bs_bs_banner

Jens Esmark’s mountain traverse 1823 the key to his discovery of Ice Ages

GEIR HESTMARK

Hestmark, G. 2018 (January): Jens Esmark’s mountain glacier traverse 1823 the key to his discovery of Ice Ages. Boreas, Vol. 47, pp. 1–10. https://doi.org/10.1111/bor.12260. ISSN 0300-9483. The discovery of Ice Ages is one of the most revolutionary advances made in the Earth sciences. In 1824 Danish- Norwegian geoscientist Jens Esmark published a paper stating that there was indisputable evidence that and other parts of Europe had previously been covered by enormous carving out valleys and , in a cold climate caused by changes in the eccentricity of Earth’s orbit. Esmark and his travel companion Otto Tank arrived at this insight byanalogous reasoning: enigmatic landscape features theyobserved close to sea level alongtheNorwegian coast strongly resembled features they observed in the front of a retreating glacier during a mountain traverse in the summer of 1823. Which glacier they observed up close has however remained a mystery, and thus an essential piece of information in the story of this discovery has been missing. Based on previously unknown archive sources, supplemented by field study, I here identify the key locality as the glacier Rauddalsbreen. This is the northernmost outlet glacier from Jostedalsbreen, the largest glacier in mainland Europe. Here the foreland exposed byglacier retreat since the Little maximum around AD 1750 contains a rich collection of glacial deposits and erosional forms. The point of enlightenment is more precisely identified to be a specific moraine and its distal sandur at 61°53026″N, 7°26043″E. In memoryof Esmark’s travelcompanion who possibly was the first to realise the analogy, it is proposed to name this moraine Otto Tank’s Moraine, a pendant to the already famous Esmark Moraine at Forsand by the sea.

Geir Hestmark ([email protected]), Centre for Ecological and Evolutionary Synthesis, CEES, Department of Biosciences, , PO Box 1066 Blindern, Oslo 0316, Norway; received 18th January 2017, accepted 30th March 2017.

The discovery of Ice Ages counts amongst the greatest • That along the coast Esmark and his companions discoveries ever made in the Earth sciences (Imbrie & observed several traces and deposits close to sea level Imbrie 1986; Rudwick 2008; Kruger€ 2013). In Norway and far from glaciers, and that at some point they this discovery was due to Jens Esmark (1762–1839), realised that these indicated the former presence of professor of mining sciences at the University of Chris- large glaciers down to sea level: (i) at Forsand, close to tiania (now Oslo), who in late April 1824 published a Stavanger and the entrance of spectacular Lysefjord, paper stating that there was indisputable evidence that they observed a long rampart across a valley, consist- Norway and other parts of Europe had previously been ing of an unsorted mixture of sand, gravel and big covered by enormous glaciers carving out valleys and boulders. This ridge is locally known as Vassrygg and fjords, in a cold climate caused by changes in the was later named the Esmark Moraine (Andersen eccentricityof Earth’s orbit (Esmark1824, 1826). Recent 1954; Worsley 2006; Briner et al. 2014); and (ii) at evaluations hail Esmark as ‘the best candidate as the the Sula archipelago at the entrance of the Sognefjord father of current ice age-theory’ (Kruger€ 2013: p. 450), a presumably glacier-polished conglomerate cliff and his ideas as ‘the first distinctly modern formulation (Hestmark 2009a). of the astronomical theory of long-term climate change’ • That during the subsequent mountain traverse (Berger 2012: p. 109). Esmark’s paper preceded by several Esmark and his companions had occasion to study years the later contributions by Swiss savants Ignaz glaciers up close and what glaciers can do to the Venetz in 1829, Jean de Charpentier in 1834, Karl landscape. Schimper and Louis Agassiz in 1837 (Rudwick 2008; Kruger€ 2013). Yet the genesis of Esmark’sIceAgepaper It has often been assumed that Esmark, faced with the has remained much of a mystery. At crucial points his curious Vassryggen at Forsand, there and then realised narrative is frustratingly parsimonious, and leaves the that it was an ancient moraine, and on the spot developed readercravingformoreinformation.Fromthepaperitself his vision of former large scale glaciations down to sea it is evident: level (e.g. Engen 1985). However, his own text suggests rather that a comparison with extant glaciers during the • That the discovery occurred on a long journey along subsequent mountain traverseprovedcrucial in releasing the south and west coast of Norway in the summer of his train of thought, because, after in detail describing 1823, followed by a traverse over the central Norwe- the Vassryggen and the sandur plain in front of it, he gian mountain chain from the west coast to the proceeds: eastern valleyof Gudbrandsdalen, and a return to the ‘Not only the rampart itself, but the whole horizontal plain too capital Christiania. delivers proofs that there has been [a] glacier, because this plain

DOI 10.1111/bor.12260 © 2017 The Authors. Boreas published by John Wiley & Sons Ltd on behalf of The Boreas Collegium This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 2 Geir Hestmark BOREAS

