Postglacial emergence and distribution of late Weichselian ice-sheet loads in the northern Barents and Kara seas, Russia

Steven L. Forman Byrd Polar Research Center and Department of Geological Sciences, Ohio State University, Columbus, Ohio 43210-1002 David Lubinski Institute of Arctic and Alpine Research and Department of Geological Sciences, University of Colorado, Gifford H. Miller Boulder, Colorado 80309-0450 Jeffrey Snyder Byrd Polar Research Center and Department of Geological Sciences, Ohio State University, Columbus, Ohio 43210-1002 Gennady Matishov Murmansk Institute of Marine Biology, 17 Vladimirskaya Street, Murmansk, Russia 183023 Sergey Korsun Vladimir Myslivets Faculty of Geography, Moscow State University, Moscow, Russia 119899

ABSTRACT straints on deglaciation, areas of maximum Reconstructions of late Weichselian glacier coverage on the continental shelves of the glacier loading, and potential global sea- Russian Arctic range from a large ice sheet terminating in northern Siberia to isolated ice level contribution from former ice sheets on caps restricted to Arctic archipelagos. This disparity in glacier reconstructions reflects the continental shelves in the Russian Arctic. lack of chronological control on glacial and deglacial landforms. We present new Holocene relative sea-level data from Franz Josef Land and northern Novaya Zemlya, Russia, that RAISED BEACHES place the thickest glacier loads in the northern Barents Sea and not over Novaya Zemlya. Raised beaches are ubiquitous on fore- Radiocarbon ages from shelf and terrestrial areas at the former ice-sheet margin support lands that border the Barents Sea and pro- deglaciation of the northern Barents Sea between 13.0 and 10.3 ka, considerably later than vide a high-fidelity record of the response of inferred from isotopic records for the Arctic Ocean. This analysis indicates that the Barents the lithosphere to glacial loading and un- Sea ice sheet was a dominant sea-level reservoir in northern Eurasia, and that glacier loading. The altitude of raised beaches was loading of Novaya Zemlya was comparatively limited during the last glaciation. measured by a Leitz digital altimeter (pre- cision of 1 m) with mean high tide as the INTRODUCTION nary ice sheets in northern Eurasia. We datum (tidal range is Ͻ0.6 m). The rate and One of the largest uncertainties in ice-vol- present new postglacial relative sea-level course of relative sea-level change is deter- ume estimates for the late Quaternary is the records from Franz Josef Land and northern mined by the radiocarbon dating of shell, areal and vertical extent of ice sheets over Novaya Zemlya (Fig. 1) that provide con- driftwood, or marine-mammal bones from the Barents and Kara seas and other shallow shelves bordering northern Eurasia. Glacial- maximum ice-sheet reconstructions range from nearly complete glacier coverage of northern Eurasia by a contiguous marine- based ice sheet (e.g., Grosswald, 1988) to individual ice sheets and ice caps centered on the Arctic archipelagos and advancing onto the adjacent shelf (e.g., Velichko et al., 1984). The discrepancy between reconstruc- tions is equivalent to a global sea-level con- tribution of 5 to 10 m. The discovery of a glacial diamicton in the Barents Sea and moraine banks on the Sval- bard continental slope leaves little doubt that the Barents Sea was covered by an ice sheet during the late Weichselian (Solheim et al., 1990; Gataullin et al., 1993). This ice sheet supported outlet glaciers that filled fjords and sounds on Spitsbergen and ter- minated at the shelf edge (Lehman and For- man, 1992; Mangerud et al., 1992). The tim- ing of deglaciation and glacial-isostatic compensation is well documented for Sval- bard, the western sector of the Barents Sea ice sheet (Forman, 1990; Lehman and For- Figure 1. Inferred isobases of emergence since 5 ka for Barents and Kara seas. Isobase pattern man, 1992; Mangerud et al., 1992). How- for northern Fennoscandinavia from Møller (1987) and J. Snyder (1993, personal commun.). Data ever, there is little chronological control on for Svalbard (a) from Forman (1990) and Møller et al. (1992). Altitude of Holocene marine limit marine and terrestrial glacial and deglacial on Kolguev Island (b; Komissarova, 1972), Izvestia Islands (c; Aller and Uhl, 1936), and Yamal Peninsula (d; Gataullin, 1988). Horizontal ruled pattern shows areas below 200 m water depth features in the eastern Barents Sea, Kara in northern Barents Sea. Inset shows inferred isobases of emergence since 5 ka for Franz Josef Sea, and other areas of the Russian Arctic, Land, Russia. Numbers indicate elevation of 5 ka shoreline on islands of archipelago. These confounding reconstructions of late Quater- observations are basis for constructing isobases.

