlands Hjerntnrsn re Afddingen for L e i ~ e i i d e Restoiircer. 3900 .Vuuk. .Vr. 23 - deceinber IYYO. 63 pp. (in Danish with English \urnmar).). RutiIc\,hi. C j . L 1Y.W Prtm)\lo rnlekopitajuscic poluo- \tro\;i Celluskin'i I z i i l i \ i i Vi ickogo. Triid! naiicii. - rsA/ed. pol/arri X . 7-MI ( i n K u s h n . not conwilted dtrcctly 1 . Sexton. D. J , Powell. K . t.. Liddle. J R. Sinrek. A . 197X: A nonoccupational outhrcak of inorganic mercury poisoning. Arch. Enuiron. Heulrh 3 , 186191. Smith. T. G . & Armstrong. F. A . J . 1975: Mercury in scals. terrestrial carnivores. and principal food items of the Inuit from Holman. NWT. J. Fish. Res. B d . Can. 32 ( 6 ) , 795-801. Stirling. I . R. Archibald. K. 1977: Aspects of predation of seals by polar bears. J. Fish. Re.\. B d . Cun. 34(X), 112&1120. Stirling. I . . Archibald. W. R . & Schwcinsburg, R . E . 1Y77: Distribution and abundance o f scals in the eastern Bcaufnrt Sea. J. Fi.\/i. Res. Bd. Curl. 2 4 . Y1WX-X. lispentki. S . M . 1979, Der Eisbur. Die Nrur Rrehm-Hucherei. A Zicrnscn Verlag. Witlcnberg Luthcrstadi. 112 pp. ( i n Ger- m a n ) . Vihc C. 19x1: Ishjmn (Ursus rnarifimus), Nanok (Polar bear). Pp. 383-387 in Salomonscn, F. (ed.): (ironlands Fauna. Fkk, Fugle. Paffedyr. Gyldcndal. KQbenhavn. 464 pp. (in Danish).