exactly resembles those plains which I found bordering the still extant glaciers between Søndfiord and Lomb in Gudbrandsdalen, Discovery comes to those prepared which I likewise travelled last summer. The resemblance is so ’ striking, that anyone who has an opportunity of making this Esmark s 1824 paper appeared to come out of the blue. comparison, must form the same ideas. I must put forward as Never beforehad he published anything about glaciers or proof of this, that Mr. O. Tank, a skilful young mineralogist, who climate change, and the available sources leave no doubt with me observed the here described rampart and afterwards travelled [with me] to the glaciers mentioned, immediately that he crafted his theory of large scale glaciations from exclaimed to me, without me giving him any pretext, that that observations and deductions made on a field trip lasting rampart by Stavanger had to be a glacier rampart’ (Esmark 1824: 3 months in the summer of 1823. However, I have p. 45; Appendix S1). discovered that by 1823 he had a long history of work This passage – the key passage of discovery strongly related to climate and glaciers. In a hitherto unknown suggests that the moment of enlightenment occurred not handwritten vitae from 1825, submitted on the occasion down at the coast but rather in front of extant glaciers of his election to the Royal Swedish Academy of Science, during their subsequent mountain traverse. Which Esmark points to a decisive episode from his youth: a brings us to the greatest uncertainty that has up to now land surveyor challenged him from the ground to esti- surrounded Esmark’s discovery: which extant glacier(s) mate the height of the bell tower of Houlbjerg church, didhe andhis companionsvisit,and whatdidtheyobserve , where Esmark’s father was rector (Esmark atthoseglaciers?IfitistruethattheideaofanIceAgewas 1825). Frustratingly he was not told the trick but only conceived in front of a Norwegian glacier, that certainly referred to a book on geometry. Entering places it amongst the most significant locations in the University in 1784 Esmark soon dropped out of theology history of the Earth sciences. But alas, here the published and into medicine, then studied mineralogy and physics record is scant, because Esmark never published more and learned to measure altitudes by barometers; in 1794 than the cited passage. This dearth of information has led he became a certified land surveyor himself. By then he to much speculation over the years, for instance that the had alsograduated from the mining academy at the silver crucial sites must be in the Jotunheimen mountain range town in Norway, a country in political union (e.g. Andersen 1992). The present paper for the first time with Denmark up to 1814. After further studies at the identifies and documents Esmark’s crucial locality, and mining academies of Freiberg (Saxony) and Schemnitz shows how a number of features of this locality proved (Slovakia) he from 1798 to 1814 served as Assessor in the particularly favourable for his serendipitous discovery. central mining administration at Kongsberg, and a lec- turerin mineralogy,geologyand physics in theassociated Material and methods mining seminar (Bergseminaret). In these years he accumulated a lot of field observations he would only This paper is based on written documents and field later re-interpret and mobilize as Ice Age evidence (see observations. Unpublished documents are available and below). The Rev. Thomas Malthus, famous for his essay can be consulted in public archives, notably the National on population and its significance for Darwin’s theoryof Archive of Norway (Riksarkivet) in Oslo, the archives of natural selection, visited Kongsberg and Esmarkin 1799, the Centre for the History of Science at the Swedish noting: ‘he means to measure all the mountains by Academyof Science, Stockholm, and the County archive the barometer’(Malthus 1966: p. 113). Indeed, in 1800 (Landarkivet) of Fyn, Denmark. To avoid footnotes, the Esmarkascended and measuredTronfjell(1666 m)anda unpublished archive sources are cited in the text similarly peakintheRondanemassif(>2000 m)(Esmark1800).In to published sources, e.g. (Esmark 1832), and listed with 1801 he summited and measured Snøhetta (‘The Snow- author, date and archival location in the References. The cap’, 2286 m a.s.l.), then believed to be the highest discovery of these unpublished sources required exhaus- mountain in Scandinavia (Esmark 1803; Hestmark tive sifting of archive material over several years, a 2009b). In 1810 he measured Gausta (1883 m) in procedure perhaps best comparable to washing tons of and the high Rjukan waterfall (Esmark 1812). rivergravelfor smallnuggets of gold. All quoted passages Inspired by Horace de Saussure’s studies on the have been translated into English by the author. The physical geography of the Alps, and Louis Ramond’s original Norwegian texts are given in the Supporting studies from the Pyrenees, Esmark now systematically Information. In the running text the quotations are measured tree lines and snowlines in Norway (Esmark identified Appendices S1, S2 etc. 1803). These limits of nature were clearly understood to The fieldwork consisted of several trips to revisit and depend on climate, and meteorology was another of document the localities visited by Esmark and his Esmark’s early passions. At Kongsberg he logged air companions during their voyage in 1823: Forsand with temperature and pressure three times a day from 1799, and the Esmark Moraine, the Sula archipelago, the Mjøsa continued this habit after moving in February 1815 to the boulder terrain, and, most notably, the site of his Norwegian capital to become the first professor of mining mountain traverse, after previously unknown and sciences at the recently established Norwegian university unpublished sources had revealed clues to its probable (Esmark 1799–1838). From January 1818 to December location. 1838 his weather observations appeared in the semi- BOREAS Jens Esmark and his discovery of Ice Ages 3 official journal Den Norske Rigstidende; thus in reality, if Enighed Copper Works, close to the entrance of the not in title, he was the first Norwegian state meteorologist spectacular Lysefjorden, east of Stavanger. State finances (Hestmark & Nordli 2016). His long time concern with were in disarray, every stone had to be turned. In 1822 he temperature variations and snowlines would have pre- had briefly visited this locality and recommended closer pared his mind for considerations on climate change, inspection (Esmark 1823b). He left Christiania (Oslo) on although this background has gone unnoticed in previous 18 June 1823, accompanied by his student Jan Theodor studies of Esmark’s Ice Agework. His vision of an Ice Age Kielland (1803–1844), who later graduated in theology would be the snowline descending to shore. and became a country parson (cf. Kielland 1823, 1897). Unrecognized too until now is the fact that even if Kielland’s participation in this illustrious voyage has not Esmark did not himself observe glaciers up close until previously been known. Their first stop was the manor 1823, except briefly at Snøhetta in 1801, he cooperated Rød at Fredrikshald (now Halden) at the entrance of the closely with a numberof local and foreign naturalists who Oslofjord. Here they picked up Nils Otto Tank (1800– studied Norwegian glaciers, providing