Low protein variability and genetic similarity between populations of the polar bear (Ursus maritimus) T H O R LARSEN, HAKAN TEGELSTROM, R. KUMAR JUNEJA AND MITCHELL K. TAYLOR Larsen, T . , Tegelstrom, H . . Kumar Juneja, R . & Taylor, M. K. 1983: Low protein variability and genetic similarity between populations of the polar bear (Ursus maritimus). Polar Research 1 n . s . , 97-105. Blood samples from a total of 460 polar bears (Ursus maritimus) from various Arctic regions, but excluding the USSR, were collected during the period 1967-1981 to study electrophoretic variation in different proteins. Two hundred and one samples from Alaska, 48 from the Canadian Arctic, 89 from Svalbard, and 21 from Northeast Greenland were collected during the period 1967-1973 and were analysed by vertical polyacrylamide gel electrophoresis to study transferrin and hemoglobin polymorphism, Thirty-one samples collected in 1974 were analysed by starch gel electrophoresis for 14 enzyme systems in serum and red blood cells. Seventy samples collected from Alaska, the Barents Sea, and Canada in 198&81 were studied by starch gel electrophoresis, and further analysed for protein variation by thin-layer isoelectric focusing, horizontal polyacrylamide gel electrophoresis, and two-dimensional electrophoresis. In all, about 75 loci were analysed for variation. The degree of protein and enzyme variation in the polar bear was observed to be relatively low. Starch gel electrophoresis revealed variation of an unidentified serum protein. The distribution of this protein indicates a closer connection between bears in Alaska and Canada compared to those in Greenland and Svalbard, but the differences were not significant. As in many large mammals, the information from protein variation in polar bears has limited use for management purposes. We could not find any simple system usable for identification of discrete populations. On the basis of protein variation as sole criterion, the populations investigated could not be separated. Possible explanations for the uniformity of blood proteins can be exchange of bears between geographical areas and/or a high selective pressure in polar bears. Thor Larsen, Norsk Polarinstitun, Box 158, N-1330 Oslo Lufthaun, Norway; Hdkan Tegelstrom, Department of Genetics, University of Uppsala, Box 7003, S-75 007 Uppsala, Sweden; R . Kumar Juneja, Department of Animal Breeding and Genetics, The Swedish University of Agricultural Sciences, S-75 007 Uppsala, Sweden; Mitchell K . Taylor, Faculty of Forestry, University of British Columbia, Vancouver, British Columbia, Canada. May 1983 (reuired June 1983). Introduction Although polar bears (Ursus maritimus) were effectively protected through the International Agreement on the Protection of Polar Bears in 1973, there is still a substantial harvest in Alaska, Canada and Greenland. There is thus increasing concern that overhunting may take place in some countries, or that industrial activities may harm polar bears in some areas. Proper management and conservation require that the borders of dis- crete populations are identified and defined, and that the exchange of bears between areas is quan- tified. Information about the extent of effective dispersal, i.e. geneflow between possible polar bear populations, is quite scarce. Pedersen (1945) claimed that polar bears belonged to one uniform population that migrated clockwise around the Polar Basin. Polar bear marwrecapture programs and satellite telemetry studies have shown that polar bears belong to several, relatively discrete, populations (Larsen 1971; Lentfer 1974; Stirling et al. 1977, 1978, 1980; Stirling & Kiliaan 1980; Uspensky & Beli- kov 1981; Vibe 1982). Many nations share polar bear populations, and there is often a migration and an exchange of bears across jurisdictions (Parovshchikov 1967; Kolz et al. 1978; Larsen et al. 1980; Vibe 1982). Manning (1971) and Wilson (1976) found geographical variation in polar bear skulls from various areas, but it is not known whether such variation is genotypic. Genetic variants of different body proteins may be usable as a powerful tool for assessing affinities between different populations of a species. Stud- ies of many species of large mammals, however, have often shown a low level of protein variability, thereby decreasing the possible use of these meth- ods. An extremely high interpopulation genetic similarity was observed in black bears (Ursus 