(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Skrifter utgit av Videnskapsselskapet i Kristiania"

SKRIFTER 



UTGIT AV 



VIDENSKAPSSELSKAPET 
I KRISTIANIA 

1921 



I. MATEMATISK-NATURVIDENSKABELIG KLASSE 
2. BIND 



^3S^ 



KRISTIANIA 

I KOMMISSION HOS JACOB DYBWAD 

A. W. BRØGGERS BOKTRYKKKRI A/S 
1922 



SKRIFTER 



UTGIT AV 



VIDENSKAPSSELSKAPET 
1 KRISTIANIA 

1921 



I MATEMATISK-NATURVIDENSKABELIG KLASSE 



2. BIND 



— <3!-^ 



KRISTIANIA 
I KOMMISSION HOS JACOB DYBWAD 

A. W. BRØGGERS BOKTRYKKERI AS 
I022 



/s 



Indhold. 



Side 
No. g. Rolf Nordhagen. Kalktufstudier i Gudbrandsdalen. (Med 44 tekstfigurer 

og 5 planchen i — 155 

„ 10. J. V. Zeh'zko. Äquivalente der untersilurischen Euloma-Niobefauna bei 

Plzenec in Böhmen. ( Mit 5 Tafeln, i Kartenskizze und i Textabbildung) 1 — 27 
„ II. Fridtjof Nansen. The Strandflat and Isostasy. (With 170 illustrations 

and maps in the text) i — 313 

^12. P. A. Øyen. Hippophaës rhamno'des L. tVa en norsk kalktuf .... 1—8 

„ 13. Hjalmar Broch. Studier over cirripedienes fylogenetiske slegtskapsforhold i— 10 

„ 14. Ivar Jørstad. Bidrag til kjendskapet til Trøndelagens rustflora ... i— 19 

„ 15. Bernt Lynge. Lichens from the Gjøa Expedition 1—7 

„ 16. Carl Siegel. Ueber den Thueschen Satz i — 12 

„ 17 Th. Skoletn. Untersuchungen über die möglichen Verteilungen ganzzahliger 

Lösungen gewisser Gleichungen i — 57 

„ 18. Th. Skolem. Bericht über die nachgelassenen Schriften L. Sylows [—7 
„ 19. L. Vegard. Die Struktur des Nordlichts und die Art der kosmischen 

Strahlen. (Mit 2 Fig. im Text) i — 16 

„ 20. Halfdan Bryn. Troms fylkes antropologi. (Med 20 figurer og karter i 

teksten). (Med tysk resumé) i — 176 



KALKTl'F STUD IHR 

I GUDBRANDSDALEN 



ROLF NORDHAGEN 



(med 44 TEKSTFIGURER OG 5 PLAN'CHER) 



(VlDENSKAPSSELSKAPETS SkRIFTER. I. MaT.-NATCRV. KlaSSE. I92I. No. 9) 






UTGIT FOR FRIDTJOF NANSENS FOND 



KRISTIANIA 

I KOMMISSION HOS JACOB D^IiWAI) 
1921 



Fi-i-mlatrt i den inat.-natur\'. 1<1 isses mot? d ti 3dje juni 1921 v.^d prof. H. H. Gran. 



A. W. BRØGGERS HOKTRYKKERI A S 



INDHOLD. 

Side 

Forord V 

Speciel del. 

1. Kalktitffcn ved Leine i Kvant i 

A. Topografi og vegetation i nutiden i 

B. Stratigrafiske undersokelser 14 

C. Oversigt over Leinetulfens stratigrafi 38 

II. Kalktuf ved Gillebti og Tingvold i Øier 54 

A. Topografi og vegetation i nutiden 54 

B. Stratigrafiske undersokelser 58 

C. Oversigt over Gillebutufi"ens stratigrafi 64 

D. Gruskeglen ved Ting\'old 69 

III. Kalktuffen ved Nedre Dal i Faaberg 73 

IV. Kalktuffer ved Onset i Biri 75 

A. Blytts undersokelser .' 75 

B. Egne undersokelser 82 

Generel del. 

I. Isavsiiieltiiingen i Gudhraiid^daUii 87 

II. iXogen beiiterkninger om den første flora og vegetation under avsnieltningstidrn 96 

III. Gudbrandsdalcns kalktuffer og Blytts teori 105 

IV. Benierkninger om Giidhrandsdnleiis plaiitcvrkst i postarktisk tid 115 

A. Den subarktiske tid 115 

B. Den boréale tid 120 

En sammenligning mellem Gudbrandsdalens og Jemtlands kalktuffer 1 23 

Hippophaës-problemet 125 

C. Den atlantiske tid 138 

D. Den sutboreale tid 139 

E. Den subatlantiske tid 140 



Fortegnelse over de fundne plantearter og snegler 142 

Litteraturfortegnelse 148 

Tekst til planchene 154 

Rettelser og tilføielser 155 



FORORD. 

Høsten 1913 omtalte professor R. Serxander Gudbrandsdalens kalktuflfer 
under sine forelæsninger ved Kristiania universitet og opfordret undertegnede 
til at undersøke dem paany. Imidlertid hadde jeg av konservator P. A. Øyen 
faat oplysninger om et par kalktufter i det trondhjemske, opdaget av hr. Mart;.\ 
MoE, Stjørdalshalsen. Og da jeg somrene 1914 og 1915 opholdt mig i denne 
del av landet, foretok jeg en undersøkelse av forekomstene, som dog gav et lite 
opmuntrende resultat. Den ene forekomst mellem Olderen og Velvang paa 
Skatvalhalvoen bestod av jordagtig tuf avsat hoit oppe i en stupbrat ur, den 
anden lokalitet ved Auran (omtalt av Øyen 1915 1. c. p. 44) opviste kun ube- 
tydelige tufmasser sterkt opblandet med bergartsmateriale. Helt uten interesse 
var undersøkelsen dog ikke. Den ene av forekomstene viste nemlig i den 
øvre tilgjængelige del et avbrud, karakterisert ved ansamling av smaa skifer- 
biter (nedraset forvitringsmateriale). Jeg har tidligere bare omtalt resultatet i 
en stipendie-indberetning til det Akademiske kollegium, men haaber at komme 
tilbake til disse kalktufter ved en anden anledning. 

Sommeren 1917 lyktes det konservator Øyen og overlærer Holme paa 
Lillehammer at paavise en ny og høist interessant kalktuf i Gudbrandsdalen, 
nemlig ved Tingvold og Gillebu i Øier herred, hvor Holme allerede for mange 
aar siden opdaget kalktufstykker i en grusavsætning. 

Konservator Ove Dahl og undertegnede assisterte senere ved bestem- 
melsen av det store materiale som Øyen medbragte fra Øier, og da det 
lyktes mig at identificere nogen eiendommelige bladavtryk med Hippophaes 
rhamnoides, var min interesse for alvor vakt. Da Øyen mente at der kunde 
være en efterhost at gjore, specielt i palæofloristisk henseende, bestemte jeg 
mig næste sommer for at reise til Gudbrandsdalen. 

Turen gik forst til Otta, hvor jeg haabet at finde en ny tuftorekomst, 
idet fiskeriinspektør A. Landmark for en del aar siden hadde iagttat hvite tuf- 
skorper og løse biter i de bratte her ved Kringen, og overfor undertegnede 
antj'det at der maatte være kalkholdige kilder paa stedet. Imidlertid viste 
det sig at det her bare gjaldt helt ubetydelige forekomster av jordagtig tuf 
og tj-nde skorper paa fjeldsiden; saadanne fandt jeg paa 3—4 forskjellige steder 
mellem Solgjem og Otta paa dalens ostside. Efter to dages ophold paa Otta 
drog jeg til Pillarviken, et stykke oppe i Otta-dalføret (Lalm), hvorfra der i 
sin tid var indleveret prøve av kalktuf til Universitetets samling. Findestedet 
laa desværre midt ute i en rugaker, men ved eierens, hr. N. Sveines. elsk- 
værdighet fik jeg anledning til at foreta en gravning, som dog bare bragte 
større og mindre mosetufbiter for dagen, av samme type som de i samlingen 
opbevarte stufFer. Nogen indgaaende stratigrafisk studie var der saaledes ikke 
adgang til paa det nævnte tidspunkt. Jeg fortsatte derfor til Sørem i Vaage, 
hvor hr. gaardbruker Sørem i sin tid hadde fundet lose blokker av en haard 
kalktuf i nærheten av en nedlagt husmandsplads (GulbrandheimenI paa Ottas 
anden bred, vis à vis Sørem. Jeg fandt ogsaa ganske rigtig stedet og et par 
tufblokker, som laa i en gammel stenmur. Men trods ihærdige undersøkeiser 



VI 

i (l;ilsi(|. Il iiinkiinj,' phulscii var det ikk<- iniilif^ iit opcJagr iiogm tuf. Knt'rn 
maa de nævnte blokker, som i sin tid ble-v fundet nede i stranden, være blit 
transportert ined elven imder isløsnin^e-n fra et eller andet sted høiere oppe 
i daliii, rlli-i (le inaa skriv s\<^ fra IjcM' t holt over Gulbrandheimen. Tuffen 
\ai' tildils bi'cccieaj^tig med smaa biter av forskjellige bergarter, men uten 
tvdcligc tossijcr. Antageligvis skriver den sig fra en lokal forekomst av ube- 
tvdclii; iitstræ-kning. 

ICflci- disse undersøkeiser, som IkI!« |- ikke var sæ-rlig opmuntrende, drog 
jeg ned i Gudbrandsdaleii ij^j'ii, liv<n- dir ialfald var tuf nok at undersøke. 
Først stoppet jeg i Kvam for al la LciiictiiJJni i oirsyn. Av forskjellige ut- 
talelser fra folk som i aarcncs li)|) hadde besokt Leine, hadde jeg faat det 
indtryk, at der nu lor tiden bare var ubetydelige rester igjen av denne tuf, 
væsentlig løse bili r i cic øvre jordlag og avfald fra Blytts store indsamlinger 
i 1891. Desto større var derfor min overraskelse, da jeg efter et par resultat- 
lose iM-øvegravninger fik op e-n groft som aabenbarte et overordentlig vakkert 
og sammenhængende profil. Resultatet var dog en smule forvirrende, idet jeg 
istedenfor én Z)rv(7.s-førende horisont, saaledes som Blytt omtaler, fandt to saa- 
danne med et mellemliggende bladtuflag. Mægtigheten av de enkelte horisonter 
var ogsaa helt anderledes end Blytts profiler viste. Av en gammel mand i 
Kvam, som i sin tid hadde hjulpet Blytt med gravearbeidet, fik jeg nøiagtig 
vite hvor disse profiler var optat, og det lyktes mig dernæst at konstatere at 
den undre del av tuffen paa dette sted var saaledes som av Blytt beskrevet. 
Imidlertid var Dryas-X'A^ftX. og den ovre del av tufïen fjernet, saa paa dette 
punkt var jeg like klok. Da jeg denne sommer fortrinsvis hadde tænkt at 
ai'beide i Øier, kunde jeg ikke ofre mere tid paa Leine, men noiet mig med 
at medføre et stort materiale fra de forskjellige lag i profilet. 

Ved Gillehit i Øier foretok jeg indgaaende sti-atigrafiske studier paa en 
række punkter og med positivt utbytte. Ogsaa Hippop/mfs's vertikale utbre- 
delse i tuffen blev nøiagtig undersøkt. Herfra medbragte jeg ogsaa en stor 
samling med haandstykker, som blev underkastet en foreløbig granskning efter 
hjemkomsten til Kristiania. 

Vaaren 19 19 bestemte jeg mig for alvor til at grave op hele Leine-tuffen 
paany, idet jeg haabet at der maatte kunne findes et mellemled som forbandt 
Blytts profiler med mit eget. Jeg bestemte mig for at grave en lang „skytter- 
grav" gjennem tuffen og opta noiagtige profiler hele veien ibr at komme paa 
det rene med om forekomsten virkelig var saadan som Blytt hadde beskrevet 
den. Jeg fandt dette absolut nødvendig, da der fra forskjellige hold overfor 
mig blev hævdet at man ikke maatte stole for meget paa en saa subjektivt 
farvet fremstilling som Blytts. Til disse undersøkeiser fik jeg likesom fore- 
gaaende aar et bidrag av Rathkes legat, hvorfor jeg her fremfører min 
bedste tak. 

I slutten av juni 1919 drog jeg atter til Gudbrandsdalen. Forst gjorde 
jeg en avstikker til Ransverk i Vaage, hvor der ifølge hr. gaardbruker Sørem 
skulde findes kalktuf nær Mysuholet sætter mellem Ransverk og Lemonsjoen. 
Sammen med hr. Jens Tronhus undersokte jeg terrænget omkring den ned- 
lagte sæter, men vi iandt kun litt drypstensagtig vakkert krystallinsk kalk, 
som var meget løs, ved foten av en bergvæg i skogen. — Derefter drog jeg 
til Leine i Kvam, hvor jeg i lopet av de følgende uker, assistert av hr. Paul 
RøEN, fik opkastet en ca. 20 m. lang og 1,5—2 m. dyp sammenhængende groft 
paa langs av bakkeskraaningen, desuten en hel del tverprofiler. Alt i alt 
blev ca. 25 saadanne nøiagtig analysert og opmaalt og et meget rikt fossil- 
materiale indsamlet. 



VII 

Arbeidet gav meget tilfredsstillende resultater og bod paa flere over- 
raskelser. Saaledes fandt jeg meget snart rester av en helt ny tufhorisont 
ovenpaa furutuften, som bragte en række nye momenter ind i diskussionen. 
Det viste sig ogsaa at de forskj eilige tuf lag hadde sit maksimum paa helt for- 
skjellige punkter indenfor avsætningens omraade. Litt efter litt lyktes det mig 
at forfølge de temmelig komplicerte stratigrafiske foriiold inden Leinetullen og 
at sammenstille disse til et sammenhængende profil. 

Efter indbydelse av professor R. Sernander tilbragte jeg hosttermincn 
1919 paa det plantebiologiske institut i Uppsala, og medbragte hele mit store 
kalktufmateriale til nærmere granskning. Institutets vakre samlinger av svensk 
kalktuf blev stillet til min disposition, likesom professor Sernander paa alle 
mulige maater bistod mig med litteratur og oplysninger om svenske tuftbre- 
komster, skaftet mig sammenligningsmateriale fra herbariene o. s. v., kort sagt 
lettet mit arbeide paa al tænkelig vis. I løpet av hosten hadde jeg den til- 
fredsstillelse at kunne avslore næsten alle de mange apokryfiske ting og kurio- 
siteter som Leinetuften indeholdt. Stor nytte hadde jeg av den vistnok forst 
av Nathorst anvendte kollodiummetode, som senere Halle har prakti- 
sert med udmerkede resultater. Paa de avtryk som skal undersøkes, dryppes 
en liten draape kollodium, som naar den er indtorket danner en fin hinde, der 
avløses forsigtig. De første avtryk er altid daarlige paa grund av forurens- 
ninger og luftblærer, men efterhaanden blir de brukbare og kan under mikro- 
skopet avsløre de fineste epidermis- og andre struktureiendommeligheter (spalte- 
aapninger, haardannelser etc.) av stor betydning for bestemmelsenes paalidelighet. 

1 januar og februar 1920 opholdt jeg mig i Stockholm og gjennemgik med 
professor Th. Halle Riksmuseets pragtfulde kalktufmontrer i den palæobotaniske 
samling, likesom professor Halle elskværdigst gav mig tilladelse til at gjen- 
nemgaa hele museets store samlinger bl. a. fra Benestad i Skaane og fra de 
jemtlandske lokaliteter. Samtidig drev jeg litteraturstudier ved Vetenskapj- 
akademiens bibliotek og fik hos statsgeolog dr. L. von Post information i mi- 
kroskopisk bestemmelse av vore skogtrærs pollen. Derefter reiste jeg atter til 
Uppsala, hvor jeg forblev til slutten av mars maaned og bearbeidet resten av 
mine samlinger. 

jeg vil her faa lov til at rette en varm tak til min ærede lærer og ven 
professor Rutger Sernander i Uppsala, som gjorde opholdet der uforglem- 
melig for mig, og for den store interesse han viste overfor mit arbeide. Likesaa 
vil jeg faa lov til at takke Växtbiologiska Institutionens amanuensis fil. lie. 
G. E. Du Rietz for hans hjælpsomhet og elskværdighet! 

Overfor professor dr. Svante Arrhenius og dr. Olof Arrhenius ved 
Nobelinstitutet for fysikalsk kemi, som paa alle mulige maater hjalp mig under 
mit besøk i Stockholm, vil jeg fremføre en dypt følt tak, likesaa overfor pro- 
fessor dr. Th. Halle og dr. L. von Post. 



Efter gjennemgaaelsen av samlingene og efterat jeg foreløbig hadde op- 
gjort mig en mening om de undersøkte kalktuffer, fandt jeg det ønskelig 
endda en gang at ta findestedene i øiesyn, for at stille min egen opfatning 
paa prøve — en fremgangsmaate som efter min mening gir de bedste garan- 
tier for resultatenes paalidelighet. 

I oktober 1920 tilbragte jeg nogen herlige hostdager i Kvam og fik bl. a. 
gravet op et profil som hvad interessant opbygning angaar, slog alle de tid- 
ligere studerte av marken. Jeg avla ogsaa et kort besøk ved Gillebu i Øier. 
I slutten av maaneden reiste jeg op til Faaberg og besaa kalktuflfen ved Nedre 



VIII 

I)al, og cftcipMa til jjiii, for al let«; (;f't<-r df.*n av Hi.vir i sin tid iiiid<rrsøkte 
tuf ved Onset. |)(svæir(- lyktcs det mig ikke med sikkerhct at gjciifinde 
lokaliteten; men en anden f'oi-ekomst, som forresten bar spor efter gravning, 
viste sig at indebære mange træk av interesse. 

Av ovenstaaende red<gjørelse vil det fremgaa at undertegnede har tat 
sin opgave meget alvorlig, og ikke skydd noget middel som kunde bringe 
klarhet over kalktulproblemene. Jeg gik til unde-rsøkelsene indpodet med 
adskillig skepsis, ikki- mindst overfor Bi.vtts studier og teorier. Og den 
mening som under arbeidets gang litt efter litt har presset .sig frem hos mig, 
er ikke resultatet av nogen paavirkning i denne eller hin retning. 

Ci,i:mi;nts hai- i et av sine verker kaldt Bi.vtts teori for et „storm- 
centrum" i nordisk kvartærforskning, og ikke uten grund. Jeg har ogsaa selv 
hat en levende folelse av, at det at delta i den kvartærgeologiske diskus.sion er 
ensbetydende med at utæske kritiken, hvis man da ikke vælger den frem- 
gangsmaate at skrive et arbeide saa upersonlig at læserne ikke paa et eneste 
punkt oiner forfatterens ansigt. Men da bør man helst ikke sætte sit eget 
navn under. 

Kanhænde vil denne bok ogsaa l)ringe vind i seilene. I)a kalktuflene 
for Hlytts bevissthet stod som vigtige støttepunkter for hans teori, vil 
man forstaa at denne avhandling bevæger sig indenfor et alt andet end nøi- 
tralt gebet. 

Kn ting er imidlertid sikker, og det er, at herefter skal man ikke behøve 
at diskutere spørsmaalet om hvorledes Gudbrandsdalens ! alktufter er op- 
bygget. Den deskriptive side av saken, som jo er den primære, er ved disse 
undersøkeiser bragt lykkelig i havn. Man kan ikke skyte sig ind under saa- 
danne paaskud som at stratigrafien er mangelfuldt undersøkt, at forholdene er 
uklare o. s. v. Allerede dette er et fremskridt. .Saa faar man stille de for- 
skjellige tolkningsmuligheter op mot hinanden og vælge den, som paa den 
mest naturlige maate besvarer alle de spørsmaal som reiser sig i det fore- 
liggende tilfælde, og som samtidig ikke kommer i konflikt med andre forsk- 
ningsomraader. Dette har jeg da efter bedste evne forsokt at gjøre. 

Til slut vil jeg faa lov til at takke konservator P. A. Øyen for den inter- 
esse han har vist overfor mine undersøkeiser, og for alt det interessante han 
paa de mange ekskursioner i aarenes løp har vist mig og mine studiekame- 
rater. Jeg har i nærværende avhandling forsøkt at ta et personlig standpunkt 
til kalktufproblemene, og min opfatning avviker paa enkelte punkter ganske 
meget fra Øyens. Det kan imidlertid aldrig være til skade for en sak at den 
blir kritisk belyst fra flere sider. 

Professor J. Schetelig har været saa elskværdig at hjælpe mig med berg- 
artsbestemmelser og litteraturoplysninger. Konservator dr. Hjalmar Möller, 
Riksmuseet, Stockholm, har bestemt en del kalksamlende moser fra Biri, og 
docent F. Økland, Aas, har bestemt mit materiale av landsnegler. Avdøde 
fil. lie. Helmer Olivecrona, assistent ved Norges Landbrukshoiskole, Aas, har 
undersøkt en del Cyanoph3'cé-prøver fra Biri. Ogsaa dr. Nils Odh.xer, Riks- 
museet, Stockholm, har hjulpet mig med det sj'stematiske bestemmelsesarbeide. 
Til alle disse videnskapsmænd vil jeg faa lov til at rette en ærbodig og hjertelig 
tak for udmerket assistance, likeledes til Videnskapsselskapet i Kristiania, som 
har bekostet trykningen av dette arbeide. 

Kristiania i mai 1921. 

Ro/f Nordhagiit. 



SPECIEL DEL. 
I. Kalktuffen ved Leine i Kvam. 

A. Topografi og vegetation i nutiden. 

I Kvam (anneks til N. Fron) i Gudbrandsdalcn gjor Laagen en skarp 
bøining, idet den forst loper ret mot ost, derefter mot syd-sydvest. Dalen 
har her karakter av en gryte eller et avlangt traug med saagodtsom al 
bebyggelse samlet paa nordsiden, og h\or elven boier om, paa ostsiden, 
hvilket dels beror paa ekspositionen mot solen, dels paa det bedre jords- 
mon paa nordsiden. 

Vra norc/ kommer elven Veikla eller \'indeaaen tossende ned mot 
hoveddalforet. Denne elv og sidebækken Borju avgrænser tilsammen en 
mægtig landtunge, som fra Tunsbergtjeldet (944 m. o. h.) falder jevnt, men 
meget brat av mot dalbunden i syd. Oppe i skraaningen ligger her de 
gamle Leine-gaarder med sine brunsorte, solsvidde tommerhus og de store, 
men tungbrukte jorder. Jordsmonnet er meget frugtbart, hvilket skyldes de 
kolossale m o r æ n e a \- 1 e i r i n g e r som i form av et kalkholdig 1er med 
storre og mindre blokker, ofte i tætpakkede lag, dækker hele skraaningen 
fra \'eikla og helt op til mellem 6 og 700 m. o. h., ovenfor Leinegaardene. 
Til trods for den voldsomt sterke insolation i sommermaanedene lider bøn- 
derne her ikke saa meget paa grund av torke som man skulde vente, netop 
takket være lermassene, som holder paa vandet. 

Disse lerbakker under Leine har været ganske skjæbnesvangre for 
bygdens befolkning. I tidenes lop har der gang paa gang gaat kjæmpe- 
mæssige skred langs Veikla, og store masser av den oprindelige moræne- 
fylding har skyllet ut over dalbunden i syd, som av denne grund er ganske 
komplicert opbygget. Den største katastrofe skedde i aaret 1789, det store 
ulykkesaar i Gudbrandsdalens nyere historie, da „ofsen" eller den store 
vandflom herjet overalt i dalene Den dag idag fortælles det om hvor- 
ledes folk reddet sig opover Leinebakkene i hui og hast for at undgaa 
skredene. De vedfoiede fotogratier gir et litet indtryk a\- hvorledes topo- 
grafien arter sig paa stedet i nutiden. Morænemassene dækker dalsiden 
temmelig langt opover \'eiklas dalfore og har over store strækninger form 



' CtV. Heli.and: Norges land og folk. Kristians amt. Hind I. 
Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 9. 



UOIJ- NOKDIIACEN. 



M.-N. Kl. 




Fis-. I. 



Kart over de undersoktc kalktuüVr i Gudbrandsd.ilen. i: 2 000000. 



av længderygger orientert lodret paa elven. Disse rygger er adskilt ved 
rendeformige smaadaler, som markerer de steder hvor skredene har gaat 
som værst. Ogsaa paa Veiklas ostlige bred gjenfindes lignende lerbakker 
og længderygger, men i mindre maalestok. Alting tyder paa at \'eiklas 
dalføre i tidligere tider har været sterkt opfyldt av morænemassene. og at 



ig2i. No. 9. 



KALKTLFSTCDIER I GL'DBRAXDSDAl.EN. 



elven etterhaanden har gravet sig ned og tbrskjovet sig i vestlig retning. 
Saaledes fortæller traditionen at man i „gamle dager" paa gaarden X'eikle 
fik vand fra kalktufkilden hoit oppe i Leinebakkene, ved hjælp av træ- 
render. Husene paa Veikle laa dengang lavere end i nutiden, men selv 
under denne forutsætning vilde det i vore dager være umulig at lede van- 
det paa den nævnte maate. Det hele forutsætter at lerbakkene hadde et 
meo-et slakkere forlop og skraanet jevnt ned mot Veikla, som antageligvis 
oo-saa o-ik mere ost og nærmere gaarden end nu for tiden. Et 
andet fænomen som peker i 
samme retning, er navnet 
„Fagervold" paa en liten plads 
ved foten av Leinebakkene 
(sees paa fig. 3). Ingen vilde 
nu for tiden falde paa at kalde 
denne eiendom med et saadant 
smukt navn; jordveien er alt 
andet end fager. — 

Ogsaa sidebækken Borju 
har foranstaltet lignende kata- 
strofer. Ifolge gaardbruker Ole 
O. Knipen, Roen, skal der mel- 
lem Leine og gaarden Krok ha 
ligget en eiendom som er helt 
forsvundet ved jordskred. Og- 
saa i vore dager gaar der 
h\er vaar i sneløsningen smaa 
skred i hakkene, hvilket for- 
ovrig ikke forhindrer folk i at 
dyrke dem op. 

I lerbakkene kommer grundvandet frem paa en række steder som is- 
kolde kilder, specielt nord for gaarden Roen og op for pladsen Knipen. 
At moræneleret er sterkt kalkholdig kan man straks se paa den rike 
vegetation i dalsiden; specielt er Gyniiiadriiia coiiopca, som i juli er sterkt 
fysiognomisk fremtrædende paa fugtig bund, en god kalkbundsindikator. I 
denne del av Gudbrandsdalen har vi adskillige kalkforende bergarter', og 
det er da disse som i forste række har leveret materialet til moræneleret 
i Leinebakkene. Man finder ogsaa kalksten blandt blokkene i leret. 

Det er for saavidt ikke merkelig, at vi her ogsaa finder flere kilder som 
i tidens lop har avsat store mængder med kalktuf. Paa det vigtigste 
tufifindested dannes der i nutiden ringe mængder med mosetuf. Dette er 
ogsaa tilfældet paa en lokalitet nede i Borjus canon; her dækker tykke 
kaker av mosetuf [Mollia (rnigiiiosa) bergvæggen over en kortere strækning. 
Ifolge BjoRLYKKE ( 1905 1. C. p. 213I skal der i denne dalside anstaa „kalk- 




Fi«;. 2. Lerbakkene ostenlbr Leine, mellem Roen og 

Knipen. Utsigt mot nord fra jordet paa Veikle. Elven 

Veikla loper i dalens bund (fra hoirel. 

Juli 1 9 19. Xordhagen tot. 



Cfr. BjoRLVKKES store arbeide ( 1905I. Aarstallene henviser til litteraturlisten bakerst i boken. 



KOI.F NORfJlIAflFN. 



M.-X. Kl. 



Iioldiî^ spai'.ii^iiiil", nllsaa en sla.ys kalksanrlstcn, og jeg skulde tro at det 
n(to|) er (Iciiiic sdiii (iaiiiK I niid« rlagct foi- den næ\'nte nK)setiir. \'ed 

pladsen Knipen, lnoi- dcy liiules store kalktiilblokker i jorden langsen bæk 
ovenfor kjoreveien, liai- jeg ikke set nogen recent tufdannelse. 

Om (Un !)( kj(n<lle l.cinetLifs opdagelse har baade Bi. '»"ri og Ovk.\ 
(1920 1. c.) i sine respektive avhandJinger skrevet saa pas indgaaende at 

jeg her ikke skal gaa nærmere 
ind paa det kapitel. Forinden 
jeg gaar over til at skildre tuf- 
fens .stratigrafi, .skal jeg imid- 
lertid omtale vegetationen 
paa og omkring findestedet, da 
dette er av stor betydning for 
de følgende avsnit og de plan- 
tegeografiske ræsonnementer 
som der er fremfort. 




Man skal i \ort land med 
dets magre jordsmon lete længe 
efter en dalside som hvad vege- 
tationens 3'ppighet angaar, kan 
maale sig med disse bakker i 
Kvam. Det næringsrike og 
jevnt fugtige substrat sammen 
med den gunstige eksposition 
(den Sterke heldning mot syd 
og .sydost) har fremkaldt en 
makelos frodighet, specielt i 
de høiereliggende lier op for 
Knipen. Man finder her ude- 
lukkende eutrafente og meso- 
file, langs bækkene mere hydro- 
file plantesamfund, bestaaende 
av løvtrær, græs og urter. L3'ngsamfund og andre oligotrafente t\-per med 
økonomisk husholdning ser man ikke antydning til. Imidleitid har vege- 
tationen paa de for omtalte længderygger og bratte lerbakker ned mot 
Veikla til en viss grad et xeromorft anstrok, paa grund av den sterke av- 
rending og den voldsomme insolation i sommermaanedene. 

Det vigtigste skogdannende træ indenfor det angjældende omraade i 
nutiden er graaôren {Aluns iiicaiia)^, dernæst bjerk (baade Bctnla odorata 
og B. verrucosa). Nordenfor Knipen er det mest lav oreskog i dalsiden 
og litet bjerk. Her er terrænget mere jevnt og ikke saa opdelt i rygger 
som længer syd, vistnok ogsaa mere fugtig. Nedenfor Leine og Roen der- 



Fig. 3. Styrtningene nedenfor gaardcn Roen Isom 
ligger nederst i Leinejordene). Orekrat, klynger av 
bjerk og xeromorfe græssamfund. Nederst sees plad- 
sen „Fagervold". Nede i dypet loper Wikla. Utsigt 
mot nordvest fra Veikle, juli 1919. Kalktufien ligger 
ovenfor billedets hoirc hjornc. Nordhagen fot. 



' Nomenklaturen er den samme som i Blvtt-D.ahl: Haandbog i Norges Flora. Kristiania 
1906. 



I92I.X0. 9- KALKTUFSTUDIER I (;L"DBRANDSDALEX. 5 

imot spiller bjerken en ganske stor rolle. Oreskog og krat er her meget 
almindelig i alle forsænkningene mellem lerryggene, mens bjerken i form 
av en meget aapen skog eller bare isolerte trægrupper ynder de mere terre 
rygger, dog kun hvor disse er stabile. De bratteste lerbakker er fuldstæn- 
dig blottet for al vegetation, og de mindre bratte bærer en hoist eiendom- 
melig og svakt xeromorf græsvegetation, som ikke er helt sluttet og 
Uten bundskikt av moser, hvilket altsammen beror paa substratets sterke 
skraaning og derav folgende instabilitet. Særlig om vaaren synes der altid 
at ske glidninger langs ryggenes overflate. Dette er vel ogsaa aarsaken 
til at bjerk og or har saa vanskelig for at vokse op og danne veritabel 
skog paa stedet. (Cfr. fig. 2 og fig. 3). Et moment som ogsaa maa tages 
i betragtning her, er den forholdsvis korte tid som er hengaat siden den 
store katastrofe under hvilken disse bakker blev utformet (1789I. Forovrig 
gjor kulturpaavirkning sig sterkt gjældende. Liene anvendes som havne- 
hager for kjorne, tildels ogsaa som utslaatter, og folk har av denne grund 
næsten overalt hugget ut og lysnet op i lovskogene. Foruten or og bjerk 
er selje (Sa/ix capræa) meget almindelig; desuten findes Salix iiigriaxiis, 
R/iaiiimis Fratigiila, Ribes nihniin og Loiiicera Xylostciiiii, men temmelig 
sparsomt. Myricaria germanica, som har stor utbredelse i K\am langs 
Laagen, og som ogsaa gaar et stykke opover Veiklas dalfore, optrær i nogen 
faa individer langs skraaningene ved veien op for Knipen. 

Naar undtages Lonicera Xylostcttm og Bctiila verrucosa, finder man mer- 
kelig nok ingen av vore mere kuldskjære levtrær og busker paa stedet ; de 
har alle sin nordgrænse længer syd i dalen. Dette er i og for sig ganske 
paafaldende, idet vi nemlig blandt græssene og urtene finder flere utpræ- 
get sydlige typer repræsentert (cfr. det folgende), hiiidlertid maa jo de 
urteagtige planter som holder sig nærmere jorden og nyder godt av vin- 
terens beskyttende snedække, ha lettere for at klare sig end trær og busker. 
Desuten opwarmes de nedre luftlag sterkest, saa de lave planter er ogsaa 
gunstigere stillet hvad sommervarmen angaar. Forøvrig kan jeg vanskelig 
tænke mig en lokalitet som bedre skulde egne sig for xerotherme relikter 
end denne dalside, hvor varmen f. eks. i juli maaned ofte er ganske uut- 
holdelig. Der er en svak mulighet for at lerskredene kan ha odelagt even- 
tuelle saadanne trær og busker, men om dette vet \i ikke noget med 
sikkerhet. 

Gran og furu mangler ogsaa fuldstændig i skraaningene ved Leine og 
Knipen. Forst i storre holder op mot Tunsbergtjeldet og længer nord i 
Veiklas dalføre ser man naaletrær, men ingen skog. Dette fænomen er 
utvilsomt ikke oprindelig, men beror paa uthugst. Ifølge oplysninger fra 
folk paa stedet har man ogsaa hist og her fundet stubber og rotter, som 
peker i samme retning. Dog vil jeg præcisere, at vi har ingensomhelst 
grund til at anta at der i historisk tid har været sammenhængende 
naaleskog i disse lerbakker. Den nu\'ærende lovtrævegeta- 
tion ( or og bjerk I paa stedet er ikke a \- saa recent natur som 



KOLK NOUnilAGKN. 



M.-X. Kl. 



man k a ii s k « lil <■\^ h <■ 14 y n d <• 1 s c skulde tro. Derom bærer kalk- 
liirtcn t\-(l(li,L;c N'idiicshyrd. 

Ved gaarden Kidk, like syd for I.eine, paa .sydsiden av Horjubækken, 
tiaifcr \i \-irkelig gi an. .kog. Dog er jordsmonnet her mere grundt og sten- 
hlanch l, tildels med opragende berg. 

Vis à vis Leine, i skraaningen ovenfor gaarden X'cikle opunder I lil- 
lingen, er (lei- i nutiden meget vakker furuskog. Men her er undergrunden 
av en helt anden natui', tor og stenet, antageligvis ut\'askede moræneav- 
leiiinger og nedraset forvitringsmateriale. 

Det ligger utenfoi- denne a\hand!ings ramme at skildre alle de asso- 
ciationstyper som udmerker dalskraaningene ved Leine. Jeg skal dog for 
ilildstaMidighetens skyld nævne en del karakteristiske eksempler. 

Om \i begynder med de mindst stabile, men vegetation s klædte 
lerbakker, saa moter vi her en eiendonimelig type, som i ferste række 
kjendetegnes ved en masseoptræden av Calantagrostis Epigcios, et Calam- 
agrostidetum Epigææ. Herav findes flere varianter, baade artsrikere og arts- 
fattigere. I synokologisk henseende udmerker typen sig ved en meget svak 
eller oftest ingen mulddannelse. Plantedækket er ikke helt sluttet; et bund- 
skikt av moser er flekvis antydet, men mangler gjerne helt. Dette beror 
sikkerlig altsammen paa substratets lite stabile karakter: overflatelagene er 
antageligvis baade vaar og host, og vistnok ogsaa efter heftige regnskyl, i 
bevægelse (cfr. fig. 3). Ökologien ligner i visse henseender forholdene hos 
den av Blomquist beskrevne „Hogbuskformation" (Svensk Bot. Tidskrift 
Bd. 5. 191 i) fra det sydlige Sverige, hvor lignende edafiske faktorer er 
virksomme og bl. a. forhindrer \-irkelig skog i at vokse op paa lokalitetene. 
Den Sterke insolation og u 1 1 o r r i n g av overflatelagene i sommertiden 
præger ogsaa vegetationen i nogen grad, og skaper en viss likhet med 
klippesamfund, f. eks. de skrindmuldete skiferbakkers i Kristianiafeltet og 
ved Mjosen. I Tyskland har man fænomenet endda mere utpræget i de 
saakaldte „pontische" eller „sonnige Hügel" (Gr aebner 1909). 

I den folgende tabel er sammenstillet to analyser, den forste fra ler- 
bakken ved Borju (10 < 10 m."), den anden fra en av de bratte skraa- 
ninger nordost for Roen (10 30 m.-) (sees paa tig. 3). Tallene angir 
artenes dækningsgrad indenfor ruten (Hult-Sernanders 5-gTadige skalal. 
den 6te oktober 1920. 



CalciDiagrostis Epigcios 
Da c/ V /is g/01 liera fa . . . . 

Ecstiica oviiia 

Agrostis vii/garis .... 



IH 1\' 
11 
1 



\\ 

111 
11 
1 



1 92 1. No. g. 



KALKTUFSTII^IER I GUDBRAXDSDALKN. 



Poa alpiiia 

Poa pratensis 

Brachypodiinn piiniatiiin .... 

Ca rex ericetcniiu 

Aiidrosace septeulrunialr .... 

Aiitliyllis viiluercrui 

Arenaria serpy/li folia 

Artemisia viili^aris 

Astragalus alpinus 

Calainintlia Aciiios 

Centanrea Scabiosa 

ClnysantlnniuniLeiicanlhenniin 

Draha incana 

Erigei-o)i aar 

Galium verum 

Hicracium umhellatum 

Hieraeium sp 

Knautia arvensis 

Lotus corniculafus 

Melandriiim ml rum 

Pimpinella Saxifraga 

Potentilla argentea 

Ranunculus polyanthenus . . . 

Rul)us saxatilis 

Rianex Acetosella 

Sedum annuum 

Silène venosa 

Stenophragma I'luxlianum . . . 

T/ialie/rum simplex 

Trifolium medium 

Trifolium prateuse 

Vicia cracca 

Viola arenaria 

Viola canina 

J^iola collina 

Mida tricolcr 

Rosa cinnaniomea 

Betula verrucosa (2 m.) 

Aluus incana (busk) 

Hypnum rugosum 

Thuidium ahietinum 



ROLF nori)HA(;k.\. 



M.-N. Kl. 



Tahi III 11 \iscr en cii ikIoimiik li;^ hl:iiHliii<; a\- plantegeografiske typer, 
sydlige ot^ noidlij^c om liinaiidcn. 

Andre associationer nujtcr os paa skraaninger som er mere stabile, og 
som derfor tillater en tættere, sluttet plantcvekst. Det følgende eks. (5 ^^ 
10 ni.^) viser en assfxialion rrcindclcs Ijc^iaaciidi- av græs og urter, men 
Ilten Cd/diiKii^i-oslls l'lpii^cios; istcdeii koiiiiiii r andre arter til, forst og frem.st 
J''('s//iC(i oviiia. Calaniagrostis Kpii^cias ynder aabenbart det mere urolige 
substrat. Med sine svære krypende rhi/onier passer den jo ogsaa ud- 
iiierket tier. 



l'cstiica ovina 
('(i/d/i/di^ros/is anindiiKiccd 
Dactylis gloiiu rala 
Brha iiinlia 
Poa alpliia 
Poa pratensis 
Agrostis vulgaris 
Achillea Millefoliimi 
Aiidrosace septentrionale 
Antennaria dioica 
Campanula rotuiuUfolia 
Carex ericetoruni 
Ccntaurca Scabiosa 
Ccrastiuni vulgatuin 
Erigeron acer 
Fragår ia vesca 
Galium verum 
Hieracium umbellatum 
Hieracium Pilosella 



11 


Kno Ilt Kl (irvensis 




11 


Lathy iiis pratensis 






Lotus corniculatus 


II 




Pimpinella Saxifraga 






Plantagii media 


II 




Piitentilla verna 






Rubus saxatilis 






Runu'x Acetosella 






Sediim annuum 






Silène venosa 






Solidago 1 Irgaurea 




II 


Trifolium nu^diuni 






Trifolium repens 






Trifolium pratense 






J 'erom'ca saxatilis 






J'icia cracca 






J lola arenaria 






Jl'ola canina 






J 'iola collina 




a 


biet i mim III — IV 





Desuten stod der en enslig Betula verrucosa (ca. 5 m. hoi) og en Alnus 
incana (2 — 3 m. hoi), samt en busk av Rosa cinnamomea indenfor proveflaten. 

Ogsaa Brachypodium pimmtum kunde være associationsdannende paa 
mere stabil bund. 

Som et eks. paa o r e k r a t fra forsænkningene mellem ryggene kan 
følgende proveflate (5 < 5 m.-) anfores: 



Alnus incana (4-^5 m.) IV 

Betula odorata (5m.) I 

Salix capreva (3 m.) II 

Acra ccvspitosa V 

Dactylis glomerata I 

Phalaris arundinacea I 



Triticum caninum I 

Carex capillaris I 

Aconitum septentrionale I 

Cirsium heterophyllum I 

Equisetum arvensc I 

Geranium silvaticum I 



1 92 1. Xo. 9. 



KALKTLFSTLDIER I C.UDBRANDSDALEX. 



Riiù/i.'^ i(f(nis I T/tssi7ai^n> Farfara 

Thalictniin simplex I Trifoliimi pratriisc 

i'huaria pentapctala 1 

Urtica dioica I 



III 



I kalktuftens høide, ca. 520 m. o. h., er der ingen antydning til ras elkr 
spor etter at der har gaat ut lerskred. Disse ophorer paa et lavere nivaa 
— heldigvis, kan man si; for ellers var der nok ikke levnet spor av denne 
merkelige forekomst. I den nævnte hoide indtar dvrket mark en meget 
bred plads og avveksler med lyse, forholdsvis la\e bjerkehager, h\'is 
bundvegetation fleresteds a\-slaaes til for. Disse aapne bjerkeskoger paa 
sterkt heldende bund er utxilsomt i sin nuværende skikkelse i hoi grad 
betinget av kulturen. De alternerer paa fugtigere bund med orekrat, som 
tieresteds opviser en bundvegetation av hoie græs og urter saa makeløst 
frodig at man kan skjule sig i den. 