Geology; February 1995; v. 23; no. 2; p. 113–116; 3 figures; 2 tables. 113 ؍ Figure 2. Height-age relation (m asl metres above sea level) for raised beaches in Nordenskiold Bay, northern Novaya Zemlya, Russia.

marine limit and the isobases of emergence increase toward the southwest; the highest marine limit is 49 Ϯ 2 m asl on Belle Island. A southwest tilt of strandlines on Franz Jo- sef Land and an eastern strandline tilt on Svalbard (Forman, 1990) indicate a thicken- ing ice sheet into the northern Barents Sea. raised beaches with known elevations (Ta- and stabilization of global sea level (Fair- Hooker and Scot Kelty islands are cov- ble 1). banks, 1989). A 35 yr tide gauge record from ered by an extensive raised-beach sequence, Previous reconstructions of a large ice Russian Harbor on the northwest coast of which provides a detailed record of relative sheet over the Kara Sea are based partly on Novaya Zemlya (Fig. 1) indicates that land is sea-level changes after deglaciation (Fig. 3). the identification on Novaya Zemlya of currently emerging at ϳ2 mm/yr (Emery Initial emergence on Hooker Island from raised beaches to heights of 100 to 150 m and Aubrey, 1991, p. 114), similar to rates ϳ10.3 to 8.0 ka was apparently modest (Ͻ8 above sea level (m asl) (Grønlie, 1924; inferred from postglacial uplift. m), reflecting a rate of uplift that just out- Zagorskaya, 1959; Kovaleva, 1974; Gross- We have identified higher raised beaches paced the rate of eustatic sea-level rise. A wald, 1988). In contrast, we found a uni- on Novaya Zemlya up to ϳ100 m altitude, radiocarbon age of 8340 Ϯ 100 yr B. P. on formly low postglacial marine limit of 11 Ϯ 1 but they are covered by a discontinuous and in situ paired valves of the mollusk Mya trun- m asl on northwestern Novaya Zemlya, be- thin glacial drift. Mollusks from raised-ma- cata from a nearby raised glacial-marine tween Nordenskiold and Foreigner bays rine deposits beneath this drift and above sediment indicates that outlet glaciers were (Fig. 1). The marine limit is isochronous on the Holocene marine limit yield radiocar- at or behind their present margins by the the northeast coast of Novaya Zemlya and bon ages of Ͼ28 ka (Table 2), indicating that early Holocene, evidence against remnant rises to ϳ18 Ϯ 2 m asl (Fig. 1), indicating a the higher beaches on Novaya Zemlya re- glacier loading dampening emergence. A ra- thickening ice sheet toward the Kara Sea. flect substantial glacier loading from a pre- diocarbon age of 775 Ϯ 65 yr B. P. on drift- Similar low Holocene marine limits have late Weichselian event(s). wood from a raised beach at 1 m asl in a been identified on Kolguev Island, southern Field studies on Franz Josef Land con- protected bay on Hooker Island indicates Barents Sea (11 m asl; Komissarova, 1972) centrated on the raised-beach forelands on that emergence is not complete; the inferred and on the northeastern (4 to 5 m asl; Aller Hooker and Scot Kelty islands adjacent to present emergence rate is 1 to 2 mm/yr. and Uhl, 1936) and the southeastern (10 m the 500-m-deep British Channel. These asl; Gataullin, 1988) Kara Sea coasts forelands are covered by a regressional DISCUSSION: IMPLICATIONS FOR (Fig. 1). Marine records from the mouth of raised-beach sequence, emplaced into an in- GLACIATION OF THE BARENTS Nordenskiold Bay indicate deglaciation of ferred late Weichselian drift. One of the old- AND KARA SEAS coastal areas prior to 10 ka (L. Polyak and est ages for deglaciation in the northern The maximum western expansion of the S. Forman, unpublished data). However, ra- Barents Sea of 10.3 ka was obtained on in Barents Sea ice sheet probably occurred late diocarbon ages of driftwood and marine- situ shells near the marine limit on Hooker in the glacial cycle sometime between 20 and mammal bones found near the marine limit Island in western Franz Josef Land (Ta- 13 ka (Forman, 1990; Mangerud et al., indicate a later transgression of the sea ble 1). Similar deglacial ages have been de- 1992); it was simultaneous with high relative sometime between 5 and 6 ka (Fig. 2), re- termined for Edgeøya, eastern Svalbard sea level (Forman, 1990) and early incursion flecting the dominance of isostatic recovery (Landvik et al., 1992). The elevation of the of North Atlantic surface waters that may