97 98 T. Larsen et al. Table 1. Summary of materials and methods used for the study of polar bear proteins. Year of Sample Method Proteins Locality collecting size used studied Alaska Canadian arctic Svalbard East Greenland Alaska East Greenland Svalbard Hudson Bay, Canada Barents Sea Alaska 1967-1973 1967-1 973 1967- 1973 1973 1974 1974 1974 1980 1980 1981 201 * 48 89 21 10 10 11 24 23 23 VPAGE VPAGE VAPGE SGE SGE SGE SGE IEF, SGE, HPAGE, 2-DE IEF. SGE, HPAGE, 2-DE IEF, SGE, HPAGE, 2-DE Transferrin and hemoglobin Transferrin and hemoglobin Transferrin and hemoglobin Transferrin and hemoglobin 14 enzymes and 3 serum proteins 14 enzymes and 3 serum proteins 14 enzymes and 3 serum proteins General proteins General proteins General proteins Killed bears americanus) (Manlove et al. 1980) and in polar bears from East Greenland (Allendorf et al. 1979). The objective of our work was to study and compare possible variation in polar bear blood proteins and enzymes from several Arctic areas. Material and Methods A total of 460 polar bear blood samples were collected and analysed between 1967 and 1981 (Table 1). Samples from killed bears were taken from the heart, a vein or an artery. Heparin was added to blood drawn from the femoral vein of the live captured bears. Samples were separated by centrifugation shortly after collection, frozen, and stored at -20°C. When a centrifuge was not available, samples were kept in a cool place for 24 hours to be separated by sedimentation before transfer to vials and freezing. Vertical polyacrylamide gel electrophoresis (VPAGE) (Raymond 1962) was performed in a Tris-borate-EDTA buffer 0.1-M with pH 9.1 and a 6% polyacrylamide gel (Cyanogum 41, Fisher Scientific Co.). After a prerun of one hour at 250 volts, plasma or serum was pipetted out in slots in the gel. Before application, a 5% solution of 1 % bromophenol blue in buffer saturated with sucrose was added to the sample. Erythrocyte samples were washed three times with physio- logical saline; water and 1 ml ether were added and the samples centrifuged. The supernatant was drawn off and the cells were frozen. Before application to the gel, sucrose crystals were added to the samples. The electrophoresis was run at 100-150 volts for ten minutes, then at 250-300 volts for about three hours. Serum or plasma proteins were stained with 9 g Amido black dissolved in 400 ml water, 400 ml methanol, and 100 ml acetic acid for 20 minutes. Destaining was performed for 12 hours in the same solution, omitting the Amido black. All samples were collected in 1974 and were later investigated with the use of starch gel elec- trophoresis (SGE). Twelve per cent starch gels were used in a Tris-citrate lithium borate buffer of pH 8.5 or a Tris-borate-EDTA buffer of pH 7.5. The separations were performed at 10 V/cm for three hours. Proteins and enzymes were stained in accordance with the techniques rou- Low protein variability in polar bears 99 tinely used (Shaw & Prasad 1970; Harris & Hop- kinson 1976). Selected samples of serum were treated according to Coppenhauer & Buettner- Janusch (1970) with neuraminidase to determine whether the observed variation in serum proteins was induced by sialic residues (Chen & Sutton 1967). The serum transferrins were also partly purified by rivanol treatment according to Boettcher et al. (1958). Seventy samples collected in the period 1980- 81 (Table 1) were studied by means of isoelectric focusing (IEF) in addition to SGE. Samples were electrofocused in a pH gradient of 4-6.5, an appropriate range for almost all polar bear serum proteins. Horizontal polyacrylamide gel electro- phoresis (HPAGE) (12% acrylamide in separa- tion gel; Tris-citrate-borate buffer pH 9.0) was done according to Gahne et al. (1977). The serum samples were analysed by a method of two-dimensional electrophoresis (2-DE) described by Juneja et al. (1981). The first dimen- sion separation in agarose gel (pH 8.6) was fol- lowed by a second dimension separation in hori- zontal polyacrylamide gel (pH 9.0). Fig. 1. Vertical polyacrylamide gel electrophoresis of polar bear hemoglobins. Samples from (A) Alaska and (S) Svalbard. Results Vertical polyacrylamide gel electrophoresis ( V P A G E ) The initial analyses of samples collected before 1974 concentrated on studies of polymorphism in hemoglobins and transferrins by protein separa- tion with VPAGE. The hemoglobin analysis showed an identical pattern, with one single