them with baro- 1864), another of Esmark’s geology students. Tank is metric reference measurements and his own handcrafted today mainly remembered as the Moravian missionary barometers to ascertain the altitude of current glacier who took up the struggle to improve the conditions of fronts. They include: German geologist Leopold von slaves in the Dutch colony of Suriname in South America Buch (1774–1853), who in autumn 1806 made a field trip where he is still honoured as ‘el libertador’ and as the to the glaciers Folgefonna and Jostedalsbreen on the founder of several Christian communities in the USA Norwegian west coast and subsequently published a (Vormeland 1999; Lampe 2001). From Rød they sailed to paper on the snowline in Norway (von Buch 1812, 1814); Stavanger and went east to the abandoned Enighed Swedish botanist Goran€ Wahlenberg (1780–1851), who Copper Works, only to find the mines unpromising after surveyed the glaciers of Sulitjelma on the border between close examination (Esmark 1823b). On his brief visit the Norway and Sweden in 1807 (Wahlenberg 1808); Nor- preceding year Esmark had however been told by locals wegian botanist Christen Smith (1785–1816) who in 1812 that the copper ore resurfaced on the other side of a steep made the first detailed study of Norwegian west coast hill further east. They now walked around this hill and up glaciers, their dynamics and landscape effects (Smith to lake Haukalivatn, at the bottom of a deep valley. The 1817); Swedish naturalist Sven Nilsson (1787–1883), who roadwent over an area of cultivated ground, then a plain. in 1816 documented moraines at the Svartisen glacier Esmark recounts: (Nilsson 1879); Norwegian country parson Niels Hertz- – ‘At the upper end this sandy plain was bounded by a glacier berg (1759 1841), who published an account of the rampart, which extended across the whole valley. As this glacier Folgefonna glacier (Hertzberg 1818); Gottfried Bohr rampart is remarkable, and, as far as I know, the only one of its (1773–1832), church musician and physicist, who pro- kind lying at sea level in a district, where you first at an altitude duced a substantial study of the Jostedalen glacier (Bohr from 2- to 3000 Rheinlandic feet find a few snowfields lying ’ – perennially in hollows of the mountain on its slopes to the north- 1819, 1826); Esmark s students Christian Boeck (1798 east, I will describe it more closely. Its length across the valley, 1877) and Baltazar M. Keilhau (1797–1858), who studied from mountain to mountain, is 2250 feet, its perpendicular and ascended glaciers in Jotunheimen and Jostedalen in height, up from the plain, is 100 feet; at the one end where it borders the mountain, it is broken through, so that its uppermost 1820 (Keilhau 1820, 1822), and German geologist Carl or highest brink is only 12 feet higher than the plain. This opening Friedrich Naumann (1797–1873), who visited the Folge- or breach is approximately 200 feet broad. The rampart itself fonna and Jostedalen glaciers in 1821–22, reporting to consists of coarse gravel and sand, mixed with many extraordi- Esmark (Naumann 1822). Thus, by 1822 Esmark was nary big boulders of gneiss, which is here the prevailing rock type in the bedrock’ (Esmark 1824: p. 45; Appendix S2). exceptionally well informed about Norwegian glaciers. That year he obtained funds to explore the mineral This description of the ridge locally called Vassryggen riches of the Norwegian south coast, and there, besides (‘the lake ridge’) and later named the Esmark Moraine is defining the rock type ‘norite’, came to wonder about the from Esmark’s paper published the following year. It is geomorphology of valleys and fjords (Esmark 1823a). however very doubtful that he on the spot recognized that He was fascinated by the clean swept valley sides and the it was a moraine produced by a glacier. His or Tank’sfield absence of any debris beneath these steep walls. Some- notes would possibly have provided a decisive answer, but thing had swept away all those detached rocks, but what? no such notes have yet been found. Today most of this Hesuggested ‘strong currents’.Buthis returntothesame moraine, dated to the Younger Dryas stage of the last landscape the next year would provide a different and major glaciation of Norway, c.12000–11 000 a BP surprising answer. (Briner et al. 2014), is unfortunately covered by vigorous pine forest planted in the 20th century, obscuring most of Moraine ridges and polished cliffs at sea level its surface features (Fig. 1B). However, due to its ancient the first crucial localities status as commons and pasture, the moraine and the plain in front remained in a more or less undisturbed, In May 1823 the Norwegian Ministry of Finance, Trade pristine condition well into the 20th century (Engen and Customs asked Esmark to evaluate the abandoned 1985). Fortunately there exists an old aerial photograph 4 Geir Hestmark BOREAS documenting this condition (Fig. 1A). In his paper of Esmark and his students then proceeded north and 1824 Esmark describes how the lake Haukalivatn behind arrivedinBergenon12August.Heretheywenttoseethe the gravel ridge fills the valley floor, and how the wall has freshly appointed bishop Jacob Neumann (1772–1848), a v-shaped gap where water has evidently at some former who had once been Tank’s private tutor and who had just time flowed out. His narrative then makes a gigantic leap: returned from a visitas to his more remote parishes, ‘From this description one will clearly be able to realize, that this including the Jostedalen valley at the mighty Jostedals rampart has its origin from masses of ice, which have filled the glacier. To the bishop, approaching the Nigards glacier whole valley and by their expansion and pressure have also tongue, no ‘revelation in nature, of the many wonderful hollowed out the floor of the valley.In all probability the water of greatscenesitpresentsin such richabundanceto the human the melted ice later burst through the rampart and had its outlet through theopeningatatimewhenits presentoutlet [onthenorth eye,havemadethedeepimpressiononmeasthesightofthis side of the lake] either did not exist or was blocked with ice and glacier. [...] tears stood in my eyes, and I have never with gravel [...]’ (Esmark 1824: p. 45; Appendix S3). deeper or holier emotion worshipped God in his works’ This is immediately followedby the passage cited in the (Neumann1824:p.540;AppendixS4).Itseemslikelythat Introduction of the present paper where the comparison the bishop communicated his impressions to his visitors. is made to the phenomena encountered at extant glaciers Yet,when Esmark & co. now went farther north along in the mountains ‘between Søndfjord and Lom’. Before the west coast, it was not to examine glaciers but to that comparison could be made, however,they had noted inspect anothercopper mine,atGrimeli (Esmark1823b), several significant features of Vassryggen: it lies across and also investigate an old claim by former Bergen the valley, it consists of an unsorted mixture of sand, bishop Erik Pontoppidan in his Natural History of gravel, rocks and big boulders (Fig. 1C), it has a breach, Norway (Pontoppidan 1755: p. 54) that at Stene Sound in an outlet where dammedwater had once flowed through, the Sula islands, there was a locality spectacularly rich in creating a sandy plain in front. fossils. No fossils were however found.