I disse bjerkehager og langs veikantene finder man en del in 
essante arter. Folgende eks. (5 ■ 5 m.-) \-il vise dette. 

Bctitla odorata 
Bracltypodiimi piniiatimi 
Calaiiiagrostis aniiidiuaaa 
Aiithoxaiithum odoratuiii 
Fcstiica oviiia 
Poa )ictuoralis 
Canx sp. (steril) 
Acliillea MillcfoUnni 
BruncUa vulgaris 
Chrysanthenuim Lciicaiithcnumi I 
Dracocephalimi Riiyscliiaiia 
Euphrasia sp. 
Fra gar ia vcsca 
Galium rerum 
Geranium silvaticum 
Hvpochoeris macula fa 



y 


Kiiautia arveiisis 


\v 


Lotus corniculafus 


III 


Origanum vulgare 




Plantago nu-dia 




Pimpiiu'lla Saxi fragå 




Ruhus saxatilis 




Solidago J 'irgaurea 




Trifolium nwdium 




Veronica serpy IH folia 


/ 1 


Vicia cracca 




Viola canina 




Jl'ola collina 




Hylocomium triquetrum 




Hypnum sp. 




Peltigera cauiua 




Cladonia sp., basalskjæl 



Kombinationen: Bracliypodium pinnatum, Calaniagrostis aruiuiinacea, 
Origanum vulgare, Dracocephalum Ruyscliiana, Trifolium )iwdiian, Viola 
collina er overmaade interessant ved sin S3'dlige, xerotherme karakter. 
Bracliypodium pimmtum har her sin nordgrænse i Skandinavien (61 '40'), o^g 
Dracocephalum Ruyscliiana er her nær sin nordgrænse (Dovre)'. I denne 
forbindelse fortjener ogsaa følgende arter at nævnes: Androsacc septen- 
trionale, Calamintha Acinos, Carex ericetorum, Centaurea Scahiosa, Dianthus 



' F"ig. 42 og fig. 43. 



KOI.F NORDIIAGFN. 



M.-X. Kl. 



di'lloidis, l'crhascnin iiii^niiii, /l'rroiiicd vrnia), riscnnn viscosa r>. fl.'. Kn 
del a\- disse er allerede nævnt tidligere i tabellcne, de øvrige optræn ogsaa 
i Leinebakkene, med undtagelse av Vo-oiiicn venia, som jeg har fundet 
ved V'eikle. Disse arter danner ikke nogen absolut ensartet floristisk gruppe, 
men er allesaninien av sydlig tyi^e og varmekjæ-re. 

I skarp kontrast til disse staar de fjeld pi an ter som vi træffer paa 
de samme lokaliteter: Astrai^alits alpiiius, Veronica saxa/i/is, I)raha iucana. 
Disse er allerede omtalt ovenfor. Men hertil kommer: Oxyfropis lapponica, 
S(ixi/)-nii(j niz()i(/(s, Poa lyrsia og ficnlidiia ///va/i's, som jeg har fundet paa et 

par steder, og Priniitla scotica, 
som Blytt iagttok i 1891^. 

Sommeren 19 19 opdaget 
jeg nogen skiferklipper i ca. 
700 meters hoide, ovenfor 
Leinegaardene, som viste sig 
at være et rent as^l for fjeld- 
planter. Fra dette sted, men 
sikkerlig ogsaa fra de høiere- 
liggende ijeldpartier i nord og 
nordvest (som dog var meget 
fattigere paa arktisk -alpine 
arter paa grund a\- haardere 
sparagmiter), skriver antage- 
ligvis de nævnte fjeldplanter 
sig. Og deres forekomst i 
lerbakkene lavere nede (4 — 
500 m. o. h.) beror efter min 
mening forst og fremst paa- de 
eiendommelige konkurren- 
ceforhold som her gjor sig 
gjældende; vegetationen er jo 
fleresteds meget aapen og 
spredt, og av og til blotlægges mindre arealer ved nve ras, hvilket gir Qeld- 
plantene en chance til at hævde sig overfor alle de andre, som paa de v a n- 
lige lokaliteter i lavlandet formaar at utkonkurrere dem. Den manglende eller 
ytterst spredte trævekst sammen med den heldige eksposition gjor vel ogsaa 
sit til at fjeldplantene, som hyppig er fotofile, trives i disse sollyse bakker. 
Hertil kommer da som et meget vigtig moment den korte avstand op til 
de nævnte skiferklipper, h\-orfra fro og frugter meget let maa kunne fores 




Fig. 4. En del av skiferklippene ovenfor Leine med 
græs-urtesamfund, hvori indgaar de i teksten omtalte 
fjeldplanter. Midt i billedet Aconihmi. Utsigt mot 
nordost, Veiklas dal tilhoire med gran- og furuskog. 
Juli 1919. Nordhagen fot. 



1 Ifolge Blytt (1. c. p. 29) findes ogsaa Ai'ciia piibesccus og Erysiinitin hicraciifoliiini 
ved Leine. 

2 1. c. p. 28. Her nævnes ogsaa Oxytropis lappoiiica og Saxifraga aizoidcs. 



1 92 1. Xo. 9. 



KAI.KTL'FSTUDIER I GUDBRANDSDALEX. 



nedover hakkene baade med vand og \-ind (f. eks. om vintereni'. Ogsaa 
substratets kalkgehalt influerer sikkert. F'lere av de antbrte planter er kalk- 
yndende og gaar ogsaa andre steder i Gudbrandsdalen ned i dalbunden 
paa kalkholdige bergarter (f. eks. ved Otta og i Sel). 

Denne merkelige blanding av sydlige xerotherme planter og arktisk- 
alpine arter er altsaa et fremtrædende træk i det plantegeografiske helhets- 
billede som meter os i disse fantastiske lerbakker. Dog maa det skarpt 
fremhæves at fjeldplantene kvantitativt set spiller en helt underordnet 
rolle. Kun .4<:frai^(i//is a/pi/iits formaar at gjore sig fN'siognomisk gjældende 
paa enkelte mindre flekker. — Som vi senere skal se, fortæller kalktuffen 
os at det motsatte har været tilfældet i længst forsvundne tider. 

Da de o\enfor nævnte skiferklipper i 700 m.s hoide og deres vegeta- 
tion vil bli trukket ind i flere av de vigtigste ræsonnementer i nærværende 
avhandling, blir det her nodvendig at fæste opmerksomheten ved enkelte 
hovedtræk i deres plantevekst. Klippene ligger ovenfor gaarden Hagen, 
omtrent 200 m. hoiere end kalktuffen, og bestaar av løse, smuldrende skifre. 
BjoRLVKKE omtaler herfra en graa, skruklet f3'llit med kvartskirder med bøl- 
gende skifrighetsfald og høiere oppe en skruklet, graagren skifer med fald 
mot X. XO. ; den forer undertiden kalkholdige sandstenslag (1905 1. c. p. 215I. 
Folgende floraliste vil gi et indtryk av h\ilken besynderlig konstellation av 
arter disse klipper huser (6te juli 1919): 



Acouitiiiu scptciitrioualr 
Alectorolophus minor 
Aiitciiiiaria dioica 
Aiithyllis vulnr varia 
Arabis liirsiita 
Arctostaphvlos in'a itrsi 
BotrycliiiiDi Liiiiaria 
Brunclla vulgaris 
Calauia grostis aniiidiiiacra 
C a II II lia vulgaris 
Carcx capillaris 
Carcx cricetoriiui 
Carcx ornithopoda 
Carcx sparsiflora 
Cotoneastcr iutcgcrrima 
Cystoptcris fragilis 
Diaiithiis dcltoidcs 



Euphrasia sp. 
Fragaria vcsca 
Festuca oviiia 
Galium borcalc 
Galium uliginosum 
Gcntiaua Amarclla 
Geranium silvaficum 
Hicraciuni Pilosclla 
Knautia arvensis 
Mclica nutans 
Origanum vulgare 
Parnassia palustris 
Pinguicula vulgaris 
Plantago nwdia 
Polygonum vivi purum 
Phcgoptcris Robert iana 
Potent i lia 'iornwutilla 



' Om denne nedvandring er av foiiioldsvis recent natur, eller har paagaat gjennem geo- 
logisk set langvarige tidsrum, kan vi forelobig ikke avgjore. Dog er det neppe tvil 
om at lerskredene og de derigjennem skapte edafiske tilstande har begunstiget ned- 
vandringen. Cfr. forøvrig den generelle del. 



12 



UOLK NORIMlAfiKN, 



M.-N. Kl. 



I 'ninilil/ii vcriKi. 
Niihiis idoiis. 
Sa.xifrai^d (uLscrndciis . 
Scdimi aiiiiitiiiii. 
Si/ciir nifh'slris. 
I \icciiiiiiiii vi tis id<va. 
] ^rrhasciini iiigntiii. 
I 'croiiica officinalis. 
I Icia cracca. 
I 'iol(j coli i lia. 
Woods 'a ri l/i (lit la. 



.liitciniaria alpiiia. 
Astra i^a li IS a lp i nu s. 
Ccrastiiiiii alpiiiinn. 
Cetraria nivalis. 
Draha liirta. 
Draha incana. 



(iintiaiia nivalis, 
(icniiana t<nrlla. 
Jiinciis trijuiiis. 
Pliyllodocc cocrulca. 
I '(Kl alpina. 
/'(III cirsia. 
.Sabina /Jiimri. 
Seloiriin Ha spiniilasa. 
I 'cronica sa.vatilis. 

Ved en l^æk i nærheten a\- klip- 
pene notertes bl. a. : 

/uncus trii^liiniis. 

.Salix (ir. arhiisciila (bastard?). 

Sali.x lapponiiin. 

Sali.x reticulata (meget sparsom). 

Sa.xi fragå aizoidcs. 

Saussiirca alpina. 

/lialictriini alpinuin. 



En saa fantastisk kombination som paa den ene side Vcrhasciini ni- 
grum, Origanum vulgare, Calamagrostis arundinacea og /uncus trifuhis, Antcii- 
naria alpina, Gcntiana tcnella og de andre tjeldplanter paa den anden, har 
jeg ikke set noget andet sted i vort land, ialfald ikke sondenfjelds. Kun 
i det indre av Nordland fylke, paa kalk og dolomit, har jeg iagttat lignende, 
tilsyneladende paradoksale artskonstellationer. 

Klippene er i vore dager fuldstændig træbare, med tyndt jorddække 
sterkt opblandet med fyllitmateriale. De synes ogsaa at være sterkt utsat 
for vinden, og bærer sandsynligvis om vinteren et ubetydelig snedække. 
Ovenfor klippene følger et mere plataaagtig omraade, bestaaende av haar- 
dere bergarter (tildels overdækket) med lyngvegetation (associationer av 
Empetrwn nigrum, Calluna, Phyllodocc, Vaccinium-arter, Arctostapliylos al- 
pina, tildels med lav /uniperiis). Her notertes Pulsatilla vernalis og Lyco- 
podiuni alpinuin. 

I 850 m.s hoide saaes folgende tjeldplanter ved en bæk: Bartscliia 
alpina, Betiila nana, Cerastiiim trigyniini, /uncus higlumis, Raimnciilus liy- 
perboreiis, Sali.x glaitca, S. hcrbacca, S. lanata, S. lapponum, S. myrsinitcs, 
S. reticulata (sparsom). Opunder Tunsbergfjeldet findes lave, spredte gran- 
og fu ru- individer, granen oftest som lave, vindtørre avlæggergrupper. 
Begge to gaar saa godt som helt op til fjeldets top (944 m. efter rekt- 
angelkartet). I 900 m.s høide vokste Turritis glabra og Sa.xifraga Cotyledon 
sammen. Ellers er Tunsbergfjeldet meget plantefattig og trivielt paa grund 
av de haarde sparagmiter hvorav det er opbygget, higensteds finder man 
en saa rik flora som paa fyllitklippene længer nede, ja en hel del arter 



ICSI. No. 9- KALKTUFSTUDIER I GUDBRANDSDALF.X. I3 

syntes udelukkende at være knyttet til disse. Antageligvis vil Torgerkampen, 
et tjeld som ligger 5 — 6 km. længer vest, vise sig rikere end Tunsberg- 
tjeldet; Bjørlvkkes oplysninger om bergartene tyder paa det. 

Mine floristiske optegnelser, som blev revidert under et nyt besok 
17 juli 1919, er i flere henseender bemerkelsesværdige. Saaledes mangler 
Diyas octopctala fuldstændig paa fyllitklippene og ogsaa høiere oppe. 
Hovedhensigten med min ekskursion til fjeldene ovenfor Leine var netop 
at opspore D/yas; men til trods for en ihærdig leting var den ikke til at 
opdage. Da kalktuften som bekjendt indeholder Drvas i kolossale masser 
i en bestemt horisont, er det nævnte faktum meget overraskende og gir 
os et tydelig vink om hvilke store forandringer vegetationen her i Kvam 
har undergaat i aartusenernes lop. Skiferklippcne og deres vegetation gir 
os vigtige holdepunkter og angrepspunkter i den følgende diskussion. Som 
Jeg senere skal forsoke at klargjore, er det ingen tvil om at Drvas den- 
gang kalktuften war tæt bevokset med denne art, ogsaa dominerte paa de 
angjældende klipper, som efter de erfaringer jeg har gjort med hensyn til 
denne dvergbusks livskrav rundt omkring i Norges fjeldtrakter, skulde S3'nes 
at være ideelle som voksested netop for Drvas octopctala. En noiere gjen- 
nemgaaelse av den anførte floraliste gir ogsaa fingerpek i denne retning. 
I Ivor langt indover Ijeldet vi maa gaa i vore dager for at flnde den nær- 
meste Dryas, vet \i foreløbig ikke. 



\'egetationen paa selve tuffindestedet er i nutiden ikke naturlig, 
men sterkt kulturpaavirket. Tuften ligger nederst i et brat jorde under 
den østligste av Leinegaardene. Kilden kommer ret frem a\- jorden under 
en stor stenrøis, som antageligvis er meget gammel (fra den tid da jorden 
i nærheten blev ryddet), og herfra fører et litet bækkeleie med spredte 
klynger av vidjebusker og bjerk nedover bakken, som en stripe mellem to 
dyrkede enger, og fortsætter gjennem en ny stenrøis over i orekrat længer 
nede. I stenrøisens omgivelser hvor bunden er mest tor, \okser et krat 
av Betitln x'crntcosa og B. odorata, samt litt Salix caprœa, S. nigricans og 
Rosa ciniiainomca. Markvegetationen langs bækken er nu ødelagt av ki^eatur- 
traakk og gravninger, likesom ogsaa den primitive kjørevei som passerer paa 
skraa over tuffindestedet, har influert paa omgivelsene (cfr. fig. 5 og 6). 
Græs og urter danner et slags dække nedover bakken, men uten syn- 
derlig orden. Følgende arter optrær indenfor kalktufomraadet: 

Aconitimi septentrionale, Acra cæspitosa, Agrostis stoloiiifera, A. 
vulgaris, ^■llcctorolopliiis minor, yllcliiniilla d/'/icinalis, ^Inthyllis vnlneraria, 
Artcuiisia vulgaris, Brachypodiuni pinnatuni, Briza media, Brunclla vulga- 
ris, Calannigrostis arundinacea, Caiiipanula rotundifolla, Carex capillaris, 
i. flava, C. panicca, C arum Carvi, Ceiitaurea Scabiosa, Cerastium vulga- 
tum, Cluysantheinum Liucanthcnnini, ( irsium heterophyllum, Crépis paludosa, 



'4 



r<Ol.t' NOKDIIAOKN. 



M.-X. Kl. 



Dddv/is i^lomrnUd, l:i/iiisiliiiii (irvriisr, A'. />r(i/riisc, l:ii/>lior/)i(i J Irlinsi n/>i(i, 
EiiphnisKi (i/JiriiKilis, /•'cslticd aviiut, i . nilira, /■ nii^drid 'orsrd,(i<i/in)ii pdliisln-, 
(•. iilii^iiiDSitiii, (iriiiii rivale, (jyiuiuuicnid coiiopcd, /lirrdciuni iiiiihclldtttni, 
Jiiiuiis hdfoiiiiis, Kiidtitid arvnisis, Ldtliyrns /)rti/rj/s/s, Lioutoda}! autnmna- 
/is, Liiiiiiii cdlhdiiiciini, Lotus aivniculdtus, Moliuid corrdlrd, l'halaris arun- 
(liiidcra, Piiiipiiiclld Sdxifraga, Pldiitdi^o incdid, l'aa ul/yina, Folygomim vivi- 
paniin, Polnitilld din^criim, Potciitilld l'oriucnlilld, /\diiiinnil!ts dcrr, R. repens, 
Riibtis sdxdtilis, Sileiir veiiosa, Suliddi^o I 'ii-i^diircd, flidlictrniii simplex, 
TrifoHuiii iiirdiuni, 7". pra/n/se, '/'rii^lfu/iin palustre, Tussildi^o P'arfdrd, 1 7- 
inaria peiitapetala, Urtica dioica, Vicia cracca, V. sepiiini. Viola tricolor. 
Forskjellige ting tyder dog paa at oprindelig en Carex pauicea-disso- 
ciation av eutrafent og hydiDlilt precg har behersket bækkens nærmeste 
omgivelser. Længer nede i bakken, i tuffens periferi, optrær nemlig saa- 
danne plantesamfund over en kortere strækning. For at belyse dette har 
jeg i nedenstaaende tabel opfort analyser av to prøveflater, hver paa ca. 4 m-, 
fra dette sted: 



1 


2 


II-III 


iv-v 


IV 


I 


I 


II 


— 


II 

I 





1 


I 


T 


1 


1 

II 


I 


II 



Carex paiiieea . . . 
/uncus laniprocarpu:- 
Aera cæspitosa . . 
Affrostis vuliTciris . 



Phalaris arundinacea . 
Phleniii prateiise .... 

Carex flava 

Juncus conipressus . . 
Triglocliiu palustre . . 
Equisetuui arvense . . 



Alecti.rolophus minor . 
Phiinella vulgaris . . . 
Cirsium heterophylluni 
Mentha sp. (steril! . . 
Parnassid palustris . . 
Polygonum viviparum 
Raininculus acer . . . 
Thalictrum simplex . . 
Tussilago Far/ara . . 
[ 'Imaria pentapetala . 



I 



1 det forste eksempel fandtes ogsaa Hypnum filicinum III med nedtil 
halvv^eis forkalkede skud. Denne moseart optrær ogsaa hoiere oppe hvor 
vandet silrer utover den blottede tufoverflate. — Hvis vegetationen fik 
anledning til at falde tilbake til naturtilstanden, vilde utv^ilsomt Almis incana 
indfinde sig saaledes som længer nede i bakken, og et orekrat om gi tuf- 
kilden, med litt bjerk paa de tørrere flekker. 



B. Stratigrafiske undersøkeiser. 

Som resultat av en række barometermaalinger i juli 19 19 fremgaar 
det at kalktuffen ligger 232 m. hoiere end Slettens landhandlen nede i 
dalbunden. Da dette maa antages at ligge et par meter hoiere end Kvam 
station (287,7 ""■• ^- h-t. skulde kalktuffens hoide over havet bli ca. 520 m. 
Blytt angir ca. 500 m. (496 m. efter barometermaalingl, men dette er 



I92I, 



No. 



KALKTLFSTLDIER I GUDBRAXDSDALEN. 



utvilsomt tor lavt. I oktober 1920 medforte jeg et andet barometer sam- 
tidig med at lufttemperaturen bestemtes \-ed hjælp a\- slyngetermometer. 
Resultatet blev 232 m. over dalbunden, altsaa noiagtig middeltallet av tjor- 
aarets maalinger. 

Lerbakkene paa tuffindestedet skraaner fra vest mot ost, dog ikke ganske 
jevnt. Heldningsvinkelen er i gjennemsnit 15 — 20 ; dog kan den lokalt 
være betydelig storre'. Hvor stor utstrækning tuften har, kan ikke siges 
med sikkerhet. Avstanden mellem mit nordligste og sydligste profil er ca. 
15 m. ; paa begge disse steder skraaner tuften indunder dyrket mark, som forbod 




Fra tuffindestedet ved Leine. Utsigt mot nordøst. Tilhoire kjoreveien. 
Foto 5te oktober 1920 (efter groftens gjenkastningl. 



videre gravning. Imidlertid skulde jeg anta at tuft"en ganske sikkert har hat en 
utstrækning i denne retning [^' tvers paa skraaningen) som er dobbelt saa stor 
som det opgitte tal. Avstanden mellem mit høiestliggende profil (indunder den 
før omtalte stenrois) og mit laveste, maalt langs efter skraaningen (3: vest — 
ost), er ca. 20 m. Opad S3'nes tuft"en at kile ut ganske snart, men nedad synes 
dens grænser slet ikke at være naadd. Under opdyrkningen av den omgivende 
mark synes store tufmasser at være brutt op av jorden; saaledes ligger 



der svære tuf blokker (hvorav en som er 1,= 



I m. X 0,30 m.l, hoved- 



sakelig bestaaende av mosetuf, paa en sfenrois længer nede i bakken. Disse 
blokker skriver sig uten tvil fra tuffens undre laç. Oesaa de lose tufstvkker 



' Faa længdepronlet (iig. 11) er heldningen i tuffens ovre del noget overdrevet for at 
undgaa altfor store knæk og bølger i lagrækken (paa grund av den forskjellige hoide- 
og længdemaalestok). 



I 6 KOM" NDKDMAf.KN. M-.N. Kl. 

som liar raiiilcl lu-ii.t;(r iicdtni r lien 014^ 'iMK*'' "^pr'fji liisl o^ lier i (•»re- 
kralld, sl<ii\cr sij^ \cl lia en clli r amlcn iiyflyrl\nin;^s|jcriofjc cllcr o[)rcnsk- 
nin^ av bækkclcict. l)ir ir iallald i din di I a\- tiirfcn som J<:g liar undcr- 
sokt, intet tegn til i"as cllcr iilglidniiiycr. 1 )og ci' det mulig at saadanne 
kan ha foregaat længer nede. 

I )(■ foiskjellige tuHag viser sig i d( t store og hele tat at være omtrent 
pai'alelle med bakkens skraaning; men der er en mængde smaa avvikelser 
i forskjellige retninger, saa stratigrafien er ganske komplicert. Den kaik- 
tufavsættinde kilde kfniinier ret ut av bakken like i tuffens overkant, og 
nogen store \ariatioiicr i (U-iis lop har netop av denne grund ikke indtraadt 
i tidenes lop. Imidlertid har de enkelte karakteristiske tuflag sin største 
tykkelse paa helt forskjellige punkter indenfor omraadet, h\ilket kunde tyde 
paa at bækken saa at si har pendlet tVem og tilbake indenfor en cirkel- 
sektor med tufkildens „dagaapning" som centrum, ialfald til en viss grad. 
Dette er bare naturlig, da jo en saadan kalkdannende kilde stadig fylder 
op sit eget leie og derved tvinger sig selv over i et nyt lop. Imidlertid 
kommer her ogsaa et andet og meget vigtig moment til, som Blvtt forst 
opdaget, nemlig at kalktufavsætningen ved Leine ikke har været kontinuerlig, 
men intermitterende; dette \'il fremgaa av den folgende utredning. Og 
det er ganske klart at en stans i avsætningen ledsaget av forvitring og 
mulddannelse i hoi grad maa virke bestemmende paa kildens fremtidige 
lop, naar den atter begynder sin kalkdannende virksomhet. — Overfor 
enkelte tufavsætninger med tydelige avbrud i avsætningen og derav følgende 
muldstriper eller forvitringshorisonter, har man gjort gjældende at dette 
kunde skyldes det forhold, at kilden periodevis har tat et helt nyt lop. 
For Leinetuffens vedkommende kan imidlertid dette ræsonnement ikke 
gjøres gjældende, da kilden som nævnt kommer ret ut av jorden i tuftens 
overkant. Det vilde ialfald være hoist ubegripelig hvorledes denne kilde, 
hvis den under de to avbrud i avsætningen som tuften viser, hadde et 
andet lop, atter kunde indfinde sig præcis i det gamle underjordiske lop 
og komme frem i dagen noiagtig paa det gamle sted, ikke bare én gang, 
men gj en tagne ganger. 

I Gudbrandsdalens tufter kan man i likhet med hvad Serx.wder har 
vist for svenske kalktufters vedkommende (191 6 1. c), adskille scdiniciitæri' 
cg sedentære lag. Hvis et lag viser sig at bcstaa av lose blader, kvister 
og andre avkastede plantedeler som vandet har bundfældt, og som senere 
litt etter litt er forkalket, eller av rent fysikalsk utfældt fossilfri tuf, faar 
man en sedimentær avsætning. I andre tilfælder derimot kan tuffens over- 
flate være dækket av en sammenhængende, rotfæstet, levende vegetation, 
som ogsaa gradvis kan forkalkes. Man faar da en veritabel au to k ton 
dannelse, en fossilificert vegetation in situ, en sedentær avsætning. I 
praksis kan det undertiden være vanskelig at holde disse to typer ut fra 
hinanden; saaledes kan visse mosetuffer og cyanotycé-tufter være saa kom- 
pakte og strukturlose at de sterkt minder om fysikalsk utfældt, sedimentær 



içai. No. 9. 



KALKTUFSTUDIER I GUDBRANDSDALEX. 



17 



tuf^ Men i almindelighet er de sedentære lag meget karakteristiske, med 
ortotropt orienterte skud og andre sikre kjendetegn. For utredningen av 
vegetationens historie er den slags horisonter meget vigtige, idet man her 
kan finde rester av urteagtige planter, som jo visner sukcessivt ned uten 
bladfældning og under vanlige omstændigheter raatner væk meget raskt, og 
som derfor helt naturlig mangler i sedimentære lag. Et par av Leinetuflfens 
lag er i denne henseende meget interessante. Meget ofte begynder en kalk- 
tufavsætning med sedentære 
lag (mosetuf) og fortsætter 
som sedimentær a\-leiring. 
Men dette er slet ikke altid 
tilfældet, hvilket vil fremgaa 
av profilene i nærværende 
avhandling. Imidlertid gir og- 
saa de sedimentære lag i 
Leinetufifen utvilsomt et ad- 
ækvat billede av vegetationen 
paa stedet, netop fordi kilden 
kommer ut av jorden og saa- 
ledes ikke kan ha transportert 
plantedeler langveisfra. Dog 
vil selvfølgelig den slags ting 
kunne ske under sneløsningen 
eller i flomtider. 

Som i forordet nævnt 
lykkedes det mig med bistand 
av en gammel mand som i 
1 89 1 assisterte Blytt, at ut- 
finde noiagtig det sted hvor 
Blytt optok sine to profiler-. 
Disse laa like i nærheten av 
hinanden i overkanten av den 

gamle bygdevei som skraar o\-er tuften. Blytts profiler kan meget smukt 
indpasses i min serie og utfylder et hul i denne. 

For fuldstændighetens sk3id, og for at man skal faa det rigtige ind- 
tryk av hvor lovmæssig tuffen er opbygget, skal jeg i det folgende beskrive 
alle de opmaalte og undersøkte profiler. Fremstillingen vil ogsaa vise hvilke 
fuldstændig misvisende resultater en mere overfladisk undersøkelse kunde 
ha ført til, idet en hel del av tuftens mere perifere profiler tilsynelatende 




Fig. 6. Den ovre del av Leinetuft'en. Kilden kommer 

frem av jorden overst mellem buskene, som skjuler 

stenroisen. Billedet er tat efter grøftens gjenkastning. 

Nordhagen foto, 6te oktober 1920. 



' Jeg agter ved en anden anledning at komme tilbake til sporsmaalet om kalktutfenes 

genesis. I de senere aar har tyske forskere tildels git vigtige bidrag til tufutskillelsens 

mekanik. 
2 Manden fortalte ogsaa at da „han kom ned til det tredje laget i tuften, fandt Blytt 

en plante som han kaldte Dryen (^- Dryas)". 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 9. 2 



HOL}' Nf)i<i)iiAf;i:N. M.-N. Kl. 



er liiniiiii K idi lorskiclli^^c tV.'i fl<' im ic ccntr.'ih- i tuffens hovedparti, mens 
(le i \ iikclii^liiu II !ii(,l;(1 liiniiionisk intjordne-r si^ under disse. 

I nedenstaacnde utredninj^ a\- i)rofil(ne er disse inddelt i flere serier 
efter sin beliggenhet i l"orhold til d« n store groft som blev gravet paa 
langs av bækken. Forst komnicr en venstre serie, optat nedenfra 
og opad bakke paa groftens venstre {^'- søndre) side. Den tar sit ut- 
gangspunkt i Blytts ])rofil, som er det nederste i serien, og ender opun- 
der stenrøisen, Inor tiiifen antageligvis kiler ut ganske snart. 

Tilsvarende hertil faar vi en hoire serie paa grøftens motsatte side, 
ogsaa nedenfra og opad. Endelig har jeg en serie paa tvers av bækken, 
tverserien, fra syd mot nord, som ender i profilet av 1920. Alle disse 
er optat ovenfor den nævnte kjorevei, som gaar paa skraa henover tuf- 
findestedet. Som profiler nedenfor v e i e n beskrives et par vigtige grøfter 
optat længer nede i bakken; de ligger i ca. 20 — 25 meters avstand fra det 
øverste profil i venstre serie. 

Længdeprofilet' (fig. 11) er konstruert paa grundlag av venstre serie 
og profilene nedenfor veien, tverprofilet (fig. 12) ved hjælp av tverserien 
og venstre serien. 

Som „Alnus-tLif" betegnes den allerøverste, tidligere ukjendte tuf hori- 
sont eller rester av denne. „Furutuf", „Dryastuf" og „bladtuf" er tildels 
de samme betegnelser som Blytt anvendte. Dog omfatter Blytts bladtut 
eller „birketuffen" ogsaa den underliggende mosetuf, som hænger sammen 
med birketufifen. Jeg foretrækker dog at betegne denne som „mosetuf". 
Ved „undre Dryashorisont" forstaaes en sedentær horisont i bladtuffens 
underkant (mosetuffens øvre del), som tidligere ikke er beskrevet fra Leine- 
tuffen; den mangler ogsaa i enkelte profiler og er av meget mindre dimen- 
sioner end den hoiere oppe i lagrækken forekommende „Dryastuf". Ved 
„rød 1ère" forstaaes den høist eiendommelige, i visse profiler skrikende røde 
1ère som optrær umiddelbart under den laveste mosetuf, og som over hele 
tufomraadet bedækker den underliggende morænelere, som benævnes 
„blaa 1ère". 

For alle profilers vedkommende gjælder at lagfølgen beskrives nedenfra 
og opad (kronologisk orden). 

Venstre serie. 
Profil I (Blytt i 89 i). 
I. Jøkeller med vandreblokker. 
II. Jernholdig 1er uten forsteninger indtil 0,03 m. 

III. JGulgraa skifrig birketuf (uten furu) 0,45 m. 

IV. [Heri er som nævnt indbefattet mosetuften paa bunden (Blytt 1. c. p. 7). 



1 Profilet benævnes her længdeprofil fordi det er optat langs med bækken (tverprofilet 
tvers paa bækken). Geologisk set er længdeprofilet i grunden et tverprofil ; men da 
stratigrafien ved Leine er ytterst komplicert, kan man hfr ikke tale om „strøk' og 
„fald" i vanlig stratigrafisk betydning. 



ig2I. No. 9. KALKTUFSTUDIER I GUDBRANDSDALEN. I9 

[Gulgraa, tildels jordagtig Dryastuf (med furu) indtil 0,03 m. 
■ 1 Gronliggraat 1er uten forsteninger 0,04 m. 
\'I. Graahvit furutuf 0,58 — 0,68 m. 
MI. Muldjord 0,10 — 0,15 ni. 

Forovrig henvises til Blytts egen beskrivelse av de forskjellige lag 
(1892 1. cl. 

Profil II. 

I. Blaa 1ère, ukjendt mægtighet. 
II. Rod 1ère ca. 3 cm. 
III. Mosetufkompleks. 

A. 10 cm. graagron, slagg- eller koralagtig mosetuf, brunfarvet paa 
undersiden, med ax'tryk og hulheter efter Eqitiscfiiiii varicgatimi. 

B. 10 cm. rodlig mosetuf. 

C. 5 — 8 cm. hvitagtig kalkgrus (mosetuf biter). 

I\'. Bladtuf, 25 cm., planskifrig, med blader av Betiila odorata, Popitlus 

tiriiiula, Sa Hees. 
V. Dryastuf, 3 cm., los og smuldrende med massevis av Z);3'(75-blader og 

stammere 
VI. Furutuf, 18 — 20 cm., sterkt for\'itret, med masser av fitniiiaaler, bla- 
der a\' J^aeeiniiiDi vi f/s idera. Øverst muldblandet. 
VII. Muldjord, 20 — 25 cm. med stener og smaa fragmenter av Alints-iuï, 
med blader av Alnus iucana. 

Sammenlignes dette profil med Blytts, som laa ca. 2 m. længer nede 
i bakken, ser man at lagene er præcis de samme, men mægtigheten er helt 
anderledes. Den slags variationer moter os imidlertid overalt indenfor tuftens 
omraade. 

Profil III. 
I. Blaa 1ère. 
II. Rod 1ère, noget grusblandet med smaa skiferfragmenter, samt en stor, 

raatten skiferblok midt i grøften. 
III. Mosetuf, ca. 10 cm., nederst med Eqiiisetiiui variegatiiiii og utydelige 
bladavtryk [Salix sp.l, pores og slaggagtig. Hænger sammen med 
overliggende. 
I\'. Bladtuf, 15 cm., grønbroget, vakkert laget, med hjerk, asp, Saliees. 
W Dryastuf kompleks, ca. 15 cm., danner nederst en skorpe ovenpaa blad- 
tuften. Tuften var eiendommelig sinteragtig, i vaat tilstand rodfiolet, 
i tør tilstand kridthvit og lite egnet til at opbevare fossilavtryk. Dog 
fandtes en række blader (13) av Dry as octopetala, fiinniaaler og et 
par blader av Betiila odorata. Overst fandtes en hvitagtig, 1er lig- 
nende sone, bestaaende av opsmuldret tuf med mindre, fastere biter, 
utvilsomt et forvitringslag, op til 10 cm. tykt. 



' Diyiis ociopctala er en dvergbusk med træagtig: stamme. 



20 



ROLF' NORDHAGEN. M.-N. Kl. 



VI. I'^iiiiiUirkonipleks, 30 cm.; nederst en veritabel forkalket furustok (grov 
gren eller fragment av en stamme), som var løs og smuldrende (raatten)! 
Ellers temmelig løs furutuf, tildels jordagtig; de faste stykker med 
umaadelige masser av fiiriiiinalcr. 

VII. Muldjord med rotter, 15 — 20 cm. 

Profil IV. 

I. Blaa 1ère. 
II. Rod 1ère, t\'pisk. 

III. Mosetufkompleks. 

A. 10 cm. hullet, slaggagtig, graagrøn mosetuf, med Eqiiisetiini varie- 
gaf uni og huller efter smaa kvister og pinder (Salicesl). Lokalt 
fandtes en liten linseformig, hvit lermasse ovenpaa denne mosetuf, 

B. 7 — 10 cm. mosetuf, øverst med den undre Dryashorisont 
ut viklet, med blader og stammer av Drvas (middels mængde), 
Equisctmn variegatiiui, bladfragment a\' Bctiila odorata, en lever- 
mos som antageligvis er en Pellia sp., et eiendommelig avtryk 
som muligens er Toficldia palustris, samt sucgler. 

IV. Bladtuf, 10 — 15 cm., meget vakker og planskifrig med de vanlige 
lavbladcr i stor mængde. 

V. Dry astuf kompleks, temmelig komplicert bygget. 

A. Underst antydning til mosetuf ovenpaa bladtuffen, 2 — 3 cm. 

B. Graagrøn Dryastuf, med uhyre masser av Z)rvf7sblader og stam- 
mer, desuten et blad av Salix reticulata og et bladfragment av 
bjcrk, ca. 5 cm. mægtig. 

C. En tynd, 1,5 — 2 cm. mægtig sedimentær horisont, grønlig av 
farve med: Dryas (temm. sparsom), ///;-// (flere store og brede 
naaler), bjcrk (bladfragmenter), Salix arbtiscula (mange blader), 
Salix capræa (i bladfragment), asp (bladfragmenter), Pyrola minor 
(et par blader), samt en stor fuglcfjær. 

D. I — 1,5 cm. hvit kalkgrus (forvitringshorisontl. 

VI. Furutufkompleks, ca. 45 cm. Nedtil tildels litt fin mosetuf, dels furu- 
tuf med mange naaler og blader av Salix arbuscula (det eneste sted 
i furutuffcn hvor denne art er fundet!). Derover vanlig furutuf, som 
opad viser antydning til en sedentær sone. Overst los og jordagtig 
(forvitret) furutut^. 

VII. Muldjord, 20 cm., med rester av Al/ius-tui' og en eiendommelig krøllet 
tuf (mosetuf?). 

Dette profil er interessant derved, at \\ her for forste gang moter en 
ganske indviklet bygget Dr^-astuf, et fænomen som de folgende profiler 
ogsaa viser. 



I92I.X0. 9- KALKTUFSTUDIER I GUDBRANDSDALEN. 21 

Profil V. 
I. Blaa 1ère. 
II. Rod 1ère, typisk. 

III. Mosetufkompleks. 

A. 5 — 7 cm. hullet slagg-tuf, normal. 

B. 5 — 10 cm. mosetuf, lysere av farve, øverst mot bladtuffen med 
en ant3'det undre Dryashorisont (Z)rv<75-blader, Equisctum varie- 
gatiDU, utydelige lovblader). 

IV. Bladtuf, 20 cm., normal. 
V. Dryastufkompleks. 

A. 5 — 7 cm. grønlig Dryastuf med masser av Dryas-xç:?X(tx, aldeles 
kaotisk. 

B. 2 — 3 cm. lagdelt horisont, med avvekslende grønlige og gulhvite 
lag, indeholdende: Dryas (flere blader), Salix arbiisaila (et litet 
blad), Pyrola minor (3 blader), ftiru (flere store naaler), bjci-k 
(mange blader, hunrakler I, asp (bladfragmenter), Carcx sp. (en liten 
plante med 3 blader). Alt dette tyder paa at tuffen har baaret 
en meget sparsom vegetation paa dette sted, og laget er halvt 
sedentært, halvt sedimentært. — Allerøverst viser denne horisont 
sig at være belagt med en ytterst tynd, hvit skorpe av kalkgrus, 
som gjerne løsner sig av idet st3'kkene hakkes los. 

C. Meget eiendommelig løs sone, kun paatrufifet tydelig i dette 
profil, 3 cm. mægtig. De smaa klidbrød-lignende tufstykker ligger 
(især nedtil) i et fint, hvitt kalksmuldr og indeholder massevis av 
3-tterst fine fu ru naaler samt mange Z)rv'f/s-blader, desuten 
et blad a\' Salix arbusciila og et fragment av bjcrk. Som nær- 
mere utredet i det følgende, er dette lag en forkalket, destruert 
førne o: sterkt opraatnet barnaalavfald, som danner raa- 
humus i skogbunden. Antageligvis er det dette lag som i visse 
profiler danner for\itringshorisonten øverst i Dr3'askomplekset 
(cfr. foregaaende profil). 

VI. Furutufkompleks, 40 cm. ; nedtil med massevis av store furunaaler og 
mange blader av Vacciniiiin vitis idæa; underflaten er altid fri og 
hænger aldrig sammen med foregaaende horisont. Der- 
over et mere løst furutuflag, optil mere kompakt tuf. 
VII. Muldjord, 25 cm., med Alnus-inï og krøllet, forvitret, grov tuf. 

Profil VI. 

Dette er i virkeligheten optat midt i grøften, men medtages av forskjel- 
lige grunde her. 
I. Blaa 1ère. 
II. Rød 1ère. 
III. Mosetufkompleks. 



22 KOi.r Noi<i)iiAf;i;\. M.-X. Kl. 

A. 5 cm. slaggagtif4 tuC, iionn.-il. 

B. 4 cm. livitlig, grovt mosctulgrus. 

C. 5 cm. mosctuf, overst med nogen blafier av JJryas, altsaa antydet 
undre I )ryashf)risont. 

IV. Bladtuf, ca. lo cm., iimiual. 
V. Dryastuf kompleks. 

A. 7 cm. grønliggraa kaotisk Dryastiil", med uiiyre masser av Dryas- 
rester, bl. a. pragtfulde staminei- med paasittende bladrester, en 
blomst i frugtstadiiim ; \idere et par furuiinalrr samt fragment av 
bjerkcblad. 

B. 3 cm. lagdelt sone med Dryas temmelig sparsom (som i foregaa- 
ende profil). 

C. 3 cm. løs, smuldrende, klidagtig tul", med fine furunaaler og litt 
Dryas (som i foregaaende profil, men mere opsmuldret og les). 

VI. I'\irutufkompleks, 40 cm., som i foregaaende profil. 
VII. Muldjord, 25 cm., med biter av A/itus-tui'. 

Dette profil er interessant idet bladtuffen her er paa vei til at kile ut 
over mot grøftens hoire side. Dette fænomen blir nærmere omtalt under 
„høire serie". 

Profil VII. 
I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetufkompleks. 

A. 7 cm. hullet slaggagtig, graagrøn mosetuf (normal). 

B. 5 cm. kalkgrus, bestaaende av mosetufstykker. 

C. 5 cm. undre Dr3'ashorisont, med Dryas-h\a.deir og løvblader. 

IV. Bladtuf, 10 cm., normal {h/'crk, asp, Salices). 
V. Dryastufkompleks, ca. 15 cm. 

A. 8 cm. ren Dryastuf, tildels et eneste smuldr av Dryas. 

B. 2 cm. lagdelt ovre sone, med furunaaler og sparsom Dryas (ofr. 
foregaaende profiler). Overst et tyndt, hvitt grusagtig belæg med 
nogen furunaaler. 

C. 5 cm. lost lag, tildels bare fint kalkgrus, men med biter av den 
samme klidagtige tuf som i de foregaaende profiler. 

VI. Furutuf kompleks, 20 — 25 cm., meget lost, øverst jordagtig og sterkt 

forvitret; ellers normal furutuf. 
VII. Muldjord med Alniis-inï, 20 cm. 

Profil VIII. 
I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetufkompleks, ca. 20 cm., præcis som i profil MI. 