com- ponent. No polymorphism could be demon- strated, and there were no visible differences between geographical areas (Fig. 1). In the serum and plasma electrophoretograms, particular attention was paid to the transferrin component. No transferrin polymorphism could be demonstrated. The serum and plasma separ- ations showed several different general protein patterns, which initially could be suspected to demonstrate transferrin polymorphism. Auto- radiography showed that only two of the bands were transferrins, a fast moving, weak compo- nent, and a slower, but stronger component. Comparison between the Svalbard samples and the Alaska samdes showed differences in the stained separations' there was only one strong and one weak transferrin Fig. 2. Vertical polyacrylamide gel electrophoresis of polar bear semm proteins. Samples from (A) Alaska and (S) Svalbard. band in the Svalbard samples, corresponding to Transfemns are indicated by dots. 100 T . Larsen et al. the pattern in the autoradiographs, there were two strong bands in the same positions in the Alaskan samples (Fig. 2 ) . Autoradiographs of the Alaskan samples could not be distinguished from the Svalbard samples. however. The visual Jif- ferences in the electrophoretograms are most likely due to sampling differences. One possible explanation is that the second strong band in the fast moving transferrin position in the Alaskan samples was caused by a haptoglobiq'hemoglobin complex. Such complexes were located close to the transferrins in the electrophoretograms, as demonstrated by benzidine stains (Fig. 3). The strong double band in the transferrin position could only be demonstrated in the Alaskan samples of killed bears. It was probably an artifact caused by hemolysis when bears were shot. Starch gel electrophoresis ( S G E ) Red blood cells and serum from 31 samples col- lected in 1974 were investigated in 1975. Fourteen Fig. 3 . Vertical polyacrylamide gel electrophoresis of polar bear serum proteins. Samples from (A) Alaska and (S) Svalbard. Autoradiographs (left) show two transferrin bands. indicated by dots. Benzidine stains (right) suggest haptoglobin/hemoglo- bin complex bindings in the same position as the lowest trans- ferrin band Fig. 4 Starch gel electrophoresis of polar bear serum proteins. Samples from (a) Alaska and (b) Hudson Bay. The variable bands are indicated by arrows. Low protein variability in polar bears 101 12 - 13 - 14 - 15 0 $16 - t" - -------------- 18 - Fig. 5. Interpretation of the variation found in polar bear serum proteins separated by starch gel electrophoresis. enzymes and three other proteins representing around 30 loci were resolved. The following sys- tems gave good activity and resolution: non-spe- cific esterase, alkaline phosphatase, acid phos- phatase. catecol oxidase, catalase, malate dehydrogenase, superoxide dismutase, aldehy- deoxidase, isocitrade dehydrogenase, glucose- phosphate isomerase, glucose-6-phosphate dehy- drogenase, 6-phosphogluconate dehydrogenase, NADP-diaphorase, hexokinase, hemoglobin, general proteins, and lipoproteins. Other systems showed only weak activity (leucine aminopepti- dase) or exhibited no activity (lactate dehydro- genase, alphaglycerophosphate dehydrogenase, alcohol dehydrogenase, retinol dehydrogenase, betaglucuronidase, and succinate dehydrogen- ase). The systems showing good resolution were investigated in the 31 bears, and of these 16 systems proved to be monomorphic. In serum, stained for general proteins, there is a system of bands showing variation i n all populations inves- tigated. Eighteen protein bands were resolved by SGE of serum (Fig. 4) where bands 11 and 12 represent transferrins. Band 5 was present in some Greenland bears only. Bands 6 to 10 show a phenotypic variation according to Fig. 5, where phenotype 1 is found in Svalbard and Greenland bears. Phenotype 2 was found in all three popu- lations and phenotype 3 was found in Greenland and Alaska bears. Phenotypes 4 and 5 were found in Svalbard bears and phenotypes 6 , 7 , and 8 were found in Alaska bears. Thus bands 8 and 10 could be found in Alaska bears only. Other bands were found in all populations. The nature and specificity of the variable serum proteins showing phenotypic variability are unknown and owing to the complexity of the variation observed, a genetic interpretation of the phenotypic variation has not been possible. Bears from Svalbard and Greenland, however, shared