A

BC

Fig. 1. A.The ancientmoraineVassryggen nownamedtheEsmarkMoraineandthesandurplaininfrontinfairly pristineconditionbeforethe moraine was planted with pine and the sandur completely cultivated. The farm Grønli in front. Aerial photograph 1940s. B. The moraine today, overgrown by pine forest planted after WorldWar II. C. Big boulders onunplanted part of moraine. Photographs (B) and (C) 28 October 2008. Site No. 3 in Fig. 8. [Colour figure can be viewed at www.boreas.dk] BOREAS Jens Esmark and his discovery of Ice Ages 5

‘To the contrary I found, that the mainland and the island Indre- the central Norwegian mountain chain, to the Gud- or Østre-Sulen separated by the sound consisted of a solid brandsdalen valley in the east, from where they could conglomerate, composed of rounded boulders, from the size of a peato that of a man’s head, yes even to the size of an alen [0.63 m] reach Christiania in a few days by carriage service on the in diameter. [...] even though the rock type, as mentioned, main road from Trondheim. The significance of this consisted of rounded boulders cemented together,these cliff sides decision cannot be overestimated. It was on this moun- were anyway quite even and smooth. [...] I cannot explain this phenomenon in another way, than by thinking this having tain traverse they got to observe extant glaciers and their happened by the great ice masses, which, pressing themselves landscape effects up close. And these observations through the sound, have abraded/polished the cliff walls running proved crucial to the discovery. But where exactly did parallel to the direction of the sound’ (Esmark 1824: p. 46; this traverse take place? Appendix S5). A letter from Esmark, until now slumbering in a This was another observation of glacier action at sea government file, provides the answer: ‘As I a number of level, far from extant glaciers (Fig. 2). It connected to a years ago travelled from Opstryen in Sondfjord€ over question Esmark had posed the year before: the mystery the snowglaciers to Lomb I hit upon a very unpleasant of missing rocks and clean swept valley and sides weather with rain and fog, so that I only after (Esmark 1823a). And it is in this context he introduces 36 hours came down into Bratedalen’ (Esmark 1834; Stene Sound in his 1824 paper: the geomorphology of Appendix S6). Thus: he started at the rural district Norway’s valleys and fjords, their carving out by ice, and Oppstryn and finished at Bratedalen, and this restricts the transportation away of eroded material by glaciers. the possible traverses very much. His letter also reveals His text on the conglomerate could again suggest that he onemajorsourceoftheconfusionthathaslingeredabout had now,at this stage of his voyage, recognized large scale his mountain traverse. Because although Oppstryn was glaciation. Yet, again these observations may have situated in the county Nordfjord & Søndfjord Fogderi, it received their ‘glacier-interpretation’ afterwards, and wassituatedintheNordfjordpart,andnotinSøndfjordas not in situ. That it was perfectly possible to observe these stated in his paper! At least he was consistent in his error, polished surfaces at Sula without seeing glacier effects is because he repeats it in another previously unknown demonstrated by the German geologist Carl Naumann letter about the same traverse (Esmark 1832). The most who had visited the site the previous year, and concluded common route between Oppstryn in the west and the that the conspicuously polished surface (glatter abges- eastern districts of Skjak and Lom was the track Hjelle- chliffener Oberflache€ ) was due to ‘destructive currents’ Folva-Sunndalen-Kamperhamrane-Rauddalen-Fram- (zerstorenden€ Fluthen) (Naumann 1824: p. 120). rusti-Bratan (Standal 1995; Fig. 3). An overlooked newspaper note by Esmark’s colleague, the university The point of enlightenment: the Rauddals glacier chemistry professor Jacob Keyser, shows that this was and Otto Tank’s Moraine indeed the path followed by Esmark, Kielland, Tank and probably a guide from Oppstryn. Keyser states that Their voyage had lasted longer and brought them farther Esmark after investigations in the Bergen area returned up the coast than initially planned. For Esmark it was ‘over Lomb, through Raudalen valley and Braaten’ now urgent to get back to the University for his autumn (Keyser 1834; Appendix S7). This explicit mention of lectures. A return along the coast could take weeks, and Rauddalen excludes some other possible routes from they had anyway ‘been there, done that’. He now made a Oppstryn to Bratan described in Standal (1995). crucial decision: he decided to go inland, crossing over They sailed northwards from Grimeli and up Nord- fjord to Stryn, from where 8 km up along the river you reach lake Strynsvatnet, and on its western shore the farms of Oppstryn. From the other side of the lake the route takes you up some 3 km up to the entrance to the narrow Sunndalen valley. The south side of this valley is lined by peaks reaching up to 1892 m, and a ‘necklace’ of hanging glaciers. Sunndalsætra summer farm at 460 m is the usual night quarter for people going east. Rising early the next morning one could reach the other side of the mountain range in the course of a long summer day. Leaving the summer farm, reaching the tree line at 600 m, the path steeply ascends towards the mountain pass Kamperhamrane at 1300 m. Although the track was here paved with flagstones so that cattle and horses couldbetowedtothepass,60-year-oldJensEsmarkmust Fig. 2. The polished Devonian conglomerate at Stene Sound. Pho- have been pretty exhausted when he reached the top. He tograph 26 April 2009. Site No. 4 in Fig. 8. [Colour figure can be viewed was about to make the discovery of his life. at www.boreas.dk] 6 Geir Hestmark BOREAS

Fig. 3. Esmark’s route from lake Strynsvatn to Bratan (red line), the mountain-glacier traverse that led to the discovery of an Ice Age. Former northward extension of the Rauddalen glacier marked with white line. Site No. 6 in Fig. 8. [Colour figure can be viewed at www.boreas.dk]