IV. Bladtuf, ca. 10 cm., normal. 



I92I.X0. 9- KALKTL'FSTL'DIER I GUDBRAXDSDALEX. 23 

V. Dryastufkompleks. 

A. 8 cm. ren Dryastuf som i profil \'II. 

B. 2 cm. lagdelt ovre sone som i p. \'1I. 

C. 5 cm. gronbrun, gruset 1ère, som gik over i foregaaende profils 
kalkgruslag (C), og som utvilsomt er et forvitringsprodukt herav. 

VI. Furutufkompleks. 

A. 20 cm. meget les og smuldrende mosetufagtig furutuf. 

B. 20 — 25 cm. furutuf, bestaaende av en række mindre lag: i. 4 cm. 
sprød, men tæt gul tuf, meget fin i bruddet, med faa furunaaler. 
Muligens av sedentær natur. 2. 1,5 cm. kulforende sort stripe. 
3. 2 cm. graagul jordagtig tuf. 4. 2 cm. sort kuljord. 5. graa- 
gul jordagtig tuf, ca. 10 cm. 

C. 15 cm. løs, sterkt forvitret furutuf. 
MI. Muldjord, 20 cm., med Ahiiis-tuf. 



Profil IX. 
I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetufkompleks. 

A. 7 cm. hullet slaggagtig tuf, normal. 

B. 8 cm. rødlig mosetuf, haardere. 

C. 4 — 5 cm. mosetuf med bladrester [bjcrk, asp, Sa/iccs). 

D. 3 — 4 cm. undre Dryashorisont, med mængdevis av Dryasblader 
og flere stammer samt en frugtstand, bladfragment av åjerk og 
Sa/i'x sp. og foliøs levermos (cfr. Pellia sp.). 

IV. Bladtuf, ca. 8 cm., meget vakkert planskifrig, typisk. 
V. Dryastufkompleks. 

A. 3 cm. typisk Dr3-astuf, kaotisk. 

B. 2 — 3 cm. laget sone med fnni, hjcrk, Dryasblader. 

C. 3 cm. løst lag, med smaa tufstykker indeholdende Dryas og 
furunaaler. 

C. 1,5 cm. hvitagtig 1ère. 
\'I. Furutufkompleks. 

A. ca. 3 cm. mosetuf med furunaaler. 

B. 5 — 7 cm. graagul, tæt, men fin og sprød furutuf, muligens seden- 
tær, med forholdsvis faa furunaaler, blader av Vacciniuiu vitis iilara, 
ledst3'kker av Equisetimi hieuiale og en kur\- av Cirsiuni hetero- 
phylhmi, samt blader av en mindre urt (utydelige). 

C. 30 — 35 cm. temmelig løs, men tydelig skiktet furutuf, delvis tuf- 
jord, med mængdevis av furunaaler, bark, flere konglcr samt 
tyttebctrbladcr. 

MI. Muldjord, ca. 20 cm., med Ahius-KxA. 



24 KOM' NOI<l)IIA<,K\. M.-N. Kl. 

I' ro fil X. 

Delte piolil ( i- (let forste som Miv iiiidersokt (1918), og det avslutter 
venstre serie op mol stenroiseii. 
I. Blaa 1ère, nkjeiidl ma/gligliet. 
II. Rod 1ère, 4 cm. 

III. Mosetufkompleks, ganske komplicert sammensal: 

A. 5 — 6 cm. luillel graagrøn mosetuf, normal, med lifjuisctimi varie- 
gatmil som vanlig. 

B. 2 — 3 cm. rødlig, haardere mosetuf. 

C. I cm. smal, red 1ère med smaa sandkorn, ligner fuldstændig den 
røde 1ère paa hunden, og er antageligvis bare en saadan som 
bækken har gravet ut litt høiere oppe og ganske lokalt skyllet 
ut over mosetuffen nedenfor. 

D. 5 — 6 cm. mosetuf, opiil mere sedimentær. 

E. 4 cm. skifrig, fossilfattig tuf, halvt sedentær, ialfald nedtil mose- 
tufagtig, med gronbrogete lag avvekslende med gulagtige; bjcrk 
(blader, rakleskjæl, Saliccs (bladfragmenter), Equisetun varicgatiim. 

F. 3 cm. opsmuldret mosetuf, øverst hvitagtig kalklere. 

G. Undre Dryashorisont, 4 — 5 cm., med mængdevis av Z)rvrtsblader, 
stammer, en blomst i frugtstadiet, Salix arbiisaila (i blad), bjcrk 
(flere store blader). 

IV. Bladtuf, 8 — 10 cm., skifrig, normal, med bjcrk, asp. Saliccs (bl. a. en 
mindre form, muligens S. pliylicifolia], desuten et par store Dryasblader. 

V. Dryastuf kompleks : 

A. 2 cm. løs, smuldrende Dryastuf, normal. 

B. 2 cm. graabrun jordagtig stripe, tildels sortfarvet av kul, over 
mot foregaaende profil mere hvitagtig 1ère, som var temmelig 
seig. Et par smaa tuf biter viste j'uniiiaalcr (tversnit). 

VI. Furutuf kompleks: 

A. 4 — 5 cm. graagrøn furutuf med mængder av barnaaler, desuten 
et par blader av Vacciiiiimi iiliginositm. Paa ett sled litt mosetuf. 

B. 4 — 5 cm. graagul, skjør og fin furutuf, tildels med avvekslende 
graagrønne og mere gulgraa lag, indeholdende huller efter kvister, 
furunaalcr, Cirsiiiiii lictcropliyllimi (4 kurver, blomsterstilker, store 

bladfragmenter), Toficldia palustris (flere vifleformige bladrosetter). 
Fragår ia vcsca (i blad), Pyrola minor (mange blader, samt en hel 
plante med paasittende blader), Equisctinn sp., moser, samt avtryk 
av en lavart, vistnok Parmclia phy sodes. Ulpræget sedentær 
horisont. 

B. I — 2 cm. kulstripe. 

C. 30 — 40 cm. furutuf, noget varierende, med /}/;7/naaler, kvister, 
bark, kongler i stor mængde, /v//r/wrblader, samt J^acciiiiiim iili- 
giitosum (flere blader). 

VII. Muldjord, 20 — 25 cm., med store biter av Almis-tnï og mindre stener. 



192I.N0. 9- KALKTUFSTl'DIER I GUDBRANDSDALEN. 25 

Profilet udnierker sig ved bladtuftens ringe mægtighet; den holder her 
paa at kile ut og viser flere avvikende træk {(. eks. sparsomme Dryasblader). 
Den undre Dryashorisont under bladtuften opnaar her sin storste mægtig- 
het, mens Dryastuften gir indtryk a\- at være sterkt forvitret og reducert. 
Endelig byr furutuften paa flere interessante træk, specielt den sedentære 
horisont med urteagtige planter, hvorav flere aldrig før er fundet fossile i 
kvartære avleiringer. 

Hoirr srric. 

Denne serie, som leper paralelt med den venstre, men paa grøftens 
anden (.i: nordre) side, viser forst fire profiler som stemmer helt overens 
med venstre-serien, men avsluttes opad med 4 utkilingsprofiler. Blad- 
tuffen og Dry as tuffen kiler nemlig ut mot nord og nordvest 
i forhold til grøftens midtlinje; men utkilingssonen ligger nærmere 
midtlinjen i tuftens øvre del end længer nede i bakken, hvor først tver- 
profilet overskjærer utkilingssonen (cfr. dette). Man kan ogsaa uttrvkke 
det samme paa den maate, at bladtuften og Dryastuften nu for tiden optrær 
i en skaalformig, men uregelmæssig omgrænset fordypning, men forsvinder 
til sidene, hvor imidlertid baade den underliggende mosetui og de over- 
liggende tuflag (furutuf og Alnus-tuf ) fortsætter, tildels med stor mægtighet. 
Bladtuffens „centrum" ligger der hvor Blytt optok sit profil, altsaa like 
ovenfor kjøreveien; her opnaar den sin storste tykkelse. 

hnidlertid er der flere sikre beviser paa at bladtuften engang i tiden 
(o : da dens dannelsestid \ar avsluttet) har overdækket et meget 
større areal end den nu gjør, men at den senere igjen er vitret 
ned i stor utstrækning, forinden furutuften begyndte at dannes. Jeg kommer 
tilbake til dette betydningsfulde punkt under beskrivelsen av tverprofilet, 
som er meget oplysende i saa henseende. 

Profil XI. 

I. Bl aa 1ère, ukjendt mægtighet. 
II. Rød 1ère, ca. 3 cm. 

III. Mosetufkompleks: 

A. 10 cm. graagrøn, hullet, slagg eller koralagtig mosetuf, brun- 
farvet paa undersiden, med avtryk og hulheter efter Eqiiisrfnni 
variegatiiDi. 

B. 10 cm. løs mosetuf, brytes let istykker. 

C. 5 cm. mosetuf med blader av Drvas; undre Dryashorisont. Hænger 
sammen med underliggende saavelsom med overliggende lag. 

IV. Bladtuf, ca. 20 cm., planskifrig og med mængder av løvblader (Bi'titla 
odorata, Popiilits tiriiiiila, Saliccs). 

\. Dryastuf. Av denne fandtes her bare spor aller øverst i bladtuftens 
o\erkant (Dfyas, Sa/ix orbuscnla (mange blader og 2 Ç-rakler), Pi/iiis 



26 KOI.K NOKlJllAriF.N. M.-X. Kl. 

silvi'stris (naalcrl, Ihhihi (irdorahi (hlaflfraf^^nicntcr), l'opnliis tnnnila 
(do.) samt (Icrc l\ ilsoiniiK- avtryk). 
VI. l''iinitiir, ca. 25 (III., Icmmcli^ k)s og opstykket. Kolossale ma.s.ser 
av Piiiiis si7vrs/ns-n:i:dcr r)g andre rester. I midten antydning til en 
mosetufagtig horisont med lovbladfragiiienter. Øverst muldblandet 
forvitret tuf. 
VII. Muldjord med rotter av den nuværende vegetation, 20 cm., og med 
med smaa biter av forvitret yl//nts-U\( med blader av Almis incana. 

Profil XII. 
I. Blaa 1ère. 
II. Rod len-, normal. 

III. Mosetufkompleks: 

A. ro cm. slaggagtig, hullet mosetuf, normal. 

B. 20 cm. lysere mosetuf. 

IV. Bladtuf, ca. 25 cm., meget lost skifrig, tildels bladet eller skjællet; 

typisk. 

V. Dryastufkompleks: 

A. 6 cm. kaotisk Dryastuf med masser av blader og stammer, des- 
uten flere furunaalcr og fragmenter av bjerkcbladcr. Lokalt var 
den fin og med klidagtig struktur (kanske fin mosetuf). 

B. 2 cm. smal lersone op mot furutuffen. 

VI. P\irutuf, her utviklet som en stor kompakt tuf blok, 20 — 25 cm. tyk, 
som syntes at ha glidd litt fremover paa grund av gravningen neden- 
for, som ophævet mottrykket fra denne kant. Overst meget forvitret. 
VII. Muldjord med A/niis-tui' rikelig, 30 — 40 cm. 

I dette profil mangler den undre Dryashorisont, i likhet med hvad til- 
fældet var i Blytts profiler. En del av profilet sees paa fig. 7. 

Profil XIII. 
I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetufkompleks : 

A. 5 cm. graagrøn slagg-tuf med Eqiiisf/iini varicgatuiii; typisk. 

B. 15 cm. mosetuf, i den øvre del med grønbrogete lag indeholdende 
bjerkcbladcr, Srr/Z-v-blader og Dry as meget sparsomt (kun 2 blader). 

IV. Bladtuf, 25 cm., overordentlig vakker og fossilrik, nedtil smuldrende, 
finskifrig. 

V. Dryastufkompleks: 

A. 5 — 8 cm. ren Dryastuf med masser av Dryas-restQV, Salix reticu- 
lata (2 blader), Salix /icrbacca (i blad), S. arbiisciila (blader), furii- 
iiaalcr (sparsomt, kun et par naaler set), desuten iiiscktrcstcr 
(bakkropsringer) og flere utydelige avtryk. 



1 92 1. Xo. 9. 



KALKTUFSTUDIER I GUDBRANDSDALEX. 



B. 2 cm. lagdelt, mere sedimentær sone med hjcrkcbladcr, asp, Sa/icrs, 
hvoriblandt S. arbitsciila og kanske S. phylicifolia, furuiiaalcr 
(sparsomt), /)rvY7sblader (sparsomt). 

C. 1,5 — 2 cm. gruset kalklere. 
W. Furutufkompleks. 

A. 4 cm. løs furutuf med masser av naaler og smaa tyttebærblader. 

B. 10 cm. sinteragtig, les, fossiltom tuf (kun med træbiter). 

C. 30 cm. furutuf, temmelig los og overst sterkt forvitret. Nedtil 
litt mosetuf. 

\\\. Muldjord, 15 cm. 

Profilet er ganske inter- 
essant derved, at det viser den 
samme antydning til tredeling- 
av Drvastufkomplekset som vi 
saa i venstre-seriens centrale 
profiler. Dryastufien var her 
særlig artsrik og pragtfuldt 
utviklet. 

Profil XIV. 

I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetufkompleks. 

A. 10 cm. slaggagtig 
mosetuf, normal. 

B. 4 cm. hvit 1ère. 

C. 10 cm. vakker mo- 
setuf 

IV. Blad tuf, 25 cm., meget 
vakker; typisk. 

V. Dry astuf kompleks. , 

A. 5 — 7 cm. kaotisk Dryastuf med mængder av Z>r\'<75rester (blader, 
stammer, en knop eller blomst i begyndende frugtstadium, „ana- 
tomiske" tversnit), Salix reticulata (3 — 4 blader), 5. arbusciila, bjcrk, 
asp, Eqnisctum variegatiim, fnrunaaler (meget sparsomt). Nedtil 
litt løs, fin mosetuf 

B. 2 — 3 cm. lagdelt sone med bjcrk, asp, Salix capræa, S. reticulata 
(6 blader), S. arbuscula, S. cfr. phylicifolia, Dryas, Pimts silvestris 
(flere store naaler). 

C. r,5 cm. hvitagtig til chokoladefarvet tuflere (gruset kalklere). 
W. Furutufkompleks. 

A. 2 cm. utpræget sone med smuldrende tuf, tætpakket av furu- 
naaler og med smaa fine blader av Vacciiiiuin vitis idera samt 
fragmenter av 5"^//.v-blader. 




Fig. 7. Profil XII. Meiselen staar nederst tast i 
den rode 1ère og markerer mosetufkompleksets t^'k- 
kelse. B — B = bladtuffen, hvis skifrighet sees ret 
over meiselen. Papirkorsene a\Tnerker Dryastuf- 
komplekset. Øverst den kompakte furutuf. Juli 19 19. 
Nordhagen fot. 



28 KOLK ,\OI<l)||Af,r.N. M.-N. Kl. 

B. 2 cm. Icrlignendc stripe. 

C. 7 — 8 cm. livitagtig sinter-tuf, fossilfattig, men mef] /?/;'/maaler, til- 
dels meget los og opsprukket. 

I). 15 em. mosetiifagtig fiirutiit. 

E. 30 cm. kompakt furutut", tre piater over hinanrle-n, øverst sterkt 
forvitret. Rik paa fururestcr. 
\'ll. Miildjorcl, 20 cm. med mange smaa biter av A/inis-tuf allerøverst. 

Profilet ei- meget fuldstændig og viser en mægtig furutuf med ganske 
komplicert bygning. 

Profil XV. 
I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetuf, 10 cm. hullet slaggagtig mosetuf, typisk. 

IV. Manuler 



- cfr. næste las:. 
V. Mangler 

VI. Furutuf kompleks. 

A. 10 — 15 cm. los sone, tildels brunlig 1ère med smaa tufstykker 
fuldproppet med furunaaler og med smaa tyttebærblader, desuten 
mosetufrester. Svarer utvilsomt til den underste del av furu- 
tuffen i foregaaende profil, som her synes at være opknust under 
vegten av den overliggende mægtige furutuf og trykket ned i en 
leragtig masse (forvitringsprodukt av bladtufFen). 

B. 30 — 40 cm. haard, sammenhængende furutuf, fossilrik, meget 
kompakt og tung. 

C. 25 cm. forvitret los furutuf, tildels tufjord. 

VII. Muldjord, 30 cm., med A///iis-tu{ og andre utydelige tufrester. 

Vi moter her det forste utkilingsprofil hvor baade bladtufifen og 
Dryastuffen er forsvundet; antageligvis er den leragtige substans et for- 
vitringsprodukt herav. Profilet kan meget let forfolges over til grøftens 
venstre side, og her kommer baade bladtuf og Dryastuf til som en kile 
fra venstre, men forsvinder indunder furutuffen tilhoire i 
grøften, saaledes som nærværende profil viser. 

Profil XVI. 
I. Blaa 1ère. 
II. Rod 1ère. 

III. Mosetuf, 5 cm. hullet slaggtuf, typisk. 

IV. 1,5 cm. hvitt kalkgrus. 

V. Kaotisk kompleks, 20 cm. ; brungron -tuflere med tufstykker, inde- 
holdende D)'yashladcr, funiiiaalcr; et stykke indeholdt bladtnfrestcr 
øverst og mosetuf nederst. Overst furutufstykker. 
VI. Furutufkompleks, 30 — 35 cm., overst løsere og forvitret, nedti! mere 
kompakt og haard tuf. 



1921. No. 9. 



KALKTUFSTL'DIER I GL'DBRAXDSDALEX. 



29 



VII. Muldjord med tufbiter, 20 cm. 

Dette profil er meget instruktivt, idet det viser os de sidste smuldrende 
rester av den utkilende bladtuf og Dryastuffen. 



Profil XVII. 

I. Blaa 1ère. 
II. Rød 1ère. 
III. Mosetufkompleks. 

A. 3 cm. gronlig, gruset stripe. 

B. 5 cm. hullet mosetuf, typisk. 

C. 5 — 7 cm. forvitret rodlighvit mosetuf; tildels gruset, hvit kalklere. 
Brungron, gruset tuflere, 8 — 10 cm., med /)n'<7*7»/'stykker (ogsaa ////7/- 
iiaalcr i disse) med mængder av Dryasrester, fragmenter av bjerk og asp 
(sparsomt)^ snegler. Furunaalene var temmelig tynde og fine. 
Her kom forst en 1er- rpncmxT 



IV. 

VI. 



VII. 



stripe, som nedtil var 
sterkt rød og optil 
blaalig, 3 — 5 cm. Der- 
efter fulgte 30 cm. furutuf, 
som i foregaaende profil, 
med /iii'iirester (naaler, 
korigler), en 5'(7//.v-rakle 
o. a. 

Muldjord. 20 cm. med 
^/;///s-tuf. 



AV^ÆTNINQ 




Fie:. 8 



Skematisk tegning av en stratigrafisk in- 
version. B = blaa 1ère. R = rod 1ère. 



Dette profil supplerer foregaaende paa en udmerket maate. Imidlertid 
er lerstripen \l (under furutufi:ens nederste del) ganske gaadefuld. Vi har 
her nemlig akkurat det omvendte av hvad vi finder paa bunden av tuf- 
fen, nemlig rød 1ère og derefter blaa. Dette fænomen kan ikke forklares 
paa anden maate end at bækken dengang furutuften skulde begynde at 
dannes, maa ha gravet sig ned til de underste lerlag paa et sted hoiere 
oppe i bakken (hvor kanske ogsaa mosetuften var vitret nedl. Den 
har da selvfølgelig forst erode rt i den rode 1ère som ligger 
øverst, og atter avsat denne længer nede, derefter i den underliggende 
blaa og placert denne oppaa den røde længer nede i bakken. Dog S3-nes 
baade dette erosionsfænomen og avsætningsfænomen at ha været ganske 
lokalt (vi moter det atter i næste profil). — Denne stratigrafiske inver- 
sion tyder paa ganske sterk vandforing i bækken i begyndelsen av furu- 
tuftens tid (cfr. fig. 81. 

Dette profil er forovrig et ukilingsprofil av det under venstre serie 
omtalte profil M, som var optat midt i groften. Fotografiet (fig. 9) med 
forklaring viser dette. 



30 



HOl.y NfJKDIIAf.KN. 



M.-X. Kl. 



I'iofil X\'III. 
P>l:ia Icrc. 

Kim ;iiU\(l(l 111(1 liorisfjiit, soin røde flaninier i overkanten av den 
hla.i Icrc. 
Mosetufkoinpleks. 
A. 5 cm. gronli,t4' iiiiilct moscliil', normal. 

1>. 5 8 cm. mosetuf- 

kalkgru.s, tildels ler- 
agtig. 
I\'. 1 5 cm. brun, gruset tuflere 
V. (med I )r3'astufstykker. 
VI. A. 5 — 6 cm. 1ère; nederst 
I — 2 cm. skrikende 
rød 1ère, øverst 4 cm. 
blaalig, sandet 1ère 
(samme fænomen som 
i foregaaende profil). 

B. 10 — 12 cm. kalkgrus, 
hvitagtig. 

C. 30 cm. furutuf, tem- 
melig løs og forvitret. 

VII. Muldjord, 20 cm., med 
tufbiter. 
Profilet stemmer helt med 
foregaaende; det danner dets 

billedet), ni = mosetuft"en, IV = bladtuffen, som kiler forlængelse opad Og grænser 

op mot stenroisen, avslutter 

altsaa høire serie. 

Bladtuffen kiler altsaa ut 

opad bakken og over til høire; 

i det øverste profil i venstre 
serie fandtes den fremdeles, men var her, som vi har set, av ringe mæg- 
tighet (8 — 10 cm.) og kiler vel ogsaa ut opunder stenroisen paa dette sted. 




Fig. 9. Billedet viser en del av prolil VI tilvenstre 
og dets utkiling over mot profil XVII tilhoire. 1 = 
blaa 1ère, 11 = rod 1ère (hvis morke farve sees paa 



ut paa det sted som er betegnet med > . Papir- 
korset markerer ogsaa bladtuffen. V = Dryastuffen; 
tilhøire er denne ogsaa avmerket med et kors. VI 
= furutuffen. Lagene skraaner sterkt mot iagttageren. 
Juli 1919. Nordhagen fot. 



Tvcrsej'icn. 

Denne begynder med profil IV i venstre serie og passerer profil XV 
i høire serie, hvortil henvises. Derefter følger endda længer tilvenstre 
(a: nordligere) følgende profil. 



Profil XIX. 
I. Blaa 1ère. 
II. Rød 1ère. 
III. Mosetuf, 5 cm. hullet slaggagtig tuf, normal. 



ig2I.No. 9- KALKfUFSTUDIER 1 GUDBRAXDSDALEN. 3I 

IV. I mangler. Isteden optrær lo cm. brunlig olivenfarvet tiißcrc med smaa 
V. jmosetufbiter. 
VI. Furutufkompleks; mægtig og a\-vekslende. 

A. 8 — 10 cm. mosetuf med fiiniiiaalcr og kong/er. 

B. 30 cm. haard, tildels flinthaard turutuf med mængdevis av furu- 
rester, desuten J^nxiniiini vitis idæa, Saliccs o. a. Især nedtil 
tydelig lagdelt og meget haard. 

C. 20 cm. los jordagtig tuf med et mere sammenhængende, spredt 
lag, som mindet om den sedentære sone i profil IX og X. 

VII. Muldjord, 20 cm. med Almis-Xwi i store st3^kker, desuten et stykke 
av en bladrik fuiaituf, som indeholdt talrike fiiriihar, blader av asp, 
bjcrk (vistnok B. vcrntcosn), Salix sp., Sorbits Aitciiparia (i blad- 
finne), tyttebær, samt nogen valkformige, glatte forhoininger, som muli- 
gens skriver sig fra blaagronne alger (cyanofycé-tuf). 
Dette profil viser de samme forhold som profil X\', med en utkilende 

bladtuf F u r u t u i'i^t n d e r i m o t o p n a a r her omtrent sin største 

mægtighet (cfr. næste profil). 

Profil XX. 

Danner en fortsættelse av foregaaende over mot venstre. 
I. Blaa 1ère. 
II. Rod 1ère, typisk. 

III. Mosetuf, 20 cm., slaggagtig, med Equisrtiini varicgatiim og thallose 
levermoser (Marcliaiitia polyniorplia); normal. 

Brun, jordagtig rand, 3 cm. med smaa flate bladtitffragiiiciitcr (Bctitla 
odorata, diverse blader og rakleskjæl), som var sterkt brunfarvet paa 
o\ersiden. Derover 5 cm. gronlig sandblandet tuflere. 
VI. Furutufkompleks, tilsammen ca. 70 cm. 

A. 10 cm. los furutuf, vistnok mosetuf, med sparsomme furunaaler. 

B. 10 cm. grønbroget, valket furutuf med masser av fururester. 

C. 15 cm. meget haard mosetufagtig furutuf med sparsom yw;7/, Se? //.v- 
blader samt /or/Vc/rAfragmenter. Tildels meget kompakt og plate- 
formig. 

D. 10 cm. furutuf, normal type, med et eneste kaos av naaler og 
andre rester av fiint. 

E. 20 cm. løs, forvitret og jordblandet furutuf. 
VII. Muldjord, 20 cm., med smaa tufbiter. 

Profilet er i flere henseender et av de interessanteste i hele tuften. 
Blad tufrestene i den brunlige jordstripe laa meget smukt paa rad 
bortover, i naturlig stilling (skifrigheten orientert paralel med bakkens 
overflate, som her næsten var horisontal over en kortere strækning). De 
er utvilsomt de sidste rester av en tidligere bladtuf, som er vitret ned 
forinden furu tuffens dannelse blev paa begyndt. Den grønlige 
1ère er ogsaa tydelig et forvitringsprodukt. 



IV. 



32 



KOl.K .\(M<l)llA',i:.\. 



M.-N. Kl. 



I'^iiniUitrin (Iciiiiiot li;ir |);i;i (l<tte sted sit maksimum hvad mægtigheten 
anganr og ikIiik rkcr sig ved en række mfd stratigrafisk forskjelligartede 
horisonlci-. l'ig. lo viser et fotografi av pmlilct. 



Profil XXI. 

Dette blev optat h(»sten 1920 og ligger hele 9 m. nordenfor profil XX. 
Mellem dem ligger der, saaledes som antydet paa tverprofilet (fig. 12) en 




Fig. 10. Profil XX. Nederst ved vandpytten sees mosetuft'en I III I. Derpaa tblger den 

brune rand (IV) med bladtufrestene, og den grønlige tiitlere (V). Øverst sees furutuffen 

(VI), som fortsætter ovenfor billedets rand. Juli 1919. Nordhagen fot. 

stenrøis og bjerkekrat. Overensstemmelsen mellem dette profil og de grav- 
ninger som jeg foretok nedenfor kjøreveien, er ganske slaaende (cfr. 
nedenfor), og de er alle sammen meget forskjellige fra tuftens øvrige 
profiler derved, at ^/;///s-tuffen her er paafaldende mægtig ut vik- 
let, mens den jo ellers kun er tilstede i form av mindre stykker øverst i 
jordlaget. Rækkefølgen i profil XXI var saaledes: 
I. Store stener i bunden. 

II. Mosetuf, ca. 10 cm., nedtil rustfarvet paa undersiden og slaggagtig, 

graagrøn med Eqiiisctiini varifgatiim. Den stemmer helt overens med 

bundlaget ellers i tuften. O verst mere rodlighvit og tæt mosetuf 

III. Kaotisk kompleks, bestaaende av graaagtig tuljord med talrike lose 

tufstykker, 8 — 10 cm. 



I92I.N0. 9- KALKTUFSTUDIER I GUDBRAXDSDALEN. 33 

A. Nederst mot mosetuften fandtes smaa bladtufstykkcr, sterkt forvit- 
ret; t^'piske. 

B. Høiere oppe fandtes storre og mindre stykker (op til 20x20x7 
cm.) a\- funituf, typisk, med masser av hor, mange store kongler, 
tyttcbcerbladcr, dverggrener av asp; de var sterkt forvitret og brune 
i overflaten (planche IV, fig. il. 

C. Jordagtig stripe, ca. 3 cm., med biter av trækul. 
IV. Alnustuf kompleks, 90 — 95 cm. 

A. 25 cm. blekt rodlig tutjord (bleke) med en del mindre fastere 
stykker, som let knuses mellem fingrene. De bestod vistnok av 
mosetuf med en del utydelige planteavtryk (græsstraa). Øverst 
fandtes en svakt antydet kulstripe. 

B. 8 — 10 cm. mosetuflag, meget spredt og porøst, med blader av 
Aluns iiicmia, Salix sp., dverggrener, vedbiter, ledstykker av 
Eqiiisetum liiernalc. I sin o\Te del \-ar dette lag sterkt rustfarvet, 
særlig i profilets hoire del, mere utN'delig i den venstre del, hvor 
de overliggende lag var haardere. Rustsonen bør muligens tolkes 
som et podsolfænomen (cfr. det følgende). 20 — 25 cm. haardere 
mosetuf fulgte ovenpaa rusthorisonten, uten avbrytelse i lagræk- 
ken. Denne tuf var grovere og mere kaotisk end foregaaende 
og stemmer helt overens med Alnus-tuf fra muldlaget i andre 
profiler; den var temmelig fossilfattig [Aliiits iiicaiia, Salix sp., 
græsblader, moser); enkelte partier sinteragtige. 

C. 20 — 25 cm. løs, blekeagtig tutjord med haardere biter, graalig- 
hvit av farve, øverst meget hvit og fin. 

D. 12 cm. muldblandet tuljord med haardere biter. 
V. Muldjord med rotter, 5 cm. 

Alnustuffens lag viste et fald paa ca. 30" mot ost-nordost og ligger, som 
fig. 12 viser, i en slags mulde ovenpaa den forvitrede furutuf. 

Profilets allerunderste del stemmer helt overens med tuffens øvrige 
profiler, og likesom i foregaaende profil træffer vi her restene av en forvitret 
bl ad tuf oppaa mosetuften. Men, man kunde fristes til at si, det sensa- 
tionelle moment ligger i den forvitrede furutuf som her møter os, og 
som aabenbarer en diskord an s mellem fu ru tuffen og Alnu s tuf- 
fen. Dette fænomen kunde man forøvrig slutte sig til paa forhaand, da 
restene av Alnustuflen i de øvrige profiler altid ligger ovenpaa furutuften, 
og fordi der aldrig er iagttat nogen overgang mellem furutuf og Alnustuf, 
som har helt forskjellig struktur, farve og fossilindhold. 

Men nærværende profil viser til evidens at paa dette sted var furu- 
tuften vitret ned til et lost kaos av muldblandet tuf forinden Alnustuften 
begyndte at dannes. Antageligvis har furutuften paa dette sted været mindre 
mægtig end i de foregaaende profiler i tverserien. Flere ting tyder paa 
at foregaaende profil markerer furutuft'ens maksimum, og at den derfra har 
skraanet noget til begge kanter (baade nordover og sydover). Dog 

Vid.-Selsk. Skr. I. M.-X. Kl. 1921. No. 9. 3 



34 i<r)i.i .\f>Ki)iiAf.(:N. M.-X. Kl. 

er det ikke niulij^ at rekonstruere furutuffens oprindelige utseende. Vi har 
for set at den i saagodtsom alle profiler er meget los og forvitret optil, 
og antageligvis har den overalt været mægtigere end nu, selv om dette 
ikkf kan siges med hesteintlut. 

Txcrprolikt viser os saaledes hvad jeg allerede indledningsmæssig 
ticinlurvct, at i Lcine-tuffen har de forskjellige horisonter nu for tiden 
sit maksiniiini paa helt forskjellige steder. Bladtuffens ligger i en mulde 
længst syd ved bækken, furutuffens ligger litt nordligere og Alnus-tuffens 
maksimum endda længer mot nord. Imidlertid vil de folgende profiler 
nedenfor veien vise at Alnusiuffcn ogsaa har et maksimum længer nede i 
bakken i en lignende mulde, hvor ogsaa furutuffen er heist reducert. 

Profilet er videre meget oplysende om selve y^/«;/s-tuflfen. Vi ser at 
den gjennemgaaende har en ytterst les, tildels blekeagtig karakter, og at 
i\vn i \ertikal retning opviser en række forskjellige lag, videre at den er 
meget mægtig sammenlignet med tuffens øvrige hovedlag. Det er sikker- 
lig oretuffens løse, smuldrende natur som er hovedaarsaken til at den gjen- 
nemgaaende er meget sterkt reducert over store arealer paa tuffindestedet. 

En speciel omtale fortjener den rustfarvete stripe indenfor mosetuflaget. 
Jeg har ovenfor antydet at dette fænomen muligens er av sekundær natur, 
en utfældning som har foregaat efter Alnustuffens dannelsesperiode. Som 
nævnt var den meget utpræget i profilets venstre del, men over til høire 
blev den mere utvisket; her var den overliggende tuf temmelig haard. 
Hadde vi her hat en autokton jernholdig tufhorisont, burde man ha ventet 
at den var lik overalt. Det er derfor mulig at rusthorisonten er et saa- 
kaldt podsolfænomen, og motsvarer den anrikningshorisont av jernforbin- 
delser som optrær i en mængde jordbundsformer under en overliggende 
saakaldt blekjord. Podsolprofil utvikles bedst i granskoger med tykt lag av 
raahumus øverst (Tamm 1920 1. c), hnidlertid pleier et kalkholdig under- 
lag, særlig hvor det er skraanende, at fremkalde en helt anden jordbunds- 
type Uten tydelig podsolering (Hesselmann 1917 1. c. p. 397 — 411). Som 
substrat for vegetationen betragtet, er jo en kalktuftbrekomst, specielt under 
perioder hvor avsætningen paagaar, meget fugtig, og man skulde ogsaa av 
denne grund ikke vente at finde podsolering; en saadan proces synes nemlig, 
efter hvad man vet, ikke at foregaa paa meget fugtig bund (cfr. Tamm 1. c. 
p. 55). Den omtalte rusthorisont i Alnustuffen ligger imidlertid netop inden- 
for en del av Leinetuffen som nu for tiden er meget tor (ca. 10 m. fjernet 
fra det nuværende bækkeleie), saa av denne grund var podsolering nok mulig. 
Dog vover jeg ikke at uttale noget herom med sikkerhet, men noier mig 
med at gjore opmerksom paa forholdet. 



ICSI. No. 9. KALKTUFSTUDIER I GL'DBRAXDSDALEX. 35 

Profiler nedenfor ve i en. 

Under selve kjoreveien og i dens nærmeste omgivelser var det selv- 
sagt umulig at foreta nogen gravninger (bl. a. paa grund av bækkens mulige 
ødelæggelser hvis den kom i et nyt leiel. Mit forste profil nedenfor veien 
ligger ca. 2 m. fra veikanten. 

Profil XXII. 
I. Blaa 1ère. 
II. Rød 1ère. 

III. Mosetufkompleks, tilsammen 20 cm., utviklet som en sammenhængende 

bænk, som nedtil var hullet og slagget ( typisk I. Overst fandtes en 
sone med talrike blader av Salix arbusciila, bladfragmenter av Bctiila 
odorata samt en 9-rakle av samme, Eqiiisetitni variegatuin. Svarer 
Uten tvil til den undre Dryashorisont i de øvrige profiler, men 
mangler altsaa Dryas. 

IV. Bladtuf, 25 — 30 cm., normal, men med talrike 5"(7//.v-blader foruten 
hjerk (blader, rakleskjæl) og asp. \'akkert skifrig, optil løs. 



V. 
VI. 



Derefter fulgte en 10 — 15 cm. mægtig løs, kaotisk sone med 



A. Bladtuf biter nederst (i bladtuffens overkant). 

B. Et vakkert stykke Dr3-astuf, meget forvitret, men med 15 bladav- 
tryk paa overflaten, desuten en stammedel og et bjerkeblad. 

C. Forvitrede furutufstykker med masser av naaler samt tyttebær- 
blader; t3-pisk. 

VII. Alnustuf kompleks, 15 — 20 cm., meget løst, tildels jordagtig, med 
større og mindre fastere stykker indeholdender ^/«//s-blader rikelig; 
typisk. 

VIII. Muldjord, 30 cm. nedtil med Aliuts-iuï, hvorav flere store stykker. 

Dette eiendommelige profil ligner i flere henseender meget det nord- 
ligste i tverserien. Bl. a. er furutuften her ogsaa tydelig vitret ned, ja 
vitringen har øiensynlig grepet endda dypere om sig, helt ned i bladtuflfens 
øvre del. Dryastufst3'kket er specielt værdifuldt; det viser nemlig at for- 
vitringsproduktene ligger ganske i sin naturlige orden paa hinanden, og 
vidner om at her ogsaa har været en veritabel Dryastuf paa stedet, saa- 
ledes som høiere oppe i tuffen. Endelig er ^///^/5-tuflfen ganske godt 
opbevaret. 

Profil XXIII. 

Optat ca. 3 m. nedenfor foregaaende. 

j Runde stener i grøftens bund. 

III. Mosetufkompleks. 

A. 15 cm. normal mosetuf med Eqiiisetiun varie gatiini. 



36 



KOI-K S(Jli\)llA(,ES. 



M.-N. Kl. 



B. 5 cm. grønbroget horisont med Sdli.x (whiiscnld, ^rcesstraa etc., 
noiagtig srnn i foregaacnde profil. 
JV. P>lnfltiif", en. 1 5 <iii. med massevis av .SV7//A--ljlad(r, vistnok mest S. ca- 

piiid, ellers l\'|)isk. ()|)lil los og forv'itret. 
V. I Los sone, J5 cm., bestaaende av kul og muldblandet tufjord med tal- 
IV. I rikc i'urutulstykker (7 8 cm. tykke), desuten biter av en merkelig kornet, 

sanimenkittet grustuf uten fossiler. Nederst plateformige bladtufbiter. 




■Mli lLliili, j£ j 



Moræne -Jere 

Hod 1ère, 

under Tno^etuffen 

Moset ufbcVnk 



"Undre Dryasliorizont, ^^3 rurutiifbcenlc, 
Ovenpaa mostituj-'fen w^^^i Tilhoirs forvitret 



2 m 



Bladiufbænk, 
"tilhöire forvitret 

Dryastul" 

Höidemaalestok 



Oretuf 

O vet uf rester 
i muldlaget 



3 -V s m -Lasn^'de 



Fig. II. Længdeprofil, Leine. B angin tverprofilets beliggenhet. 

VII. Alnustufkompleks, 25 — 30 cm., graagul blekeagtig tuf med haardere 
stykker, især nedad. 



VIII. 



luldjord, 30 cm., med store biter av Aliiiis-X.\\{ nederst. 



Sammenholdes di.sse to profiler, som er temmelig like, med tverseriens 
nordligste profil, gir det hele et ganske interessant billede av av Alniis- 
tuffens forekomstmaate. Den har sin største mægtighet der hvor 
fu ru tuffen er mest forvitret, og hvor der av denne grund er mulde- 
agtige fordypninger i tuffens ovre del. Det er da ganske naturlig at den 



1 92 1. No. 9. 



KALKTUFSTUDIER I GUDBRAXDSDALEX. 



37 



her er bedst bevaret ; men kanske har den ogsaa paa disse steder 
oprindelig været mægtigere end paa alle andre punkter 
inden omraadet. Hvor meget der atter er vitret bort over største- 
parten av tuftens areal, er det umulig at avgjøre; de i muldlaget fore- 
kommende tufrester er av flere t3'per (^/«»5-rikere og -fattigere, mosetuf- 
agtige, krøllet-cyanofycé-lignende o. s. v.) og er tydelige vidnesbyrd om, 
at selv paa de punkter hvor denne øverste tufhorisont nu er 3'tterst 
reducert, har der engang været flere vertikale facies. 




Moræne-lere 



iTj Rödlere.undeT ^^^^ Iryastuf, 
mosstufftn ^i=^;5&fM 



Bladtufbænk ^sset^ Furutufbænk 
Oretuf 



'ili'';i,f I HcsetufJDsenk 

i:i;L;!!i ! iii ' ' i 



under furut uf Sl^Si i muldlaget 

^ni Koidem aale stok 

S 



c- m Længde » 



Fig. 12. Tverprofil, Leine. A angir længdeprofilets beliggenhet. 



Alle disse tre perifere profiler, som ved første øiekast ser hoist para- 
doksale ut, og som for en iagttager der ellers var ubekjendt med tuftens 
stratigrafi, vilde være ganske problematiske, indordner sig i virkeligheten 
meget smukt i helhetsbilledet. Saalænge det ytterste profil i tverserien var 
ukjendt, kunde man nok staa tvilende overfor litt av hvert. Men da over- 
ensstemmelsen er saa slaaende i disse ytterpunkter av tufomraadet, baade 
nordover og østover, hvilket fig. 11 og fig. 12 tydelig viser, blir det ende- 
lige resultat av denne stratigrafiske detaljstudie, at kalk tuffen ved 
Leine har en strengt lovmæssig, klar og oversigtlig opbyg- 
ning, som ikke kan misforstaaes. 



38 KOl.l XOKDIIAf.F.N. iM.-N. Kl. 



C. Oversigt over Leinetutt'ens stratigrafi. 

Kiler at ha skildret alle de inaii^c enkeltprofîler, skal jeg nedenfor gi 
en samlet oversigt over tutten ' og diskutere de enkelte horisonter nærmere. 

De II hhia Irrr. 

Under kapitlet „isavsmeltningen i Oudbrandsdalen" vil spørsmaalet om 
morænemassenes oprindelse og alder bli nærmere berørt. En række av 
kjendsgjerninger taler for at de er avsat av en bræ fra nordvest, antage- 
ligvis under en periode med lokalglaciation i det centrale Norge (Port- 
iandia-nivsiaet). 

De sneglefornier som det lykkedes mig at fremfinde i den blaa leres 
øvre del, Vitriua pellucida (Müll.), Comilus fiilvus (MCll.) samt Limncea 
truncatiila (Müll.)^ er ikke særlig oplysende i klimatologisk henseende. Det 
er altsammen temmelig ubikvisite former, som baade kan tyde paa et 
alpint og et subalpint klima. 

Dcii rode Icrc. 

Denne horisont er saa eiendommelig at den fortjener den største op- 
merksomhet fra geologenes side. Den er i gjennemsnit 3 cm. tyk, under- 
tiden tykkere, men ofte ogsaa tyndere. I sin mest typiske form er den 
meget sterkt rod, næsten skrikende, og forbauser altid ved sin tilsynekomst 
under den laveste mosetuf. I fugtig tilstand er den noksaa plastisk, under- 
tiden sees litt tilblandet sand (blaalige kvartskorn bl. a.). I motsætning til 
den underliggende blaa leres overflatelag indeholder den nu ingen opbe- 
varte snegler. 