the same bands in different combinations in con- trast to the bears from Alaska. General proteins were also investigated in the 70 bears from the period 198G1981. The same general pattern of variation as in the 1974 samples was found. There was a greater similarity between the Hudson Bay and the Cape Lisburne samples compared to the samples collected in the Barents Sea. Graphic demonstration starch gel electropho- resis variation is shown in Fig. 5. Isoelectric focusing (IEF) Analysis of serum samples yields from 43 to 63 protein bands in pH gradient 4-6.5 (Fig. 6). Individuals from different populations were com- pared in order to estimate differences between the three populations. Comparisons between sep- arated proteins were performed according to Fig. 6. General protein patterns of polar bear serum separated by isoelectric focusing in a pH gradient of 4-6.5 (lowest pH at top of the gel). Samples from Hudson Bay, Barents Sea and Alaska. 102 T . Larsen et al. Tegelstrom et a l . (1982). Identity values between individuals from the same popdation range from 0.81 to 1 .OO. Intrapopulational identities esti- mated for the different populations are: Hudson Bay 0.95 ? 0.03 (n = 8), Cape Lisburne 0.95 * 0.03 (n = 8), and Barents Sea 0.92 * 0.06 (n = 8). There is less identity within the Barents Sea material and also a greater statistical variation indicating a more heterogeneous population. Comparisons of individuals from different popu- lations gave t h e following interpopulation ident- ities: Barents Sea versus Hudson Bay 0.93 ? 0.05 (n = ll), Barents Sea versus Cape Lisburne 0.94 * 0.05 (n = 12), and Cape Lisburne versus Hudson Bay 0.96 * 0.04 (n = 13). The identity between populations is very high, and not stat- istically different from the identity between ani- mals from the same population. There is no stat- istical identity difference between the three investigated populations. There is, though, an indication of a higher identity between bears from the North American continent compared to the animals sampled in the Barents Sea. With the higher resolution achieved by I E F compared to standard gel e~ectrophoresis, we could not find any protein diagnostic for the populations inves- tigated, although there is variation within populations. Nonspecific esterases were also investigated by I E F in a p H gradient 3-10, where 11 bands could be resolved. There was no difference between populations. Horizontal polyacrylamide gel electrophoresis ( H P A G E ) Polar bear serum yields from 20 to 27 protein bands (Fig. 7). Variation is present in the different populations, but a genetic interpretation for the observed variations is unlikely. The band which probably represents the a-foeto-protein is divided into two distinct bands in some bears in all popu- lations. The most probable explanation is that the extra band is the result of age and not a genetic difference between individuals. There was vari- ation in a fast moving protein where an extra band was observed in two individuals from the Alaska population. This variation was also observed when the proteins were separated by two-dimensional electrophoresis (Figs. 7 and 8). Fig. 7. Horizontal polyacrylamide electrophoresis of polar bear serum proteins. Samples from Alaska Bands with a presumed genetic variation are indicated. These were also resolved by two-dimensional electrophoresis. Fig. 8 Two-dimensional electrophoresls of polar bear serum proteins. Samples from Alaska. The bands with a presumed genetic variation found in HPAGE are illustrated with arrows. Two-dimemiona[ e/ectrophoresk (2-DE) Patterns Of some samples are shown in Fig. 8. Low protein variability in polar bears 103 The prealbumin system with probable genetic variation, also found in HPAGE, is indicated. N o other variation was revealed by this technique. All together around 30 enzyme or protein loci were investigated by SGE, and approximately 35 additional proteins and 10 enzymes in serum were resolved by IEF, representing around 45 loci. HPAGE and 2-DE probably revealed some additional proteins not observed by the other two methods. Thus around 75 loci were analyzed all together. Discussion Efforts have been made in recent years to evaluate the potential of genetic information in the man- agement of wildlife populations. Levels of genetic variation have primarily been estimated by means of electrophoresis. The method, however, has often proved less useful owing ta the usually low level of variability found in large mammals. Selander & Kaufman (1973) pointed out that vertebrates generally show lower levels of elec- trophoretically detectable variation than inver- tebrates. They propose that large and highly mobile animals could show lower levels of vari- ation than smaller and less mobile animals. The level of genetic variation in terms of protein polymorphism seems to be substantially reduced in natural populations of large compared with small mammals (McDermid et al. 1972; Bonell & Selander 1974; Allendorf et al. 1979; Bruce & Ayala 1979; Ryman et al. 1980; Simonsen 1982; Simonsen et a l . 