A big glacier tongue apparently blocked their path: big boulders lined up as giant fences, evidently pushed up Rauddalsbreen (The Red Valley Glacier), the northern- the slope by the glacier, roughly 100 m above valley most tip of the mighty Jostedals glacier (Fig. 4). Today bottom. Polished cliff surfaces everywhere emerge from this glacier tongue has almost completely melted away, the gravel and boulder fields, swept free of debris, with leaving a large, exposed foreland called Merradalsbotn. sculpted surface forms unrelated to the structure and At its maximum extent during the Little Ice Age (LIA) c. veins in the (gneiss) bedrock. Here and there are rows of AD 1750, Rauddalsbreen crossed the valley and reached small moraines, with a mixture of boulders, pebbles and roughly a hundred metres up on the north side. By 1823 sand. These landscape features would all be noted as the glacier had receded somewhat from this LIA max- evidence of former glaciation in presently non-glaciated imum, some 150–200 m to judge from other dated areas in Esmark’s paper. sequences in the same area (Winkler et al. 2003), leaving The track now continued over the flat glacier itself. a rim of recently exposed areas, exhibiting many of the They stepped onto the ice walking eastwards a little more phenomena associated with glacial action. Esmark had than a kilometre to the other side (Fig. 5). Stepping often read about such phenomena; now he could observe down into the small foreland on the eastern side they hit them up close. Most prominently the sharp border of the upon a large ridge of gravel and boulders parallel to the foreland itself, between vegetated and unvegetated land, glacier front a terminal- and side-moraine (Fig. 6A). they stepped from grass to gravel in a single step (Fig. 4). Deposited at the LIA maximum it has an uncanny This highlights the glacier as a dynamic body of ice, resemblance to the Esmark Moraine (Vassryggen) they advancing and retreating. At this western border of the had seen close to sea level at Lysefjorden (cf. Fig 1A), foreland several very large rounded single boulders rest although scaled downto25%size– itis 4–5 mhighrather directlyon clean-swept, polished cliff surfaces. They look than 20–30 m. Like the Esmark Moraine it runs at a right remarkably detached, like the big boulders (erratics) angle to the valley direction and consists of an unsorted scattered all over Scandinavia (Fig. 4). At the foreland mixture of boulders, pebbles, sand and clay a diamic- limit in the north slope of thevalleytheycould see rows of ton pushed in place by the glacier. A v-formed gap in the moraine lets water through from the melting glacier behind, resembling the gap they had noted in the Esmark Moraine. Given Otto Tank’s exclamation about the similarity toVassryggen cited above,it seems appropriate to name this moraine at Rauddalsbreen Otto Tank’s Moraine in honour and memory of Esmark’s enthusi- astic young student. While much has been made of the significance of the Esmark Moraine in previous analyses of Esmark’s discovery (e.g. Strøm 1950; Andersen 1992; Worsley 2006), it has been largely overlooked that Esmark puts equal weight on the sandur in front of the moraine. The outwash plain east of Otto Tank’s Moraine is indeed a beautiful example of this feature created by glacial streams, with braided rivers depositing their light parti- cles in the aptly named Ytste Leivatnet (Upper Clay Fig. 4. The western approach to the Rauddalsbreen glacier and its Lake) at 1015 m a.s.l. (Figs 5, 6B). The sandur is app- foreland. Green vegetation in front is outside LIA maximum extent. roximately 350 m across, covering the entire valley Big boulders deposited by the ice resting on smooth polished bottom, and 1.1 km long. bedrock surfaces. The suggested extension of the glacier around 1823 is indicated by the white line. Photograph 19 September 2008. Site No. 6 Here then, at this moraine and outwash plain on the 0 in Fig. 8. [Colour figure can be viewed at www.boreas.dk] east side of the Rauddalen glacier at 61°53 26″N, BOREAS Jens Esmark and his discovery of Ice Ages 7

Fig. 5. The Rauddalsbreen glacier foreland. White line indicates approximate position of glacier in 1823, and arrow points to the distinct LIA maximum extent. White arrows show the probable track of Esmark and his companions over the glacier. To the right black arrows point to Otto Tank’s Moraine, and the sandur plain with braided glacial rivers running towards lake Ytste Leirvatnet. Remnant of glacier at bottom of page, northernmost tip of Jostedalsbreen. Base photograph Copyright: Google 2013. Image CNES/Spot Image 2014. Site No. 6 in Fig. 8. [Colour figure can be viewed at www.boreas.dk]

7°26043″E 1020 m a.s.l., is the point of enlightenment, to sea level at Stavanger. Implying? A climate signifi- the moment where with certainty we can say that Esmark cantly colder than the present. Here a scientific revolu- and his companions, reasoning by analogy, realised that tion took place: acompletely new readingofthe landscape. the rampart Vassryggen at sea level at Lysefjordenwas an That it was perfectly possible to walk over the ancient moraine, a ridge of graveland boulders deposited Rauddalen glacier and its moraines without getting such by a glacier. And the implication? Enormous glaciers visions is again attested by German geologist Naumann must previously have covered Norway and reached down who made the same traverse in the opposite direction the previous year, describing the moraine and the glacier (Naumann 1824: p. 210–223). Indeed Naumann only reluctantly became convinced of the realityof Ice Ages in A the late 1840s (Naumann 1848). From the sandur the track continued to the lake Rauddalsvatnet, the summer farm Framrusti, and over a small ridge to the farm Braten (Fig. 3). From there they followed the valleys to lake Mjøsa, and took the main road along the east side. Here they passed through areas B where very large erratic boulders are scattered over the forest floors, particularly prominent between Løten and Tangen. These boulders received particular note in Esmark’s glacial paper of 1824, and it is thus probable that he and his companions discussed them in relation to what they had just seen of boulder transport by the glacier (Fig. 7A, B). This locality is far east, close to Sweden, and would indicate glaciation not only on the west coast but all over Norway. Returning to Christiania in mid-September their voyage had lasted 3 months (Fig. 8).