114 GEOLOGY, February 1995 limits (10 to 50 m asl) on Franz Josef Land and Novaya Zemlya are similar to the Aust- rheim area in southwestern Norway, within 150 km of the margin of the Fennoscandi- navian ice sheet (Emery and Aubrey, 1991, p. 86–87; Hafsten, 1983). The similarity in the present and postglacial emergence rates between southwestern Norway and Franz Josef Land and Novaya Zemlya may reflect equivalent ice-sheet loads, assuming similar earth rheology and deglacial history. The re- constructed ice-sheet thickness during the late Weichselian over southwestern Norway is ϳ1500 m (Tushingham and Peltier, 1991; Elverhøi et al., 1993), which is our initial estimate of the thickness of the Barents Sea have nourished glacier growth (Hebbeln et The new relative sea-level data from ice sheet over Franz Josef Land and north- al., 1994). Marine and terrestrial records Franz Josef Land and Novaya Zemlya show ern Novaya Zemlya. The Barents Sea ice from the western margin of the Barents Sea maximum emergence for the past 5 ka and sheet—having an inferred maximum thick- ice sheet show initial retreat sometime be- earlier centered over the northern Barents ness of 2500 m over the northern Barents tween 12.5 and 13.5 ka with glaciers at or Sea (Fig. 1). We infer from the regional iso- Sea and thinning to ϳ1500 m over Franz behind present margins before 9.5 ka (Hald base pattern that Franz Josef Land and No- Josef Land and Novaya Zemlya—was a sub- and Vorren, 1987; Forman, 1990; Svendsen vaya Zemlya sustained a thinner glacier load stantial sea-level reservoir during the last et al., 1992; Lehman and Forman, 1992). than at the hypothesized center of the ice glaciation, accounting for ϳ6 m of global Radiocarbon ages from forelands on Franz sheet over Kong Karls Land, Svalbard, sea-level rise (cf. Nakada and Lambeck, Josef Land indicate full deglaciation of in- where Holocene raised beaches have been 1988; Tushingham and Peltier, 1991). terisland channels by or before 10.3 ka. Ter- identified to 110 m asl (Salvigsen, 1981). The The 1500 m estimated ice-sheet thickness restrial and shelf records from the Barents pattern of postglacial emergence indicates over Novaya Zemlya is, however, at least Sea area indicate a later deglaciation than that the Barents Sea ice sheet was the dom- 40% thinner than previous estimates of 2500 inferred from a light ␦18O isotopic interval inant load, and glacier coverage of Novaya to 3000 m (Grosswald, 1988; Tushingham between 16.0 and 13.0 ka in Arctic Ocean Zemlya was comparatively limited. Varying and Peltier, 1991; Elverhøi et al., 1993), im- marine cores (Jones and Keigwin, 1988; directions of glacial striae on Svalbard (In- plying a thinner Kara Sea ice sheet. A 40% Stein et al., 1994). This light isotopic signal golfsson et al., 1992) support the presence of thinner ice sheet over Novaya Zemlya would may not necessarily reflect ice-sheet col- an ice divide in the northern Barents Sea; ice reduce the global sea-level contribution lapse, but may signal drainage of ice-mar- was funneled through fjords and sounds on from the Kara Sea ice sheet to Ͻ4 m, less ginal lakes on Siberia or increased discharge Svalbard. than previous estimates of ϳ7 m (Tushing- of the Ob, Yenisey, and Lena rivers. The disintegration of the marine-based ham and Peltier, 1991; Nakada and Lam- Barents Sea ice sheet may be coincident