1982a and b). Several explanations have been proposed (Levins 1968; Selander & Kaufman 1973; Ohta 1974; Valentine 1976). The low variation found in the present study of 460 polar bears is in agreement with two pre- vious reports on protein and enzyme variation in bears (Manlove et al. 1980; Allendorf et al. 1979). Manlove et al. (1980) investigated 19 proteins in 233 black bears from six localities covering the species range. They found six polymorphic pro- teins exhibiting two alleles. The low levels of genetic variability in these populations are towards the lower end of the range for mammals. Allendorf et a l . (1979) investigated 12 enzymes by starch gel electrophoresis of sera and red blood cells in 52 polar bears from eastern Greenland and found no variation. The degree of polar bear variation in blood proteins and enzymes was found to be low in this study in spite of the use of high resolution tech- niques. Even proteins usually found to be variable in most species (such as transferrins, GC-vitamin D binding serum protein and non-specific ester- ases) are monomorphic in this species. Of the 75 loci investigated, SGE revealed an unidentified variable protein, and HPAGE another system. The SGE variation indicates a closer genetic relationship between bears from Alaska and Can- ada, compared to Greenland and Svalbard. This protein system could not be found in the IEF, HPAGE, or 2-DE, and has not been further investigated. It may be a tool for further identi- fication of polar bear populations. Extensive maryrecapture programs on polar bears have been made in all Arctic countries since 1966. None of the bears marked in North America have been killed or recaptured in the Eurasian Arctic, or vice versa. Nor have any of the radio-instrumented bears migrated between Europe and North America. The common con- clusion of these studies, observations of abun- dance and distribution, and studies of cranial variation (Manning 1971; Wilson 1976), is that polar bears stay in distinctively different areas for many years. Research suggests the following: Svalbard, Frans Josef Land, Northern Novaja Zemlja and adjacent ice covered seas include the range of one population, which has connections across the Greenland Sea and Northeast Green- land (Larsen et al. 1980; Larsen 1981; Parovsh- chikov 1967). The central East Greenland fjords may have their own small population (Vibe 1981, 1982). In the Canadian Arctic, there may be ten or more populations (Schweinsburg et al. 1982; Stirling et al. 1977, 1978, 1980; Stirling & Kiliaan 1980; Taylor 1982). Bears in the northeast Cana- dian Arctic have connection with bears in north- west Greenland (Schweinsburg et al. 1982; Stirling et al. 1978). In Alaska there may be two different populations (Lentfer 1972, 1974), of which one, in the Chucki Sea, may have connections with bears in the Soviet Wrangel Island and the adja- cent Sibirian coast (Kolz et al. 1978; Taylor 1982). Bears in the central Soviet Arctic probably belong to a separate population (Uspensky & Belikov 1981). But the discreteness of these populations has been questioned (Prevett & Kolenosky 1982; Stirling et a l . 1978; Taylor 1982). Observations from many sources show that bears are often encountered in the central Polar Basin, outside their normal range (Lentfer 1970). Such bears may come from, or migrate to, any Arctic region. 