Fig. 6. A. The present is the key to the past. Otto Tank’s Moraine The paper,the proofs and an astronomical theory seen from east towards west, 1020 m above sea level, deposited at LIA of climate change maximum c. AD 1750. Photograph 19 September 2008. Compare Fig. 1A. B. The sandur east of Otto Tank’s moraine. Arrow indicates Esmark’s loans at the University Library reveal that he, the moraine. Ytste Leirvatnet lake in front with fine-grained only days after his return, set to workon a paper he could sediments deposited by the glacial river. In the background the not have imagined when leaving town a few months remnants of Rauddalsbreen glacier, the northernmost tip of the Jostedalen glacier. Compare Fig. 1A. Photograph 19 September 2008. before he took out works on Swiss glaciers, on erratic Site No. 6 in Fig. 8. [Colour figure can be viewed at www.boreas.dk] boulders and on the Antarctic seas a model for an 8 Geir Hestmark BOREAS

A snow, namely his observations in the summer of 1823, although he evidently also searched his back-list mobilizing observations made much earlier reinter- preting them for his new reading of the landscape. Precisely this seems to be his greatest achievement: his subsumption of these different observations under a new unified theory of glaciation, his deduction of a common cause. No one before had united all these phenomena into one consistent story, as proofs of glaciation. His first observations indeed go back to his childhood in the Danish countryside, where he now claimed that the mysterious big boulders, Kampestene, scattered all over northern Europe had been transported there by glaciers. B The boulders of Alpine granites found on the limestone Jura ridge had to be explained likewise, he wrote. From Norway he pointed to the boulder locality at Mjøsa passed through in 1823 (and in 1800 and 1801). Big boulders found on mountain peaks could similarly be explained ‘by extraordinary immense masses of ice’ (Esmark 1824: p. 41; Appendix S8). He noted that he had

Fig. 7. Similar phenomena may have similar causes. A. Big boulder resting on polished bedrock at western approach to Rauddalsbreen glacier foreland. Photograph 19 September 2008. B. Big boulder in forest, deposited at the end of last Ice Age 10 000 years ago on polished bedrock in the Mjøsa boulder terrain. Photograph 15 September 2009. Sites No. 6 and 7 in Fig. 8. [Colour figure can be viewed at www.boreas.dk]