beck, 1988). with the rapid initial rise in global sea level The pattern of postglacial emergence between 13.0 and 11.5 ka (Fairbanks, 1989) places the maximum ice-sheet load over the and not a response to maximum isostatic de- northern Barents Sea and thinning over No- pression (Jones and Keigwin, 1988). The vaya Zemlya. Modest glacier thicknesses rise in global sea level may have destabilized over northern Novaya Zemlya call into the ice sheet, particularly in the deepest part question the presence of a thick (Ͼ1500 m) of the Barents Sea, the troughs bordering ice sheet over the Kara Sea that terminated the Barents Sea, and the interisland chan- in northern Siberia (Grosswald, 1988; Tush- nels on Franz Josef Land. The recent iden- ingham and Peltier, 1991). We contend that tification of iceberg scour traces at 300 to much of Siberia was distal to moisture 600 m water depths on the Yermak Plateau sources, like northern Alaska, and thus gla- (Vogt et al., 1994) supports the presence of cier extent possibly resembles limited valley- a thick (Ͼ1000 m) ice sheet in the northern glacier and ice-cap expansion characteristic Barents Sea during the late Weichselian. of the Brooks Range, Alaska (Hamilton, The iceberg scour traces imply an ice sheet 1986). Many studies of the eastern Kara Sea decoupled from the sea bed as it entered (Astakhov, 1992), the Laptev Sea, and ad- water depths of Ͼ500 m in the troughs bor- jacent lowlands (Vigdorchik, 1980; Velichko dering the Arctic Ocean. Elevated summer et al., 1989) provide equivocal evidence for insolation values (Berger and Loutre, 1991), coverage by a large ice sheet, but support particularly later in the deglaciation be- the presence of a refugium with large Pleis- tween 12.0 and 9.0 ka, probably accelerated tocene mammals and possibly humans sur- the demise of land-based outlet glaciers of viving throughout the late Weichselian into the Barents Sea ice sheet (Svendsen and the Holocene (Mochanov, 1978; Verescha- Mangerud, 1992). gin and Baryshnikov, 1984; Velichko et al., Figure 3. Height-age relation for raised beaches of Hooker Island and Scot Kelty The current low rates of land emergence 1989; Makeyev et al., 1993; Vartanyan et al., Islands, Franz Josef Land, Russia. (1 to 2 mm/yr) and the relatively low marine 1993).

GEOLOGY, February 1995 115 This analysis of relative sea-level data biostratigraphical methods and radiometric Møller, P., Hjort, C., and Ingolfsson, O., editors, from the northern Barents Sea potentially dating: Norsk Geografiska Tidsskrift, v. 37, 1992, Weichselian and Holocene glacial and reduces the global sea-level contribution p. 63–79. marine history of east Svalbard: Preliminary Hald, M., and Vorren, T. O., 1987, Foraminiferal report on the PONAM fieldwork in 1991: from ice sheets on Russian Arctic continen- stratigraphy and environment of late Weich- LUNDQUA Report 35, 224 p. tal shelves to between 6 and 10 m, consid- selian deposits on the continental shelf off Nakada, M., and Lambeck, K., 1988, The melting erably less than previous estimates of 16 m Troms, northern Norway: Marine Micro- history of the late Pleistocene Antarctic ice (Nakada and Lambeck, 1988; Tushingham paleontology, v. 12, p. 129–160. sheet: Nature, v. 333, p. 36–40. Hamilton, T. D., 1986, Late Cenozoic glaciation Olsson, I., 1980, Content of 14C in marine mam- and Peltier, 1991). The lower estimate on of the central Brooks Range, in Hamilton, mals from northern Europe: Radiocarbon, sea-level contribution from former ice T. D., et al., eds., Glaciation in Alaska, the v. 22, p. 662–675. sheets in the Russian Arctic increases the geologic record: Anchorage, Alaska Geolog- Salvigsen, O., 1981, Radiocarbon dated raised apparent discrepancy between eustatic sea- ical Society, p. 9–50. beaches in Kong Karls Land, Svalbard and level depression and grounded