104 T. Larsen et al. Over long periods of time, there may, there- fore, be an exchange of bears and hence genetic material sufficient t o justify the early statement by Pedersen (1945) that bears belong to one com- mon population, at least in a genetic sense. The possibility of recent ‘bottlenecks’ caused by drastic reduction in the number of animals and therefore resulting in lowered levels of genetic variation cannot be rejected, but seems improb- able. There is no evidence of any radical reduction in the number of polar bears in any Arctic region in historical time. It should also be remembered that the polar bear is a relatively young species in the evolu- tionary sense, and is probably not more than 250,000 years old (Kurten 1976). Low blood pro- tein and enzyme diversity and lack of differences between populations may be explained by high selective pressure in a species which lives in a difficult environment and a highly specialized niche. Our conclusion is that the lack of polymorphism in polar bear blood proteins and enzymes reflects the generally low level of protein variability that has been observed in other large mammals. Even if there is little exchange of individuals between polar bear populations, such exchanges may be sufficient to create genetic uniformity over time. This fact, combined with the possible high selec- tive pressure described above. may explain the lack of polymorphism observed in this study. Acknowledgements. - This study was sponsored by the Vor- wegian Polar Research Institute and by grants from the Nilsson-Ehlc and Erik Philip-Sorensen foundations. Blood samples were kindly provided by E . Born, C . Jonkel. J W. Lentfer, and C. V i b e We are grateful for help and support from polar bear projects under the Alaska Department of Fish and Game. US Fish and Wildlife Service. and the Canadian Wildlife Service. We received help from the Sysselmann (Gov- ernor) in Svalbard. from the crew and assistants o n the Nor- wegian Polar Research Institute Svalbdrd expeditions, and from the organizers of the Swedish ‘YMER 80’ expedition. Thanks arc expressed to those who aided our research and who par- ticipated in the practical work in the laboratories: Britt-Marie Bergquist, D r . U. S Seal. and Per-Ivan Wyoni. We thank Dr. David Clapman for revising the English text and Professor Karl Fredga. Dr. Johan B. Steen, and Dr. Nils Chr. Stenseth for valuable comments on the manuscript. References Allendorf, F. W , Christiansen, F. B.. Dobson. T.. Eanes. W. F. & Frydenberg, 0. 1979: Electrophoretic variation in large mammals. I The polar bear. Thalarctos maritimus. Hereditas 91 I 19-22. Boettcher, E . W.. Kistler. P. &? Nitschmann, H . 19%: Method of isolating the &metal binding globulin from human blood plasma. Nature 181, 4 W 9 1 . Bonell, M. L . & Selander, R. K . 1974: Elephant seals. Genetic variation and near extinction. Science 814, 908-909. Bruce, E. J . & Ayala, F. J . 1979: Phylogenetic relationships between man and the apes. Electrophoretic evidence. Evo- lution 33, 104CL1056. Chen, S.-H. & Sutton, H . E. 1967: Bovine transferrins: sidic acid and the complex phenotype. Genetics 56,425-430. Coppenhauer, D . & Buettner-Janusch, J . 1970: Transferrin in Cercopithecinae. Folia. Primat. 13, 2 S 3 4 . Gahne, B . , Juneja, R . K. & Grolmus, J. 1977: Horizontal polycrylamide gradient gel electrophoresis for the simulta- neous phenotyping of transferrin. post-transfertin, albumin and post-albumin in the blood plasma of cattle. Anim. Blood Grps Biochem. Genet. 8 . 127-137. Harris, H. & Hopkinson, D. A . 1976: Handbook of Enzyme Elecnophoresh in Human Genetics. North-Holland Publish- ing Company. Juneja, R . K . , Lundin, L . G. & Gahne, B . 1981: Genetic polymorphism of an a-protease inhibitor in mink plasma. Hereditas 94, 249-252. Kolz. A. L., Lentfer, J. W. & Fallek, H . G . 1978: Polar bear tracking via satellite. Int. ISA Biomed. Sci. Imtrumenration S y m p . 15. 137-144. Kurtkn. B. 1976: The Caoe Bear Story. Columbia University Press. New York. 163 pp. Larsen, T. 1971: Capturing. handling and marking polar bears in Svalbard. 1. Wildl. Manage. 35 (I), 27-36. Larsen. T. 1981: Distribution, numbers, and population charac- teristics of polar bears in Svalbard. Report to the 8th meeting o f t h e IUCN Polar Bear Specialist Group. Mimeo. 14 pp. Larsen, T . , Jonkel. C. & Vibe, C . 1980: Satellite radiotracking of polar bears between Svalbard and Greenland. In Meslow, C. (ed.). Bears - Their Biology and Management. Bear Biology Assor. Conf. Ser. 4 . In press. Lentfer. J . W. 1970: Polar bear research and management in Alaska 1 9 6 b 6 9 . IUCN Publications New Series Suppl. Paper 29, 45-66. Lentfer. J . W . 1972: Alaska polar bear research and manage- ment 197C-71. IUCN Publication New Series Suppl. Paper 35, 29-39. Lentfer. J . W . 