IceAge?(NationalLibrary,Oslo,Utlansprotokol1823–26). He had observed and recognized the effects of the ice; he now sought an explanation of a former colder climate. Screening the scientific literature he found a proposal by Newton’s successor as professor at Cambridge William Whiston (1667–1752) that Earth had originally been a comet, with a comet’s elongated elliptic orbit (Whiston 1696). This would periodically bring the planet so far from the sun (in aphelion) that water froze. Esmark thus tried to keep Ice Ages within the lawful reign of celestial Newtonian mechanics. The return of comet P2/Encke in 1822 only the second after P1/Halley’s comet to have a Fig. 8. Map of southern Norway, with white line showing Jens confirmed periodic orbit around the sun may have Esmark’s voyage from June to September 1823. 1 = Christiania inspired him to pursue this theory, as well as reading (Oslo); 2 = Rød, the home of Otto Tank; 3 = Enighed Copper about the comet of 1811 for which in his paper he cites Works and the site of the Esmark Moraine close to sea level; William Herschel who had found it to have ‘a planet-like 4 = Steinsund, with polished Devonian conglomerate at sea level, ’ Sula Islands; 5 = Grimeli Copper Works; 6 = Northern tip of core (Herschel 1812). Changes in the eccentricity of Jostedals glacier, Rauddalsbreen with Otto Tank’s Moraine; Earth’s orbit form an essential part of the causal theories 7 = Mjøsa boulder terrain. Observations of moraines and boulders of Ice Ages later presented by James Croll and Milutin at the sites Østerdalen and Numedal are cited in Esmark’s 1824 Milankovich (Berger 2012). Esmark now presentedwhat paper, and were visited by him decades earlier. The letters refer to ‘ ’ high peaks ascended and measured by Esmark between 1800 and he considered incontrovertible proofs that during 1810: A = Tronfjell 1800; B = Rondane 1800; C = Snøhetta 1801; aphelion Earth formerly has been shrouded in ice and D = Gausta 1810. [Colour figure can be viewed at www.boreas.dk] BOREAS Jens Esmark and his discovery of Ice Ages 9 first encountered such peak boulders in Numedal an entists to become generally convinced even of the reality area he visited when a student at the Kongsberg mining of Ice Ages, although new impetus came in the 1830s academy in 1789–91 and repeatedly from 1798 to 1814 as with the work of Venetz, Charpentier, Bernhardi, Assessor. Similarly boulders on ridges and peaks in Schimper, Agassiz and others (Rudwick 2008; Kruger€ Christiansand’s Stift, the southernmost tip of Norway, 2013). visited in 1804, 1818, 1822 and 1823. Old observations Although this story supports the claim that discovery were also resurrected from Østerdalen valley, visited in comes to those prepared, it is worth noting the excep- 1800, 1801 and 1810, where ‘the slopes of the mountains tional coincidence of favourable circumstances that led towards the valley are covered with large unattached to Esmark’s serendipitous discovery: his visit to Enigh- rocks, mixedwith a great amount of sand and gravel, and eds copper works in 1822; government support for that this cover stretches to a significant altitude above the further prospecting in 1823; his pondering the ‘missing bottom of the valley.. [...] One finds too that these rocks’ in 1822; stumbling over the Esmark Moraine at mixtures exactly resemble those, which the glaciers push Forsand; the decisions to extend the trip north to in front of them’(Esmark 1824: p. 41 and 43; Grimeli, and then make a mountain traverse; choice of Appendix S9). The observed facts could not be explained the route across the Rauddalen glacier; the LIA creating by ‘anyother waythanbyassumingice fillingintheentire the fresh foreland and glacier effects; the uncanny valley’ (Esmark 1824: p. 42; Appendix S10). As his ‘most similarity between the Esmark and Otto Tank Moraines powerful proof, that there really have existed such as well as their outwash plains; Esmark’s support of immense ice masses in Norway in places, where now no colleagues studying glaciers; his own previous observa- perennial ice is more to be found’ (Esmark 1824: p. 43; tions. And what if the talented free spirit Otto Tank had Appendix S11) he gave a detailed description of the notagreedtoaccompanyhim...?Orcomet P2/Enckenot Esmark Moraine by Stavanger and the sandur in front, flared the sky in 1822...? and Tank’s exclamation: Acknowledgements. – The author gratefully acknowledges economical ‘As I think thus to have proven sufficiently, that the Norwegian support from University of Oslo, the comments of two reviewers, and mountains have been enveloped in ice all the way down to the sea, accommodation and transport from the owners of Pollfoss Fjellstue, and that as an effect of this, the sea itself must have been an ice sea Skjak, before his own mountain traverse in late September 2008, and (Iishav), we may now easily explain the cause of the frequently Dagrun Haukalid Djuv for permission to use the photograph Fig 1A. steep, yes mostly perpendicular form of the Norwegian moun- Eva Bjørseth and Jan Mangerud helped in creating Fig. 8. tains on the sides that face towards the valleys, as well as in the valleys that lie above the surface of water, as in those, on which floor the water penetrates (in the Norwegian fjords). Ice and glaciershaveundoubtedlybytheirextraordinarygreat expanding References force caused this change in their original form, in that they have continuously sled down from the higher mountains to the lower Andersen, B. 1954: Randmorener i Sørvest-Norge. Norsk Geografisk areas, and bythis forward movement carriedaway large masses of Tidsskrift 14, 273–342. rocks, which they have detachedfrom the mountains.I can hereby Andersen, B. 1992: Jens Esmark – a pioneer in glacial geology. Boreas easily explain to myself why one below these precipices finds no 21,97–102. trace of the detached rock masses. As these steep mountain walls Berger, A. 2012: A brief history of the astronomical theories of are often 3, 4 to 5000 feet high, and the breath of the valley paleoclimates.InBerger,A.(ed.):ClimateChange,107–129.Springer, towards which they face similarly is several thousand feet, one Wien. must be astonished by thinking valleys filled with ice for a length Bohr, C. 1819: Om Iisbræerne i Justedalen og om Lodalskaabe. of several miles; this ice must in the lower areas have projected Blandinger 2, 289–317. itself far out into the sea; by melting large masses of ice have Bohr, C. 1826: Account of a visit to the Glaciers of Justedal, and to the detached themselves and have floated away, just as we see it still Mantle of Lodal. The Edinburgh New Philosophical Journal 2, 225– happening in the Polar areas. I dare to assert, that this is how it is, 261. when I take into account the remnant unmistakable traces of the Briner, J. P.,Svendsen, J. I., Mangerud, J., Lohne, Ø. S. & Young,N. E. immense effects of the ice masses’(Esmark 1824: p. 45–46; 2014: 10Be chronology of the south-western Scandinavian Ice Sheet Appendix S12). history during the Lateglacial period. Journal of Quaternary Science 29, 370–380. Here then was the causal explanation of fjords, and von Buch, L. 1812: Ueber die Granzen€ des ewigen Schnee’s im Norden. – the solution to the mystery of the missing volumes of Annalen der Physik 41,1 52. von Buch, L. 1814: On the Limits of perpetual Snow in the North. rocks: glaciers as a major geomorphological erosional Annals of Philosophy 3, 210–220, 338–348. force. Esmark then described the polished conglomer- Engen, S. 1985: Forsandboka. Gards- og Ættesøge, Volume 2. 573 pp. ate at Stenesund as further evidence of glaciers at sea Forsand, Stavanger. ‘ Esmark, J. 1799–1838: Handwritten meteorological observation pro- level, rounding off: I could still adduce more proofs of tocols from Kongsberg 1799–1810 and Christiania/Oslo 1816–1838. my adopted assertion, but I presume, that any com- National Archives (Riksarkivet), Oslo, S-1570. Fa. Materiale etter petent [person] by the adduced examples will feel himself professorer. convinced of its correctness’ (Esmark 1824: p. 48; Esmark, J. 1800: Letter to N. Hofman Bang, Kongsberg 7 November 1800. Stamhuset Hofmansgaves arkiv qb031. Landsarkivet Fyn, Appendix S13). This, however, proved overly optimistic, Denmark. despite Esmark’s paper appearing in English translation Esmark, J. 1803: Udtog af en Skrivelse fra Oberbergamts-Assessor J. a couple of years after the original paper (Esmark 1826), Esmarck paa Kongsberg. Collegial-tidende for Danmarkog Norge 45, – – it took several decades after Esmark’s paper for geosci- 716 719, 722 728. 10 Geir Hestmark BOREAS