At den rode lerhorisont er et f o r v i t r i n g s p r o d u k t av den blaa 
morænelere, er hævet over enhver tvil. Og det skyldes kalktuffen, som har 
lagt sig som et beskyttende skjold over den, at den her er bevaret, mens 
den ellers overalt er forsvundet. 

I jordbundslæren adskiller man som bekjendt flere typer av forvitring 
efter de klimatiske forhold. I et relativt humid klima faar man jordbunds- 
typer karakterisert ved en utpræget opløsning og bort- 
førelse av opløselige f or vi tri ngs produkt er fra de øverste 
markskikt, altsaa en utvaskning og absolut ingen ophobning i overflate- 
lagene. I motsætning hertil staar den aride type, hvor der netop paa 
grund av klimatiske forhold sker en ansamling av f o r v i t r i n g s p r o- 
dukter i overflaten. Hovedforskjellen mellem disse to klimatiske typer, 



• ØYEN har i sit arbeide om kalktuflene kommentert Blytts oprindelige beskrivelse paa 
en række punkter. Imidlertid er disse supplerende oplysninger tildels noget mis\-isende, 
hvilket nærværende stratigrafiske oversigt godtgjor. De grunder sig paa studiet av 
Blytts efterlatte samlinger alene, ikke paa nye iagttagelser paa findestedet. 

2 Den sidstnævnte art er bestemt av dr. Nils Odhner, Stockholm. 



1 92 1. No. 9. 



KALKTUFSTL'DIER I GUDBRAXDSDALEX. 



39 



som selvfølgelig ikke er skarpt adskilt, ligger i forholdet mellem nedbør 
og fordampning; i et humid klima mottar marken mere vand end den av- 
gir ved fordampning, og overskuddet S3'nker ned og bortfører ustanselig 
visse forvitringsprodukter. I et arid klima derimot er der likevegt, eller 
oftest en stor deficit paa grund av den intense fordampning, hvilket be- 
virker den nævnte akkumulerende virksomhet i overflatelagene (cfr. Ramaxn 
1911, Hesselmann 1917 p. 397). 

Alting tyder paa at lerbakkene ved Leine i tidsrummet efter den sidste 
isbræs tilbakervkning har været gjenstand for en intens oksvdation og 



Tegnforklaring. 




ami 



Kartskisse over nordre Gudbrandsdalen. vidende de vigtigste forekomststrok av saltbitteriord. I .750000 

Fig. 13. Tallene angir nedborhoiden i mm. L = Leine i Kvam. iFive 191 i). 



anden kemisk omdannelse i overflaten under ganske særegne forhold. A t 
klimatet har været vtterst kontinentalt, kanske likefrem 
arid, er meget sandsynlig. 

Det er i denne forbindelse meget interessant at den ovre del av Gud- 
brandsdalen i nutiden horer med til de nedbørfattigste egne i hele det svd- 
lige Norge, med under 400 mm. aarlig nedbørhøide. De stationer som 
ligger nærmest Kvam, er Botn i Sel med 332 mm. og Steinfinsbø i He- 
dalen med 325 mm. I Skjaak gaar nedborhoiden ned til 254 mm. (Metorol. 
Institut 1. c.). 

Som vistnok Bjorlvkke først har gjort opmerksom paa, har vi i disse 
strøk ogsaa i nutiden en virkelig arid jordbundstype, nemlig „salt- 
bitterjorden" i Vaage, Skjaak, Dovre og Lesje (1911 1. cl. Denne er spe- 
cielt studert av I. Five, som har skrevet en meget interessant opsats om 
saltbitterjorden, dens utbredelse og økonomiske betydning (191 1 1. c). 



40 KOI.K NOKnilAf.KN. M.-N. Kl. 

I 

„Saltbittcret", skriver Fivk, „som viser sig paa jordoverflaten i den varme 
og torre aarstid, enten som skorpe, klumper eller kruster, hcsUiar hoved- 
sa^r/ii;' ni' i^^i/>s (svovlsm- k.ilk); iii' ii imli liolde-r ogsaa adskillig hittersalt 
(svovlsur magnesia), som gjor, at det smaker bittert. I)esuten kan det være 
smaa mængder av alka/irsuljatcr, samt en del klorider. ^^ „Saltbitteret holdes 
op/ost i jordvandet, som l)evæger sig i jordens porer. Denne bevægelse vil 
i dni vanne og regnfattige forsonnncr (tildels ogsaa utover hestparten) 
væsentlig vair i opadgaaende retning. De foran nævnte salte følger med i 
opløst tilstand helt til jordoverflaten. Naar saa vandet dunster væk ut- 
fa'ldes sallene (sallbittf rctl i en av de for næ\'nte former" (1. c. p. 23I. Ved- 
føiede kartskisse er hentet fra Fivf.s arbeide og viser saltbitterjordens ut- 
bredelse samt nedbørkurver. Kalktuffens beliggenhet . er avmerket med 
L. i kartets sydøstre hjørne. — Aarsakene til denne merkelige jordarts 
dannelse maa ifølge Five sokes i de klimatiske forhold: „et utpræget 
indlandsklinia med varm sommer og kold vinter — og Uten nedbør, specielt 
i vaarmaanederne" (1. c. j). 9, hvor en række tabeller belyser forholdene). 
BjøRLYKKK omtaler ogsaa aride „hardpan" -dannelser og saltjorder fra de 
øvre deler av vore østlandske dalfører i „Norges jordbundsprovinser og 
klimatiske hovedstrøk" ^ 

Den „fossile jordbund" under kalktuffen er efter min mening et tydelig 
og værdifuldt vidnesbyrd om at de kontinentale forhold var i endda 
høiere grad potensert i disse længst forsvundne tider. 
Kanske sterke føhnvinder kan ha bidradd til at paaskynde oksydations- 
processene. Desuten har utvilsomt den heldige eksposition og derav føl- 
gende sterke insolation og opvarmning stimulert de kemiske omsætninger. 
Imidlertid sier det sig selv at en saadan forvitringshorisont ikke opstaar i 
en haandvending, men kræver tid til sin dannelse. — Nogen anden for- 
klaring paa fænomenet kan for tiden ikke gives. Saavidt jeg vet, foreligger 
der i litteraturen ingen oplysninger om tilsvarende fænomener fra andre 
strøk i Skandinavien, som kan gi angrepspunkter for en mere indgaaende 
diskussion'-. 

Om lerbakkene ved Leine under denne periode var vegetationsklædt 
eller ikke, kan ikke siges med bestemthet, da alle tydelige organiske rester 
mangler. Undertiden synes den røde 1ère at indeholde mere brunlige par- 
tier, som muligens er destruerte levninger av plantevekst; dette kunde jo 
tyde paa en meget spredt vegetation. Et andet moment som fortjener op- 
merksomhet, er det faktum, at den underste, slaggagtige mosetufhorisont er 
brunfarvet paa undersiden og her fleresteds bærer tydelige avtryk efter 
kvister og mindre grener (Salices?). Men disse kan ogsaa forst ha ind- 



' Naturforskermotets forhandlinger 1916. Kristiania 1918, p. 522 — 523. 

2 Den rade 1ère minder baade om „terra rossa" og lateritdannelser; men da disse op- 
staar i varme klimatomraader, er det vanskelig at trække nogen mere indgaaende 
sammenligning. 



ig2I.No. 9- KALKTUFSTL'DIER I GUDBRANDSDALEX. 4I 

fundet sig paa stedet da kalktufkilden beg3'ndte at sprudle frem av 
bakken ^ 

Flere ting tyder med bestemthet paa at den røde 1ère ved Leine ikke 
er et rent lokalklimatisk fænomen, men av mere generel natur. Kalk- 
t uf fen ved Gillebu iOier længer syd iGudbrandsdalen viser 
nemlig det samme fænomen. Her er, som vi senere skal se, tuffens 
underlag meget sterkt oksydert og rikt paa jernforbindelser, hvilket bl. a. 
ogsaa har hat tilfølge at selve tuffens bundlag tildels er mørkt brunt eller 
chokoladefarvet, mens lagene høiere oppe har den vanlige graagule tuffarve 
(cfr. Oven 1920, p. 285). 

Mosctufkoiiiplcksct. 

Den første kalktufhorisont som meter os ved Leine, er en m o s e t u f. 
Og den underste del av denne, som er meget karakteristisk, gjenfindes i 
samtlige profiler, har altsaa en sammenhængende utbredelse over hele 
omraadet. 

Denne mosetuf er hullet, poros og uren, og kan betegnes baade som 
slaggagtig og koralagtig. Mosetufstrukturen er meget utvisket paa grund 
av forkalkningen, men aabenbarer sig ikke sjelden paa brudflater, hvor 
mosens blader ofte sees. Farven er i fugtig tilstand eiendommelig graa- 
grøn, og temmelig mork. Hox-edmassen utgjøres muligens av en A)}iblystc- 
giimi-zrt, men det er ikke mulig at konstatere dette med nogen sikkerhet. 
Dog kan det ikke være A. falcatitiu, da denne art, som høiere oppe i tuffen 
er almindelig, er meget grovere. Tuften viser en paafaldende likhet med 
stuffer fra subarktiske lag i tuffen ved Hemrike i Lerdala, Västergötland 
(Serxander 1916, 1. c. p. 141, Växtbiologiska Institutionens kalktufsamling). 

Som for nævnt er mosetuftens underside rødfarvet av den underlig- 
gende 1ère. Men ogsaa selve den først dannede kalktuf er fleresteds meget 
sterkt jernholdig og rødfarvet, desuten forurenset med kvartskorn. Paa 
undersiden, men ogsaa høiere oppe, er der avtr3'k og hulrum efter flere 
middels grove træpinder (muligens trærøtter), hvorav en del var i cm. i 
diameter; et grovt, men mere tvilsomt, halvcylindrisk avtryk var 3,5 cm. i 
diameter. Muligens skriver de sig fra Salix-husker, da jeg andre steder 
i tuffen har fundet utN'delige smaa SaZ/.v-blader (som minder litt om S. phy- 
licifolia eller S. arbitsatla). Ellers egner denne porøse mosetuf sig meget 
daarlig for opbevaring av blader og andre urteagtige plantedeler. — Gjennem 
næsten hele mosetuftens masse findes nogen smaa hvite, tynde rørlignende 
fragmenter, som skriver sig fra Characéer (planche I, fig. 10). De viser 
den største likhet med saavel svenske characé-holdige tuff'er som recent 
characé-kalk fra bunden av gotlandske sjøer (Växtbiologiska Institutionens i 
Uppsala samling av organogene kalkdannelser). Nogen artsbestemmelse er 



1 Kilden har tydelig\'is virket eroderende i begyndelsen og skyllet væk en del av den 
røde 1ère (cfr. mosetuffens underste, jernholdige parti) og fjernet opløselige salter i denne. 



42 KOLK NOKDIIAf.KN. M.-N. Kl. 

ikke iiiiilii;, iicppc engang nf)g(n sikker slegtsbestemmelse; dertil er frag- 
iiRiitcne altfor lite utpræget. I Skandinavien findes der baade Cham og 
Ni/r/Ia-'tWtcr som gaar ganske hoit over havet (Rai{KNI{Ohst 1. c). 

Ellers er /'.([iiiscliiiii v(\yi(\!^(üuiii Nllcrsl karaktd-islisk for denne- underste 
mosetuf. Overalt træffer man avtryk og hulheter efter dens stængler, ofte 
i tætte masser. Av landsnegler fandtes Coniiliis fulviis Müll. 

Alt i alt niaa man anta at lerbakken ved Leine ved tufavsætningens 
begyndelse har været klædt med en svulmende, meget fugtig mos matte 
indenfor det av bækken overrislede omraade. I denne mosmatte har der 
vokset Marchnuiia polymorplia (cfr. profil XX), Characeer^ og Equisetum 
variegatiiiir, nuiligens var der et krat av Saliccs (ialfald har trær eller 
busker været tilstede), eller kanske spredte vidjebusker. I Jesuten kan man 
være sikker paa at en række med græs og urter ogsaa har holdt til paa 
stedet (f. eks. subalpine Cariccs, EpilohiiiiiKwitiv o. s. v.), saaledes som til- 
fældet er med den subalpine kildevegetation i nutiden. 



Ovenpaa dette aller underste sedentære tuflag kommer der i flere pro- 
filer mere kompakte, grovere mosetuffer, som hovedsakelig synes at være 
dannet av Ainblystegiiiiii falcatimi (Blytt 1892 1. c). Andre steder findes 
mere sprøde smuldrende mosetufpartier. Farven er gjennemgaaende rodlig 
eller gulhvit i motsætning til den underste graagronne bænk (renere kalk). 
I de øverste profiler IX og X i venstre serie møter vi et ganske kompli- 
cert bygget mosetutkompleks. Her kommer nemlig øverst en halvsedimen- 
tær, av avvekslende gronlige og gule lag bestaaende sone med tydelige 
bladrester av Bctitla odorafa, Popitliis trciiiula og Saliccs, hvilket viser at 
disse trær og busker har været tilstede næsten helt fra tufavsætningens 
begyndelse. At de mangler i det allerunderste lag (cfr. dog kvistavtrykkene), 
kan som ovenfor antydet bero paa at denne sedentære, porøse horisont 
ikke egner sig for opbevaring av myke eller flate, avfaldne plantedeler. 



En ganske stor overraskelse møter os i dette sedentære mosetuf kom- 
pleks's alier øverste parti, nemlig den undre Dryashorisont. Denne 
mangler i flere profiler, specielt i den del av tuften hvor Blytt foretok 
sine undersøkeiser, hvilket jeg selv ved mine gravninger har erfaret. Og 
nedenfor kjoreveien finder vi heller ikke Djyas paa dette sted i lagræk- 
ken, men en smal horisont med anrikning av Salix arbiiscula, som dog 
utvilsomt korresponderer med den undre Dryashorisont hoiere oppe i bak- 
ken. Dette nivaa er for saavidt av en anden valør end den høiereliggende 
„Dry as tuf", som for det første er meget mægtigere og markert, og som 
desuten er av helt generel natur. 



' Eventuelt i mindre kulper med aapent vand. 1 naturparken ved Sylene findes Nitclla opaca 
paa denslags lokaliteter i 700 m. hoide. 



I92I.N0. 9- KALKTUFSTUDIER I GUDBRAXDSnALP:X. 43 

Den undre Dryashorisont er dog i flere henseender meget interessant. 
Den lærer os for det første at arter som Dryas octopctala, Salix arbuscula 
og andre Saliccs, Bctitla odorata, Eqiiisctitiii variegatuiu og PcUia sp. vokste 
ved Leine paa dette tidspunkt, altsaa en blanding av alpine og subalpine 
arter. 

Blytt fremhæver i sit arbeide ganske sterkt at de arktiske planter 
mangler i den undre hahpart av tuften, netop fordi, som vi har set, den 
undre Dryashorisont ikke findes i hans profiler. Og for ham maatte da 
de arktiske planters pludselige masseoptræden i Dry as tuffen synes end 
mere paafaldende. Dog har han tydeligvis studset over forholdet; han 
skriver nemlig om Dryas og de ovrige arktiske arter: „de mangler ikke 
blot i den overliggende furutuf, men merkelig nok ogsaa i den underlig- 
gende tuf — — — " (1892 1. c. p. 6). 

Nu vet vi imidlertid at saa ikke var tilfældet; \-i maa derimot anta at 
lerbakkene under mosetu iiç: ns tid har hat en b 1 a n d i n g a \- a 1 p i n e 
og subalpine arter. 

Imidlertid er Dryas, som jeg i det folgende kommer til at diskutere 
nærmere, en dvergbusk med xerofil b^-gning, som ynder tor bund, specielt 
løse, smuldrende skifre og kalkbund. Den undre Dryashorisont er derfor, 
saavidt jeg kan forstaa, et tegn paa en mindre oscillation i t u f k i 1- 
dens vandforing. Enkelte partier av omraadet er blit mindre fugtige 
end før, og her har bl. a. Dryas kommet ind. Dette synes specielt at ha 
været tilfældet allerøverst i venstre serien, nær kildens utspring (profil IX 
og X). Andre partier har ikke gjennemgaat nogen forandring paa dette 
tidspunkt, men opviser den samme s\ulmende, vanddrukne mosmatte som 
før. Nedenfor veien synes Salix arbuscula at ha dominert lokalt. 

Mosetufkompleksets sedentære natur potenseres saaledes opad, hvor det 
kulminerer i den undre Dryashorisont. Men derefter S3-nes tufkilden at 
ha svulmet ganske voldsomt op, og den folgende tid \'iser os utpræget 
sedimentære tuflag. 

Bladtuffcn. 

Dette utpræget sedimentære kompleks er saa karakteristisk at man kan 
kjende det igjen selv i ganske smaa biter. Skiktningen er meget markert, 
ofte er tuften likefrem skjællet-skitrig. Planterestene ligger tydelig klappet 
paa hinanden; det ser ut som om de alle sammen er blit tvunget over i 
horisontal stilling (eller rettere paralel bakkens overflate) og har lagt sig 
paa hinanden etagevis ^ Bctula odorata, Popuhts treinula, Salices [S. capræa, 
S. glauca, S. hastata, S. cfr. nigricans, S. cfr. phylicifolia) forekommer her 
i umaadelige kvantiteter; enkelte partier av tuften bestaar ikke av andet 



' Om dette simpelthen skjaldes en rolig bundfældning i vandet eller dets raske strømning, 
kan nappe avgjøres med sikkerhet. Bladene synes ialfald at ha hat tid til at „ordne 
sig" inden de blev forkalket (cfr. Gillebu-tuffen, hvor forholdet er et andet I. 



44 



Koi.i- .\oi<i)iiA(.i;.\. M.-N. Kl. 



end forkalketc bladrcstcr. Dcsiit« ii r.indi I'.i ^ i i i sin tid ot blad av Ribes 
ruhrmu, samt tvilsommc bladfraf^iiK iiU i- ;i\ Alniis incana. Kfter opdagelsen 
av Alnus-tiit'fcn ved Lcinc bar jc^ et likboldij^ materiale av Alnus incana- 
blader i alle iniili,!;!' slorrclscr, og sammenlignes de av Blytt omtalte frag- 
menter med de sikre ore l)ladcr, ser man at overensstemmelsen er meget 
svak. Jeg tior at del kim ei- fragmenter av nogen store bjerkeblader; der- 
om vidner baade bovednervene og det fra orebladene avvikende anasto- 
môse-net mellem disse. Makroskopiske furure.ster mangler totalt; hundrede- 
vis av baandstykker er untlersokt, men med negati\t resultat. Derimot 
indebolder bladtuften furu-pollen, som forst av Dr. G. iloLMsp:N paavist 
i baandstykker fra Bi.vtts samling'. Da jeg i en anden forbindelse kom- 
mer til at diskutere spørsmaalet om furuens første optræden i Gudbrands- 
dalen ^, skal jeg ber noie mig med at præcisere, at vi vel maa anta at Pinus 
silvcslris under denne periode levet . i dalen, men at den av klimatiske aar- 
saker har været lovtrærne helt underlegen-^. Naar tre forskjellige tufdan- 
nelser paa rad nedover dalen fra Kvam til Lillehammer (cfr. Gillebu og 
Nedre Dal) viser noiagtig samme fænomen, saa kan man ikke længer tale 
om tilfældigheter. Man staar da overfor en lovmæssighet som kræver en 
tilfredsstillende forklaring. Spørsmaalet om pollenets langflugt kommer jeg 
senere tilbake til; hvor stor vegt man vil tillægge denne feilkilde, er fore- 
løbig i boi grad en trossak. — Forekomsten av furupollen i bladtufbænken 
er imidlertid et interessant fænomen, som fortjener vor største opmerksom- 
het. Det forandrer imidlertid ikke opfatningen av bladtuffen som et distinkt 
palæofloristisk ledenivaa i Gudbrandsdalens kalktuffer. 

De forvitrede bladtufrester som det lykkedes mig at fremfinde i utkilings- 
profilene baade i boire serie og i tverserien, viser med al ønskelig tyde- 
lighet at bl ad tuffen engang har overdækket saagodtsom 
hele om r a ad et. Dog er det mulig, at dens mægtighet altid har været 
størst der hvor den nu har sit maksimum, og at tykkelsen har avtat fra 



' Jeg har senere gjort pollenanalyser av kalktuf og iagttat forskjellige ting som er av 
interesse. For at undgaa feilkilder har jeg vasket tufstvkkene meget omhyggelig og 
finknust dem og vasket tufpulveret paany. Dette er absolut nødvendig, idet saa- 
godtsom al tuf er porøs og indeholder forurensninger i alle hulrummene (avsat av ned- 
sivende vand). Man har ingen garanti for at disse forurensninger 
skriver sig fra det samme tidsrum som selve kalktufs ubstansen; 
de kan være yngre. Ved Leine er hele tufavsætningen gjennemfiltrert av vand, og 
under gravearbeidet har man stadige ubehageligheter av kilden. — Jeg har fundet at 
efter en meget omhyggelig knusning og utvaskning indeholder selve kalktutfen bare 
smaa mængder av pollen, tildels minimale (oplosning i fortyndet saltsyre, efterpaa 
kokes bundfaldet med kalilut). I enkelte prover finder jeg ikke spor av pollen. — 
Kalktuflfen ved Gillebu er ogsaa uhyre let tilgjængelig for nedsivende regnvand og for 
strømninger i grundvandet. 

2 Cfr. den generelle del. 

3 Forekomsten av Hippophaës ved Gillebu og de biologiske fænomener som knytter sig 
til denne, peker ogsaa i samme retning. Cfr. den generelle del. 



ig2I.No. 9- KALKTUFSTL'DIER I GUDBRAXDSDALEX. 45 

dette punkt og til sidene. — I det øverste profil i venstre-serien indeholdt 
bladtuften et og andet /)n'«sblad, men uhyre sparsomt. Dette beviser at 
planten ogsaa under bladtuftens tid har eksistert langs tuftens overkant. 
Ellers er bladtuften paafaldende ensformig og artsfattig uten variationer, 
hverken i vertikal eller horisontal retning. Kun nedenfor veien er der 
antydning til en ophobning av SaZ/A'-blader (S. cfr. nigricans). 

Som helhet betragtet tyder bladtuft'en paa stor vandforing i bækken. 
Dette gjælder sedimentære lag i sin almindelighet. Noget brud mel- 
lem det underliggende mosetufkompleks og bl ad tuffen eksi- 
sterer ikke. De danner tilsammen en serie med den sterkeste tufdan- 
nelse til slut; dog er der som tidligere omtalt antydning til mindre oscilla- 
tioner i fugtighetsforholdene i periodens begyndelse. 



Diyashtß'ni. 

At denne merkelige og utpræget sedentære horisont maatte vække 
Blvtts store interesse da han første gang fik øie paa den, er let forstaae- 
lig. Og dog synes Drj-astuften i Blytts profiler ikke at ha været saa 
pragtfuld som fleresteds i de ovenfor beskrevne serier, hvor man fakdsk kan 
ta ut det ene store st3'kke efter det andet som ikke bestaar av andet end 
Dryasrester. 

Dryastuften har en meget karakteristisk graagron farve ', specielt i fug- 
tig tilstand. En undtagelse fra denne regel møter vi i profil III, hvor tuften 
var rødlighvit eller svakt fiolet og eiendommelig sinteragtig (ujevn, vortet 
og knutet og med paralelle tynde, bølgende lag). Muligens har her cyano- 
fycéer lokalt været virksomme ved kalkutskillelsen. Paa enkelte punkter har 
Dryastuften karakter av en mosetuf (cfr. Blytt 1892 p. 1 1 ; her omtales 
en recent mosetuf av Hypniiui filiciniini fra Bardodalen i Tromsø amt, som 
ifølge Blytt viser stor likhet med visse partier av Dryastuften). 

Tuffindestedet ved Leine maa under denne periode ha frembudt et 
ganske enestaaende syn. Bakken maa paa forsommeren ha været et eneste 
hav av blomstrende Dryas. Det vedføiede fotografi av Dr3'as i fuldt flor 
fra Hesjehompan i Saltdalen gir et indtryk av h\"orledes markvegetationen 
saa ut. Nu er Dryas en utpræget xerofil plante, og da kalkkildens nær- 
meste omgivelser utvilsomt har været fugtigere end den øvrige del av bak- 
ken, ialfald til visse aarstider, maa man nødvendigvis anta at Dryas har 
været likesaa frodig, ja sandsynligvis endda frodigere rundt omkring tuf- 
findestedet, hvor man jo overalt har den kalkholdige jordbund. 

De forholdsvis sparsomme rester av bjerk, asp og andre løvfældende 
lignoser samt de spredte furunaaler tyder paa en meget spredt og aapen 
trævekst. Dryas er en utpræget lyselskende plante, som ikke taaler over- 



' Dette skyldes tuffens urene karakter. Furutuffen er derimot paafaldende ren kalk 
og gir tildels intet residuum ved opløsning i salts\'re. 



46 



ROLF NOI<I)IIAf;i:N. 



M.-.\. Kl. 



sky/^nintc, f"K fl«^""'' masseoptrædcn forutsættcr med bestcmthct at der ikke 
liar cksistcrt no^'-en tæt sammenhængende skog paa stedet, saaledes som 
linder l)l;i(lliill( lis lid. Smii vi senere skal se, utdor J)r\'as ganske tydelig 
ved Leine som en fol^c a\- ("iiniskfj.i^ens tillaj.;cnde tæthet i den efterfol- 
gende lid. 

Efter all Inad \i \'el om ariens iilhredelse i skandinaviske fjeldtrakter 
i nutiden, maa Drvns or/a/yc/a/a karakteriseres som utpræ-get xerofil 
og kalkyndende (clV. A. Ci.kvk 1901 I. c, Scifrofter: Das Pflanzenleben 




Fig. 14. Dryadetum octopetalae paa kalkberg. Hesjehompan, Saltdalen i Nordland. 
Juli 1920. Nordhagen fot. 



der Alpen 1908 p. 182). Som Blytt altid fremhævet i sine plantegeogra- 
fiske arbeider, er den knyttet til løse, smuldrende skifre (særlig kalkholdige) 
og kalkbergarter. Den elsker en varm og i Vegetationsperioden tør jord- 
bund og optrær her associationsdannende. Svenske forskere adskiller ofte 
en lavrik og en mosrik „ Dry as-hede " ; men dette skille er temmelig tvil- 
somt og lar sig ikke altid gjennemfore i praksis, ialfald i visse strok av 
den skandinaviske Qeldkjæde (ofr. Samuelsson 1916, hvor forholdene ved 
Finse omtales; i naturparken ved Sylene har jeg konstatert det samme). 
I det nordlige Norges kalk og dolomitstrok, f. eks. mellem Ranen og Suli- 
tjelma, er Dryadeta octopetalae meget utbredt og synes her at være mere 
ekstremt xerofile end sydpaa (ofr. ogsaa Fries 1913, Texgwall 1920 1. c). 
Ved Leine har man hat det særs}'n, at et Dryadetum 
optrær som under vegetation i en meget lysaapen skog 



I92I.N0. 9- KALKTUFSTUDIER I GUDBRANDSDALEN. 47 

(om samfundet som helhet fortjente navnet skog, er et sporsmaal hvorom 
der kan disputeres). Sidste sommer iagttok jeg en motsvarighet hertil ute 
i naturen, nemlig i Junkersdalsuren i Salten, Nordland. Som en folge av 
eiendommelige edafiske forhold (stadige jordras og ophobning av talusma- 
teriale fra de stupbratte styrtninger i uren) er skogen her fleresteds meget 
aapen og indskrænker sig til smaa trægrupper (furu og løvtrær, særlig 
bjerk). Store matter av Drvas dækker her flekvis den kalkrike, ujevne 
bund '. 

Av de ovrige i Dryastufifen fundne arter indgaar Sa/ix reticulata, S. 
hcrhacca, Pyrola uiiiior og Eqiiisctuin varicgatiini allesammen i de av mig 
statistisk undersokte Dryadeta i naturparken ved Sylene (i Sor-Trondelag 
fylke). Salix reticulata er, som Blytt fremholdt (1. c. p. 5), en plante „som 
paa vore fjelde trofast pleier at ledsage Dry as" ; man burde ogsaa ha ventet 
at ThalictruDi alpimmi og Carex riipestris hadde været tilstede. Disse er 
imidlertid ikke fundet fossile. 

Sammenlignes nu denne sedentære horisont med den underliggende 
bladtuf, er kontrasten sterkt ioinefaldende-. Under Dry as tuffens tid 
maa bækkens vandforing være svundet betydelig ind, og 
antageligvis har bækken været helt utterret i sommertiden. 
Den tætte, xerofile Dryasmattes tilstedeværelse kan ikke forklares paa anden 
maate. Periodevis har vel bækken silret utover marken og avsat kalktuf, 
men ikke til stadighet. 

Undersøker vi de profiler nærmere hvor Dr3'astufkomplekset har sin 
største mægtighet, finder vi flere komplikationer. Saaledes tyder den i en 
række profiler forekommende ganske tynde, lagdelte og halvsedimentære 
horisont som er avsat ovenpaa Dryasmatten, paa en m i n d r e opsvulmning 
i bækkens vandføring, dog ikke stor nok til at fordrive den sedentære 
vegetation [Dryas, Salix reticulata, Pyrola uiiiior, Carex sp.). 

Den allerøverste del av komplekset er atter meget karakteristisk. 
1 saagodtsom alle profiler m ©ter vi her en ops m uldret, for- 
vitret, tildels leragtig sone med enten utydelige tufrester eller smaa 
biter indeholdende furu og Dryas. I Blytts profiler fandtes et virkelig 
grusblandet kalkler (4 cm. mægtig), utvilsomt et forvitringsprodukt; de i 
palæontologisk museums kalktufmonter utstillede prover fra Blytts samling 
viser dette. Av undertegnedes profiler er f eks. nr. III og X alleroverst 
ved stenroisen meget oplysende. Paa sidstnævnte sted fandtes en 2 cm. 
tyk graabrun jordagtig stripe, som over mot profil IX antok natur av seig 
1ère, ovenpaa den opsmuldrede Dryastuf Under oprenskning av dette pro- 
fils nordligste del viste Dryastuffen sig at være aldeles brun og forvitret 



' Merkelig nok syntes Carex riipestris, en av Dryas's trofaste ledsagere, at taale beskyg- 
ning meget bedre end Dryas. Den optraadte i Junkersdalsuren fleresteds i skogbun- 
den (paa meget torre steder! i store masser. 

2 Dog maa det specielt fremhæves at der mellem bladtuften og Dryastuften ikke er spor 
av brud eller diskontinuitet i lagfolgen i de fuldstændige profiler. 



48 ROLF NOrU3HAGi:.\. M.-N. KI. 

paa ovcrflatcn. Alle iilkiliiii^sproüh tie i hoire sorie og i tvcrserien er 

i denne fnrl^indelse ogsaa meget vaidiluldc . I hi- viser Dryastuffen 
sig at være ytterst forvitret eller mangler helt, samtidig 
med at ogsaa den underliggende bl ad tuf er vitret ned til 
ubetydelige rester. Isteden finder man brunlige eller olivenfarvete 
lerstriper paa disse tufhorisonters plads i lagfnlgen. 

Dette faktum kan ikke forklares paa anden maate end ved den anta- 
gelse, at under en periode forut for furutuftens dannelse og efter bladtuf- 
fens tilblivelse, har kilden svundet saa sterkt ind at der over visse arealer 
av tuffens overflate ingen avsætning har foregaat, men tvertimot en for- 
vitring av den tidligere dannede kalktuf Utkilings pro filene og de 
øvrige „fuldstændige" serier taler her et samstemmig sprog, 
som ikke kan fortolkes paa mere end én maate. Om der 
til en begyndelse blev avsat Dryastuf over hele tuffens omraade, vet vi 
ikke med sikkerhet. Det er meget mulig at der f eks. i profil XX og XXI 
i tverserien, hvor ingen Dryasrester er opbevaret, aldrig har været nogen 
tufavsætning under denne periode, men kun forvitring. — Det karakteri- 
stiske Dryastufstykke som blev fundet nedenfor kjoreveien (profil XXII), viser 
at der ogsaa foregik kalkdannelse længer nede i bakken. Men alt i alt lar 
grænsene for Dryastuft'ens oprindelige horisontale utbredelse sig ikke fast- 
sætte med sikkerhet. 

Der er, som vi har set, antydning til en tredeling i Dryastufkomplekset. 
Men det allerøverste lag er gjerne vitret ned, undertiden ogsaa det mid- 
terste, halvsedimentære; dog er det ikke sikkert at dette har været avsat 
overalt ^ 

Profil V i venstre serie og dets naboprofiler opviser enkelte træk av 
speciel interesse. Her er nemlig det alleroverste lag i Dryaskomplekset 
bedst opbevaret og ytterst eiendommelig. Man finder her en ca. 3 cm. 
mægtig overordentlig fin og sprod horisont, som jeg ovenfor har betegnet 
som „klidagtig", fordi den har en viss likhet med det saakaldte klidbrød 
(som indeholder hveteklid). Den er meget porøs og aabenbarer ved noiere 
eftersyn en utallighet av fine kanaler. Disse skriver sig fra meget tynde 
furunaaler, saa tynde og korte at hvis man ikke saa det karakteristiske 
halvmaaneformete tversnit og den typiske gruppering i naalepar, kunde man 
staa tvilraadig overfor deres natur. Da furunaalene baade i furutuften og 
i det underliggende lag var store, brede og kraftige, fik jeg den tanke, at 
de ovennævnte fine avtryk maatte skrive sig fra sterkt opraatnede furu- 
naaler, hvor grønvævet var forsvundet og kun den centrale karstrengagtige 
del med omgivende skeder var tilbake. ' Ved undersøkeiser i furuskogen 
ovenfor gaarden Veikle i Kvam, kunde jeg konstatere at denne tydning var 



' ØYEN skriver (1920 1. c. p. 253) at Dryastuffen er tredelt med et egte Dryaslag i midten 
og et overgangslag nedad mot birketuflfen og et opad mot furutuften. Dette er, som- 

mine profiler viser, ikke ganske rigtig. 



I92I.N0. 9- KAI.KTL'FSTL'DIER I GUDBRANDSDALEX. 49 

rigtig. Skogbunden var her tæt dækket av nedfaldne recente naaler (saa- 
kaldt „fald-forna" et\er Sern anders terminologi 1918 1. c). Disse \-ar ufor- 
andret, store og brede. Men dN'pere nede ble\- de sukcessivt forandret, 
mere og mere myke og tynde, og til slut fandtes en raahumus bestaaende 
av tæt sammenfiltrede eller sammenklæbede fine naaler, h\-or bare de for- 
vedede elementer endnu holdt stand mot forraatnelsen. Naalene var ofte 
brukket o\-er paa tvers i kortere stykker, og hele massen 
h a d d e den mest s 1 a a e n d e 1 i k h e t med den i k a 1 k t u i'(e n f o r e- 
kommende horisont. Drvas octopctalas blader er ogsaa som bekjendt 
læragtige og „eviggronne" og motstaar meget længe forraatnelsen, hvilket 
man kan iagtta paa gamle eksemplarer ute i naturen, som altid har en 
mængde indtorrede, gamle blader sittende paa stammene. 

Denne merkelige lille tufhorisont utdyper saaledes paa en udmerket 
maate billedet av naturforholdene paa stedet under den tid da kilden var 
meget reducert. Den viser os at furuskogen maa ha tx'knet til; tjeldplan- 
tene gaar sin undergang imote under den tiltagende beskygning; de sidste 
Dry as bl ad er mot er os her. Og kalkavsætningen synes at ha fore- 
gaat ytterst trægt; barnaalene har hat tid til at raatne op og danne raa- 
humus og er efterhaanden langsomt for\-and!et til en sprod, smuldrende og 
poros tuf. Antageligvis markerer profil V og de tilstotende profiler, som 
dog er mindre utpræget, det eneste sted indenfor tufomraadet hvor en fort- 
lopende, men mere og mere utpint avsætning hai" fundet sted. 

Dryastuften indeholder de forste makroskopiske fururester og tilkjende- 
gir dette træslags indvandring paa stedet. Dette beror utvilsomt paa en 
tiltagen i sommertemperaturen sammenlignet med bladtuffens tid. Under 
Dryastuffens tid synes lerbakkene \'ed Leine at ha været torre og varme 
i sommermaanedene. Antageligvis har vintrene været strenge. Om Leine 
dengang laa ved furuens hoidegrænse over havet, saaledes som Blytt 
antok, eller om træet gik endda hoiere tilljelds, kan ikke avgjores med sik- 
kerhet. Blytt fremhæver at furunaalene i Dryastuften er korte og smalere 
end i furutuften, og tyder dette som et tegn paa relativt ugunstige kaar. 
Imidlertid viser den halvsedinientære horisont (midt i Dryaskomplekset) store, 
kraftige og brede furunaaler (flere 3,6 — 4 cm. lange), saa jeg tror ikke at 
man kan komme til nogen sikre resultater ad denne vei. At furunaalene i 
den sedentære Dryasmatte er smaa, kan muligens bero paa forraatnelse (cfr. 
det Gverste, klidagtige lag). Forøvrig viser furunaalene i en recent skog 
meget store \-ariationer, alt efter deres stilling paa grenene og efter be- 
sk3'gningen. 

Efter at ha diskutert sporsmaalet om tufiagenes geologiske alder 
kommer jeg i et følgende kapitel tilbake til de arktiske planterester ved 
Leine og deres bet^'dning for sporsmaalet om fjeldplantenes utbredelse i 
Skandinavien i tiden for den postglaciale varmetid. 

Blytt anfører i sit arbeide ogsaa Cotojicaster vulgaris fra Dryaslaget 
ved Leine, dog under t\il ; desuten opføres Betitla nana og B. intermedia 

Vid.-Selsk. Skr. L M.-X. KL 1921. Xo. 9. 4 



50 KOI.F NORDMAGKX. M.-N. Kl. 

saml .li-(l()^l(i/ili\'/()s (i/JiciiKilis som meget usikre. Jeg har selv hat an- 
ledning til at se disse a\tryl< og linder dem ogsaa meget problematiske og 
utydelige. Og vi gjor derfor rettest i ikke at ta dem med i det plante- 
geografiske ræsonnement, hvorpaa de forovrig ikke influerer i nogen av- 
gjorende grad. l*'<)rresten er det slet ikke usand.synlig at de kan ha vok.set 
paa stedet '. 

Av landsnegler er folgende arter iagttat i DryastufTen: Cochlicopa lu- 
hrica MCu.., Cni/ti/iis fiilviis Müll., iJyaliiiia radiatiila Alder og Pyrami- 
(fil lu nuicrata Stl'dkr. 

Funifiiffrii. 

Da tufkilden i den efterfølgende tid atter begyndte at vælde frem med 
stor kraft, var stedet dækket av en dyster furuskog, for at anvende Blytts 
ord. Selv de allerunderste lag viser en enorm mængde med store furu- 
naaler, som danner et eneste lost, forkalket kaos. I et par profiler er der 
fundet antydning til litt mosetuf underst, men gjennemgaaende har furu- 
tuffen et sedimentært præg helt fra bunden av. Den i profil XVII og X\'III 
forekommende lerstripe (rod og blaa 1ère), som er skyllet ovenpaa Dry as- 
tuffen, tyder vistnok paa sterk vandfoiing helt fra begyndelsen av (i likhet 
med det sedimentære præg). Furutuften er ellers meget ren og rodlig 
av far\'e. 

Ellers viser profilene adskillige variationer i \-ertikal retning, uten at 
egentlig bestemte gjennemgaaende horisonter kan paapekes. Og furutuften 
er paafaldende ensformig; store masser av naaler, barkstykker, grener og 
en mængde furukongler gjenfindes overalt, desuten blader av J^icciniiiin 
vitis idæa. Den anden art V. iiligiiiosiiin er bet3-delig sparsommere i sin 
optræden. Man fa ar et levende indtryk av at en tæt, dyster 
ur sk og av furu har behersket Leinebakkene under denne 
periode. 

I furutuffens ovre del gjenfindes, ialfald i en del av de overste profiler 
ved stenroisen, en sedentær horisont, som frembyr flere træk av inter- 
esse. De opbevarte rester a\- urteagtige planter (Cirshim hctcrophylliiiii, 
Fragaria vcsca, Pyrola luirior, Toficldia pnhistrii^) samt Parniclia pliysodcs 
tyder paa meget rask tufax'sæt n i ng. De blote blader av Cirsiimi har 
ikke engang faat tid til at raatne op, ja de opbevarte kurver bærer tildels 
forkalkede rester av blomstene i sit indre (planche II, fig. i). Horisonten 
er som nævnt meget fin og sprod, tildels mosetufagtig, og fossilfattig, men 
er antageligvis til en \iss grad fysikalsk utfældt tuf. Fossilene tyder paa 
en mere englignende \eg'etation i bækkens nærhet under lagets avsætnings- 



' Naar Oyen i sit sidste arbeide opforer Biiiila nana, Coloncastcr og Arctostapltylos 
officinalis for Dryastufi'en uten spor av reservationer, er dette efter min mening meget 
misvisende. Naar Blytt, som kjendte sine arter ut og ind, satte sporsmaalstegn efter 
best^mmelsene, har vi ingensomhelst grund til at si jife dem! 



ig2I.No. 9- KALKTITSTUDIER I GL'DBRAXDSDALEN. 5I 

tid. Høiere oppe i lagfolgen nioter vi imidlertid den samme kaotiske furu- 
tuf som længer nede. 

Fragaria vesca Ibrtjener særlig opmerksomhet, da den er av en plante- 
geografisk set, sydlig type. Dog gaar den selv i nutiden paa lokalklimatisk 
særlig gunstige steder over looo m. o. h. i tjeldtraktene (Resvoll-Holmsen 
1920 1. c. p. 255). 

Toficldia palustris, som er en typisk tjeldplante, viser at seh' under 
denne periode har der vokset arktisk-alpine arter i Leinebakkene '. Dry as 
synes dog at være helt forsvundet. 

Blytts fund av Mniiini ptiiictatum og Pcltigcra caiiiiia „midt i bænken" 
(1. c. p. 4) tyder paa forekomsten av en lignende sedentær horisont længer 
nede i bakken. 