global ice Hebbeln, D., Dokken, T., Andersen, E. S., Hald, their consequences for the glacial history of M., and Elverhøi, A., 1994, Moisture supply the Barents Sea area: Geografiska Annaler, volume to more than 30 m of global sea- for northern ice sheet growth during the last v. 63A, p. 283–292. level equivalent at the time of the last glacial glacial maximum: Nature, v. 370, p. 357–360. Solheim, A., Russwurm, L., Elverhøi, A., and Ny- maximum (cf. Andrews, 1992). Ingolfsson, O., Røgnvaldsson, F., and Sejrup, land-Berg, M., 1990, Glacial flutes, a direct H. P., 1992, The glacial history of western evidence for grounded glacier ice in the ACKNOWLEDGMENTS Kongsøya, Svalbard, in Møller, P., et al., eds., northern Barents Sea: Implications for the Supported by National Science Foundation Weichselian and Holocene glacial and ma- pattern of deglaciation and late glacial sedi- grants DPP-9001471, OPP-9222972 and OPP- rine history of east Svalbard: Preliminary re- mentation, in Dowdesell, J. A., and Scourse, 9223493 and Office of Naval Research grant port on the PONAM fieldwork in 1991: J. D., eds., Glaciomarine environments: Pro- N00014-92-M-0170. We thank the crew of R/V LUNDQUA Report 35, p. 25–50. cesses and sediments: Geological Society of Dalnie Zelentsy, T. Jull at the University of Ari- Jones, G. A., and Keigwin, L. D., 1988, Evidence London Special Publication 53, p. 253–268. zona for AMS 14C ages, and J. Nagy for drafting. from Fram Strait (78ЊN) for early deglacia- Stein, R., Nam, S.-I., Schubert, C., Vogt, C., Fu¨t- tion: Nature, v. 336, p. 56–59. terer, D., and Heinemeier, J., 1994, The last REFERENCES CITED Komissarova, T. S., 1972, Holocene tectonic move- deglaciation event in the eastern central Arc- Aller, G. D., and Uhl, G. F., 1936, The Izvestia ment of Kolguev Island: Leningrad University tic Ocean: Science, v. 264, p. 692–696. Islands: Leningrad, Komintern and Shkola Bulletin, v. 18, p. 108–112 (in Russian). Svendsen, J. I., and Mangerud, J., 1992, Paleocli- FZUim KIM’a, 80 p. (in Russian). Kovaleva, G. A., 1974, Modern movement of Ad- matic inferences from glacial fluctuations on Andrews, J. T., 1992, A case of missing water: miralty Peninsula (northern island of Novaya Svalbard during the last 20,000 years: Cli- Nature, v. 358, p. 281. Zemlya), in Geotectonic conditions for the mate Dynamics, v. 6, p. 213–220. Astakhov, V., 1992, The last glaciation in west search for resources on the shelf of the Arctic Svendsen, J. I., Mangerud, J., Elverhøi, A., Sol- Siberia: Sveriges Geologiska Undersøkning, Ocean: Leningrad, Scientific Research Insti- heim, A., and Schuttenhelm, R. T. E., 1992, v. 81, p. 21–30. tute for the Geology of the Arctic, p. 87–93 The late Weichselian glacial maximum on Berger, A., and Loutre, M. F., 1991, Insolation (in Russian). western Spitsbergen inferred from offshore values for the climate of the last 10 million Krasnozhen, A. S., Baranovskaya, O. F., Zark- sediment cores: Marine Geology, v. 104, years: Quaternary Science Reviews, v. 10, hidze, V. S., Lev, O. M., and Malyasova, p. 1–17. p. 297–318. E. S., 1982, Upper Quaternary deposits of Tushingham, A. M., and Peltier, W. R., 1991, Elverhøi, A., Fjeldskaar, W., Solheim, A., Nyland- the southern Island of Novaya Zemlya, in ICE-3G: A new model of late Pleistocene Berg, M., and Russwurm, L., 1993, The Bar- Blinkov, A. A., ed., Stratigraphy and paleo- deglaciation based upon geophysical pre- ents Sea Ice Sheet—A model of its growth geography of the late Cenozoic of the Arctic: dictions of post-glacial sea level change: and decay during the last ice maximum: Qua- Leningrad, Scientific Research Institute for Journal of Geophysical Research, v. 96, ternary Science Reviews, v. 12, p. 863–874. the Geology of the Arctic p. 40–52 (in p. 4497–4523. Emery, K. O., and Aubrey, D. G., 1991, Sea