Esmark, J.1812: Bemærkninger,gjorte paa en Reise til Gousta-Fjeldet i Naumann, C. 1824: Beytrage€ zur Kenntnis Norwegen’s gesammelt auf Øvre-Tellemarken. Topographisk-Statistiske Samlinger 1, 175–196. Wanderungen wahrend€ der Sommermonate der Jahre 1821 und 1822, Esmark, J. 1823a: Om Norit-Formationen. Magazin for Naturvidensk- Volume 2. 407 pp. Bey Adolph Wienbrack, Leipzig. aberne.Forste€ Aargangs forste€ Bind, 205–215. Naumann, C. 1848: Uber€ die Felsenschliffe der Hohburger Porphyr- Esmark, J. 1823b: Noget om det nedlagte Enigheds Kobberværk i berge unweit Wurzen. Berichte uber€ die Verhandlungen der Koniglich€ Ryfylke. Budstikken, Fjerde Aargang, No. 49–52, 5 June. Sachsischen€ Gesellschaft der Wissenschaften zu Leipzig 1, 392–410. Esmark, J. 1824: Bidrag til vor Jordklodes Historie. Magazin for Neumann, J. 1824: Bemærkninger paa en Reise i Sogn og i Søndfjord Naturvidenskaberne. Anden Aargangs forste€ Bind, Forste€ Hefte, 28– 1823. Anden Afdeling. Justedals-Reisen. Budstikken. Femte Aar- 49. gang, 67–73, 529–584. Esmark, J. 1825: Vitae, 25 October 1825. Handwritten. Archives of the Nilsson, S. 1879: Dagboksanteckningar under en resa fran sodra€ Sverige RoyalSwedishAcademyofScience,CentrefortheHistoryofScience, till Nordlanden i Norge 1816. 256 pp. Fr. Berlings Boktrykkeri och Stockholm. Stilgjuteri, Lund. Esmark, J. 1826: Remarks tending to explain the Geological History of Pontoppidan, E. 1755: The Natural History of Norway: Containing, A the Earth. The Edinburgh New Philosophical Journal 2, 107–121. particular and accurate Account of the Temperature of the Air, the Esmark, J. 1832: Letter to Collegium Academicum, Christiania 17 different Soils, Waters, vegetable, Metals, Minerals, Stoned, Beasts, March 1832. National archives (Riksarkivet), Oslo, Archives of the Birds,andFishes;togetherwiththeDispositions,Customs,andManner University of Oslo, Collegium Academicum, Journalsaker Box 23. of Living of the Inhabitants: Interspersed with Physiological Notes 1832, No. 152. from eminent Writers, and Transactions of Academies, Volume 1. 206 Esmark, J. 1834: Letter to Professor Jacob Keyser, Christiania 14 June pp. Printed for A. Linde, Bookseller to Her Royal Highness the 1834. National archives (Riksarkivet), Oslo, Ministry of Finance Princess Dowager of Wales, London. archives (Finansdepartementet) Journalsaker Box 0088, 1834, No. Rudwick, M. J. S. 2008: Worlds Before Adam. The Reconstruction of 532. Enclosed. Geohistory in the Age of Revolution. 648 pp. The University of Herschel, W. 1812: Observations of a Comet, with Remarks on the Chicago Press, Chicago. Construction of its different Parts. Philosophical Transactions of the Smith, C. 1817: Nogle Iagttagelser, især over Isfjeldene (Gletscher), paa Royal Society of London 102, 115–143. en Fjeldreise i Norge 1812. Topographisk-Statistiske Samlinger 2,1–62. Hertzberg, N. 1818: Noget om Sneebræen Folgefondi Søndhordlehnog Standal, R. 1995: Vegar før vegar var I. Farleier over fjell og bre i HardangersFogderie.Budstikken.FørsteAargang.No.90and91,13 Nordfjord. 156 pp. Eige Forlag, Volda. April, 713–722. Strøm, K. M. 1950: Early discoverers IV.Esmarkon glaciation. Journal Hestmark, G. 2009a: Sula-konglomeratet – fra fossilparadis til istids- of 1, 388–391. bevis. GEO 12,32–35. Vormeland, J. 1999: Otto Tank. Pioneren fra Rød Herregard. 122 pp. Hestmark, G. 2009b: “Her ligger Sneen evig”. Da Dovre falt – for Genesis forlag, Oslo. Esmarks barometer. Historisk Tidsskrift 88, 231–249. Wahlenberg, G. 1808: Berattelse€ om matningar€ och observationer for€ att Hestmark, G. & Nordli, Ø. 2016: Jens Esmark’s Christiania (Oslo) bestamma€ lappska fjallens€ hojd€ och temperatur vid 67 graders polhojd;€ meteorological observations 1816–1838: the first long term contin- for€ attade€ ar 1807. 58 pp. Carl Delen, Stockholm. uous temperature record from the Norwegian capital homogenized Whiston, W. 1696. A New Theory of the Earth, from its original, to the and analysed. Climate of the Past 12, 2087–2106. consummation of all Things. Wherein the Creation of the World in Six Imbrie, J. & Imbrie, K. P. 1986: Ice Ages. Solving the Mystery. 224 pp. Days, the Universal Deluge, and the general Conflagration, as laid Harvard University Press, Cambridge. downin theholyscriptures,areshewntobe perfectlyagreeabletoreason Keilhau, B. 1820: Nogle Efterretningerom et hidtil ubekjendt Stykke af and philosophy. With a large introductory discourse concerning det søndenfjeldske Norge. Budstikken, Andre Aargang, No. 49 and Genuine Nature, Stile, and Extent of the Mosaick History of the 50, 21 September, 385–398. Creation. 388 pp. Printed by R. Roberts, for Benj. Tooke at the Keilhau, B. 1822: De høieste Fjelde i Norge. Hermoder 30, 16 Middle-Temple-Gate in Fleet-Street, London. November, 235–240. Winkler, S., Matthews, J., Shakesby, R. & Dresser, P. 2003: Glacier Keyser, J. 1834: Untitled, but occasionally cited as: Kort Beretning om variations in Breheimen, southern Norway: dating Little Ice Age en af ham i 1834 foretagen Reise til Lomb i Gudbrandsdalen for at moraine sequences at seven low-altitude glaciers. Journal of Quater- undersøge omElvenedernufortidenmaatteføre Platinaog Guldmed nary Science 18, 395–413. sig. Morgenbladet 361, 27 December. Worsley, P. 2006: Jens Esmark, Vassryggen and early glacial theory in Kielland,G.K. 1823:Lettertohis brother CasparKielland,Christiania Britain. Mercian Geologist 16, 161–172. 20 June 1823. Lyngdal Bibliotek. Kulturhuset. Kiellandsamlingen. I. 2. Mappe Nr. 14. Kielland, A. 1897: Familien Kielland med dens kognatiske ascendents.I Kommission hos Jacob Dybwad, Christiania. Supporting Information Kruger,€ T. 2013: Discovering the Ice Ages. International Reception and Consequences fora Historical Understandingof Climate. 556 pp. Brill, Additional Supporting Information may be found in the Leiden. online version of this article at http://www.boreas.dk. Lampe, A. 2001: Mission or Submission? Moravian and Catholic Missionaries in the Dutch Caribbean during the 19th Century. 244 pp. – Vandenhoeck & Ruprecht, Gottingen.€ AppendicesS1 S13.Textsinoriginal Danish-Norwegian Malthus, T.1966: The Travel Diaries of ThomasRobert Malthus. 348 pp. language. Cambridge University Press, Cambridge. Naumann, C. 1822: Einige Bemerkungen auf Ausflugen€ in die norwegischen Schneegefilde. Isis [von Oken] H. 6, 641–664.