Av løvfældende trær optraadte folgende under furutuftens tid: Bctiila 
odorata og B. verrucosa, Populus Iren ni la, Salix capræa (og muligens flere 
arter) og Sorbus Ancuparia. Ingen av disse arter er i klimatologisk hen- 
seende særlig opl3'sende, uten kanske Betula verrucosa. Bjerkeblader er 
vanskelige at bestemme paa grund av artenes store variationsvidde. Dog 
tror jeg at det av Blytt fremfundne dobbelttandete blad (1. c. p. 12) og et 
lignende, som undertegnede fandt i et furutufstykke i profil XIX, maa hen- 
føres til Betula vci'rucosa. Denne art staar paa grænsen til vore kuldskjære 
lovtrær og gaar i almindelighet ikke hoiere end 4 — 500 m. o. h. (Blytt- 
Dahl: Haandbog i Norges Flora 1906). Nordpaa gjor den et stort sprang 
fra Inderøen til Salten, hvor konservator Ove Dahl og undertegnede fandt 
den i Junkersdalsuren sommmeren 1920. Desuten kommer den igjen i Øst- 
Finmarken ^, hvor den er indvandret fra Finland. Betula verrucosa og 
Fragaria vesca tyder begge paa at klimatet maa ha bedret sig betydelig 
siden bladtuflfens og Dryastuftens tid. Da Betula -verrucosa, som Blytt 
fremholder, i nutiden er nær sin høidegrænse ved Leine (ca. 600 m. o. h.), 
maa man anta at klimatet under furutuffens avsætningstid ialfald ikke kan 
ha været strengere end i nutiden'^. Vi har her en minimumsgrænse som 
er ganske værdifuld. 

Furutuffen har i likhet med bladtuffen oprindelig dannet et sammen- 
hængende lag over hele omraadet. At dens nuværende maksimum (pro- 
fil XX i tverserien) betegner det sted hvor den ogsaa oprindelig var 
mægtigst, er meget sandsynlig, men kan ikke bevises. 

Furutuffen hviler, som vi har set, diskordant paa Dryastufkomplekset 
(muligens med undtagelse av profil V, hvor der som ovenfor nævnt er svak 
antydning til kontinuitet), og viser ingensomhelst tegn til nogen overgangs- 



' Et eiendommelig avtryk i den undre Dryashorisont tilhorer muligens ogsaa denne art 

(profil IV). 
- Her er den HN'lig opdaget av forstmænd. 
3 Hvor hoit Leine dengang laa over havet, kan neppe siges med bestemthet, dog vistnok 

ikke lavere end 4 à 450 m. 



52 ROLF NOKDHAGEN. M.-X. Kl. 

sone mot y7/////.s-Uilï(n. jc,^ skuldr ogsaa være tilb^jiclig til at tro at furu- 
tuffen ingensteds indenlor (l( t inidi isokte omraade opviser sin oprinde- 
lige mægtighet; den er o\(i-all iiicgct forvitret øverst, og man savner paa 
en niaate en logisk n\s]iiliiing av komplekset opad (f. eks. i form av en 
sedentær sone). - Som IuIIk t l)etragtet maa furututfens tid ha været en 
lixlig kalkdar.nende periode; dcrum \idiHr den betydelige mægtighet. 

Trækulrestene og de lokale kulstriper indenfor furutuffen paa forskjel- 
lige punkter viser, som Blytt ogsaa nævner, at lynnedslag ikke var nogen 
sjeldenhet under denne periode. Ogsaa i Alnus-tuften fandtes kulrester 
(profil XXI). Disse katastrofer synes ikke at ha infhicrt paa kalkavsætningen 
i nogen paaviselig grad. 

Av landsnegler er følgende arter fundet i furutuffen: Coi/ii/iis fulvits 
Müll., Hvalinia radiatula Alder og llliiiid prlliicida MCli,. 

Aliiiis-tiijfcn. 

I likhet med bladtuften og furutuffen er denne horisont saa karak- 
teristisk i petrograhsk henseende at den kan gjenkjendes selv i smaa frag- 
menter. 

Farven er gjennemgaaende brunlig, og tuffen meget grov med store 
hulrum, bølget og valket og uten tydelig skiktning. Fleresteds har den 
karakter av grov mosetuf. Folgende arter er med sikkerhet konstatert i 
oretuffen: Bctiila odorata, Aliiits incana (blader, talrike Ç-rakler), Poptihis 
trennt la, Salix capræa, Piints silvrstris (nogen faa, men tydelige naaler), 
Eqitisctitiii liiciiialf (talrike ledstykker), samt grove blader og straa av græs 
og Cariccs og flere sorter mos. Enkelte av de lose stx'kker fra muldlaget 
viser stor likhet med recente Çvc7/;o/3YV''-tuffer '. 

Tuffen gir gjennemgaaende indtryk av at være dannet temmelig raskt 
(ofr. mangelen paa utpræget skiktning). Profil XXI og de to profiler 
nedenfor veien, hvor Aliitts-Vxx^îtn var bedst be\aret, viser at kalken tildels 
har været utfældt som jordagtig tuf, der alternerer med fastere sedentære 
horisonter. Paa dette punkt stemmer Leinetuffen med en flerhet av de 
svenske tufforekomster, hvor de alleryngste tuflag meget ofte er 3-tterst 
sprøde og sterkt forvitret (Sern ander 19 i6 p. 129 og 160). Da rester av 
oretuffen er fundet over hele det undersokte areal, viser dette at kilden 
ogsaa under denne tufavsættende periode har været ganske betydelig og 
har overrislet hele arealet. Jeg har for paapekt at ^/////5-tuffen antageligvis 
har hat .sit maksimum i de muldeagtige forsænkninger (op til 95 cm. i pro- 
fil XXI), selv om dette ikke kan siges med sikkerhet. 

Pollenundersøkelser fra en række med haandstykker har bragt for 
dagen fttru, or, hjcrk og 5V///.v-pollen, d e r i m o t ikke a n t }• d n i n g e r til 



' Folgende snegler er iagttat i Alnus-tuffen : Cochlicopa lubrica Müll., HyaUnia radiahila 
Alder, Pvraiuidida ruderata Studer, Vertigo alpcstris Alder, samt Hydröbia Steint 
V. Mart. 



I92I.X0. 9- KALKTUFSTLDIER I GUDBRAXDSDALEX. 53 

gran. Da Leinetuffen under denne periode har været dækket av et 
fugtig orekrat med en englignende, h3'drofil eller mesofil bundvegeta- 
tion (cfr. den recente prøveflate som er beskrevet p. 81 og saaledes maa 
ha frembudt stor likhet med de nutidige forhold i Leinebakkenes fugtigere 
partier, kunde man ogsaa vente at finde rester efter gran, som likeledes 
optrær i den samme dalside (ialfald ikke mere end i — 2 km. vestenfor 
tuffindestedet) i nutiden. Men undersøkelsen har altsaa git et negativt re- 
sultat, hvilket kunde tyde paa at kalktufa vsætningen var op- 
hor t allerede f o r i n d e n granen \- a r i n d \' a n d r e t til K \- a m. 
Imidlertid utelukker ikke dette negative resultat en mere sporadisk fore- 
komst av gran i bygden. Det er ogsaa mulig, at de partier av 
Al nus-tuffen som nu t^'deligvis er forsvundet, kan ha inde- 
holdt spor efter granen. 

Alnus-tuffen er meget interessant derved, at den bygger en bro over 
til nutiden. Som i indledningen omtalt, er Aliiits iiicana fremdeles et meget 
vigtig skog- og kratdannende træslag i dalskraaningene under Leine og 
langs Veikla. Man kunde ved forste oiekast tro at dette forhold hoved- 
sakelig var fremstaat som et resultat av kulturpaavirkning (uthugst og op- 
rydning). Men kalktuffen viser os ganske tydelig at nutidsvegetationen har 
sine retter langt tilbake i tiden. 

Alnus-tuffen utfylder et hul mellem furutuftens tid og nutiden; og det 
helhetsbillede som vi nu faar av Leinetuffen, er ganske anderledes har- 
monisk end tilfældet var tidligere da oretuffen ikke var kjendt, og da furu- 
tuffen saa at si braastanset, uten spor av o\-ergangsled til stedets recente 
vegetation. 



Hvad der ligger mellem furutuffen og Alnus-tuffen, og hvorledes vege- 
tationen har artet sig paa voksestedet i dette tidsrum, er foreløbig h3-llet i 
mørke. Diskordansen og den sterke nedbrytning av furu- 
tuffen i tuffens nordreog sydostlige del (nedenfor veien) 
viser at tufkilden har stoppet op; der er ingen antyd- 
ning til overgang mellem furutuf og oretuf Forvitringen har 
til og med grepet om sig endda dypere; saaledes er i profilene nedenfor 
veien baade Dryastuften og bladtuffens øverste del forvitret. Dog foreligger 
her en anden mulighet, som ikke bør oversees, nemlig at baade Dr3'astuffen 
og de øverste lag i bladtuffen paa dette sted kunde være vitret ned alle- 
rede i slutten av Dryastuffens tid (saaledes som i profil XIX og XX); de 
to forvitringshorisonter skulde altsaa møtes. Forøvrig er den første even- 
tualitet slet ikke udelukket; ti furutuffen kan paa dette sted ha været meget 
mindre mægtig end hoiere oppe i bakken (det samme ræsonnement gjælder 
ytterprofilet i tverserien (nr. XXII). 

Jeg skal foreløbig ikke opholde mig nærmere ved denne interessante 
diskordans, men isteden fæste opmerksomheten ved et andet forhold, som 



54 ROLF NOROHAGKN. M.-X. Kl. 

ogsaa er av Ix l\ (Iniii!^, iK-mli^ y//////.s-liiircn.s t'oivitring. iJct er ganske 
klart at de une maa ha foregå at mellem tuffens avsætnings- 
periode og nutiden. I vore dager er tufkilden forholdsvis beskeden 
og danner som for na>\nt ubetydelige mængder av mosetuf hist og her. 
Vi har altsaa atter en oscillation i denne merkelige kildes kalkdannende 
virksoinhet fra Alnus-tiiffeiis tid Dg til nutiden, ledsaget av tydelig for- 
vitring. Og denne forvitring har ikke alene forvandlet oretuflfen til ube- 
tydelige rester overst i det stadig voksende muldlag over store deler av 
tufomraadet, men har ogsaa begyndt at angripe fuinituffen, altsaa atter to 
forviti'ingsperioder som griper over i hinanden! 



Jeg haaber ved denne utredning, som ikke indeholder et eneste 
postulat, men som er en fuldstændig nøktern diskussion av Leinetuft'en ut 
fra stratigrafiske, genetiske og plantegeografiske synspunkter, at ha vist at 
denne tuf horer med til vore mest interessante kvartærgeologiske av- 
leiringer. Opbygningen er helt lovmæssig og klar, og de problemer som 
knytter sig til den, fortjener den allerstørste opmerksomhet og alvorlig 
drøftelse. 



II. Kalktuf ved Gillebu og Tingvold 1 Øier. 

A. Topografi og vegetation i nutiden. 

I Øier, som ligger ca. 60 km. s3'dost for Leine, gjør Laagen en boi- 
ning, som viser en paafaldende likhet med forholdene i Kvam. Den danner 
nemlig et utpræget S. Paa det sted h\-or elven løper ret vest — ost, ligger 
Tingvold og Tolstad, og her kommer en liten bæk nedover skraaningen 
fra nord — nordvest og løper ut i Laagen. Bækken render langs en for- 
sænkning mellem den temmelig bratte Skarskampen (hg 15 tilvenstre) og 
Skjønsbergkampen (tilhoire). Terrænget er her tæt bevokset med naaleskog 
og tilhører gaarden Gillebu, som ligger straks ostenfor. G i II e b u t u f f e n 
ligger omtrent halvveis oppe i lien i ca. 50 meters høide over landeveien 
ved Tingvold og i alt ca. 240 m. o. h., ifølge Ovexs maalinger (1920 1. c). 

Som i indledningen berørt blev tuften opdaget av Oven og Holme 
sommeren 1917, efterat Holme allerede for mange aar siden hadde frem- 
fundet løse kalktufstykker i et grustak ved Tingvold, like i kanten av lande- 
veien. Disse to forekomsters beliggenhet og relation til hinanden vil frem- 
gaa av fig. 16, som gir et meget skematisk billede av forholdene. 

Da mine undersokelser paa stedet i flere henseender gav betydnings- 
fulde resultater, som utvider kjendskapet til tuftens stratigrafi, skal jeg her 
gi en utførlig fremstilling av denne. Redegjorelsen blir i det hele tat en 
logisk fortsættelse av den beskrivelse som Oven tidligere har publicert 
over denne interessante tuff"orekomst (1920 1. cl. 



1 92 1. No. 9. 



KALKTL'FSTUniER I GL'DBRAXDSDALEX. 



OD 



Den omtalte bæk, som nu er ganske ubetydelig i sommertiden, har, 
som fig. 16 viser, ved tilbakeskridende erosion gravet sig et markert leie i 
dalsidens løse dække, som bestaar av bundmoræne og bræelvgrus. Kalk- 
tuft'en optrær nu kun som rester paa begge sider av bækken, og er like- 
som den ovrige del av skraaningen bevokset med n a a 1 e s k o g. Træ- 
bestanden i denne utgjores av gran og furu i blanding, dog mest av den 
første. Bund\'egetationen er fysiognomisk præget av Hylocomiiini parictiuiiDi 
og H. proUfrntm. Særlig den forste danner pragtfulde gronne matter 




Fig. 15. Grustaket ved Fingvold og Gillebuskogen hoiere oppe. Bækken kommer ut av 
skogen mellem huset og telegrafstolpen og render over til venstre, bakenfor skigarden. 

August 1 91 8. Nordhagen fot. 



mellem stammene. Associationens konstitution \-il fremgaa av neden- 
staaende analyse, der omfatter 10 prover à 0,25 m.-; de 5 forste er fra 
bækkens ostside, de 5 sidste fra vestsiden. Tallene angir dækningsgraden 
(Hri.T-SERXAXDERS S'Sradige skala). 





I 


2 


3 


4 


5 


6 


7 


8 


9 


10 


l'acciiiiitni i'itis idaa 


3 

*2 

1 
I 

5 


3 

I 

I 
5 


2 
2 
1 

1 
3 


3 
2 

1 

1 

3 


3 
3 

I 

5 


3 
2 

I 
I 

3 


3 
2 

1 

I 

5 


2 
2 

1 
I 
I 


3 
2 

2 

I 
5 


3 
I 
2 


Liuiura boiralis 

Ca/ninagro^/is aniii(Uiiacra 

I^rdi^dria l'csca 


Hy/oconiiiini parictiiiiim 


5 



56 



ROLF NORDMAGEN. 



M.-N. Kl. 



1 


■- 


■'. 


t 


5 


6 


7 


V, 


9 



I lyldCoiitiiiiit /irali/cniiii . . . 

JJicraiiiiiii nndulaium 

Agroslii^ vuli^aris. 

Caiiipa)iiil(i pcrsicifnlid . . . 

Cairx digiliita 

Fcsliica oviiia 

Uicraciuui sp 

Jiiiiipcnis cniiiiiiiiiiis (lilcnl 

Lotus ciiniiciilalits 

Lifiula pilosa 

Melainpynini silvaticiiiii . . . 

Oxa/is ^■Icrfosr/Ia 

Plicgoptcris D/yop/rris . . . 
Phegopfcris Rohcrtiaiia . . . 
PiuipincUa Saxifraga . . . . 

Piroln chloraullia 

Riibits idiVHS 

Ruhiis saxatilis 

Silcnc veiiosa 

Sorbits AucHparin (littn) . 

StcUaria graminca 

Vacciniiim MyrtiUus 

Veronica officiiiaUs 

\ ^icia cracca 

V^iola coUina 

Viola Riviiiiaua 

] 'iola arenaria 

Cladonia spp 

Cladina silvatica , 

Dicranuni nia/iis 

Dicranuni scopariuni . . . . 

Pcltigcra apldosa 

Polyirichiini sp 



Tabellen viser os en tæt uiosniatte med meget spredt bevoksning av 
tyttebær, Linna'a, /ordbær og Calantagrostis arnndiiiacea. Xx de ø\"nge 
arter kan merkes Oxa/is Acetosella, Carex digitata og Pirola elilorautha. 

Bækkedalens skraaninger var meget lose og lite stabile. Naale- 
skogen gaar tildels nedover bratten, men gjennemgaaende ser man her en 
aapcn vegetation, som er temmelig \'ekslende og ganske artsrik: 



1921. Xo. g. 



KAl.KTUFSTUDIER I GUDBRAXDSDALEX. 



57 



Aiitciiiiaria dioica 
Arabis liirstita 
Arenaria scrpyllifolia 
Calaiuiiitha Aciuos 
Ec/iii/ospcrin/iiii (ifflcxiini 
Epilohiiiiii nioiitaiiiini 
Erigcron accr 
Fcsfitca oviiia 
Gnlcopsis tetrahit 
Galiinn ve ruin 
Hieraeiitni Pi/ose lia 
Hieraeiiiin spp. 
Kiiaittia arveiisis 
Latliyriis pratensis 
Linaria vulgaris 
Lotus cornieulatus 



Melica nutans 
Moeh ringia irincrvia 
Myosotis arvensis 
Origanum vulgare 
Pliegopteris Rohertiana 
Phegopteris Drvopteris 
Piinpinella Saxi/raga 
Seduni alhuni 
Seduni niaxinnini 
Stellaria graminea 
Trifolium nwdium 
Trifolium pratensc 
Verbaseuni nigrum. 
Verbascum Thapsus 
l 'ieia eraeea 
J 'iola eollina 



Artslisten minder deh'is om den for Leinebakkene antorte og viser en 
række sydlige typer. 

Bækkedalens bund var tildels ganske fugtig. Her vokste spredte 
trær og busker: Alnus incana, Picea excelsa, Prunus Padns, Rliannnis 
I-^rangula. Bund\'egetationen bestod av mesofile, tildels hvdrofile græs 
os: urter. De kvantitativt fremherskende arter var folgende : 



Oxalis Acetosella 
Pteridium Aquilimnn 
Raniuiculus repens 
Rumex Aeetosa 
Stachys palustris 
Stellaria nenuirum 
Taraxacum officinale 



^ Aconitum septentrionale 
Acra cœspitosa 
Agrostis vulgaris 
Brunella vulgaris 
Circœa alpina 
Impatiens noli tangere 
Knautia arvensis 
Leontodon autunmalis 

Trifoliuni repens 

Tussilago Farfara 

Ulmaria pentapetala 

I ^aleriana sand>ucifolia 

Desuten fandtes nedenstaaende arter i mindre mængde: Achillea Mille- 
folium, Alchimilla vulgaris, Anthriscus silvestris, Artemisia vulgaris, 
Athyrinm Filix femina, Caltha palustris, Campanula rotuiuii folia, Carduus 
crispus, Carum Carvi, Ccrastiwn vulgatuni, Chrysantheimim Lencanthenmm, 
Cystopteris fragilis, Dactylis gloinei-ata, Efilohium niontamim, Festuca rubra, 
Fragaria vesca, Galeopsis tetrahit, Galium palustre, G. uliginosuni, Geum 
rivale, Glechoma hcderacea, Hieracinm Auricula, Mentha sp. (steril), Phego- 



58 



FU)I.K NOKDHAf.KN. 



M-.N. Kl. 



p/cris /xj/v/xjtfin/t/i s, /'/fU///i/^o i!i(t/(ir, I', iiirdia, I 'od ik iiinnilis, l'on pmlnisis, 
Polygonum i'ivi/uiniui, lùuiiiiicii/iis arcr, Nuiiirx Acrtosrlla, Rtihns idæus, 
Silriic vciiosd, Solmntin I )ii/((iiii(ini, Slr/lurid media, Stnithioplrris ger- 
iiiaiiicd, Tro/li ns niropaits, I rlicn dioicd, l'crhasciim nigrum, Veronica 
ofjiciudlis, I 'icid sipiiiiii, / 'iolii chiiuki, I '. F^iviniana, V. nmhrosa. Av nio- 
senc i hunden \ai" ('/iiiiuriiim (Irmlroiilrs vn av de mest fremtrædende. — 
Flere av de opregnede arter er iitvilsonit indkommet paa stedet med 
beitende kreaturer; skogen anvendes nemlig tildels som havnehage. 

P^orøvrig er planteveksten 
i Oier lite kjendt. De bratte 
styrtninger under Skarskampen 
like vest for kalktufifen opviste 
ganske mange interessante arter, 
tildels de samme som ovenfor 
er omtalt fra bækkedalens sider. 
Carcx miiricntn, Grum urhanum, 
Tiirritis glabra, Viscaria vi- 
scosa, 7'ri/iciim caniniim og Po- 
lygonum (liimelortim anføres her 
som et supplement til oven- 
nævnte liste. Forekomsten av 
Polygonum dumetorum \-ar gan- 
ske overraskende, da denne art 
tidligere ikke var kjendt længer 
nord end paa Ringerike i det 
østenfjeldske Norge. Langs 
kysten optrær den meget sparsomt inde i fjordene til nordsiden av Trond- 
hjemsfjorden (Fosen 63^50'). 

Efter disse orienterende bemerkninger om vegetationen paa kalktufifen 
og i dens nærmeste omgivelser, skal jeg gaa over til at beskrive de strati- 
grafiske forhold. 




Wà^&ê^f'î'^^a^i^^^^^S^^^^.: 



S^^-«=rr 



Fig. 16. Kalktuffen ved Gillebu (K) og dens forhold 
til gruskeglen ved Tingvold (G). Kl — Kl = kalktut' 
transportert ovenfra av bækken. Meget skematisk. 



B. Stratigrafiske undersøkeiser. 

Kalktuffen i Gillebuskogen er som ant\-det paa fig. 16 opbevaret paa 
begge sider av bækken i ca. 50 meters hoide over landeveien. Hoved- 
massen ligger i nutiden paa \-e.'itsiden av bækkedalen. Imidlertid maa fore- 
komsten i tidligere tider ha været meget storre; man finder nemlig tufrester 
høiere oppe i lien (antvdet paa fig. 16) og længere nede. Paa det nederste med 
K betegnede sted paa figuren ligger der dog bare lose blokker, saavidt jeg 
har kunnet se; disse kan tænkes at være fort ovenfra og nedover under 
en eller anden flomkatastrofe. Avstanden mellem den alleroverste fore- 
komst og den sidstnævnte noget tvilsomme blev ved maaling med staal- 
traadline fundet at være ca. 90 m., og avstanden mellem de mest perifere 



I92I.N0. g. KALKTUFSTL'DIER I GUDBRANDSDALEX. 59 

tufrester paa bækkens vestside og dens østside (niaalt tvers paa bækkens 
længdeakse) 30 m. 

Efter tuftens opdagelse har adskillige personer besokt findestedet, og 
efter ytringer som de har latt falde, s^'nes enkelte at ha dannet sig den 
opfatning av forholdene, at „der er avsat litt tuf paa begge sider av 
bækken". Imidlertid vil den folgende fremstilling vise at det her ikke er 
tale om to eller flere selvstændige tufavsætninger, men om restene 
av én, stor sammenhængende k a 1 k t u f, saaledes som O vex allerede 
fra forste stund av var klar over. Dog maa man selvsagt finde sig i at 
der blir avkrævet én absolute, haandgripelige beviser naar man fremsætter 
en saadan antagelse. I denne henseende er nedenstaaende profil av av- 
gjorende betydning. 

Profil I paa bækkens ostside. 

O VEX anforer herfra et profil, som o\"er morænegruset i bunden viser 
tre tufforende horisonter. I det overste lag fandtes et stykke med furu- 
naaler, men „paa grund av et antal meterstore blokker stikkende op av 
morænegruset, blir stillingen av serie 2 og enkelte lag av serie 4 tildels 
noget dubios, idet de begge antar paa sine steder karakteren av koncentriske 
kalktufkruster omkring blokkeoverflaten" (Øyex 1. c. p. 287). For Ovex og 
Holme var dette profil meget avgjorende, idet de nemlig paa grundlag av 
de i gruskeglen ved Tingvold (cfr. fig. 16) fundne lose tufstykker nødven- 
digvis maatte trække den slutning, at baade et bladtufkompleks og et furu- 
tufkompleks maatte eksistere in situ hoiere oppe i dalsiden. Bladtuffen fandt 
de med engang, men av furutuften saaes bare det ene stykke som er om- 
talt ovenfor. 

Den opgave som undertegnede stillet sig, var da for det forste at 
grave op et stort og helt tydelig profil paa denne side av bækken, og 
dernæst at lete efter mulige iurutufrester. Heldigvis var der like for min 
ankomst til Oier opkastet en groft paa stedet for at faa konstatert om 
tuften var saa betN-delig at den hadde nogen økonomisk betydning. Med 
utgangspunkt i denne grøft fik jeg oparbeidet et meget vakkert profil, som 
hadde folgende utseende: 

1. Grovt morænegrus med større og mindre blokker av kvartsit, 
skifre etc. Midt i profilet laa 2 store avrundede blokker. Gruset var 
brunt og gjennemgaaende sterkt oks y der t. 
II. Bladtufkompleks, tilsammen 45 — 50 cm. 

A. Underst kom en 10 — 20 cm. sterkt jernholdig, rustbrun til choko- 
ladefarvet tuf, som var tN'delig lagdelt og temmelig los og sprød. 
Paa tuftens underflate masser av Hi/>pop/iacs-h\ader, ellers hjcrk- 
og vSr/Z/A'-blader i mængde. 

B. Kaotisk, foldet og valket graagul bladtuf, 20 — 25 cm., med 
mængder av hjcrkcbladcr, poros og lakunos og meget los. 



6o 



KOM-- .\oKi)H.\f,(:N. 



M.-N. Kl. 



C. Utprægct lai^ïlclt, fossiHatti,!^, ,i;iil lul, 5 8 cm.; tcnimclig kompakt 
med smiiki loldidc la^, men sprnrl oi;; let at trænge igjennem. 
Lagene kan let hj.snes Ira liiiiandi n og viser sparsomme bjerke- 
blader fß. odorala) paa llatcn. 

III. Jordstripe, 5-8 cm., hcstaaende a\- binnlig, fin tutjjlandet jord uten 
haardeiT kliim[:)er. Optil \'ar den iiiidcitidcii mere graahvit av farve. 

IV. I' II r u t u Ik o m p 1 e k s, ca. 30 cm. 

i\. j ,5 3 cm. smukt laget, gul, fossillVi tuf; knækkes let over med 




Fig. 17. Profil fra bækkens ostside, Gillebu. Nederst moræne, derover bladtuf. Korsene 
angir jordstripen. Overst furutuf og muldlag. (Cfr. beskrivelsen i teksten». August 19 18. 

Nordhagen fot. 



fingrene. Over denne fandtes en uhyre tynd (0,5 cm.) brunlig 
stripe. 
B. 25 — 27 cm. furutut, nedtil eiendommelig konglom er at agtig med 
bittesmaa gronblaa, avrundede kvartsitbiter, sammenkittet med smaa 
tufbiter, samt furunaaler, der var fossilificert som tynde ror eller 
dobbeltror (naalepar). Derover kom meget los furutuf med 
masser av naaler, en kongle, blader av tyttebær (sparsomt) samt 
bjerkeblader. Tuffen var tildels bolget-knudret og drN'pstensagtig, 
altid meget let, poros og sprod. Paa de fleste punkter var 
komplekset s a m m e n \- i t r e t til et lost, smuldrende 
kaos a \' t u f s 1 3' k k e r. 
V. 15 cm. muldjord; øverst 3 cm. humus av Hylocomier. 



1 92 1. No. g. 



KALKTLTSTUDIER I GL'DBRAXDSDALEN. 



6i 



Fig. 17 viser en del av profilet. Den underste papirremse avmerker 
„jerntLifFens" grænse mot det underliggende morænegrus; den utfylder 
ogsaa mellemrummet mellem de to store blokker og kitter disse fast 
sammen. Papirkorsene er fæstet i jords tri pen mellem bladtuften og 
furutuften. Mellem kors nr. 2 og 3 fra venstre markerer en tynd horisontal 
papirstripe den „knækkelige" tuf underst i furutuf komplekset. 

Av billedet og den skematiske figur 18 ser man at alle tuflagene 
hel der os to ver, hvor de desuten blir smalere. Desuten skraaner 
alle lag ganske sterkt ret i mot iagttageren. Disse stratigrafiske 
eiendommeligheter beviser at tuffen umulig kan være avsat av en 
kilde fra ost eller nordost. Den eneste mulige tvdning er den, at 




T<7 



. - //TV. 



Fig. 18. Pro'il fra bækkens ostsid;?, Gillebu. I = I\Iorænegrus med blokker. II = bladtut- 
kompleks. III = jordstripe. IV = furiitufkompleks. Overst raahumus og muld. 



lagene engang har fortsat s a m m e n h æ n g e n d e o \- e r t i 1 \- e n s 1 1" e, 
urglasformig, men senere er blit bortcrodert av bækken. Dermed er 
sammenhængen med tufmassene paa bækkens vestside git, og vi kan med 
sikkerhet paastaa at Gillebutuffen oprindelig har ligget som en skjoldformig 
masse over det sted hvor nu bækken har sit lop. 

Profilet viser ogsaa at furutuften er ytterst forvitret og daarlig opbe- 
varet. Dette fænomen er endda mere fremtrædende paa bækkens vestside. 

Profil II. \'estsiden. 

Profilet er optat xis à vis 1, men noget mere mot nord, og like i 
randen av bækkedalen. 
I. Grovt oksydert grus. 
11. Morkebrun, jernholdig tufjord, 4 cm. Denne sone manglet hvor der 

laa storre stener i bunden. 
III. A. Jerntuf, ca. 36 cm. Xedtil \ar den sterkt brunfarvet og hadde 
en slaaende likhet med utbræ^ndt koks. Mange blader av Hippo- 
pliars, Biiiila oaorafa, I'(p/////s trnnula, Sali.x sp. 



62 ROLF NORDHAGKN. M.-N. Kl. 

r>. Ilaard hladliif, 15 ein., i midti n iioksaa uregelmæssig, nicn optil 
o^r ncdtil siniikl la,t,Mt. Hen \ai" j(\nt toiijiind« t incfl forcgaaende. 

IV. Los tuf jord, ca. 8 cm., med lose biter av en skjnr^ laget tuf. 
V. Muld og raahumus, 15 cm. 

I dette |)i"()lil ci- hladluffcn \akk(i"t iil\iklet og stemmer udmerket med 
foregaaende prolil. I'"ui"iitiirfeii drriinot er helt frjrsvundet paa dette sted. 

Profil III. Vestsiden. 

l'rofiKt ligger ca. 8 111. fjernet fra foregaaende og ret i vest for dette. 
1. Grovt grus, med knytnævestore avrundede stener, sammenkittet av 
kalktuf til en samnienhaMigende meget haard masse. Det kunde for- 
folges til 40 cm. dyp. 

II. Derpaa fulgte et kuriost brunt lag, bestaaende av tæt sammenfiltrede 
levende og dode rotfibre. Det kunde Hækkes av som en filt, 0,5 cm. 
tyk. Aarsaken til lagets forekomst maa sokes i det underliggende 
ugjennemtrængelige lag, som tvinger den recente vegetations rotter 
til at torgrene sig horisontalt langs lagets overflate. 

III. Bladtufkompleks, tilsammen 20 — 25 cm. 

A. Los jordagtig tuf, 5 cm. 

B. Bladtuf, los og sprod med bjerkeblader i stor mængde, 10 — 
15 cm. 

C. Overst 3 — 5 cm. lagdelt sone, som ligner II. C i profil I. 

IV. Tufholdig jord, 15 — 20 cm., med talrike tuf biter fra nottestore til 
{15 :^^< 10 ;< 7) cm.^, med sparsomme funiiiaalcr, især nedtil, dog i et 
par stykker i mængde. Desuten rester av en laget, knækkelig og 
sprød tuf Laget hadde fleresteds karakteren av los jord fra overst 
til nederst. 

V. Muldjord, 10 — 15 cm., med skiferbiter. Overst 3 — 4 cm. raahumus 
av Hylocomier. 

Profilet var helt igjennem temmelig løst og gjennemsat av trærotter. 
Interessant var det at finde tlirutufstykker paa dette sted; de var dog meget 
forvitret og smaa. Også a bl ad tuffen svnes at kile ut vestover. 

Profil \\\ Vestsiden. 

Omtrent 1,5 m. nordvest for foregaaende profil kiler tuffen helt 
u t. Her saaes folgende serie : 

I. Grovt, stenblandet grus med a\-rundede stener, brunlig av farve 

(oksydert). 
II. Brun tufjord, 5 cm. Fin og jevn. 

III. Furutuf, 3 — 4 cm.; dannet en fast plate med tydelig lagning. 

IV. Løs tufjord, 15 cm., med spredte furutuf biter. Rikelig med furunaaler 
samt et aspeblad. 

V. Muldjord med skiferbiter, 25 cm. Overst som vanlig raahumus. 



1 92 1. No. 9. 



KALKTL'FSTL'DIER I GUDRRAXDSDALEX. 



63 



Dette utkilingsprofil er meget interessant. Bladtutfen er her forsvundet, 
og underst findes bare brun tuf jord. Derimot er furutuffen meget tydelig 
opbevaret i sammenligning med andre steder i nærheten, og viser at av de 
to hovedlag i tuften, har furutuften transgrediert længst utover til siden. 
Profilet viser (i likhet med de ovrige) at furutuften hviler diskordant oven- 
paa det underliggende; ti der er ingen overgang mellem bladtuf og furu- 
tuf. Den brune jord paa bunden er muligens forvitret bladtuf, men svarer vel 
snarere til jordstripen i profil I. 

Profil V. Vestsiden. 

Dette er optat omtrent 

3 m. sondenfor profil III og 

like i kanten av bækkedalen. 

I. Grovt grus med storre 

og mindre stener, oksy- 

dert. 

II. Bladtufkompleks. 

A. Jerntuf horisont, 3 — 
4 cm., mørk av farve, 
tildels sandet. Læn- 
ger vest i grøften var 
den 10 cm. og jevnt 
laget. 

B. Eiendommelig laget, 
haard tuf, 100 cm., 
avsat skalformig om- 
kring storre og min- 
dre stener. Cfr. fig. 19. 

III. Muldjord med stener og 

smaa forvitrede tufbiter, 

23 cm. 

En speciel interesse knytter sig til lag II. B. Det synes at bestaa av 

fysikalsk utfældt, næsten krystallinsk tuf, avsat kuleskalformig omkring runde 

stener, og viste en meget fin lamellert veksling av rodbrune og gulagtige 

lag, som Uten tvil er uttryk for en viss periodicitet i avsætningen. Det er 

mulig at vi her befinder os i nærheten av det oprindelige leie for den kalk- 

avsættende bæk, eller rettere sagt: der hvor strommen i denne hai- været 

sterkest. 





IV*;' 



■K 



Fig. 19. Profil \' paa ü.Lkk^iis xisiside. Den eien- 

dommelige konglomeratagtige tuf, avsat skalformig 

omkring stenene, sees midt i billedet. August 19 18. 

Nordhagen fot. 



Profil \'I. Vestsiden. 

Længer vest i samme groft, men samtidig noget lavere i terrænget, 
saaes følgende lagrække : 

I. Stenblandet grus som i foregaaende profil, med fiere svære blokker. 



64 F<OLF NORDHAGEN. M.-X. Kl. 

II. lil.uluil'kniiipli ks, tilsainmcn ca. 50 cm. 

A. ji iiiliif, 10 cm., typisk, u]r(\ / ///'/)n/)//(irs og hjcrkrhldd« r. 
V). Laget tuf, 20 cm., jcvnt IVnijiindf 1 med forcgaacnde. 

C. Los Ijladtuf, 15 20 cm., tildels rent kaotisk med masser av bjerke- 
blader, desiiteii sparsomt I lipfuifïhdi's og Sn/i'.v capraa. Mellem B. 
og C. fandtes ganske lokalt en 2 cm. tyk stripe a\' tuflere; den 
forsvandt til begge, sider. 

D. Øver.st fandtes en ganske tynd (0,5 cm.) lagdelt sone belagt med 
chokoladefarvet overdrag. 

111. Tuf|ord med lose tuf biter, hvori en og anden furunaal, desuten 

stykker av en noget lagdelt, sandet tuf. 
1\'. Muldjord, 20 cm. 

Profilet er alt i alt helt normalt, •^: det stemmer med de foregaaende. 
Det gik kontinuerlig over i det sidst beskrevne (Vj, saa horisonten IL B i 
dette er bare at opfatte som en speciel horisontal facies av bladtuf- 
komplekset. 

C. Oversigt over Gillebutuffens stratigrafi. 

Fig. 20 viser et svakt skematisert tverprotil omtrent lodret paa bækkens 
længdeakse, altsaa V — O. Det er basert paa profilene I, II, III og I\'. 

Moraiic gruset. 

Tuffens underlag bestaar av et stenblandet grovt grus, som undertiden 
forer storre blokker. Det gir indtryk av at være vandslitt og noget ut- 
vasket. Ellers frembyr det ingen specielle træk av interesse. Det maa 
oprindelig ha dannet en sammenhængende masse der hvor bækkedalen 
senere er skaaret ned (1 paa fig. 20). 

I sin overste del er gruset meget sterkt oksydert, rustbrunt til 
chokoladefarvet; det maa altsaa ha ligget aapent for forvitring i tidsrummet 
mellem sin avsætning og dannelsen av de forste tuflag. I flere profiler \ar 
gruset overst sammenkittet av utfældt kalk til en konglomeratagtig 
masse. 

Bladtnfkniuplcksci. 

Som ovenfor antydet, er de forst avsatte tuflag ytterst karakteristiske. 
De er mørkfarvede, rustbrune til chokoladefarvede, undertiden svakt blaalige, 
hvilket beror paa kalkens sterke forurensning med jernforbindelser. „Jern- 
tuffen" viser altsaa at da kildene paa stedet begyndte sit lop, indeholdt 
jordbunden store mængder av oksvdations- og forvitrings- 
produkter, som utfældt es sammen med den kul sure kalk. Jern- 
tuffen ser i sin mest ekstreme form ut som gammelt rustent jernskrap eller 
koksslagger, og blir op til 30 cm. mægtig. Den er tydelig skiktet og danner 



1 92 1. No. g. 



KALKTLTSTUDIER I GUDBRAXDSDALEX. 



65 



hyppig kruster omkring morænegrusets blokkei*, og \iser derfor en meget 
valket overflate. 

Naar man bryter tuften los og faar væltet stykkene om, saaledes at 
undertlaten blir fri, blir man overrasket ved at Hnde avtryk av Hippop/iars- 
blader hobet paa hinanden i store mængder. Jerntuffen viser os et 
avtryk a \- bakkens o \- e r f 1 a t e \- e d t u f a v s æ t n i n g e n s begyn- 
delse. Desuten sees blader av Brtiila odorafa og Saliccs, hvorav en del 
vakre avtrvk vistnok tilhorer Salix pliylicifolia, men ogsaa andre arter 
synes at \-ære repræsentert, f. eks. 6". capræa. Imidlertid er S<^7//.v-blader 
overordentlig vanskelige at bestemme i fo35il tilstand. Talrike mellemformer 
av h\-brid natur kjendetegner jo specielt denne siegt, som til og med s^p^- 



M 



L 4 ^. i # 




i- 3 t r 



Fig. 20. Tverprofil av Gillebutuifen. I = morænegrus. II = bladtiifkompleks med jerntuften 
nederst. III = furutut". Høiden er overdrevet to ganger i forhold til horisontalutbredelsen. 



cialistene har vanskelig for at mestre. Det er indlysende at artsbestemmelsen 
for de fossile Sali ces' vedkommende maa bli endda mere dubios, da 
alle de saakaldte „habituelle" kjendetegn (farve, behaaring, \'oksemaate, bla- 
denes konsistens etc.) er meget utvisket eller mangler. 

Jerntuftens underside har undertiden en hoist eiendommelig traadet 
struktur, som minder svakt om mosetuf, men som muligens ogsaa kan 
skyldes blaagronne alger. Nogen tydelig mosetufbænk saaledes som ved 
Leine foreligger ikke; avsætningen har helt fra begyndelsenav sedimen- 
tært præg paa de fleste steder, og turde være fysikalsk utfældt i stor ut- 
strækning. 

Over Jerntuften folger oftest vekslende tuflag av vanlig graagul farve, 
hyppig sterkt kaotiske, med lovblader fossilificert i alle mulige planer. 
Mange blader viser sig at være forkalket i helt fri stilling, o: det oprinde- 
lige blad er helt erstattet av kalk, som fuldstændig har antat bladets form, 
nervatur og tanding. Visse partier a\- bladtuften maa derfor sandsynligvis 
være dannet meget raskt, idet man \anskelig kan tænke sig en saadan for- 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 9. 5 



66 ROLF NORDHAfir.N. M.-N. Kl. 

kalket ii()i(l( iillii; Madiiiassc, (\l\- \is',r ricn storstc liklut merl netop ncd- 
faldt-nt, rntct Im- om hosten, som en sekulær avsætning'. I denne blad- 
masse finfler man hist o«^ her mindre, lose moset ufparti er, over- 
ordentlig vakre, merl frie, oitotif)pe skud (op til 5 cm. lange). Ogsaà disse 
lag synes at være dannet tcimiiclig raskt. 

0\rM\ avskittes komplekset som oftest, og specielt i de mindst for- 
vitrede deler av tuffen, med en ut præget lag'delt sone, som viser 
tydelig og vakker lamellering, et sikkert vidnesbyrd om periodicitet i av- 
sætningen. Lagene er tildels meget tynde, og denne ovre sone synes i 
motsætning til den underliggende lose tuf at være avsat gjennem længere 
tidsrum med noget svakere vandforing i bækken end tidligere. — Hippophaës 
rliainiioidcs Hndes spredt gjennem hele bladtufkomplekset og er temmelig 
almindelig, men bladaxtrykkene er ingensteds saa talrike som paa jern- 
tuffens underside. Som helhet betragtet er bladtufkomplekset uhyre trivielt 
og artsfattig i palæofloristisk henseende, mens kvantiteten derimot er desto 
mere fremtrædende. I denne henseende er overensstemmelsen med Leine- 
tuffen slaaende. Betula odorata, Hippophacs }-/iain/ioidrs, Popiihts tromda, 
Salix capræa, S. phylicifolia, Salix cfr. nigricans samt en eller et par mos- 
arter — dermed er materialet uttomt. Chancene for ved nye gravninger 
at finde flere arter er uhyre smaa; ti baade Øyen og Holme og under- 
tegnede har undersokt hundredevis av haandstykker uten at bringe nye 
former for dagen. Heller ikke de paa sekundært leiested (i Tingvoldgruskeglen) 
fremfundne typiske bladtufrester gir nye oplysninger om floraen paa den 
tid. Oyen anforer herfra (1. c. p. 273) bl. a. Vacciiiiinu vitis idcea og Almis 
iiicaiia efter Dahls og undertegnedes bestemmelser. Jeg har senere gransket 
disse stykker nærmere og tror at vi gjor rettest i at sætte den sidstnævnte 
art ut av betragtning, da bestemmelsen nu, efter at jeg har hat anledning 
til at gjennemgaa mit eget store materiale og svenske samlinger, forekom- 
mer mig ganske tvilsom. 