1974: Discreteness of Alaskan polar bear popu- lations. International Game Biol. 1 1 , 323-329. Levins. R . 1968: Evolution in Changing Environments: Some Theorertcal Explorarions. Princeton University Press, Princeton. N.J. 120 pp. Manlove. M. N . . Baccus. R.. Pelton. M . R . , Smith, M. H. & Graber. D. 1980: Biochemical variation in the black bear. Pp. 3 7 4 1 in Martinka. C. & McArthur, K. L. (eds.), Bears - Their Biology and Management. Bear Biol. Assoc. Conf. Ser. 3. 375 pp. Manning. T . H . 1971: Geographical variation in the polar bear. Canadian Wildlife Service Report Series N o . 13. 27 pp. McDermid. E. M.. Ananthakrishan, R. & Agar, N . S. 1972: Electrophoretic investigation of plasma and red cell proteins and enzymes of Macquarie Island elephant seals. Anim. Blood Groups Biochem. Genet. 3 , 85-94, Ohta, T. 1974: Mutational pressure as the main cause of mol- ecular evolution and polymorphism. Nature 252, 351-354. Parovshchikov, V. Y . 1967: The polar bear in Frans Josef Land. Problems of rhe North 1 1 , 149-159. Pedersen, A . 1945: Der Eisbar. Verbreitung und Lebemweise. E . Bruun & Co., Kebenhavn. 166 pp. . Low protein variability in polar bears 105 Prevett, J. P. & Kolenosky, G. B. 1982: The status of polar bears in Ontario. Le Naturaliste Canadian 109(4), 933-939. Raymond, S. 1962: Convenient apparatus for gel electropho- resis. Clin. Chem. 8 , 44-470. Ryman, N . , Reuterwall, C., Nygren, K. & Nygren, T. 1980: Genetic variation and differentiation in Scandinavian Moose (Alces alces): are large mammals monomorphic? Evolution 34, 1037-1049. Schweinsburg, R . E., Lee, J. L. & Latour, P. B. 1982: Distri- bution, movement, and abundance of polar bears in Lancaster Sound, Northwest Territories. Arctic 35, 159-169. Selander, R. K. & Kaufman, D. W. 1973: Genetic variability and strategies of adaptation in animals. Proc. Natl. Acad. Sci. 70, 1875-1877. Shaw, C. & Prasad, R . 1970: Starch gel electrophoresis of enzymes - a compilation of recipes. Biochem. Genet. 4 , Simonsen, V. 1982: Electrophoretic variation in large mammals. 11. The red fox, Vulpes uulpes, the stoat, Mustela erminea, the weasel, Mustela niualis, the polar cat, Mustela putorius, the pine marten, Martes martes, the beech marten, Martes fiona, and the badger, Meles meles. Hereditas 96, 299-305. Simonsen, V., Allendod, F. W., Eanes, W. F. & Kapel, F. 0. 1982a: Electrophoretic variation in large mammals. 111. The ringed seal, Pusa hispida, the harp seal, Papophilus groen- landicur, and the hooded seal. Cystophora cristata. Hereditas 97, 87-90. Simonsen, V., Born, E. W. & Kristensen, T. 1982b: Electro- phoretic variation in large mammals. IV. The atlantic walrus, Odobenus rosmarus rosmarus (L). Hereditas 97, 91-94. Stirling, I . , Jonkel, C., Smith, P., Robertson, R. & Cross, D. 1977: The ecology of the polar bear along the western coast of Hudson Bay. Canadian Wildl. Serv. Occas. Paper 33, 64 PP. 297-320. Stirling, I., Schweinsburg, R. E., Calvert, W. & Kiliaan, H. P. L. 1978: Population ecology of the polar bear along the proposed Arctic Islands Gas Pipeline Route. Final report to the Environmental Management Seruice, Department of the Environment, Edmonton, Canada, 93 pp. Stirling, I . , Calvert, W. & Andriashek, D. 1980: Population ecology studies of the polar bear in the area of south-eastern Baffin Island. Canadian Wildl. Serv. Occas. Paper 44, 33 pp. Stirling, I. & Kiliaan, H. P. L. 1980: Population ecology studies of the polar bear in northern Labrador. Canadian Wildl. S e w . Occas. Paper 42, 21 pp. Taylor, M. K. 1982: The Distribution and Abundance of Polar Bears in the Beaufort and Chucki Seas. Ph.D. thesis, 2 vols. Tegelstrom, H . , Wyoni, P . 4 . & Ryttman, H. 1982: Thin layer isoelectric focusing as a taxonomic tool. Ann. Zool. Fenici. 19, 20?-207. Uspensky, S . M. & Belikov, S. E. 1981: Polar bear research and management in the USSR 1979-80. IUCN Publ. New Ser. Suppl. Paper. In print. Valentine, J. W. 1976: Genetic strategies of adaptation. pp. 78-94 in Ayala, F. J . (ed.), Molecular Evolution, Sinauer Ass. Vibe, C. 1981: Miscellaneous information relating to polar bears in Greenland. Proceedings of the 8th working meeting of the IUCN/SSC Polar Bear Specialist Group. In print. Vibe, C. 1982: Grenlands hvide bjerne. Naturens uerden 1982, Wilson, D. E. 1976: Cranial variation in polar bears. Pp. 447-453 in Pelton, M . R., Lentfer, J . W. & Folk, E . G . (eds.). Bears - Their Biology and Management. IUCN Publ. New Ser. 40, 467 pp. 353 pp. 41-56.