levels, Russian). Vartanyan, S. L., Garutt, V. E., and Sher, A., land levels and tide gauges: New York, Landvik, J. Y., Hansen, A., Kelly, M., Salvigsen, 1993, Holocene dwarf mammoths from Springer-Verlag, p. 237. O., Slettemark, O., and Stubrup, O. P., 1992, Wrangel Island in the Siberian Arctic: Na- Fairbanks, R. G., 1989, A 17,000 year glacio-eu- The last deglaciation and glacimarine/marine ture, v. 362, p. 337–340. static sea-level record: Influence of glacial sedimentation on Barentsøya and Edgeøya, Velichko, A. A., Isayeva, L. L., Oorshkin, D. B., melting on the Younger Dryas event and eastern Svalbard, in Weichselian and Holo- and Faustova, M. A., 1989, The last glacia- deep-ocean circulation: Nature, v. 342, cene glacial and marine history of east Sval- tion of Eurasia, in Herman, Y., ed., The Arc- p. 637–642. bard: Preliminary report on the PONAM tic seas, climatology, oceanography, geology Forman, S. L., 1990, Post-glacial relative sea-level fieldwork in 1991: Lund, Sweden, Lund Uni- and biology: New York, Van Nostrand history of northwestern Spitsbergen, Sval- versity Department of Quaternary Geology, Reinhold, p. 729–758. bard: Geological Society of America Bulle- p. 61–84. Vereshchagin, N. K., and Baryshnikov, G. F., tin, v. 102, p. 1580–1590. Lehman, S. J., and Forman, S. L., 1992, Late 1984, Quaternary mammalian extinctions in Gataullin, V., 1988, Upper Quaternary deposits Weichselian glacier retreat in Kongsfjorden, northern Eurasia, in Martin, P. S., and Klein, of the western coast of the Yamal Peninsula west Spitsbergen, Svalbard: Quaternary Re- R. G., eds., Quaternary extinctions: Tucson, [Ph. D. thesis]: St. Petersburg, Russia, All- search, v. 37, p. 139–154. University of Arizona Press, p. 483–516. Russian Geological Institute, 230 p. (in Makeyev, V. M., Pitul’ko, V. V., and Kasparov, Vigdorchik, M. E., 1980, Arctic Pleistocene his- Russian). A. K., 1993, The natural environment of the tory and the development of submarine per- Gataullin, V., Polyak, L., Epstein, O., and Ro- De Long Archipelago and ancient man in the mafrost: Boulder, Colorado, Westview Press, manyuk, B., 1993, Glacigenic deposits of the late Pleistocene and early Holocene: Polar 286 p. Central Deep: A key to the late Quaternary Geography and Geology, v. 17, p. 55–63. Vogt, P. R., Crane, K., and Sundvor, E., 1994, evolution of the eastern Barents Sea: Boreas, Mangerud, J., and nine others, 1992, The last gla- Deep Pleistocene iceberg plowmarks on v. 22, p. 47–58. cial maximum on western Svalbard: Quater- the Yermak Plateau: Sidescan and 3.5 kHz Grønlie, O. T., 1924, Contributions to the Qua- nary Research, v. 38, p. 1–31. evidence for thick calving ice fronts and a ternary geology of Novaya Zemlya, Report of Mochanov, Y. A., 1978, Stratigraphy and absolute possible marine ice sheet in the Arctic the scientific results of the 1921 Norwegian chronology of the Paleolithic of northeast Ocean: Geology, v. 22, p. 403–406. Expedition to Novaya Zemlya: Oslo, Nor- Asia, according to the work of 1963–1973, in Zagorskaya, H. G., 1959, Novaya Zemlya, in Qua- way, 367 p. Early man in America from a circum-Pacific ternary geology of the Soviet Arctic: Mos- Grosswald, M. G., 1988, An Antarctic-style ice perspective: Edmonton, Alberta, Canada, cow, Nauka, p. 20–36 (in Russian). sheet in the Northern Hemisphere: Toward a Archaeological Researches International, new global glacial theory: Polar Geography p. 54–67. Manuscript received June 27, 1994 and Geology, v. 12, p. 239–267. Møller, J. J., 1987, Shoreline relation and prehis- Revised manuscript received November 9, 1994 Hafsten, U., 1983, Shore-level changes in south toric settlement in northern Norway: Norsk Manuscript accepted November 16, 1994 Norway during the last 13,000 years traced by Geografisk Tidsskrift, v. 40, p. 77–84.

116 Printed in U.S.A. GEOLOGY, February 1995