Ovenstaaende skildring vis£r det typiske bladtufkompleks' utformning. 
Imidlertid moter der os adskillige horisontale facies. Av disse er den kon- 
glomeratagtige tuf i profil \' tidligere omtalt. De eiendommelige sfæriske 
kalktufskaller med den fine lamellering og den likefrem krystallinske struktur 
tyder paa fysikalsk utfældning i strømmende vand-. Midt inde i saadanne 
agat-lignende kuler har jeg paa lagflatene fundet bladavtryk av Hippophacs 
og hjcrk. Hvordan den gjentatte veksling av rodbrune og gulagtige lag skal 
fortolkes, tør være tvilsomt. Man kunde tænke paa vaarflom i bækken led- 
saget av sterkere avsætning (gulagtige lag) og en trægere utfældning ut- 
over sommeren (brunlige lag). Men lagene varierer saa sterkt i tykkelse at 
jeg ikke vover at paastaa noget med bestemthet. — Ogsaa i andre profiler 



• ClV. Leinetuffen, hvor bladtuffens karakter overalt var en helt anden. 

2 Da de runde stener i laget ligger helt omgit av tuf, kunde man være tilboielig til at 

tro at bækken har revet stener med sig ovenfra i flomtiden og avsat dem længer nede. 

Ogsaa dette moment peker hen paa at \i her befinder os nær bækkens centrale del- 



1 92 1. No. 9. 



KALKTUFST'CniER I GL'DBRAXDSDALEX. 



67 



meter man antvdning til lignende forhold, f. eks. paa det alier overste sted 
i bakken hvor der anstaar tuf. Her er lagene sterkt forvitret, men den 
samme kuleskalstruktur (omkring runde stener! er her smukt utviklet (fig. 21). 

I skraaningen paa bækkens vestside, like nedenfor profil II, hadde jern- 
tuffen en besynderlig breccieagtig struktur, idet den bestod av et utal 
av smaa tufbiter, som atter var sammenkittet av utfældt tuf. Hvordan dette 
fænomen genetisk set skal forklares, er meget usikkert. Man kunde tænke 
sig glidninger, opknusninger og forskyvninger i den forst avsatte tufmasse, 
kanske ogsaa sprængninger foraarsaket av isdannelse. Enkelte partier 
av bladtuften minder ogsaa 
sterkt om cyanofycé-tuf. Men 
desværre vet vi endda saa 
overordentlig litet om sam- 
menhængen mellem kalktuf- 
fenes dannelsesmaate og 
deres struktur, at en gene- 
tisk klassifikation forelobig 
er utelukket. Serxaxders 
interessante „genetiska bi- 
drag" (19 1 6 1. c.) er i vir- 
keligheten det eneste sikre 
som hittil er fremkommet i 
skandinavisk litteratur ved- 
rorende dette overordentlig 
\igtige emne. \'i maa derfor 
forelobig finde os i at an- 
vende mere eller mindre ufuld- 
komne, morfologisk-petrogra- 
fiske analogibetegnelser som 
„cinter-agtig", „drypsten-ag- 

tig", „breccie-agtig" tuf o. s. v.^, som i virkeligheten ikke sier noget ab- 
solut sikkert om dannelsesmaaten. 

Paa den skematiske figur 16 er ogsaa avmerket det aller nederste sted 
i bækkedalen hvor kalktuf er paatruftet. Her fandtes bladtufblokker i 
jorden, som var optil 25 cm. tykke og sterkt forvitret i overflaten. Tuffen 
\ar optil smukt skiktet med talrike store blader a\- Sn/ix caprcco, desuten 
bjirk og Hippophacs, men \ar dehis mere kompakt og haard end ellers. 
Kn blok viste tydelig mosetufstruktur paa undersiden. Som tidligere antydet 
foreligger der den mulighet, at disse blokker her befinder sig paa sekundært 
leiested og egentlig horer hjemme hoiere oppe i lien. 




Fig. 2 1 . Fint skiktet kalktuf, avisât skalformig omkring 
en sten. I Det undre lamellerte stykke har sittet uten- 
paa en anden sten, horer altsaa ikke sammen med 
det ovre.l. ' 1. B. Larssen fot. 



Jordsfripr/i. 
Denne er bare tydelig opbevaret paa bækkens ostside, Inilket hænger 
sammen med at furutuffen her ogsaa er bedst vedlikeholdt. I de ovrige 



1 Cfr. ØYEXS „konkretionsagglomerat" etc. (Naturen 19 181. 



68 KOrj- NOKDIIAf.KN. M.-N. Kl. 

prolilci' ir liiiiiliirti 11 s:i;i loiviiid ;il iii.tii oj>p;i;i Madtuftcn kun (inder et 
tykkere elli r tyndere jordlaj^ inc-fl stumper og stykker av furutuften; og det 
er da ganske klail al dt n oprindelige lagrække: bladtuf jordstripe — furutuf 
er utvisket. 

Pi-()lil I er derldi- sarlig \a-i-dit'ii!ili, idel det \-iser os en tydelig dis- 
kordans nullein de to hovedlag. Der er ingen overgang at spore. Lik- 
heten med Leinetiift'en er for saavidt slaaende. 

Imidlertid linder \i ved Gillebu ingen spor efter nogen Dryasforende 
horisont. Og der blii' lullei- ikke, som profil I \iser, nogen plads for en 
saadan, idet baade bladtuffens avslutning og furutuftens begyndelseslag er 
meget karakteristiske og let kjendelige. Oven antvder i sit arbeide (1. c. 
p. 284) at der muligens kan ha været avsat Dryastuf ved Gillebu, da han 
i det for omtalte furutufstykke paa bækkens østside fandt et blad der mindet 
om Sa/ix rrficit/afa. Det er senere ødelagt under transporten, da det var 
meget skjort. — Imidlertid har jeg selv baade i bladtuffen og furutuffen 
fundet lignende smaa rundagtige 5'<'r//.v-blader, som ikke er Sa/i.v reticulata, 
da nervevinklene og strukturen er helt anderledes end hos denne art. Des- 
uten var de altfor tynde og sprede ; Salix reticulata er altid meget solid 
og grov selv i fragmentariske a\tr\k, paa grund av sine tykke læragtige 
blader. Jeg har som sagt ikke set spor a\- denne art eller Dry as i nogen 
av de profiler som jeg meget omhyggelig har studert. Og profil I viser 
at der ikke er plads for nogen Dryastuf, da jordstripen kommer ind i dens 
sted. Som vi senere skal se, er Dryastuften som saadan et lokalfænomen 
eiendommelig for Leine. Det generelle moment d e r i m o t ligger i 
diskord an sen mellem bl ad tuff en og furutuffen. 

Fiinitiiffcii. 

Denne tuf er paafaldende los og sprod, hvilket utvilsomt er aarsaken 
til at den er saa sterkt forvitret, hnidlertid har selvsagt forvitringen bi- 
dradd sterkt til at gjore tuften endda mere skjor og pores end den var 
oprindelig. 

Komplekset begynder med en ganske specifik dannelse, nemlig den 
fossilfrie, knækkelige tuf, som utgjør en egen- plate underst. Rester av 
denne er ogsaa lundet paa bækkens vestside. — I profil I er som nævnt 
furutuffen tildels konglomeratagtig med tilblandede smaa runde stener, hvilket 
kan tyde paa sterk vandforing i bækken. Enkelte partier var sterkt knudret 
og valket, tildels avsat som cylindriske ror omkring kxister og pinder, der 
dannet t3'nde kanaler gjennem tufmassen. 

Artslisten er meget beskeden: Piiius silvestris (naaler, en kongle, 
grener, bark etc. I, hjcrk (mange fragmentariske blader, hvoriblandt sikker 
B. odorata), Poptilus trciiiiila, Salix sp.', / 'acciiiiimi vit is ida-a (sparsomme blader). 



' ( )yf,n anforer Sa/i'.v atnita fra uirututst ykker paa sekundært leiested ved Tingvold, 
efter Dahls og undertegnedes forelj'oige bestemmelse. Denne er muligens rigtig, men 
ikke sikker. 



I92I.X0. 9- KALKTIFSTUDIER I GL'DBRAXDSDALEX. 69 

Fra gruskeglen ved Tingvold har Oven og Holme plukket frem en 
mængde lose furutufst3-kker, som jeg ogsaa har, hat anledning til at under- 
søke. Men de bringer ingen nye arter for dagen. Et interessant 
faktum er det at Hippophaës ikke er i agt tat i et eneste fu ru- 
tufs tykke, hverken fra Gi 11 eb u tuffen eller Ti ngvoldgr us- 
tak e t. Den s \' n e s at \- æ r e u t d o d p a a stedet i tidsrummet 
mellem bladtuffens avsætning og furutuffens begyndelse. 
Jeg kommer senere tilbake til dette vigtige punkt. 

Profilene paa vestsiden synes at antyde at furutuften har transgrediert 
utover til sidene noget længer end bladtuften. Antageligvis har hele 
komplekset oprindelig været ganske betydelig ogsaa i vertikal retning. 

Hvad der folger efter furutufkomplekset, om der oprindelig har været 
avsat nok en tufbænk i Gillebuskogen, saaledes som vi saa ved Leine, 
derom kan vi forelobig ikke opgjore os nogen mening. Det er imidlertid 
klart at bækken maa ha skaaret sig ned gjennem kalktuflfen først efter at 
furutuften var dannet. Nye holdepunkter for bedømmelsen faar vi først 
efter at ha undersokt gruskeglen ved Tingvold. 



D. Gruskeglen ved Tingvold. 

Da grustaket ved Tingvold er indgaaende undersokt og beskrevet av 
Oyex (1920 1. c), skal jeg her bare gi en kortfattet oversigt over de 
stratigrafiske forhold paa dette sted. 

Av den skematiske tegning fig. 16 og fig. 15 faar man et nogenlunde 
korrekt indtryk av gruskeglens beliggenhet i terrænget. Den har sit ut- 
gangspunkt i skogbrynet tilvenstre for det lille huset paa fig. 15 og brer 
sig herfra \-ifteformig ut over det lavereliggende terræng. Avstanden ned 
til skigarden er ca. 80 m. Bækken har nu gravet sig et litet leie norden- 
for grustakets centrale del, og selve keglens overflate er i stor utstrækning 
opdyrket. 

Fig. 22 viser et snit gjennem lagrækken like i overkanten av chausséen 
ved Tingvold landhandleri. Mægtigheten er her tilsammen 8 à 9 m. over 
veiens nivaa. 

I. Paa bunden ligger et tykt lag av morænegrus med større og 

mindre blokker. 
II. Deroxer kommer en mægtig fluvioglacial grusav leiring med 
skraatstillede lag, som helder 10 — 15^ mot ost (Oven I. c. p. 275 fig. 4). 
III. Diskordant avsat herpaa folger en' gruskegle . med svævende 
lagstilling, ca. 3 m. mægtig. Oven adskiller her en hel suite 
med underavdelinger, som dels har karakteren av utvaskede blokke- 
lag, dels mere normal gruskarakter. Avsætningen gir forøvrig et 
noksaa rotet indtrN-k. I dens ovre halvdel var det at Holme oprin- 
delig fandt de løse tufstykker. Man finder her en ganske 
distinkt horisont, som i keglens centrale del har karak- 



70 



ROLF NORDMAGEN. 



M.-N. Kl. 



tcrcn av el d o h li i 1 1 1 a j.^, iiu.d talrikc storrc n<^ iiiindre tufstykkcr 
og los tiit'jord 1111(1 iill)laii(l<l niiildjonl (K K paa figuren). De en- 
kelte tuflag som \i stiftet hekjendtskap med i Gillebutufien, er rikt 
repræsentert. Dit Kkkedes Ovkn og Hoi.mk at fremfinde ikke mindre 
end 67 mindic hlnkker a\ f 11 r 11 1 11 f fe 11, hvilket gir et tydelig vink 
om at ogsaa denne oprindelig niaa ha været ganske mægtig in situ. 
— Hosten 1920 viste sig at være ganske heldig for studiet av denne 
horisonts karakter i den nordvestlige del av gruskeglen, idet der var 
foretat storre gravninger og kjort \'æk betydelige grusniasser. Vro- 
filet saa her saaledes ut ovenfra og nedad (cfr. fig. 23 1 : 





- 


'^":<s. 


;o/. 


.0 


0.. 


-- 


C5 




-0 

a 
0. 










c 






Fig. 22. Snit gjennem grustaket ved Tingvold. I = moræne. II = fluvioglacialt grus. III = 

gruskegle med svævende lagstillinf. K — K = gruslag med kalktufstykker og muld. Skematisert. 

(Efter Øvens og ea;ne undersokelserl. 



A. Muldjord og raahumus, 20 cm. 

B. Gruslag, 80 cm., noget varierende. 

C. Kalktufførende lag, tilsammen 20 — 25 cm. (tig. 23). 

1. Smal mørkebrun muldlignende stripe. 

2. Et par cm. blekgraa tutjord. 

3. 5 — 8 cm. brungraa tutjord. 

4. 5 — 8 cm. tufgrus, med masser av smaa tufstykker og kalk- 
smuler. 

5. 8 cm. brunlig tuf^jord. 

D. Grovt stenblandet grus, overst noget sammenkittet av utfældt kalk, 
fortsatte vistnok ca. 1,5 m. nedover. De dypere lag var over- 
dækket. 

Paa dette sted var altsaa den kalktufforende horisont ganske kom- 
plicert bvgget, men ikke særlig mægtig. Forholdene stemmer for- 
øvrig med Oyens beskrivelse; han omtaler nemlig at horisontens 
dobbeltkarakter utviskes til sidene 11. c. p. 281 1. 



I92I.N0. 9- KALKTUFSTUDIER I GUDBRAXDSDALEN. 7I 

Over kalktullaget folger, som profilet viser, atter gruslag. Fleresteds 
var der antydning til en podsolagtig rustjordstripe akkurat i overkanten av 
tuflaget; det saa ut som om jerntbrbindelsene var utfældt netop i grænse- 
sonen (cfr. fig. 23 tilvenstrel. Længer ost i grustaket var dette mindre iVem- 
trædende; dog var der her ogsaa tydelige podsoleringsfænomener i de ovre 
gruslag. 

Denne gruskegle er et overmaade \-ærdituldt supplement til Gillebutuften, 
og tilsammen gir disse to geologiske dannelser et meget interessant billlede 
av forholdene. 

y\ ser hvorledes bækken ved tilbakeskridende erosion litt efter litt 
har arbeidet sig ned i dalsidens dække og avlastet materialet længer nede 
i det flatere terræng i form av svævende lag ovenpaa det fluvioglaciale grus. 
Sluttelig har den ogsaa begyndt at angripe kalktuften, og har i tiden 
efter fu ru tuffens dannelsesperiode ikke alene evnet at sage sig 
ned gjennem de forskjellige tuflag og tjerne disse o\-er store strækninger, 
men har ogsaa gra\et sig dypt ned i det underliggende morænegrus, som 
vi gjenfinder alier overst i gruskegiens svævende lag. — Nu for tiden er 
bækken helt ubetydelig og synes ikke engang i flomtiden at foranstalte 
nogensomhelst katastrofer. I torre somre torker den næsten helt ind. 
G r u s k e g 1 e n s dannelse f o r u t s æ 1 1 e r en ganske anderledes 
stor vandforing i bækken. 

Naar alle momenter tages i betragtning, kommer man for bækkens ved- 
kommende til det konsekvente resultat, at den likesom kilden ved Leine har 
været intermitterende, med tre tydelige ops vul mingsperiode r 
av sekulær natur: i) under bladtuftens dannelsestid, 2) under furutuffens 
tid, 3 1 under kalktuffens nedbrytning og bækkedalens utformning paa finde- 
stedet. Det er ialfald et ganske respektabelt arbeide som bækken har præ- 
stert i tiden efter furutuffens avslutning (cfr. fig. 2 ol. 

Mellem bladtuften og furutuften ligger der, som vi for har set, et tydelig 
a\brud i bækkens virksomhet (Jordstripen). At der mellem furutuffens tid 
og den sidste „erosionsperiode" ligger et lignende avbrud, er meget sand- 
synlig. O YEN fremhæver forekomsten av betydelige mængder muldjord 
i gruskeglens kalktuftbrende horisont som et tegn paa at furutuften efter 
sin dannelse maa ha været mulddækket og faat tid til at hærdne (1. c. p. 272'. 
Ialfald har vi her et moment som maa tages i betragtning under diskussionen. 

Gillebutuftcn mangler oiensynlig en tredje tufhorisont (tilvarende til 
^/««s-tufien ved Leine). Lalfald er dei- hittil ikke fundet spor efter en 
saadan. Men til gjengjæld har man her en erosions- og akkumulations- 
periode. Dette forhold kan muligens bero derpaa, at bækken ved Gillebu 
har været stridere og vandrikere end kilden \ed Leine. Imidlertid kan 
aarsaken vel ogsaa ligge i en viss forskjellighet i sehe det kalkydende 
substrat. Bækken ved Leine faar sin kalkgehalt fi-a moræneleret, og dette 
er endda ikke uttomt. \'ed Gillebu derimot kan neppe det utxaskede dal- 



72 



Roi.K .\()i<i)ii.\f;i:.\. 



M.-N. Kl. 



sidegrus iiiidcr turtcn og hoiirc opi«- ha Icvcrt sari store knlkmængder som 
Gi]lel)iitiit"fcns opstaaen lorutsætter. Ikekkcii har vel snarere erhvervet sin 
kalkholdighct under passage gjennein kalkrikc bergarter h(;iere oppe i dal- 
siden, og det er tænkl)ait at den til slut li a r saget sig ned gjennem 
disse og saaledes for f'rciiiliden er blit berovet den oprindelige tilgang paa 
kulsur kalk. En noiagtig undersokelse av bækkeleiet mellem Gillebuskogen 
og de li< >iereliggende Kindalssætre vil selvfolgelig kimne gi sikre holde- 




Fig. 23. Den kalktutforende horisont i gruskeglen ved Tingvold (nordvestlige del av grustaket). 
Nordhagen lot. 6te oktober 1920. 



punkter i denne sak. — For tiden er der ingensomhelst antydning til tuf- 
dannelse langs bækkedalen. 

Da Gudbrandsdalen saaledes som tidligere omtalt har været herjet 
sterkt av flomkatastrofer helt op imot nutiden, kunde man kanske til en 
begyndelse være tilboielig til at sætte gruskeglens dannelse i forbindelse 
med denslags begivenheter. Men den kalktufforende horisonts jevne forlop 
og de overliggende jevne gruslag tyder ikke paa nogen pludselig opstaaen. 
Desuten er der flere ting, som med sikkerhet viser at keglens dannelse 
ligger langt tilbake i tiden. Som av Oven paavist, loper nemlig den æld- 
gamle kjorevei i Gudbrandsdalen, „kongeveien", henover gruskeglens over- 
flate. Hvor gammel denne kan være, vet vi ikke med bestemthet; den 
litteratur om norske veier ' som jeg har hat anledning til at undersoke, gir 



' Yngvar Nielsen: Om Norges veier for 1814, Historisk tidsskrift, bd. IV. Joh. Schou- 
G.\.\RD : Det norske veivæsens historie. 1899. 



1 92 1. Xo. 9. KALKTUFSTUniER I GLDBRANDSDALEN. 73 

ingen absolut sikre oplysninger herom. Imidlertid har hovedfærdselsaaren 
gjennem dalen altid gaat paa nordsiden av Laagen, og da „kongeveien" 
ved Tingvold synes at ha et hensigtsmæssig og naturlig fbrlop, tør det 
nok hænde at den gaar helt tilbake til middelalderen. Under veilegemet 
gjorde Oven og Holme desuten en anden interessant opdagelse; de fandt 
nemlig restene efter en kulmile (ofr. fig. 22 1. Denne maa nødvendigvis 
være endda ældre, men da der ingen redskaper er fundet paa stedet, kan 
kulmilen ikke dateres med sikkerhet. Ovex har fremsat den formodning, 
at den kanske skriver sig fra jernalderen. 

\'i ser saaledes at i lys av disse kjendsgjerninger rykker gruskeglens 
dannelsestid allerede langt bakover i tiden. Og én ting kan ialfald betragtes 
som fastslaat: kalktuftens nedbrytning og avsætningen av de svævende 
gruslag ved Tingvold skyldes ikke nogen recent katastrofe. Sandsynlig- 
hsten taler for at \i her har med forhistoriske fænomener at gjore. 

Mens vi ved Leine, takket være det ovre tuflag [A/mis-tuffen], kunde 
opspore nutids vegetation ens oprindelse og binde denne sammen med 
din fossile, er vi for Gillebutuffens vedkommende ikke i den samme hel- 
dige situation. Den sidst dannede tufbænk paa dette sted vidner om tæt 
fur us kog som fremherskende plantesamfund. I nutiden indgaar imidlertid 
granen som en væsentlig bestanddel i det skogsamfund som omgir bækken. 
Men om granens indvandring og vegetationens forandringer op imot histo- 
risk tid, derom fortæller disse interessante avleiringer i Oier desværre intet. 



III. Kalktuffen ved Nedre Dal i Faaberg. 

Denne kalktuf er indgaaende beskrevet av Blvtt (1892 1. c). Den 
ligger i 225 m.'s hoide over havet paa Gudbrandsdalens sydside, men med 
østlig eksposition, og adskiller sig fra baade Leine- og Gillebutuften ved at 
mangle distinkte tufbænker. 

Tuften findes i en brat li nedenfor det store opdyrkede jordet paa Nedre 
Dal. Her kommer en liten bæk irem av bakken og rinder nedover mellem 
et virvar av s'.ø/re og mindre stener og klippestykker. I jorden mellem 
stenene ligger der k alk tufbl okker, hvorav en del er ca. 0,5 m. i dia- 
meter, andre mindre. De er forresten nu tjernet og spaltet op i stor ut- 
strækning. Terrænget er i det hele tat meget uryddig og brat, og jeg 
tror for mit personlige vedkommende at her engang i tiden maa ha gaat 
et ras. Det store bratte jorde like ovenfor findestedet bestod, saavidt jeg 
kunde se, av lerholdig moræne, og der er intet iveien for at der her, 
kanske i forhistorisk dd, har gaat ut et skred. 

Blvtt fandt to stratigrafisk og palæofloristisk helt forskjellige nivaaer 
repræsentert ved Nedre Dal, en bl ad tuf med bjerk, asp, Sa/iccs uten 
makroskopiske fururester, og en furutuf, fuldstændig typisk og overens- 
stemmende med Leinetuften (og Gillebutufifen, som Blytt ikke kjendte). 



74 ROM .NDKDHAOF.N. M.-N. Kl. 

Men lilokk( 1- a\- (lisse to tulsl.'if^ laa \((l s i fl c n a \- hinaiiflcn paa stedet, 
aldiii; l\(lili,!4 nvcnpaa liiii;!n<l( n snin ha-nkcr'. Nogen o v e rga n gs t u f 
k Li n d c; h a ii ikke lindi, liilhr ikke iio,t;en Dryastiif. lilokker.e 
tilhørte alle saniiiicii «nicn d( n < iic < ll<i- drn aiidcn sort. Bi.V'iT antok der- 
for med rette at de skrev sig tra tr) lorskjellige tider, saaledes som ved 
Leine, lian skri\ei- ogsaa (1. e. p. qI at han anser blokkene for at ligge der 
„h\(>r tie l)le\ dannede"; men han hai" ikke levert noget bevis for at de 
befinder sig in situ i strengeste iorstand. 

P)i.vTT tænkte sig en hel del sniaa lokale tufdannelser langs bækken, 
hxilket altsaa resulterte i blokkenes opstaaen. Og aarsaken til at furutuffen 
ikke laa oxenpaa bladtuften, tænkte han sig niaatte ligge i det forhold, at 
det gamle \andlnp til en \iss grad hadde skiftet retning da det igjen be- 
gyndte sit lop under furutuffens tid (1. c. p. 9). 

Jeg avla hosten 1920 et besok \ed Nedre Dal og grov op en del tuf- 
blokker paa stedet. Saavidt Jeg kunde se, laa pragtfulde blad- 
tufblokkker like op til typiske furu tu fsty kker („side om side" 
Blytt 1. c), og jeg har vanskelig for at ti'o at Bi.vtts tolkning er rigtig. 

At tufl)lokkene skriver sig fra to forskjellige tider, er hævet over 
enhver t\il; derom vidner deres helt forskjellige fossilindhold og karakter 
og den paafaldende overensstemmelse baade med Leine- og GillebutufFens 
to horisonter. Men i motsætning til Blytt og Oyex, som slutter sig til 
Blytts opfatning eller ialfald refererer denne uten. kommentar (1920 1. c. 
p. 256 — 258), antar jeg at blokkene befinder sig paa sekundært 
leiested, og at der o pi" in delig hai^ eksistert en kalktuf ved 
Nedre Dal med to tufbænker oven paa hinanden, men som 
senere er od el agt ved utgl idn inger. Ialfald har jeg ikke paa 
nogen anden tufforekomst i vort land set og heller ikke i litteraturen fundet 
omtalt tui^dannelser av en saadan natur og med en saadan genesis som av 
Blytt antat. 

Bladtuften ved Nedre Dal er vakkert planskifrig og næsten smukkere 
utviklet end bladtuften ved Leine; og jeg har meget vanskelig for at fore- 
stille mig, hvorledes en saadan struktur skulde kunne komme istand hvis 
tufdannelsen var koncentrert paa en række adskilte punkter og uten hori- 
sontal kontinuitet, ialfald over en viss strækning. En saadan abrupt blokke- 
dannelse som Blvtt tænkte sig, vilde neppe kunne skape en saa planskifrig 
og regelmæssig struktur. Det moment som imidlertid veier mest i denne 
forbindelse, er furutuf blokkenes optræden side om side med bladtufstykkene, 
i samme dybde i jorden — et Ibrhold som umulig kan være op'rindelig, 
men som niaa bero paa sekundære forstyrrelser. Den antatte utglidning 
behøver paa den anden side ikke at ha været a\- særlig store dimensioner. 



' Oyen skriver i sit sidste arbeide (1920 1. c. p. 2561 at man ved Dal har to „tufbænker". 
Dette er altsaa ikke ganske korrekt. Blytt bruker selv meget konsekvent uttrykket 
„blokke" om tuffen ved Dal. 



I92I.N0. 9- KALKTL'FSTL'DIER I GUDBRANDSDALEN. 75 

I bladtuffen ved Nedre Dal fandt Blytt foruten de vanlige lovblad- 
rester ogsaa Pntniis Padiis, Sa/ix nigrica/is og av snegler J'itriiia pcUucida, 
Pupa unisconiii og Hclix arbiistonmi, som supplerer artslistene fra de oxrige 
forekomster i Gudbrandsdalen. Forovrig stemmer de alle sammen paafaldende 
godt overens. Det samme gjælder furutuffen, hvor Blytt ogsaa fandt et 
blad a\- Liniuva horcalis. 

I nutiden dominerer Urtica dioica i faretruende tætte bestander paa 
findestedet; dette har som nævnt tildels karakteren av en skraanende, svær 
stenrøis. Lon [Acer plataiioidrs), rogn og hassel er temmelig almindelige 
paa stedet; men der er ingen sammenhængende skog, bare smaakrat og 
træklynger. 

Kalktuffen ved Nedre Dal er, til trods for de noget dubiose stratigra- 
fiske forhold, som jeg her har forsokt at tolke paa en naturligere maate, 
meget interessant og værdifuld; den viser os nemlig at de to perioder som 
hoiere oppe i Gudbrandsdalen \ar karakterisert ved bladtuf- resp. furu- 
tufdannelse, har været av noiagtig samme karakter i floristisk og fysiografisk 
henseende ogsaa helt nede i Faaberg ved dalens begyndelse. Ellers er den 
historie som tuffen ved Dal fortæller os, adskillig mere fragmentarisk end 
h\'ad tilfældet er med Leine- os" Gillebutuffene. 



IV. KalktufPer ved Onset i Biri. 

A. Blytts undersøkeiser. 

I 1892 undersokte Blytt en kalktufforekomst ovenfor gaarden Onset 
i Biri ved Mjosen Icfr. kartet p. 2I. Han har selv aldrig publicert noget 
om sine fund; dog omtaler han i det efterlatte manuskript som blev trykt 
i Bergens Museums aarbok 1909 (1. c. p. 15), Onsettuffen med folgende ord: 
„furutuf og birketuf i særskilte blokke fra to forskjellige tidsrum fandtes 
ogsaa i Biri ved Mjosen". 

Blytts efterlatte samlinger opbevares paa Geologisk Museum i Kristi- 
ania, og samlingene fra Biri er vedlagt folgende veiledning, skrevet av Blytt 
selv: „Al tuf i denne kasse er fra Biri, samlet af A. Blytt i 1892. Overst 
ligger stvkker af furutuf, taget i uren ved \-eien mellem Eriksrud og Kræm- 
merodden ^ Al furutuf er fra dette sted undtagen et stykke 
som har etiket, og er fra Undset-. Fra Undset er al den øvrige tuf 
med Brtiila, Sa/ix cfr. capræa etc. \'ed Undset ligger furutuffen i klumper 
i løs jord over birketuffen, som danner storre blokke." Derefter anforer 
han følgende protil: 



' Denne lokalitet ligger nordentbr Onset. R. X. 
2 Uthævet her. 



76 HOI.F NORDIIAflKN. M.-X. Kl. 

Kalkjord. 

Siii;i;i tul klum])<i' iii'd I'iiiiis. 

Kalkjoi'd. 

Storrc blokker med lîctiila, Salix. 

Kalk|.)i-(l. 

Glaciallcr. 
DesLiten har Bi.vtt fcict til: „Mdhni Kriksrud og Kræmmerodden 
kun t'urutuf i smaa kliiiiipcr lese i uren." (ClV. Ovk.n 1920 1. c. p. 259). 

I samlingen lindes altsaa kun et eneste stykke furutuf fra Onset; 
resten er fra uren mellem Kriksrud og Kræmmerodden, altsaa en helt 
anden lokalitet. Den etiket som Blytt -omtaler, bærer folgende indskrift: 
„Furutuf fra Undset. Smaa klumper i los kalkjord o\-er den i større blokke 
forekommende birketuf. Maaske er denne klumpformede tuf ækvivalent med 
Dryastufifen ved Leine." 

Desværre er det nu ikke mulig med sikkerhet at avgjore hvilket st3'kke 
i samlingen denne etiket tilhorer. Dog er der ikke mere end to stuffer som 
det her kan være tale om. Den ene av disse er nu spaltet i fire smaa- 
deler med en meisel og viser en meget fin mosetufstruktur samt en del 
hulheter koncentrert paa et bestemt sted, intet andc;t. Disse hule avtryk 
ser ut som furunaaler ved første oiekast; men flere av dem er helt c3-lindriske, 
og ingen av dem viser det typiske og letkjendelige halvmaaneformete tver- 
snit som udmerker furunaalsavtryk. Jeg tror snarere at de skriver sig fra 
et græs. 

Det andet stykke er meget litet (ca. 4 cm. paa h\er kant) og delt i to. 
Det viser tydelige fuiunaaltversnit (6 — 8 naaler) og har ellers en grynet 
struktur, som vistnok skyldes moser. Ingen av de angjældende stykker 
viser andre tydelige plante-avtryk (f. eks. av lovblader). Det sidste lille 
st3'kke har tydeligvis ligget i jorden, da det er chokoladefarvet og forvitret 
i overflaten. Det stemmer ellers paafaldende godt med de smaa tufbiter 
som jeg selv opdaget paa en ny forekomst ved Onset (cfr. det følgende). 
Furutufstykkene fra Eriksrud — Kræmmerodden er helt forskjellige fra 
begge de nævnte stuffer, og bestaar av lakunes, koralagtig eller cinteragtig 
tuf med utvisket mosetufstruktur. De bærer alle sammen tydelige spor efter 
at ha ligget oppe i dagen i den av Blytt omtalte ur; der vokser nemlig 
moser i smaa gruber paa dem alle sammen, likesom de er mørke og for- 
vitret i overflaten. Et av stykkene er etikettert av Blytt selv. 

Jeg omtaler dette saapas utforlig fordi konservator Oye\ i sin sidste 
avhandling beskriver Onsettuffen meget indgaaende, men vistnok uten at 
være opmerksom paa disse forhold. Oven omtaler nemlig furutuffen 
ved Onset paa folgende maate : „Furutuffen er derimot utviklet som en i 
los kalkjord liggende klumptuf over birketuffen, der danner større blokker 
eller en mere sammenhængende tufbænk ^ Den graagule furutuf, der som 



' Dette er, som nedenfor omtalt, ikke tilfældet. 



J 92 1. No. 9. 



KALKTLTSTUDIER I GUDHRANDSDALIA'. 



regel er fast, viser dog ogsaa i den nedre del niosetuf a\- drypstensagtig 
karakter med blader av Sa/ix capræa, Popnlits trcniula og Bctula odorata, 
men her allerede med iblandet Piuiis silvcstris; senere blir den omend til- 
dels vekslende med noget jordagtig tuf. tildels noget graa, breccieagtig med 
skiferbiter, og her tandtes ved siden av de nu allerede nævnte ogsaa blad- 
avtryk av Vacciniinn MyrtiUiis L. Paa mange steder ser man dog i den 
cinteragtige, graabrungule tuf kun spredte furunaaler uten spor av ind- 
blanding av lovtræblade, ja det samme træk gjenfindes paa sine steder 




Fig. 24. Breccieagtig kalktuf med bergartsfragmenter. 
Bii-i; Blvtts samling ('2). B. Larssen tot. 



selv der, hvor tuffen har noget mere præget av en mosetuf. Furutuffen 
er saaledes i det hele tat \-el skilt fra de underliggende tufforekomster" 
(1. c. p. 260 — 261 1. Da der, som Blytt uttrykkelig gjor opmerksom paa, 
bare findes et eneste stykke fur u tuf fra Onset i hans efterlatte sam- 
ling fra Biri, og dette stykke maa være et av de to som jeg ovenfor har 
beskrevet, maa denne Ove.ns karakteristik \-istnok bero paa en mis- 
forstaaelse. 

De nævnte stykker indeholder ingen lo\-bladre5ter, hverken av Salix 
capnxa, Popiilns trciiuila eller Bctula odorala. Og Oyexs uttalelser om at 
furutuffen senere blir „tildels noget graa, breccieagtig med skiferbiter, og 
her fandtes ved siden av de nu allerede nævnte ogsaa bladavtryk av ]^ac- 
ciiiiiiin Myrtilliis" , er for mig ganske uforstaaelig. 1 samlingen findes nemlig 
et solid stykke bestaaende av en m.ængde skiferbiter tæt sammenkittet av 
haard kalktuf, som er totalt forskjellig fra alle de øvrige stykker i Blvtts 



yS ROIJ" .\or<i)iiAr;i;.\. M.-N. Kl. 

sanilin.i; fra Onset. I)(l har iiii;* n oii^^inalclikil; hlaaharhlaflct er ogsaa 
meget iityddi.t; ".U lu 'ist t\ilsotnt. I museets samling ligger et andet stykke 
Ira „Biri" saiiili t a\- Ri rs( ii, og dette viser en slaaende likhet med oven- 
nævnte, idet cUt er j)i()|)t"Lildt av skiferfragmenter. Antageligvis stammer 
begge to fra samme sted og er sikkert opstaat paa den maate, at kalkholdig 
\and hai" passert en spræk med lorviti"ingsmateriale, som kalken har kittet 
sammen, eftersom den ble\' uttældt (ofr. næste avsnitl. Naar (Jyen uten 
videre indfletter det nævnte tufstykke, som har en helt avvikende struktur, 
i sin omtale av furutuffen ved Onset og forseker at gjore det til en 
egen „breccieagtig" facies av denne, saa er dette blot og bart et postulat. 
Det er for det forste hoist tvilsomt om stykket er fra Onset; det er ialfald 
hverken litet eller klumpformet, og har intet med furutuffen at gjfire. Det 
maatt(- da i tiUælde være fra Eriksrud — Kræmmerodden; men det ligner 
heller ikke de typiske furu-mosetufstykker fra dette sted. — Og Øyens 
sidste bemerkninger om furutuffen: „Paa mange steder ser man dog i den 
cinteragtige graabrungule tuf kun spredte furunaaler uten spor av indblanding 
av lovtræblader, ja det samme træk gjentindes paa sine steder selv der, 
hvor tuften har noget mere præget av en mosetuf", er mig likeledes ganske 
uforstaaelig. Denne beskrivelse passer nemlig aldeles udmerket paa styk- 
kene fra Eriksrud — Kræmmerodden (cfr. ovenfor). Man faar uvilkaarlig det 
indtryk, at Oven bygger sin karakteristik av furutuffen ved Onset paa et 
rikholdig materiale. Det er ialfald en levende umulighet at utlede hele 
denne utforlige beskrivelse av det ene lille furutufstykke som Blytt har 
medbragt fra Onset og etikettert. At der her foreligger en misforstaaelse, 
er givet. Da det ikke er andre end Blytt som har set forekomsten, maa 
man nødvendigvis holde sig strengt til Blytts egne veiledende opl3'sninger, 
og de er ikke til at ta feil av. 

Naar Oven desuten mener at kunne rekonstruere folgende sammen- 
hængende profil fra Onset : 

1. I bunden en tildels breccieagtig eller ogsaa ofte en konglomerat- 
agtig tuf. 

2. Derover en mosetuf, der som en egen avdeling dog nærmest grup- 
perer sig under birketuffen og o\-erst fører Brtiila odorata sammen 
med den for avdelingen karakteristiske Hypnimi sp. 

3. Birketuf. 

4. Noksaa tæt mosetuf, ogsaa tildels med furu under h\is sone den ogsaa 
nærmest horer som underavdeling. 

5. F unit uf, 

saa maa jeg desværre erklære mig helt uenig ogsaa i denne tolkning. Det 
faktiske utgangspunkt for ræsonnementet, nemlig Bly'tts materiale, er her 
absolut avgjorende. 

O YEN forsoker altsaa at gjore gjældende at Onset-tuffen er en autochtcn 
in situ-dannelse, og foretar desuten en opdeling i stratigrafiske 



I92I. No. 9- KALKTUFSTUDIER I GUDBRANDSDALEX. 79 

u n d e r a \- d e I i n g C r, noget som \\ ikke f i n d e r a n t \- d e t engang 
i Blytts no tiser. 

Furutuften har jeg allerede omtalt utforlig ovenfor. Jeg har ogsaa 
gransket Blytts b 1 a d t u f m a t e r i a 1 e meget kritisk og omhyggelig, og det 
viser sig at dette lar sig inddele i forskjellige grupper av haandst3-kker, 
som utvilsonit horer sammen. 
I. En hel del vakre stufter (15 — 20 stykker) bestaar av en mosetuf- 
grundmasse, ofte meget grov, desuten med talrike lovbladavtrvk av- 



1^. 



t^^ 






-m. 








Fig. 25. Bladtufstykker fra Onset med valket overdate (Cyanofyce-tut l',2ll. 
Blytts samling. B. Larssen fot. 

\"ek3lende hermed. Deres o ver fl atest r uk tu r er vtterst eiendom- 
melig og viser en ytre, paafaldende glat og valket, fint lamellert sone, 
som skyldes kalkutskillelse ved blaagronne alger (cyanofvcé-matter) 
(fig. 25 1. Disse stykker er fremkommet ved opspaltning av en eneste 
større blok, idet de lar sig sætte sammen som en mosaik. Denne blok 
har ligget los i jorden ; overflaten er nemlig mork og forvitret. 
15 storre og mindre stykker med tildels den samine mosetuf-bladtuf- 
karakter, dog dehis mere kompakte, bærer visne paasittende recente 
moser i den for\-itrede overflate ; de m a a a 1 1 s a a f a k t i s k ha 
stukket op av jorden. Det er altsaa klart at enkelte av Blytts 
bladtufblokker har ligget overst i jordlaget, tildels helt frit. 
En mængde stykker er morkt chokoladefarvet i overllaten og har ligget 
lose i muldjorden (cfr. mine egne fund, som omtales i det folgende. 



8o kOLK NORDIIAGKN. M.-X. Kl. 

4. I'^iidclif; lindes der en scii«- med vakrt; stiitVcr Ijestaaende av avvekslende 
mosetufparticr ot; loxhladrikc [).irti(r, sr)iii nicn t\il <.-r spaltet løs fra 
det indre av en elier et pai- blokke r. I)c bærer nemlig tildels friske 
brudflater. Et par av dim indilmldi r skiferbiter. 

5. Et |)ai- vakre mosetufstykker er t\'deligvis halvrecent tuf, idet de bærer 
halv f o r k a 1 k e d e, recente moser i overflaten. 

Denne gjennemgaaelse gir det bestemte resultat, at „birketuffen" ved 
Onset er en m oset ufagtig, bladrik tuf {Aliiiis incana m. aim., des- 
utc-n l^<ipii/iis Inniiild, Salix caj^rdd og Salix ciV. iiii^iucaiis, Bctuln odorata 
alm.'l. Den har, som l>l.^■| 1 anfører, ligget los i jorden som større og mindre 
blokker. Vertikale faciès (Ovens „breccieagtige (?) eller ogsaa ofte 
konglomeratagtige tuf" (?) i bunden, derover „mosetuf" som en egen av- 
deling under den egentlige „birketuf") kan ikke paa\-ises med nogen- 
somhelst sikkerhet. Tuffen er tvertimot paafald ende mose- 
tufrik helt ig j en nem og temmelig ensartet. 

Jeg har lagt meget arbeide paa denne gjennemgaaelse, fordi jeg 
mener at vi naar det gjælder disse omdisputerte stratigrafiske problemer, 
maa være saa kritiske som mulig og eliminere alle dubiøse 
eller hypotetiske momenter. F'remfor alt maa \i ikke la os forlede 
til forhastede slutninger og korrelationer paa grundlag av et usikkert 
materiale. 

At der her er antydning til en overensstemmelse med Gudbrands- 
dalens tuffer, er ganske klart. Men ingen kan \ite om ikke de smaa furu- 
tufstykker ved Onset, hvorav Blytt bare har medført en liten prøve som 
fortæller svært lite, er av forholdsvis recent natur. I et av Blytts stykker 
har jeg fundet nogen naaleformige avtryk, som ut\ilsomt skriver sig fra 
gran {Picea rxcc/sa), og som altsaa maa \ære av geologisk talt yngre 
datum. Dette stykke har vel ligget øverst i jorden sammen med de smaa 
furutuf klumper. \' i mangler e t h \- e r t sikkert middel til at t i d- 
fæste disse lose tufbiter. Det samme gjælder furutuffen fra Eriks- 
rud— Kræmmerodden-, som forresten heller ikke ligner den \-ii-kelige 
typiske „furutuf" i Gudbrandsdalen. 

Desuten kommer her et meget vigtig moment til, som Oven ikke er 
opmerksom paa, men som jeg allerede har berørt under omtalen av fore- 
komsten ved Nedre Dal i Faaberg: Ligger tufblokkene der hvor 
de blev dannet, eller befinder de sig paa sekundært leie- 
sted? Er Onsettuffen autochton eller ikke? Under mit besok i 
Biri høsten 1920 lykkedes det mig at fremfinde en antageligvis for Blytt 
ukjendt tufforekomst ved Onset. Denne er i flere henseender interessant 
og kaster ogsaa lys over det ovenfor opstillede sporsmaal. 



• Pollenanalyser bragte sparsomt pollen av Abats, Bctiila og Piiiiis silvestris for dagen. 
2 I likhet med forholdene paa den av mig opdagede nye tuft'orekomst ved Onset antar 

jeg at disse Blytts „lose stykker i uren" er faldt ned ovenfra hammeren (cfr. det 

folgende). 



192 1. No. 9. 



KALKTLTSTUDIER I GUDBRAXDSDALEX. 



Forinden jeg gaar over til at omtale mine egne fund, maa jeg imid- 
lertid gaa ind paa nok et forhold av betydning, som Oven har trukket ind 
i diskussionen, og som ogsaa hænger sammen med Blytts efterlatte sam- 
linger. ØYE.N" mener nemlig at ha fundet avtryk av Dryas octopctala i et 
litet tufstykke, som Blytt medbragte fordi det indeholdt en snegl [Hyalinia 
petronella). Oyex skriver, at dette lille stykke minder om det furutufstykke 
fra Onset som Blytt har etikettert, og antar, om end under nogen tvil, at 
der ved Onset ogsaa har været en „Dryastuf" 11. c. p. 259 — 260 1. Stykket 






Fig. 26. Øverst „ Dr rastufstvkket " fra Onset fotoorafert i to tbrskjellige belysninger. Billedet 

tilhoire viser tvergaaende tbrbindelsesnerver mellem de tre hovednerver. Nederst fragment 

av et tvdelig Alnus-h\-xå fra Onset til sammenligning. ' I. B. Larssen fot. 



skal ifolge Oyen ogsaa indeholde et par furunaaler. Xu er det imidlertid at 
merke at det angjældende lille stykke ligger i en pakke som Blytt selv har 
etikettert paa folgende maate: „Snegler fra birketuffen' ved Onset, be- 
stemt af frk. B. Esmark", og sammen med det ligger en del smaa lignende 
tufbiter med snegler. 

Jeg har selv gransket stykket meget indgaaende og er ogsaa kommet til 
et bestemt resultat. \''ed første eiekast kunde man muligens ta avtr^-kket for 
at være tre grove Dryasblader paa rad, med midtnervene helt paralelle 
og med diffuse randpartier (cfr. fig. 26 1. Men ved nøiere undersokelse ser 
man sranske tvdelis: at disse tre „midtnerver" er forbundet ved 



' Uthævet her. 

Vid.-Selsk. Skr. I. M.-N. KL 1921. No. 9. 



82 ROr.F NORDIIACF.N. M.-N. Kl. 

et t \ r rs^a ae n (I I- a ii a s lo m os(.- n c t ; (!<■ Iiorcr altsaa sammen. 
Og ved samni« iili.nnin.u med (I likholdij^ lovl^ladmateriale er jeg blit over- 
bevist om at (l( tic Litydelige „I )r}asljlader" kun er et bladi'ragment av 
y]///iis iiicaiia, som er meget almindelig i bladtufblokkene fra Onset. Vi 
har her tre av de karakteristiske paralelle sidenerver og forbindelsesnerver 
lodret paa disse, hvilket netop særkjender orebladene (fig. 26). iJryas hører 
med til de lettest kjendelige av alle fossiler, og jeg anser det for helt utelukket 
at \i h(i- liai- med blader a\- denne art at gjore. Nogen sikre furunaaler 
Hndes heller ikke i stykket, kun et tvilsomt hulrum, som like godt kan være 
noget andet. Ifolge Blytt skal jo stykket ogsaa være fra „birketuffen" 
ved Onset og ikke fra furutuffm. Ktiketten „snegler fra birketuffen" er 
utvilsomt rigtig. 

Oyen har selv karakterisert „Dryastuffen" ved Onset som noget tvil- 
som; jeg er selv overbevist om at den ikke har eksistert. Blytt, som 
kjendte Dryas saa godt, har heller ikke opdaget noget blada\ tryk av denne 
art. Og han har sikkert gransket materialet meget omhyggelig. 



B. Egne undersøkeiser. 

Gaarden Onset i Biri ligger ved Mjesen i 40 — 50 meters høide over 
denne. Like bak gaarden stryker et langsgaaende høidedrag i retningen 
SØ — NV; det bestaar av den eiendommelige „Birikalk", som her danner 
steile bergvægger, tildels med urdannelse under styrtningene. 

Den av Blytt studerte tufforekomst har efter de oplysninger som 
jeg indhentet paa stedet, antageligvis ligget et steds ved foten av dette 
høidedrag i nærheten av en forfalden husmandsplads, som ligger et par 
hundrede meter sydvest for gaarden. Ret op for pladsen i den skogbe- 
vokste skraaning under hammeren lykkedes det mig bare at finde et eneste 
løst tufstykke, som jeg antok maatte være kommet ovenfra. Jeg klatret da 
helt op til bergvæggene, og her fandt jeg ganske rigtig en ny tufforekomst. 
Stedet laa ca. 60 m. høiere end pladsen og denne omtrent 20 m. høiere 
end Onset, altsaa i alt ca. 75 m. over Onset. Mjosen angives at ligge 124 
m. o. h., og lokaliteten skulde efter dette ligge ca. 250 m. o. h. Dette tal 
er muligens noget for hoit. 

Under de lodrette, tildels overlutende kalkklipper fandtes der her en 
meget brat skraaning med gran og lovtrær. Jordbunden bestod av sten- 
blandet muld med opragende fast fjeld og blokkemasser. Skraaningen var 
ca. 45^, tildels brattere. Ved gravning fandtes en mængde smaa tufstykker 
i jorden, de fleste sterkt forvitret og brune eller chokoladefarvet i overflaten. 

Profilet hadde folgende utseende: 
I. Muldjord, 20 cm. 

II. Lose tufbiter i sort muldjord, især optil en mængde smaa 
biter, nedtil noget grovere stykker. Tilsammen ca. 50 cm. De 



1 92 1. No. g. 



KALKTL'FSTUDIER I GL'DBRAXDSDALEX. 



83 



f 1 e S t e s t u t t'e r var helt t"o s s i 1 1 o m m e, tildels flintagtig' haarde 
og kompakte, otte med smaa bergartsfragmenter indesluttet i kalken 
(breccieagtig). \'idere fandtes flere slags mosetuf, tildels med utyde- 
lige thallose levermosavtryk. 1 et storre stykke saaes en nott av 
hassel iCorv/iis Avcllaiia^), blad av Salix sp. (?) samt smaa fiirii- 
tiaalcr. 1 et andet fandtes utydelige lovblader vistnok av b/'crk. 
Endelig indeholdt et par smaa tut^ragmenter nogen avtryk, som an- 
tageligvis skriver sig fra 
graiiiiaalcr [Picea cxccl- 
sa); disse fandtes hoit 
oppe i laget. 

Fælles for alle tuf- 
stykkene \ar mangelen 
paa skifrighet og deres 
haarde konsistens. Flere 
biter hadde cyanofycé- 
struktur. 
III. Derefter fulgte sten- 
blandet muldjord 
Uten skarp avgrænsning 
fra foregaaende lag, med 
en del mindre tufstyk- 
ker; disse avtok stadig 
i antal nedad. Her saaes 
ingen bestembare fossi- 
ler. Tuffen var gjennem- 
gaaende flintagtig haard 
og kompakt. Dette „lag" 
kunde forfolges til 50 
cm. 's dyp under det fore- 
gaaende, men fortsætter 
sikkert dypere ned. 

Nogen tydelig stratigrafi kunde altsaa ikke paavises, heller ikke 
nogen skarp grænse mellem lag II og III. Men der er altsaa en anrikning 
av tuf biter optil, mot overflatens muldlag. 

Merkelig nok bar lokaliteten svake spor efter gravning, og jeg antok 
derfor forst at jeg hadde fundet Blytts lokalitet. Men folkene paa Onset 
gaard fortalte at efter Blytts besok i 1892 hadde der været flere oppe 
og rumstert under fjeldet, vistnok for at se efter kalk til at forbedre aker- 
jorden med. Blytt angir desuten „glacial-ler" i bunden av sit profil, 
hvilket tyder paa at det er optat et steds længer nede ved foten av skraa- 
ningen. Overlærer Holme, som ogsaa har a\lagt besok ved Onset, antar 




Fig. 27. Klipper av Birikalk ovenfor Onset med to 

mørke soileformige kalktufmasser nederst. Disse er 

dækket av moser og cyanofycéer. Nordhagen fot. 

20de oktober 1920. 



' Denne busk fandtes paa tuffindestedet ogsaa i nutiden. 



84 ROI,F NOHDHAGEN. M.-N. Kl. 

at JJl.^■■ns loi'ckomsl iiiaa \aic ndclaj^^t v(-fl ras, livilkct er ni*;gct vel 
tænkelig. 

Men Inni-fra skriver- nu alle de lose tufstykker sig? En nærmere 
granskning a\' sehe liainnu icn \-ii op for lindestedet gav opløsning paæ 
gaadcn. lier l'andtes nemlig et par meget interessante, hal\- 
receiile tu Tinasser a\sat sk al form ig utenpaa de lodrette 
eller over! u tende k al k k 1 i ]:> pe r. Kalkholdig vand syntes at komme 
frem langs nogen revner i fjeldet, og langs disse hadde der bygget sig op 
søileformige eller ribheformige tufansamlinger (ofr. fig. 27 og 28), som nu 
var klædt av en tæt pels med moser og cyanofycéer. Disse kalk- 
samlende organismer syntes at forkalkes langsomt og kontinuerlig. Da jeg 
besøkte stedet (20de oktober 1920), \ar mosene fugtige, men der var ikke 
rindende vand tilstede. Foruten disse pilarlbrmige masser laa der under 
hammeren en svær tufblok, som tydelig nok var løsnet fra bergvæggen 
og faldt ned paa underlaget. Ogsaa paa denne var der svak mosetuf- 
dannelse (fig. 28). 

Den vigtigste tufdannende mos var Mollia ænigi/iosa, som dannet 
svære kaker, men av et yderst sykelig utseende; den var meget sterkt 
foi'kalket, kun med faa opstikkende blader overst. De øvrige arter dannet 
i grunden en serie med hensyn til sin kalksamlende evne: Sivartzia mon- 
taun, Mollia tortiiosa, Lccrsia contorta, MyurcUa apicitlata, M. julacca og 
M. tciicrrima følger nærmest efter ovennævnte art. Ditricliiiiii flcxicaiilc og 
Amblystegium protensnin var meget svakere forkalket, Leucodon sciiiroides 
uhyre svakt. Foruten disse 10 arter saaes Hypmun sericeiim, H. Bambergi 
og Solorina saccata paa den nedfaldne store blok; men disse arter vokste 
helt tørt og var uten spor av kalkansamling. 

Mellem mosene og paa disse fandtes store kvantiteter av kalksam- 
lende blaagrønne alger. Mollia æntgiiiosa var ofte saa overtrukket av 
filtete traadmasser av Scytoneina uiirabilc (Borxet) at den var helt sortbrun 
og lignet morke, korte tafser'. Foruten denne cyanofycé-art, som ogsaa 
dannet rene, smaa, filtagtige puter, var Chroococcus turgidus (Naeg.) rikelig 
tilstede samt flere andre Chroococciis-dirier. Merkelig nok optraadte ogsaa 
en svakt likenisert Collcina-avt som brunlige gelé-klumper mellem mosene 
og cyanofycéene. Blandt Scytouciiia uiirabilc forekom ogsaa enkelte traader 
av Pctaloncum alatiiin (Berkeley). 

Denne liste over kalktufdannende moser og cyanofycéer supplerer den 
av Sern ander publicerte fortegnelse fra svenske forekomster (1. c. p. 165 — 
179). Av Sernanders arter har jeg i Norge kun gjenfundet Ainblystcgiitm 
filiciintni ^ og Sivartzia luoiitaiia som recente tufdannere. De øvrige 9 mos- 
arter fra Onset er hittil ikke iagttat som kalksamlere i Sverige, hvor til 
gjengjæld en 5 — 6 andre arter er fundet. 



• Bestemt av assistent H. Olivcoroxa. 
2 Cfr. Leine p. 14. 



1 92 1. No. 9. 



KALKTL'FSTL'DIER I GL"DBRA\DSDALEN. 



Av recente kalktufdannende cyanofycéer anfører Serxaxder 3 arter: 
Riviilaria lioouatitcs (C. A. Ag.I, Petalonciua cntstaceitiu (C. A. Ag.I og Di- 
plocolcon Heppii (Naeg.). Desuten nævner han Cliroococcus variiis (A. Br.) 
og Chr. tiirgidus (Naeg.) som usikre kalksamlere. 

Til denne forholdsvis sparsomme skandinaviske liste kan vi altsaa nu 
foie Scytoiiema inirabile (Borxet) og Pctaloncina alatiim (Berkeley). Des- 
uten bestvrker mine fund den av Sernander antatte kalktufdannelse hos 
Cliroococcus tur^idiis. 






»^ /•^\, 



mm 



Serxaxder antar ogsaa at 
C3'anofycéene leverer bet3'de- 
lige bidrag til kalkansamlingen 
i mostuene, at disse organis- 
mer altsaa supplerer hinanden. 
For Onsetforekomstens ved- 
kommende maatte man, hvis 
man anskuet forholdene plante- 
sociologisk, tale om en tuf- 
dannende Mollia ccritgiiiosa — 
Scytoiiema iiii?'abi7e-a.ssociat[on. 
De øvrige arter forekom ind- 
sprængt i denne eller dannet 
mindre bestand. Interessant 
var det at gjenfinde Mollia 
æriiginosa som recent tut'dan- 
ner ved Borju-bækken nær 
Leine i Kvam (cfr. ovenfor 
p. 3). Denne synes vistnok at 
være almindelig paa norske 
tuflforekomster. — Som en kuri- 
ositet vil jeg nævne, at den 
Collcnia-art som optraadte mel- 
lem mosene ved Onset, ogsaa 
syntes at bevirke kalkutskillelse, 
idet den flekvis var belagt med en kornet kalkmasse. Imidlertid var denne 
lavart som ovenfor omtalt svakt likenisert, eller rettere sagt: den Nosloc-art 
som utgjør alge-komponenten i laven, var kun svakt omspundet av sop- 
hyfer, saa kalkansamlingen blir allikevel let forstaaelig. 

Hvad Lcucodoii sciuroidcs og Ditrichimi flcxicaidc angaar, som begge 
var ytterst svakt kalksamlende ved Onset, og hvorav ialfald den første i 
almindelighet er knyttet til træstammer, maa man vistnok anta at kalk- 
utskillelsen omkring deres basaldel skyldes de tilstedeværende blaagrønne 
alger (cfr. Serxaxder 1. c. p. 179I. 

Tufmassene ved Onset syntes at udmerke sig ved en temmelig træg 
vekst; ialfald var mosene svært lite frodige og tildels ganske kvalt av 




Fig. 28. En stor kalktufblok, som har løsnet sig av 

fra klippevæggen. Paa avlosningsstedet sees merke 

polstere av Mollia aeruginosa og Scytonenia mirabile 

(øverst tilvenstret. Hammeren ved Onset. 

20de oktober 1920. Xordhagen fot. 



86 KOI.K NOKDIIAOKN. M.-N. Kl. 

algene. }v^ har dciioi- \al^t ril kalde disse tufpilarcr og kakcr for h a 1 v- 
rc CC 11 te, idet Jeg tror at d< r har incdgaat ganske lang tid til deres 
dannelse. 



Ret ned tor disse kalkklip|)cr, kun 2 — 3 meter ia\ere, laa den oven- 
for beskrevne masseforekomst a\' iiiiiidrc tufstykker i Jorden. Og det er 
ingen t \' i 1 om at disse s k r i \- c r sig fra k a 1 k k 1 i p p e n e; de har 
faldt ned ovenfra i tidenes lup. Deres karakteristiske konsistens, 
mangel paa skifrighct, kompakte stiuktur etc' faar ogsaa herigjennem sin 
naturlige forklaring: det skyldes altsamnien den eiendommelige dannelses- 
maate. Disse tufstykker i Jorden tyder paa at der har paagaat tufdannelse 
langs hammeren i betydelige tidsrum. Men da stykkene desværre er saa 
uhyre fossil fattige, fortæller de os saare lite om utviklingen paa stedet. 
Den merkclige ophobning av smaa tufbiter overst i Jorden under muld- 
laget betyr utvilsomt et eller andet. Men om fænomenet beror paa for- 
vitring og sterk avløsning av tuftragmenter langs den ovenfor værende 
klippevæg (under et tort klima) eller skyldes rask tufdannelse paa dette sted, 
derom tor Jeg ikke uttale mig. 



I lyset av disse fænomener, som knytter sig til denne ny^e forekomst 
ved Onset, blir den av Øven omtalte tuf ganske dubios. Hvem vet om 
de av Blytt fremfundne lose blokker befinder sig in situ? Det er meget 
mulig at ogsaa deres dannelsescentrum oprindelig har ligget et eller andet 
sted oppe i skraaningen eller under hammeren, og at de senere er faldt 
ned. Et par av Blytts bladtuf blokker har som tidligere nævnt cyanofycé- 
struktur i overflaten, flere stykker indeholder ogsaa bergartsfragmenter, 
hvilket betegner likhetspunkter med den av' undertegnede opdagede fore- 
komst. I a 1 fa 1 d er der h i 1 1 i 1 ikke 1 e v e r t noget bevis f o r at 
Blytts profil skriver sig fra en autochton tuf in situ. Indtil 
dette er gjenfundet og undersøkt paany ute i naturen, kan tuften ved 
Onset ikke tillægges nogen avgjorende betydning. Ræsonnementet blir 
nemlig staaende paa altfor svake føtter. 



• Likeledes forekomsten av bergartsfragmenter i tuften. 



I92I.N0. 9- KALKTL'FSTUDIER I GUDBRAXDSDALEN. 



GENEREL DEL. 

I den specielle del har jeg fremlagt hele det rent deskriptive materiale 
som vi nu raader over fra Gudbrandsdalens kalktuffer. Jeg har ogsaa forsokt 
at paavise likheten og forskjellen mellem de enkelte forekomster, og like- 
ledes gjennemgaat de forskjellige tuflag og diskutert de stratigrafiske, 
genetiske og biologisk-plantegeografiske problemer som knytter sig til dem. 

I avhandlingens generelle del skal jeg gaa næmiere ind paa under- 
søkelsens almindelige resultater og forseke at besvare folgende spørs- 
maal: hvorledes stemmer disse tufter med andre i litteraturen beskrevne 
forekomster? Er de stratigrafiske fænomener som tuftene aabenbarer, av 
rent lokal eller mere generel natur? Hvorledes stemmer de plantegeografiske 
eiendommeligheter som Gudbrandsdalens tufter opviser, med de resultater 
som tor\myrforskningen og andre palæobotaniske undersøkeiser har git os? 
Alt dette er centrale sporsmaal, som maa besvares forinden man kan gaa 
til en tidfæstelse av de enkelte lag. 

For at faa en tidsbestemmelse bakover i tiden blir det nødvendig at 
omtale isavsmeltningen i Gudbrandsdalen forst. Jeg har ogsaa fundet det 
hensigtsmæssig at gi en oversigt over hvad man for tiden vet om den 
første flora og vegetation under avsmeltningstiden, da dette er av betyd- 
ning for de efterfølgende avsnit. 



I. Isavsmeltningen i Gudbrandsdalen. 

Desværre foreligger der i norsk geologisk litteratur ingen samlet og 
fuldstændig fremstilling av isens avsmeltning i denne del av landet, kun 
spredte iagttagelser publicert av forskjellige forfattere. 

For Kristiania-fjordens omgivelser og traktene op til Mjøsen er for- 
holdene nu klarlagt paa en tilfredsstillende maate. Man adskiller her en 
række stadier i avsmeltningen, som vi med O yen betegner som: 

j Smaalenene — larlsberg-trinnet 
Rastadiet in, tt '• 

( Moss — Horten-trinnet 

, , . , Aas-trinnet 

Aasstadiet \ „, . 

bki-tnnnet 



KOI.K NOKUlIACiKN. M.-N. Kl. 



A k c I-Stad i et 
l\ o 111 (• ri k s ta cl i c; t 



Nyckils-tiiriiKt 
Maridals-trinnct 



Ski (l^iiio-liinnct 
1 li r^cr-iriiinet 

(O^KN 1911 1. C, IjJOKL^KKK J913 1. cl. 



At disse dobbeltrækkcr er os ci 11 a ti on sm oræ n c r (uven 1904), 
dannet under fremstot av isranden, fremgaar av de intramorænale fore- 
komster av niai-int 1er tildels med marine fossiler, som man nu har kunnet 
paavise i saagodtsom alle de nævnte morænerækker. Disse mange sving- 
ninger i brærandens stilling under tilbakerykningen er av stor teoretisk 
betydning for dt-n absolute kronologi ; det er nemlig vanskelig at avgjore 
hvor store tidsrum disse oscillationer repræsenterer. Man har her en ganske 
væsentlig feilkilde som maa tages med i beregningene. 

Nordenfor Romerik-stadiet moter vi en vældig morænerækkc foran 
Mjøsen — Hurdalsvand — Randsfjorden — Spirillen, som man oprindelig betegnet 
som „indsjotrinnet" eller det „epiglaciale" trin (tig. 29I. De fænomener som 
knytter sig til denne række, er omstridte. 

ØYEN har i denne moræne fundet svære blokker av fint skiktet hav- 
ler op til 4 — 5 m. i diameter (191 5 1. c. p. 310), hvilket viser at bræene her 
har rykket frem over en tidligere havbund og plöiet op lerlagene foran sig. 

I lermassene paa den romerikske slette som er avsat foran denne 
bræfront, gjorde Oyen sine opsigtvækkende fund av den yngste Po/i/andia 
arcfica-ïsium (1903 1. c). Mens man kun har den ældre Porf/aiic/ia iYo/c/ia)- 
fauna (ældre og yngre Yo/dia-ler efter Brøggers terminologi 1900 — 1901} 
utenfor raene eller i nær tilslutning til disse ^ har man faktisk paa et langt 
senere stadium en fornyet opblomstring a\- Portlaiidia arctica (i en eiendom- 
melig mindre varietet) paa Romerikssletten. Man har villet forklare dette 
fænomen ved at anta at ishavsmuslingen trak sig tilbake fra Vänernbassinet 
og nordover og optraadte som relikt paa Romerike (De Geer, Geol. Foren. 
Förh. 1913 p. 308 — 309). Men dette er ingen tilfredsstillende forklaring 
paa det faktum, at Portlaiidia arctica her har en ny blomstringstid. 
ØYEN har nu fundet den paa en mængde lokaliteter helt op til Elverum og 
Solør (Norsk Geol. Tidsskr. 19 13 p. 2). En relikt, som tydelig avhænger 
av temperaturforholdene i vandet, opviser ikke uten videre en saadan paa- 
faldende leved^'gtighet og spredningsevne. 

En hel række andre kvartær-geologiske fænomener har fort Oyen til 
at opstille det saakaldte „Portlaiidia-iiivaa" som en egen avdeling i vort 
lands kvartærhistorie. 

En ældre og en yngre Portlaiidia a rctica-ïorendQ avdeling var allerede 
tidligere kjendt fra Trondhjemsfeltet ved Oyens undersokelser, og viste sig 
at være ikke alene petrografisk forskjelligartet, men var ogsaa adskilt ved 



' Cfr. dog ØYEN 19 1 5 1. c. 



1921. No. 9. 



KALKTUFSTUDIER I GL'DBRAXDSDALEX. 



89 



\ ^^ "^ 


\ ^^^ VV_. J^^"^ \^ 


1 \W ^^A \ 




: 




^ /Kß ^\ ) ..o°°7o'' ■■' 


1 /^yKi ■^v:X''»V''''°°/ i 


^«\\ --oo.^"^ y 


Y> "oToo- s 




) r\L. 


y ^^rk ^ lU^/ 


T ^ #'^^i^">^€ 


^7 0. 4 i) 



Fig. 29. Kart over de vigtigste moræncrækker i det sj-dostlige Norge. 
Sydligst Rastadiets dobbeltrække, derefter Aas-stadiets to trin, nordenfor 
dem Aker-stadiets dobbeltrække og Romerik-stadiet. For" Mjosen — Hiir- 
dalsvand — Randsfjorden o. s. v. „Indsjotrinnet". (Efter Øyen 191 i 1. cl 



go ROI.K NOKDIIAfiKN. M.-X. Kl. 

avsa'tniii,t;<i" Inis I'aima liaddi- il iioi;« I mi Id c re [)ræg. Og da <^^)vK.\ 
gjcnfandt noget tilswarcndr i Kri-.lianial'<ll(t og fKsiitcn opdaget den nierke- 
lige tempererte og rike il^v'//7//.s-fauna \(d Kristiania (1902) ved selve den 
marine grænse, maatte man ii\ill<aarlig stiidsc 'Fi ;illf disse fænomener 
stemte ytterst daailig mid del liillede som man indtil da hadde opgjort sig 
av axsmcltningstidens natur. 

Paa den iiie side liadde man altsaa den tempererte Myfi/iis-fauna 
{M. (■(/////s, Mya Intiicata, Macotim haltica, Saxicava pholadis, Balanits 
crciialiis^) i 221 meters iioidc \ed Kristiania, paa den anden side den 
yngste Po/i/a/K/ia-i'aunn i lik blomstring paa Komerike i et noget lavere 
nivaa, som dog korrespondere)- med en ha\stand som gaar op mot den 
marine grænse (som her forovrig er vanskelig at fiksere). 

Naar saa hertil kom at allerede Blvtt (1. c. 1892) mente at finde spor 
i vore tjelddaler efter en 3ngrc bræfremrykning (efter den egentlige 
avsmeltningstid), og Oven ved sine tidligere studier i det centrale Norge, 
specielt Atnedalen, kom til det resultat, at store dalbræer her hadde be- 
væget sig nedover dalforene efter selve storbræens ophor (1899 1. c), vil 
man forstaa at det laa noksaa nær at forsøke paa at kombinere alle disse 
fænomener. 

Dette har da Oven gjort. Han antar at der efter Romerikstadiet fulgte 
en mildere periode, „Mytilits-nivaact , hvorunder bl. a. de rike banker ved 
Kristiania i 221 m. hoide blev dannet. Under denne tid trak bræene sig 
langt tilbake, men i hvor stor utstrækning indlandsisen avsmeltet, kan ikke 
avgjøres med bestemthet. Oven antar videre at havet nu gik ind i Mjesen 
helt op forbi Lillehammer. Denne periode paralelliseres med Blvtts „ark- 
tiske" tid. 

Efter denne mildere periode fulgte atter en depression av klimatet 
under „Portlaiidia-iiivaaets'^ tid. Bræene skjot fra det centrale Norge ned- 
over hoveddalforene og helt ned til enden av de store indsjoer. Da hadde 
havet allerede sunket noget. Bræene ploiet op lerlagene fra Mytiliis- 
nivaaets tid, og disse blev indpresset i de store endemoræner foran Mjosen, 
Hurdalsvandet o. s. v. Denne periode paralelliseres med Blytts „sub- 
glaciale" tid. 

Forholdene ved Lillehammer synes at stemme godt med denne opfatning. 
Allerede i 1895 omtaler nemlig Rekstad nogen merkelige avleiringer ved 
Aaretta i km. syd for Lillehammer, ved N3'flot nær Losna o. fl. steder. Paa 
den første lokalitet saaes en veritabel moræne op paa horisontale 
sandlag av betydelig mægtighet (1895 1. c. p. 14). „Disse sand- 
og lerlag maa rimeligvis være avsat under svingninger av bræstanden i 
slutningen av den sidste istid." Det hele tyder paa „at mildere klima er 
bleven fulgt av glaciale tilstande" (1. c. p. 14). Senere har overlærer Holme 
iagttat lignende fine sand- og lerlag overleiret av glaciale avsætninger, til- 



' Flere av disse arter optrær i sydlige varieteter. 



1921. No. 9- KALKTl'FSTL-DIER I GUDBRAXDSDALEX. 9I 

dels ogsaa hvilende paa ældre moræner paa en række steder ved Lille- 
hammer og nordover til Myre st. Oven, som ogsaa har undersøkt for- 
holdene her, tolker alt dette som A/vfi/iis-nWaaets dannelser overleiret av 
Porf/a/idia-nivs^aet (Oyex 1915 1. c. p. 307— 310I. Bjørlykke har ogsaa 
iagttat lignende merkelige stratigrafiske forhold ved Svanefoskanalen og 
ved Lena-elven paa Toten (de mellem morænene liggende lerlag var her 
3,5 m. tykke) (Bjørlykke 1. c. p. 144I. 

Den vngste Portlaiidia ri'/r//Vr?-fauna paa Romerike faar herigjennem 
sin naturlige forklaring. Ishavsmuslingen har hat gunstige livsvilkaar under 
denne nve kuldeperiode. Man mener ogsaa at ha fundet il/\'//7//s-førende 
1er under Po/i/aïKÙ'a-i'orende ved Jesseim; men dette er endda ikke bevist. 
Bjørlykke mener at det motsatte er tilfældet (Oyex: Nogle bemerkninger 
om klimatforandringer. Vid.-Selsk. Forh. 1904. Bjørlykke 19 i 3 1. c. p. 151). 

ØYEX har ogsaa senere fundet brudstykker av Portlandia arctica \ed 
Skaadalen st., ca. 16 m. lavere end de ovenfor nævnte il/v/////5-banker (1909 
1. c. og 19 15 1- c. p. 290). Han antar at drivis under Poii/aiK/ia-nWaaets tid 
har stott mot stranden paa dette sted og trykket og knuget ældre sandlag 
med J/r//7//.s-faunaen ind i lerlag med Poii/a/KÙ'a-i'aun^ien. De forstyrrede 
lag som blev fundet her, er ialfald vanskelige at forklare paa nogen anden 
maate. Saavidt vites har ingen anden norsk geolog forsokt at utrede denne 
forekomsts natur. 

Endelig fortjener det at nævnes at den store bræ i Atnedalen, som fra 
Rondemassivet har skutt ned mot Osterdalen (ogsaa beskrevet av Werex- 
SKioLD 191 i), ifølge Oyex maa ha dæmmet op for Glommen og tvunget 
denne til at rende ostover mot Rendalen. Under denne periode skal „Jutul- 
hugget" være blit utformet'. Pashoiden over hugget korresponderer med 
de laveste seter i Osterdalen (ofr. Holmsex 19 i 5 p. 143), som likeledes 
tolkes som minder fra denne opdæmning. 

Nu, alle disse fænomener som Jeg her har referert, er tildels meget 
omstridt, og flere norske geologer stiller sig skeptisk, tildels avvisende 
overfor Mytilus- og Port/a/idia-nivaaet saaledes som disse teoretisk er ut- 
formet av Oyex. Paa Vesdandet har Kaldhol (1909 og 1912 1. c.l, delvis 
ogsaa KoLDERUP (1911) fundet forhold som stemmer med Oyexs opfatning. 

Lnidlertid kræves der endnu omfattende undersøkeiser for at klargjøre 
forholdene. Vi kjender endda ikke sikkert lagfølgen paa den romerikske 
slette; vi vet heller ikke, hvor langt bræene trak sig tilbake indover mot 
det centrale Norge forinden de atter rykket frem til enden av de store 
indsjøer. Hvor stor rest av indlandsisen overlevet i]/v/////5-nivaaet, vites 
ikke; vi kjender heller ikke dette intervals varighet og hele oscillationens 
amplitude. Oyex har forsøkt paa grundlag av sine studier over ler- 
sedimentene i Trondhjemsfeltet at tidfæste bræsjo- og sete-perioden i det 



' Smlgn. ogsaa Schetelig (foredrag, referert i Norsk Geol. Tidsskrift B. r. p. 44), og 
Reusch I1917 L c.l, hvor Jutulhugget omtales. 



92 



UOI.F NOHDHAGKN. M.-X. KI. 



ostcnljcldskc Norj^c. I i.iii .iiil.u- al drnnc av.sluttcdcs hovcdsakclig for 
i1/i7/7//.s-nivaa(ts lid (1915 I. <". p. 294 297). Laiirliscn riiaatte altsaa paa 
denne tid ha været sterkt decimeret. I Iolmskn iiiencr derimot at en del 
av isrestcn i est overlevet denne iiiild( re periode, og at den atter vokste 
noget under /'r>i/-/A///r//cMiivaaets tid (sainlidig med lokalglaciationen), da sne- 
linjt n blev forskjøvet nedad. (G. Holmsen: Forskyvninger i snelinjens høide 
under avsmeltningsperioden. 1918 1. r.l. 

Dette tvistepunkt er overordenllig \igtig. S\en.ske kxartærgeologer 
hæx'der nemlig i motsætning til Ovf.n, at indlandsisen i ost vedvarte et godt 
stykke utovei- i postarktisk tid, ja man ser gjennemgaaende paa svenske 
karter ogsaa indlagt en vedvarende pølseformet isrest langt ind i Gud- 
brandsdalen. Dette punkt kommer vi tilbake til i de følgende avsnit. Kalk- 
tuffene er i denne forbindelse meget betydningsfulde. 

Ogsaa fra andre kanter i Skandinavien er der beskrevet eiendomme- 
lige avleiringer og fænomener som i flere henseender synes at stemme 
overens med Oyexs tolkning. Dette gjælder først og fremst den saakaldte 
Allerød-oscillation i Danmark', som nu maa betragtes som viden- 
skabelig fastslaat, idet alle de indvendinger som har været reist herimot, 
har vist sig at være ugrundet (Naturforskermotets forhandlinger, Kristiania 
19 18, p. 418 — 421). „Allerød-tiden" har Nordmann og Øven paralellisert 
med 71/v/////5-nivaaets tempererte periode („senglaciale klimatoptimum" Jessen 
1920 1. c.(, „den yngre Dryas-tid" med Porf/audia-nivaaet At der i Dan- 
mark har skedd store forandringer i vegetationsdækket under denne oscilla- 
tion, er en videnskabelig kjendsgjerning. 

I Sverige har Enouist i flere arbeider behandlet hvad han kalder 
„lokalglaciationen". Han opdaget forst I1910 1. c.) at visse bræer i Herje- 
dalen — Jemtland tidligere har været betydelig storre end nu; saaledes har 
Helagsfjeldets bræ engang skutt 1500 m. længer frem end i nutiden. Lig- 
nende fremstot markert ved endemoræner har han iagttat flere steder (i 
Norge har Holmsen omtalt en endemoræne fra Hummeltjeldet i Tolgen^l. 
Senere mener Enouist at ha fundet spor efter intens lokalglaciation paa 
en række steder i det nordvestlige Skandinavien (19 18 1. c). Eiendommelig 
nok synes Enquist ikke at kjende til den diskussion som man i norsk lit- 
teratur har ført om lokalglaciation (f. eks. Blytt 1892, Oyen: „Kontinental- 
glaciation og lokalnedisning" 1899 o. s. v.). Hans opfatning av hele fæno- 
menet er ogsaa en anden end Øyens. Enouist antar nemlig at de av ham' 
studerte fænomener skriver sig fra den sidste istids avsmeltningsfase, specielt 



' Litteratur om Allerod-perioden: Hartz og Milthers 1901 1. c, Hartz 1903 1. c, 
Johansen 1904 1. c, Nordmann 1912 1. c, Øyen 1915 1. c, De Geer 1916 1. c, Nord- 
MANN 191 8 1. c, Jessen 1920 1. c. 

2 1 915 og 1918 1. c. Enquist anforer ogsaa i et senere arbeide (1918 1. c.) (ifolge 
H. Smith), at der paa Knutsho, Dovre, skal findes en endemoræne som tyder paa lokal- 
glaciation. Antageligvis er dette den sidemoræne i 1500 m. hoide som Holmsen har 
omtalt 1 191 5 1. c. p. 1 89 1, og som ifolge ham skriver sig fra indlandsisen. 



ICSI. No. 9- KALKTUFSTL-DIER I GLDBRAXDSDALEX. 93 

fra dennes forste del, da det sammenhængende isdække forsvandt fra 
visse omraader og gav plads for lokale bræer. Hans kart over isdækkets 
utbredelse under denne tid er imidlertid meget lite overbevisende og staar 
for Norges vedkommende i strid med hvad vi vet om isavsmeltningen i det 
trondhjemske, hvor en lang række med morænetrin er beskrevet og delvis 
sammenstillet med de østlandske stadier. At lokalglaciationen netop skal 
svare til det tidspunkt, da bræene i det sydlige Norge stod ved raene', 
er ogsaa i grunden bare et postulat. — De av ExouiST omtalte fænomener, 
specielt fra Herjedalen — Jemtland, kan, saavidt jeg forstaar, ogsaa meget 
godt tages til indtægt for Ovens lille atpaa-istid [Portlandia-iiivaact), som 
han sammenstiller med Bühl-X'orstoss i Alperne og med amerikanernes 
PostAVisconsinperiode. 



Jeg har med vilje omtalt Mytiliis- og Po/i/a//(/ia-nW3.aene ganske ind- 
gaaende, fordi de sporsmaal som knytter sig til dem, er ganske centrale i 
norsk kvartærgeologi. Saalænge der ikke opnaaes enighet paa dette punkt, 
kommer man ikke av flekken. Man maa haabe paa at disse problemer i frem- 
tiden blir underkastet en kritisk og indgaaende behandling og belyst fra for- 
skjellige sider; ellers kan der jo ingen diskussion bli. Mange svenske geologer 
negligerer desværre helt disse vigtige sporsmaal. Jeg for min del tror at 
dette i forste række skyldes det forhold, at der i norsk litteratur ikke fore- 
ligger nogen tilfredsstillende klar og koncis fremstilling a\- disse aktuelle 
problemer. Polemiske og krigerske utfald mot anderledes tænkende torer 
her aldrig til maalet, men kun en kritisk og saklig utredning. 

Saalænge situationen ligger saaledes an, er det vanskelig at faa istand 
en overensstemmelse mellem de nyere svenske oversigter over isdækkets 
avsmeltnine: osr de norske forhold-. 



Skal man gi en fremstilling av isavsmeltningen i selve Gudbrands- 
dalen, vil denne i hoi grad axiiænge a\- den opfatning man har av Mytilus- 
og Por/Iandia-nivaaene. Man finder den ene morænerække efter den anden 
fra Lillehammer og helt op til Otta, hvilket utvetydig viser at bræene her 
i den sidste avsmeltningstid har bevæget sig nedover dalen (NV — SØ), 
og at deres utgangspunkt har været i nærheten av landets hoideakse {Rek- 
STAD 1895, 1896 og 1898I. Isskillet har ligget ved grænsen 
mellem Do vre og Sel herred. De kartografiske fremstillinger som 
man finder i litteraturen, hvor en polseformet isrest placeres over S. Fron, 
Ringebu eller Oier, er altsaa ikke rigtige. (Cfr. Oyexs undersokelser ved 
Rosten 1899 1. c. p. 54 — 56 og Holmsexs stripekart 19 15 1. c. p. 22). Man 



' Disse er helt feilagtig indtegnet paa Enquists kart. 

2 Cfr. Øyen: Naturen 1916 p. 223 og Hoel i Geol. Foren. Förh. Stockholm 1916 p. 484. 



94 



ROLF NOKOHAGEN. M.-N. Kl. 



Ii;ii- i (Iciiiic (1(1 ;i\ ( 'iii(lbraii(l-.(l;iliii (i\ ciIkkN l iiii( l ic;^ii lil ;ii hræbcvægelsen 
paa noget slid h.ir \a id i( ll( t dpovcr d.-ilforct, saaledcs som f. eks. i 
Osterdalstraklciic '. 

II\i.s in;m nii akccplcrcr / 'nii/n//r///i-\nvaaQt som en egen avdeling 
karakteriseit ved lokalglacialion i det centrale Norge, maa man anta at alle 
disse morænerækker i dalen er sammenskjovet av de sidste bræer i disse 
strøk under Pur//(i//ff/<i-/ih'(ia(is a vsl u tn i ngsfase, og at de markerer 
temporære stilstande ellei- mindre fremrykninger under den sukcessive av- 
smeltning. Ældre avleiringer er ogsaa tildels bevaret, men det er de sidste 
bræer som har sat de tydeligste merker efter sig. 

Anerkjender man derimot ikke /^ort/aridia-n'ivaaet som nogen egen 
periode, blir morænerækkene at henføre til selve indlands-isrestens av- 
smeltningstid. Av morænenes placering og skuringsmerkene fremgaar det 
da at der her har været nedgaaende dalbræer fra isskillet nær landets 
hoideakse og ingen pølseformet isrest over stroket Fron — Øier. 

Da jeg selv ikke har arbeidet med disse trng, skal jeg her bare nøie 
mig med denne orientering i problemstillingen^. 

I indledningen omtalte jeg de merkelige moræneavleiringer som op- 
bygger Leinebakkene i Kvam. Disse har som nævnt engang sikkert fyldt 
op hele den nedre del av Veiklas nuværende dalføre, idet der anstaar 
rester av fyldningen ogsaa paa elvens østside. Den har sandsynlig\'is 
gravet sig ned gjennem disse og flyttet sig vestover. Moræneleren indeholder 
talrike blokker, fleresteds i tætpakkede lag (skyldes ogsaa rasl. For om 
mulig at komme paa spor efter isbræens bevægelsesretning, medtok jeg en 
del bergartsprøver, som professor J. Schetelig har været saa elskværdig 
at undersøke. Det viste sig at være forskjellige spar agm i ttN'per, 
kalks andsten er, fylliter, desuten saaes blokker av en grov, rød 
i egn eis. Alle disse bergarter angives av Bjørlykke for traktene 
nord og delvis nordost for Leine; den kalkrike morænelere forutsætter 
kalkholdige bergarter, og saadanne optrær ogsaa i nærheten. Oiegneis an- 
staar i foten av den lille top Gnedden nord for Leine og i Formokampen 
nord for Otta (Bjorlykke 1905 1. cl. 

Ved Kolloen nær Sjoa og Sandbovangen (fig. 30) litt høiere oppe i dalen 
findes der store endemoræner, beskrevet av Rekstad, og disse angir en bræ- 
bevægelse NV — SO (ut Hedalen og Ottadalen). Morænemassene ved Leine 
er ældre og vistnok avsat av en bræ fra nord — nordvest; men om dette 
har været hovedbræen fra Ottadalen, som har presset sig sydøstover 



' En unciersokelse av blokkene i de tindre moræneavleiringer paa de ste(der hvor man 
har intramorænale san(d- eller lerlag (cfr. ovenfor), maa kunne gi oplysninger om man 
her nogen gang har hat blokketl^-tning opover (dalforet. (Cfr. Frödins utredning av 
Frösöfyndet. Geol. Foren. Förh. B. 38. 1916). 

2 Jeg har her ikke omtalt Dovretraktens bræsjo og de omdisputerte sete-dannelser aller 
overst i dalen. Tidsbestemmelsen vil her ogsaa avhænge av det standpunkt man ind- 
tar til Por/laiidia-nWaaet. 



1 92 1. No. 9. 



KALKTUFSTLDIER I GUDBRANDSDALEN. 



95 




96 ROLF NORrjHAf;EN. M.-N. Kl. 

'rorj^crkaiiip-pl.itaacl ', < ll( r om in mere nordostlij^ bræ fra I< ondemassivet — 
]'\irusj()cii har arlxidc t sig sydover mol K\am, vet vi f'orelobig ikke. Blok- 
kene i leret .i^ii" iiii^cn sikker besked herom. MoræneU-ren gaar op til 6 à 
700 m. o. h., altsaa temmelig- hoit. Wkrenskiold omtaler imidlertid side- 
moræner i S. l'i'on o[) til 600 m. hoide (191 1 1. e.). ^ 

Avsætningene ved Leine maa iit\ilsomt kunne tidfæstes i forhold til 
moræner baade sondenfor og nordenlor K\'am; men geologene har hittil 
merkelig nok ikke olret dem nogen opmerksf)mhet eller diskutert deres alder. 



II. Nogen bemerkninger om den første flora og vegstation 
under avsmeltningstiden. 

1 ældre palæobotaniske oversigter finder man opfort en saakaldt „Dryas- 
tid" for saagodtsom hele Skandinavien som den forste epoke efter isens 
avsmeltning. Imidlertid har opfatningen paa dette punkt ændret sig ganske 
betydelig i de senere aar, eftersom forskningen er skredet fremover. 

I Danmark maa det nu ansees for fastslaat at man har hat en oprindelig 
„ældre Dryastid" med fuldstændig arktisk præg. Derefter folger den eien- 
dommelige „Allerød-oscillation", som synes at udmerke sig ved aapne krat- 
skoger av bjerk, asp og vidjer (Jessen 1920 I. c. p. 2i8(. Under den „3'ngre 
Dryastid" synes atter de arktiske planter at ha hat overtaket. 

Oprindelig antok man at en lignende arktisk vegetation var fulgt efter 
isen nordover gjennem Sverige til Norge. Og i det sydlige Sverige, hvor 
Nathorst gjorde sine klassiske undersøkeiser, har nok dette været til- 
fældet. Men jo længer nordover man kommer, desto mindre markert 
synes den arktiske floras blomstringstid at ha været, desto mere opblandet 
er den med subarktiske arter (G. Andersson 1906 1. c. p. 60). Rigtignok 
er arktiske planterester iagttat paa en række steder helt nord til Laxå i 
Mellem-Sverige, og ved Kristiania har Oyen bl. a. fundet Sa/ix polaris i 
lerlag paa flere steder. Men forholdene stemmer i almindelighet, saavidt 
man kan se, ikke ganske med de i Danmark paaviste. Og en række bota- 
nikere og geologer synes at være av den opfatning, at da isen hadde trukket 
sig saa langt tilbake som op i Mellem-Sverige, hadde de klimatiske forhold 
ændret karakter; de \-ar ikke længer arktiske, men tempererte. Man antar 
at en blanding av arktiske og subarktiske arter her tok landet i besiddelse 
(Andersson 1. c., Wille 1. c. p. 325). 

For at kunne forklare den skandina\iske tjeldfloras tilstedeværelse har 
man grepet til forskj eilige teorier, som desuten forsøker at løse en række 
gaader som knytter sig til det samme flora-element. Blytt (1893 1. c. 
p. 21) og Sernander (1896 p. 117) har allerede for lang tid siden hævdet, 
at vi i Norge maa ha hat plantekolonier som paa isfrie nunatakker 



' Skuringsstripene paa Gudbrandsdalens ostside i stroket Sel kirke — Otta st. — Bredevangen 
tyder paa dette (cfr. Rekstads kart 1896 1. c). 



I92I. No. Q. KALKTl'FSTUDIER I GUDBRAXDSDALEX. 97 

overlevet den sidste istid. A\' nyere datum er teorien om den i s tv i e 
kystrand under den sidste istid, med overlevende haardfore arter (Han- 
sen 1904, Wille 1905, Fries 1913, Tengwall 1913I. Denne teori kom- 
bineres nu ofte med nunatak-teorien, som jo ikke er væsensforskj eilig fra 
den forste (Fries 1913, Tengwall 1913I. 

Fund av arktiske planterester i det trondhjemske og langs Norges 
vestlandskvst taler til gunst for denne opfatning, selv om de strengt tat 
ikke beviser andet end at isranden her i vest og nordvest blev efterfulgt 
av arktiske planter, saaledes som i Danmark og Syd-Sverige. 

Av stor teoretisk interesse er Ovens fund i det trondhjemske, specielt 
naar disse stilles i relation til isens avsmeltning i disse strøk. I sit arbeide 
om Trondhjemsfeltet (1915) paralelliserer Oyen den store „Ørlandsbanke" 
ved mundingen a\- Trondhjemsfjorden med „Stagnations-trinnet" i Danmark 
— med andre ord: vi skulde i O ri an ds ban ken ha grænsen for 
den sidste istids (Mecklenburgian) brædække i denne del av 
landet (ofr. Ussings kart). Denne opfatning, som er basert paa de eien- 
dommelige faunistiske forhold som udmerker disse gamle avleiringer, der 
stemmer med tilsvarende avsætninger i Danmark', medforer meget vigtige 
konsekvenser; der aapner sig da muligheter for en isfri, om end 
smal kystrand son denfor Trondhjemsfjorden, i al fa Id mulig- 
heter for opragende fjeldpartier. 

Ved Ytterland fandt Oyen blader av Sa/ix polaris (1901), hvilket viser 
at der paa Oiiandet eller i dets nærhet har været en arktisk vegetation. 
Senere har Oven kunnet forfolge denne arktiske Hora eftersom den trak 
sig længer indover det trondhjemske. Sommeren 1900 fandt han nemlig 
Sa/ix polaris ved Nidaros og Reitgj erdet teglverk nær Trondhjem og i 
1901 Sa/ix rcticti/ata og Dryas odopeta/a i 164,4 'i^- lioide ved Sandsæter- 
volden nær HommeKik (Mytilus-ni\aaet. 1915 1. c. p. 291 1. I 1909 fandt 
BjøRLVKKE Sa/ix rcticti/ata ved Reitgjerdet. 

Alle disse fund er fra en langt senere periode end Orlandsforekomsten 
og maa ikke sammenstilles med denne, saaledes som Enquist har gjort paa 
sit kart (191 8 1. c. p. 82)-. De interessante fund av arktiske planterester 
som H. Smith i den nyeste tid har gjort i Jemtland-Herjedalens fjeldom- 
raader (191 7 1. c., 1920 1. c.), og som utvilsomt er meget gamle, danner paa 
en maate en fortsættelse østover av de ovenfor nævnte forekomster i Trond- 
hjemsfeltets marine avleiringer. Dog maa man medgi at en sikker tid- 



' Cfr. Portlatidia arctica, Maconia Torelli, Macoiiia Lovcni, Biiccimiiii terrae novae, Siplio 
virgatus og 5. Verkriitseui, Cylichna scalpta, L'tricitlns pcrtemiis, hoiarktiske br^'ozoer 
etc. (ØYEN 19 15 1. c. p. 177). 

2 Her fremstilles alle norske fund av arktiske planterester tilsyneladende sans façon 
som synkrone. Ja, raene trækkes helt op til Kristiania, vistnok fordi Salix 
polaris er fundet i byens nærhet, og fordi disse findesteder helst „burde" ligge uten- 
for raene! Man maa virkelig ha lov til at sporre hvad meningen er med alle disse 
manipulationer? Naar man finder det umaken værd at indtegne findestedene paa et kart, 
bør man ogsaa studere originaloplysningene og citere forfatteren. 
Vid.-Selsk. Skr. L M.-N. Kl. 1921. No. 9. 7 



gS ROLF NORDKAGKN. M.-N. Kl. 

fæstelsc av arktiske plaiili IiiikI i (I linit'jcldsoinraadc cr forljiiiulct med ad- 
skillige vanskclighetcr. Smiiiis tolkniiii; virker dog meget tiltalende, og 
det maa vel nu betragtes som fastslaat at der har gaat en overordentlig 
vigtig indvandringsstrøm av fjeldplanter fra Norge i vest og østover ind i 
det centralsvenske luiifjeldsomraade. 1 <!( Itiindlag som S.Mnn har analysert, 
fandtes der ikke spor av pollen hverken av bjerk eller furu. Dette stemmer 
med den av Andf.rsson og Bir(;ek fremsatte anskuelse (1912 1. c. p. 142): 
„Vi anse det således sannolikast atl tallskog icke funnits eller atminstone 
icke spelat nagon roll waster diu isdelaren sa länge nagon afsevärd inlands- 
isrest ännu kvarlag." Andersson antar ogsaa at bjerkeregionens arter 
senere vandret den samme vci østover ind i Sverige fra Norge i de an- 
gjældende trakter (1912 1. c. p. 141). Dette forutsætter med bestemthet en 
„bjerk-asp-periode" ialfald i visse deler av Norge (nordvestlige og centrale 
del) et punkt av stor vigtighet, som vi senere kommer tilbake til. 

De mange interessante fund som Holmüok, Kekstad og Kolderup 
har gjort paa Vestlandet, viser at ogsaa her har en rent arktisk vegetation 
holdt til langs isranden under avsmeltningstiden; men tidfæstelsen i forhold 
til bestemte avsmeltningsstadier er her meget vanskelig. 

Th. c. E. Fries har i sit interessante arbeide fra det nordlige Sverige 
(i 9 1 3) trukket frem en række nye plantegeografiske fakta som taler til 
gunst for en „overvintrende Mecklenburgo-glacial flora" paa visse stræk- 
ninger i Norge (cfr. hans inddeling av ijeldplantene i bicentriske arter, 
vestarktiske arter o. s. v. 191 3 1. c. p. 330). „Ob das eisfreie Land eben 
an der Küste gelegen war, lasse ich dahingestellt sein. Es sprechen in- 
dessen gewisse pflanzengeographische Verhältnisse dafür daß? die eisfreien 
Nunatakken nicht an der Küste sondern weiter in das Land hinein gelegen 
waren" (1. c. p. 315K 

Jeg hadde sidste sommer anledning til at gjore nogen iagttagelser fra 
et av de omraader i det nordlige. Norge, hvor baade Feries og Tengwall 
tænker sig at der har været overvintrende Mecklenburgo-glaciale plante- 
kolonier, nemlig i stroket Saltdalen — Sulitjelma. Jeg vil her benytte an- 
ledningen til at nævne litt om forholdene i disse strok; de er nemlig ad- 
skillig mere problematiske end vistnok mange har tænkt sig. 

Jeg skal her kun opholde mig \ed forekomsten av Carex scirpoidea 
og Saxifraga Aizooii, to aiter som baade av Fries og Tengwall regnes 
med til de interglaciale overvintrere. 

Carex scirpoidea har sine eneste europæiske voksesteder ved Solvaag- 
tind i Junkersdalen (Dyring 1900 1. c. p. 277). Den vokser her over en 
kortere strækning paa fjeldets S3'destlige side i 760 m. hoide omtrent 150 m. 
over skoggrænsen. 1 191 6 blev arten ogsaa fundet paa vestsiden av Sol- 
vaagtind paa skraaningen av fjeldet Trækta av Dr. E. Hayrén (Finlandias 
Årsbok 1919 p. 57). Dette lave fjeld (912 m.) danner den vestligste be- 
grænsning av Solvaagtinds skraaning mot Saltdalen. Det nve findested 
ligger 435 km. fjernet fra det ældste kjendte. 



192 1. No. 9. 



KALKTUFSTUDIER I GL'DRRAXDSDALEX. 



99 



Under et besok paa stedet sommeren 1920 kunde jeg konstatere at 
fjeldskraaningene paa Solvaagtinds ostside til og med i et høiere nivaa 
end findestedet for Carcx scirpoidca var belagt med erratiske blokker (t\-ndt 
morænedække). Og a\- hele topografien og isskuringen kan man se at bræ- 
massene her har presset sig fra Junkersdalen nordvestover netop i skaret 
















A. 







> Uff 



1^^* 



/.?< 






L 



^c 



^-^if 










\ 



.ï'IîiAt 



\ 



Fig. 31. Kart over stroket Solvaagtind — Baatfjeld i Junkersdalen. Voksestedene for Car ex 

scirpoidca er avmerket med +. 



mellem Sohaagtind og det ostenfor beliggende Baatfjeld. Naar man har 
hat anledning til at se Solvaagtinds besynderlig formede top fra forskjellige 
kanter, faar man indtr\k av at den er tilsiepet som en vridd kam av 
isen, der har glidd forbi paa østsiden og vestsiden. En ting er ialfald 
sikker: Carcx scirpoidca kan umAilig ha \'okset paa de nuværende vokse- 
steder under den sidste istid. Planten synes ogsaa at være noget fordrings- 
fuld i valget av voksested. Den optrær paa smaa skraanende græsmyrer 



KOLK NOKDIIAGK.N. 



M.-X. Kl. 



iiicd (diiw /uir<i//(/ii, ( . (d/u/liiri-^, ( . (ilj>ni<i, ( . rlt^iiln, Sii/i.x nticiilnld, 
Si/riic acaii/is, Sai/ss/in n (iljiiiiti, l'olys^utniin -.'ivi/xiniiii, linrl.siliid nl/>iii(i 
o. fl. samt »11 il<l ciiliarciUi iiHi-^cr i huiidcn. I loicrc oppe paa skraaningen 
av tinden lindes i\Ln ikke, oi; di mx s nakiic trip merl f|c steile styrtninger er 
aldeles uskikket for planten i niiliilen, og endda mindre under et strengere 
kliiiial. 1 I \' i s ('(fjy'x sci/'/ut/ffra har overlevet den sidste istid i 
disse s trek, m a a det ha været paa et andet sted. Det lar sig 
ogsaa tænke at denne eneste og merkelige forekomst av arten i Kuropa 
er de sidste rester av et tidligere storre utbredelsesielt. Specielt maa man 




Fig. 32. Utsigt fra Rosnevarre og s^-iJuvcr mut Sukljuldcl og Skaitidalen. Lave, avrundede 
klipper. 12te juli 1920. Nordhagen fot. 



tænke sig at den postglaciale varmetid nued sin vældige forskyvning av 
skoggrænsene opad paa fjeldene har utryddet mange arter og indskrænket 
deres utbredelsesfelt (Hansen 1904). Det er ogsaa forskjellige ting ved 
denne Solvaagtindlokalitet som synes at staa i forbindelse hermed. I nu- 
tiden er der netop paa østsiden av tjeldet og i skaret ved Solvaagvandet en 
voldsomt sterk vindvirkning, som presser skoggrænsen nedad. Denne 
„avtrækkanal" til og fra Junkersdal, som specielt i gamle dager, da ride- 
veien til Evenesdalen og Salten forte forbi her, var ber3'gtet blandt befolk- 
ningen, har utvilsomt altid paavirket vegetationen paa stedet', og sikkert 
ogsaa under den hoiereliggende skoggrænses tid forhindret skogen i at 



' I Juli 1920 maatte man krype paa alle fire for overhovedet at komme gjennem passet 
mot Solvaagvandet. 



igsi. No. 9. 



KALKTUf^STUDIER I GUDBRAXDSDALEX. 



sætte sig fast paa disse avblaaste skraaninger. Carcx scirpoidcn maa her 
ha hat specielt gunstige betingelser for at kunne o\-erleve varmetiden og 
dens utrvddelseskrig mot Ijeldplantene. 

Problemet Cairx scirpoidca og dens „overvintring" er vanskeligere at 
lose end vistnok mange har tænkt sig. Dette gjælder i endda høiere grad 
Saxifraga Aizooii. Denne art har efter Læstadius', Schlegel og Arnells 
og i nyere tid overingenior Karlsons undersokelser et litet utbredelsesfelt 
i det nordlige Norge nordost for Carcx scirpoidca, nemlig paa begge sider 
av Balvandet syd for Sulitjelma. Sidste sommer kunde konservator Ove 
Dahl og undertegnede konstatere at planten fremdeles findes paa Balvandets 
østside (Læstadius' lokalitet 

„Balvandsryggen"), hvilket man I 

har betvilet, idet den flere 
ganger har været eftersokt, 
men med negativt resultat. 

Baade paa Rosnevarre i 
nordvest for Balvand og paa 
Balvandsry ggen (ved grænsen) 
optrær Saxifraga Aizoon i 
forholdsvis lave nivaaer. Fjel- 
dene er ikke hoiere end 8 à 
850 m., og planten overstiger 
neppe 800 m. Disse to fjeld- 
partier er paafaldende like. 
De er jevne og glatskuret av 
isen, som her er kommet fra 
SØ (cfr. Rekstad og Holm- 
SENS kart 191 7 1. c). Overalt 
paa disse avrundede berg og 

jevne skraaninger finder man et tyndt b u n d m o r æ n e d æ k k e, saa tyndt at 
det væsentlig bestaar av utstrodde blokker, som ofte ligger placert paa de be- 
synderligste steder (cfr. billedene fra Rosnevarre og Bahandsryggen). Berg- 
grunden bestaar begge steder av kalkholdig glimmerskifer og bærer gjennem- 
gaaende en sparsom og aapen vegetation paa grund av manglende los- 
materiale. Her vokser Saxifraga Aizooii ganske talrik paa flate berg eller 
svake Ijeldskraaninger i smaa sprækker og fordypninger, ikke paa hvider 
eller steile avsatser, som man gjerne tænker sig paa forhaand. Voksestedet 
er i det hele tat meget karakteristisk og eiendommelig. 

Det er nu ganske klart, at da planten kun optrær her i relativt lave 
nivaaer paa steder som har været isdækket lang tid uto\er i avsmeltnings- 
tiden, kan den umulig ha overlevet den sidste istid paa dette 
sted. Den synes at ha lette konkurrencevilkaar paa disse glatskurte 
klipper, idet den xerofile vegetation her er meget aapen. Dette er kanske 
en av aarsakene til at den holder sig netop her. Desuten kommer det 




Fig. 33. Balvandsryggen nær Kvebilokskaret. Isskuret 
tjeld med erratiske blokker. Saxifraga Aisooii-colsiere 
i forgrunden. Det kamel-lignende tjeld i bakgrunden er 
Nord-Saulo. Utsigt mot nordost. Den lille top tilhoire 
er i virkeligheten en lang hoideryg, som her sees i profil. 
Ilte juli 1920. Nordhagen fot. 



KOM" Noi<i)HAf;i:.N. M.-N. Kl. 



samme moment til si mi jeg omtaite l'or Carrx scir/x/if/ca, ncmlij^ ^^^^ii' 
grænsen nndei" varmetiden. 

Ilvei'ken paa de anjijældende deler av Rcsnevarre eller lialvandsryKf^on 
kan der ha x'okset virkelig skog under varmetiden, som kan ha utkonkur- 
reret tjeldplantene paa stedet (f. eks. ved beskygningl. \'i har her kanske 
en av aarsakene til at Bahandsi-yggen ogsaa ellers rummer saa mange 
sjeldenheter. 

Men h\'orlra er da Snxi/'rnj^ti ylizoon kommet? Typiske nunatakformer 
(Turmskulptur, Knqüist 1918 1. r.) tindes ikke i nærheten, og partiet syd for 
Balvand (Ai-galadei Salefjeldet o. s. v.) synes at ha været et bræcentrum. 
Planten forekommer heller ikke i store heider. Man maa vistnok anta, at 
den har fulgt elter isen da denne trak sig tilbake fra Skjærstadtjorden og 
indover landet, og at den oprindelig hadde et videre utbredelsesfelt end 
nu. Dens nuværende isolerte forekomst er oiensynlig betinget av vokse- 
stedenes eiendommelige natur. 

Baade for Carrx scirpoidca og Saxifrai^a Aizooii er spørsmaalet om 
„overvintring under den sidste istid" ikke saa kurant som man kanske til 
en begyndelse skulde tro. Ovenstaaende utredning viser hvor uhyre van- 
skelig det er av en plantearts mere eller mindre isolerte Ibrekomst i nutiden 
at trække slutninger bakover i tiden. Det nuværende alpine utbredelsesfelt 
er nemlig bestemt eller saa at si resultanten av en række faktorer, forst 
og fremst: i) plantens utbredelsesfelt før den postglaciale varmetid, 2) ind- 
skrænkninger og utdoen under den høiereliggende skoggrænses tid, 3 1 even- 
tuelle nyere spredninger efter det klimatomslag som rykket skoggrænsene 
ned til sit nuværende leie, og som følgelig utvidet regio alpina. 

Jeg har her villet omtale disse fænomener fordi slagord som „nunatak- 
kolonier" og „isfri kystrand" i de senere aar stadig dukker op i litteraturen, 
men uten at vi er rykket problemene nærmere ind paa livet. Special- 
undersøkelser mangler nemlig totalt. Og det nytter efter min mening 
ikke længer bare at ræsonnere mere eller mindre abstrakt over disse pro- 
blemer; ialfald kommer vi ikke videre ad den vei. 



I likhet med Hansen (1904) og Wille (1905) antar Fries at en del 
av den skandinaviske fjeldflora er indvandret fra øst i senglacial tid'; des- 
uten regner han med en indvandring fra syd. 

Tengwall har fæstet opmerksomheten bl. a. \ed Kobrcsia bipartita, 
Pedicttlaris Orden', Caiiipainila barbata, Geiitiana piirpiirra og Ranimculits 
platauifoliits, som han kalder sydlige fj eld pi an t e r-. For disse antar 
han en sydlig indvandringsvei; men da de ikke har formaadd at sprede 



1 Fries fordeler imidlertid de pstlige indvandrere paa flere grupper ifjeldarter, bjerk- 0£ 
naaleskogarter). Hans inddeling er saaledes en anden end de nævnte forfatteres os 
problemstillingen mere specialisert. 

2 Til disse slutter Nipritclla in's'rci sig. 



1921. No. 9. 



KALKTUFSTCDIER I GL-DRRAXDSDALE.\. 



103 



sig utover hele den skandinaviske fjeldkjede (de fire forste findes bare i det 
sydlige Skandinavien), antar Tengwall at den sydlige indvandringsvei har 
været mindre væsendig. 

Man maa utvilsomt gaa ut ifi-a at der virkelig har indvandret fjeldplanter 
til det centrale Skandinavien fra syd (Alperne og Mellem-Europa) via Dan- 




Fi"-. 34. Kobresia bipartita'^ utbredelse i Skandinavien. 
Etter Tengwall (Enquist 19 18 1. c). 

mark og Sverige. Hvorledes skal man ellers forklare forekomsten av Ko- 
bresia hipartifa, Gciitiaua purpurea etc. i Norge? Mens man oprindelig an- 
saa denne indvandringsstrom for den vigtigste, slaar man nu over i den 
motsatte yderlighet. Nu skal nærsagt alle fjeldplanter enten være inter- 
glaciale eller osdige. Det ene behover ikke at utelukke det 
andet. Naar Dryas og andre fjeldplanter har kunnet vandre helt op til 



I04 KOI.K NOKDHAGF.N. M.-X. Kl. 

Laxa (iiidciilor dcii inorirncia-kkc som svarer til racn«), har de nok ogsaa 
evnet at komme sig op til Krislianiatraklin. Og i anledning av Daniels- 
SENS interessante fund a\- arktiske planterester ved Kristiansand (1908, 1910, 
19T2 1. c), liggel- det siKihlende nai' at anta en spredning over Skagerak 
tVa Jylland, f. eks. med vinlerstoiiiieni' lieno\er det tilfrosne hav (Warming 
1904 p. 13 og 18). Man antar jo mi al der i postarktisk tid har ind\'andret 
planter direkte fra ICngiand ' og Danmark til Norge IWii.le, Salmonsens 
leksikon XIII p. 503 og 1915 p. loi), og at der i nutiden paagaar en ind- 
\andring til Danmark i motsat retning (Wakmi.no 1904 p. 15-17). Saa- 
danne spredningsmuligheter har seKfolgelig a 1 1 i d ' været tilstede og ikke 
bare været begrænset til noget enkelt tidsrum. 

At Ijeldplanter fra sydost eller sydvest er vandret helt op til Kristiania, 
fremgaar av Ov):xs fund av .SVeZ/.v polaris i Åker og Asker fra Portlandia- 
nivaaets tid^. Nordmarken ved Kristiania har topper som nu gaar op til 
over 700 m. o. h., og som var ca. 500 m. dengang havet stod ved den 
marine grænse. Da isen forsvandt herfra efter Aker-stadiet og Romerik- 
stadiet (ofr. ovenfor), har utvilsomt en arktisk vegetation hat gode be- 
tingelser paa disse koUer og høidedrag. Oyens fund av Dryas octopetala 
og Betiila iiaiia fra yngre avleiringer (som dog er vanskelige at tidfæste) 
ved og i Kristiania by, viser at fjeldplantene ogsaa har holdt sig her en 
stund utover i postarktisk tid (O yen 19 i 5 1. c. p. 35). 

Vi maa anta at den skandinaviske fjeldflora er blit rekrutert ogsaa fra 
syd, og det er forelobig meget vanskelig at avgjore om enkelte arter har 
været interglaciale overvintrere, eller om de er sydlige indvandrere. V^ort 
kjendskap til den vegetations beskaftenhet som fra S3^d fulgte efter det av- 
smeltende isdække opover mot det centrale Norge, er imidlertid uhyre 
mangelfuldt. De fleste er vel nu tilbøielige til at anta at den var mere 
blandet arktisk-subarktisk end egentlig arktisk, og at vi i Norge kun har 
hat en egte „Dry as-tid" i de deler av landet som allerførst blev isfrie (det 
trondhjemske, vestlandets tjordegne med tilstotende tjeldtrakter, kystranden 
nordover), og at de lavere landsdeler i syd og ost som blev blotlagt senere, 
og som laa aapne for den sydlige indvandringsstrom, ikke har hat nogen 
veritabel Dryastid av hoiarktisk præg. Her har ogsaa store strækninger 
ligget under hav i det kritiske tidsrum. 

Imidlertid er der her en mulighet som ikke bor oversees, og som 
Oyen har fæstet opmerksomheten ved. Hvis vi nemlig i Norge har hat 
en saadan senglacial klimatoscillation som Allerodperioden i Danmark [My- 
//7//s-nivaaet) og betydelige deler av landet under denne tid var isfrie, maa 
dette ha været en gylden tidsalder for fjeld vegetationen (Oyex 1915 1. c. 
p. 296), som dog atter i betydelig grad maatte bli decimert, ialfald i visse 



• Norman (1855) er vistnok den forste som har fremsat denne mening. 

2 ØYEN 1907 B. 



ig2I.No. 9- KALKTUFSTUDIER I Gl'DBRAXDSDALEN. I05 

strok, under Por//ci'//c//VMii\aaets nye brætremrykning. Vi ser altsaa at ogsaa 
paa dette punkt er spørsmaalet om Mytilus- og Po li /au di a -niv aa.ets natur 
og specielt disse perioders klimatologiske valor av indgripende betydning. 



Det bor efter ovenstaaende utredning ikke virke særlig overraskende 
naar \i i bunden av Gudbrandsdalens kalktufter ikke finder rester av 
arktisk-alpine planter. Imidlertid har der i den forste tid efter isens ende- 
lige avsmeltning i disse strok ikke været gunstige betingelser for kalktuf- 
dannelse. Som tidligere omtalt viser baade Leine- og Gillebutuften at den 
første tid efter isens forsvinden har \-æret meget kontinental, kanske arid. 
Og den forste flora og vegetation har derfor ikke efterlatt sig noget spor 
i tuftene. Det kan for den saks skyld godt ha været tjeldplanter, det kan 
ogsaa ha været en mere blandet vegetation. \'i vet heller ikke hvor længe 
dette tidsrum, som er indhyllet i morke, har varet. 



III. Gudbrandsdalens kalktufPer og Blytts teori. 

Blytt tok kalktuffene til indtægt for sin bekjendte teori om de veks- 
lende tørre og fugtige perioder efter den sidste istid (1876), ja han be- 
tragtet dem som vigtige støttepunkter for sin teori. 

Om denne har stridens bølger altid gaat meget hoit. Man har kritisert 
Blytts torvmyr undersøkels er, og med en viss berettigelse, idet de 
ikke tilfredsstiller nutidens krav. Men saa hadde Blytt heller ikke det 
erfaringsgrundlag at bygge paa som senere forskere. — Blytts teori frem- 
stilles ogsaa undertiden i litteraturen som et slags „skema", som vel i sin 
tid hadde heuristisk betydning, men som nu forlængst har passert henover 
tidens slipesten. Imidlertid vil enhver der ser objekti\-t paa forholdene, 
maatte indrømme at dette ikke er tilfældet. Den av Blytt hævdede op- 
fatning er modificert paa mange maater; en hel del av hans meninger har 
tiden ogsaa visket ut. Men studerer man opmerksomt moderne skandinavisk 
kvartærgeologisk litteratur, skinner Blytts teori igjennem paa en række 
punkter. 

Dette gjælder saaledes læren om s t u b b e 1 a g e n e i t o r \- m y r e n e. 
Ved n^'ere svenske, danske og norske (Holmsex 19 i 9 og 1920 1. c.) under- 
søkeiser har det vist sig at ialfald det yngste av Blytts stubbelag har 
generel betydning i Skandinaviens myrer. Under den hydrofile „yngre 
Sphagnumtorv" optrær en markert „uttørringshorisont", som oftest i form 
av et stubbelag, hvis tilstedeværelse ikke kan forklares paa anden maate end 
ved den antagelse, at der dengang laget blev dannet, fremhersket et tørt 
klimat, saaledes som Blytt antok. Disse to 3'ngste lag i myrene benævnes 
nu almindelie: den s ub a 1 1 an t i s k e torv oq den subboreale uttork- 



I06 ROI.K NOKIMIAGKN. M.-N. Kl. 

n i n ,i(sh O r i SO n t '. Dctti- fænomen (fl(;n siihatlantisk-subboreale kontakt) 
er lin b(ski-(\(l Ira saa inanj^c lokaliteter at man ingen grund har til at 
l)( t\ile dets generelle karakter. (CtV. Bi.vtt.s, Sf.rna.nders, von Post's og 
deres elevers arbeider, jicssrcNS nye studier i Danmark, IIoi.msens i Norge). 
Under det subboi-eale stubbelag optrær der i myrene gjennemgaaende 
sterkt hydrolile tor\(lannelser, som av de forskere der akcepterer Bi.vtt- 
Sern ANDERS opfatning, benævnes „atlantisk torv". Og under denne kommer 
i mange tilfælder attei- et avbrud, „det boréale stubbelag", som forovrig 
ikke synes at være saa sterkt niarkert som det sul>boreale (\'o.\ Post i 918 
1. c, Jessen 1920). Disse fire hovedled i torvmyrenes opbygning (oven- 
fra nedad): 

subatlantisk torv, 

subborealt uttoi-kningslag, 

atlantisk torv, 

borealt uttorkningslag, 

gjenfinder man i alle de nyeste oversigter inden skandinavisk torvmyr- 
litteratur. Svenske forskere har ogsaa trukket frem en mængde andre 
lakta (bl. a. hydrografiske studier o\-er avlopslose sjøer i boreal og sub- 
boreal tid^) som bekræiter ovennævnte periodiske vekslinger. 

Hvad der ligger under det boréale lag, er derimot meget omdisputert. 
Blytt antok oprindelig at der her fandtes et „subarktisk torvlag" ; senere 
paastod han til og med at der under dette torvlag atter fandtes et ældre 
stubbelag (i de myrer som laa holest over havet), og han forandret da 
terminologien derhen, at torvlaget kaldtes „infraborealt" og stubbelaget 
„subarktisk" (sensu stricto). Imidlertid er det paafaldende, at ingen av de 
forskere som i nyere tid har arbeidet med torvmyrene, tydelig har gjen- 
fundet disse to horisonter. De „præboreale" lag er forovrig lite bearbeidet 
(cfr. Sernanders uttalelser 1916 p. 136 — 138); men hvis forholdene hadde 
været saa markerte som Blytt mente, vilde disse lag utvilsomt ogsaa faat 
sin rette plads i den kvartærgeologiske oversigt. De fleste palæobotanikere 
synes vistnok at helde til samme anskuelse som Sernander, nemlig at be- 
tegnelsen „subarktisk" bor anvendes om alt det som er ældre end det 
boréale lag (altsaa Blytts oprindelige betegnelse), og at dette navn kun 
er et uttryk for mere eller mindre usikre fysiografiske forhold („glidende 
utviklingsmomenter"). 



Allerede Blytt forsokte at faa istand en korrelation mellem tor\-myr- 
forskningens resultater og de marine n i vaaforandri nger i post- 
arktisk tid; dog hadde han et hoist utilstrækkelig materiale at bygge paa 
(1882 1. c. p. 8). 



' Webkrs „grænsehorisont" (1910 1. cl, som udmerker Nordtysklands mjTcr. 
2 Sernander 1910 (uttorlig litteraturfortegnelse) p. 205, 213 — 214. 



I92I.X0. g. KALKTUFSTUDIER I GUDBRA.XDSDALEN. I07 

For Norges vedkommende har senest Øyex sekt at sammenstille resul- 
tatene fra disse forskningsgrener. Og det eiendommelige er, at mens i torv- 
mvrene de „præboreale" lag er daarligst kjendt, er i grunden det motsatte 
tilfældet med de norske marine avleiringer. \'ed ældre geologers og i nyere 
tid Brøggers og Øyens vidtløftige studier er de avleiringer som ligger 
mellem den marine grænse og Tapes-ni vaaet, meget godt kjendt 
(Brøgger 1900 — 1901, Øyex (talrike avhandlinger)). 

Da bræene for sidste gang stod ved de store indsjoer i det østen- 
tleldske Norge, hadde landplaten allerede begyndt at hæve sig en smule. 
Et klimatomslag begyndte nu at gjøre sig gjældende. Den yngste Portlandia- 
fauna dor ut, og isteden faar vi Litforiiia-iiivaaets tempererte fauna'. Dette 
nivaa er meget utpræget baade i Kristiania- og Trondhjemsfeltet og ud- 
merker sig ved fremtrædende skjælbanker (littorale avsætninger), derimot 
ved svak akkumulation, hvorfor Øven tolker denne periode som relativt 
kontinental. 

Landet fortsatte imidlertid at hæve sig, og vi kommer over i en periode 
hvis avsætninger følgelig ligger lavere end de foregaaende. Øven kalder 
dette karakteristiske nivaa for Pliolas-iiivoact. Det udmerker sig ved fore- 
komsten av betydelige lerterrasser i Kristianiafeitet, altsaa en akkumula- 
tionsperiode. Det maa utvilsomt ha været en tid med rik vandforing og 
store slammængder i elvene, en relativt fugtig periode, hvis tauna var av 
sydvestlig præg og betydelig mere varmekjær end den foregaaende periodes. 
Endelig opfører Øyen Mactra-nivaaet som det tredje led i rækken; det 
udmerker sig atter ved manglende akkumulation, ved skjælbanker og ved 
en varmekrævende fauna. Efter iV/ac/rrz-tiden synes der at ha skedd en 
tvdelig sænkning av landplaten, som dog er liten baade ved Kristiania 
og Trondhjem, men som maa paralelliseres med „Littorina-sænkningen" i 
Sverige og Danmark. Denne forer os over i Tapes-iiivaaet (sensu stricto), 
den fugtige atlantiske tid, atter med akkumulation og dannelse av lerterrasser. 
At Tapes-iiivaaet og den „atlantiske periode" falder sammen, derom 
synes nu alle kvartærgeologer at være enige. Under denne periode 
skal da det atlantiske torv lag i myrene være dannet. Og 
ØVEN paralelliserer da helt naturlig 

Littorina-nivaaet med Blytts subarktiske tid (sensu stricto) 

Pholas-nivaaet — „ — infraboreale tid 

Mactra-nivaaet — „ — boréale tid. 

\'\ ser altsaa at de norske marine avleiringer stemmer med Blytts 
opfatning av de præboreale lag, og Øyen finder derfor anvendelse for 
Blytts engere opdeling av den oprindelige „subarktiske" tid i to torskjel- 
lige avsnit. 

ØYENS nivaa-inddeling for disse ældre postarktiske tidsrum er fast 
underbygget og anerkjendes nu av saagodtsom alle norske geologer (cfr. 



• Dette nivaa maa ikke forveksles med „Littorina-tiden" i Sverige og Danmark. 



I08 KOI,F NORDMAGEN. M.-N. Kl. 

UjoKiAKKi: IQ13, Bkix.ckk 1Q14). Saa iiuj^ct er ialfald sikkrrt, at hvad 
<!<■ iii.iiiiic ;i\ Iciriiit^cT aiit;aar, saa kan vi i Norge meget vel karakterisere 
avsnittct iiullciii den marine grænse og Tapes-nivaaet, saavel faunistisk 
som klimatologisk. Men da kjendskapet til torvm^'renes præboreale lag er 
mangclluldt, er det al grund til at \ ære forsigtig og ikke generalisere 
ovennævnte resultater uten \ idere. k'renitiden faar avgjore om vi ogsaa i 
præborcal tid har hat pciioder synkrone for hele Skandinavien. 

De a\Ieiiingcr som er yngre end Tapes-nivaaet, er ogsaa meget inter- 
essante baade i Kristiania- og 'IVondlijemsfeltet. i)\KS har her indført be- 
tegnelsen „Trivia-iiivaact efter et „faunistisk kompleks" [Trivia europæa, 
Lima loscoinbi og Cniiiiliis uiillrgraims], som udmerker skjælbankene i et 
nivaa lavere end Tapes-nivaaet. Av faunaens utpræget sydlige og varme- 
kjære karakter, den manglende akkumulation etc. udeder Øyex at Trivia- 
nivaact betegner den aller v a r m este og aller tørreste del av den 
postglaciale varmetid. Dette gjælder tildels ogsaa første del av det 
efterfolgende nivaa, som Oven kalder Osfræa-iiivaart, som falder i et ældre 
og et yngre avsnit. Skillet mellem disse avsnit ligger ved en havstand 
av ca. II m., og Ostræa-nivaaets yngste avdeling synes at ha været 
relativt fugtigere og noget k joligere end det ældre avsnit mellem 1 1 m. 
og 22 m. o. h.'. 

Nu skulde man tro at Oyen vilde paralellisere sit Trivia-nivaa og 
forste del av Ostræa-nivaaet med Blytts „subboreale" periode, som jo 
karakteriseres som \'arm og tor, og den anden del av Ostræa-nivaaet med 
den „subatlantiske periode". Imidlertid har Blytt i den korrelationsover- 
sigt som han utarbeidet, sat grænsen mellem disse to perioder ved en 
havstand 9,4 m. — 13 m. 

subboreal 15,7 — 13 m. o. h. 
subatlantisk 9,4 — o m. o. h. 
(Blytt 1882 1. c. p. 8, 9, 12 — 13 og 1892 p. 44 — 45). 

Og til trods for at Oyens nivaaer hvad strandlinjens beliggenhet angaar, 
ellers stemmer meget daarlig med Blytts (paa et mangelfuldt materiale 
baserte) beregninger ^, lar han her Blytts angivelser gjøre utslaget og 
paralelliserer kun Ostræa-nivaaets f r s t e del med Blytts subboreale 
og det samme nivaas senere del (under 11 m.) med Blytts subatlantiske 
periode. Paa denne maate blir Tri\'ia-nivaaet staaende igjen uten relation 
til nogen av Blytts perioder, til trods for at det er mere markert konti- 
nentalt og varmt end Ostræa-nivaaets første del. Trivia-nivaaets strandlinje 
gaar nemlig helt op til 47 m. o. h. 

Følgen av det hele blir da den, at Oven blir nodt til at indfore to 
nye betegnelser: „neoboreal" for Trivia-nivaaet og „neoatlantisk" (= sub- 



' Ostræa-nivaaets ældre avsnit skal dog; ifolge Øyen være en smule mere maritimt be- 
tonet end Trivia-nivaaet. 
2 Cfr. ØVEN 1915 1. c. p. 99. 



I92I.X0. 9- KALKTUFSTUDIER I GL'DBRAXDSDALEN. 1 09 

boreal -^ subatlantisk) for Ostræa-nivaaet for at faa orden paa tingene. Men 
disse nve betegnelser, hvorav den sidste „neoatlantisk" likefrem bryter med 
hele traditionen i Blytts nomenklatur, idet her ordet atlantisk (a: fugtig, 
mai-itiin) anvendes som fællesbetegnelse baade paa en tor (subboreal) og 
en fugtig (subatlantisk) periode, bidrar bare til at eke nomenklaturvanskelig- 
hetene. Det store sporsmaal blir da om baade Trivia-nivaaet og forste del 
av Ostræa-nivaaet skal paralelliseres med Blytts „subboreal". En kombi- 
nert studie av skjælbanker og lerlag mellem Tapes-ni\aaet og nutidsstrand- 
linjen paa den ene side, og torvmyrer med utviklet subboreal — subatlantisk 
kontakt i det samme spatium paa den anden side, kan her alene si det av- 
gjorende ord. Hvis nemlig Trivia-nivaaet, saaledes som Oyex antar, har 
været den varmeste og mest kontinentale periode efter isens avsmeltning, 
maa torvmyrene utvilsomt bære vidnesbyrd herom i form av et stubbelag. 
Dette bor være meget markert, ialfald vel saa markert som den horisont 
som Ostræa-nivaaets første avsnit har efterlatt sig („subboreal" i Oyexs 
mening). Men hvis nu Trivia-nivaaet og Ostræa-nivaaet I begge har været 
hovedsakelig av kontinental natur, burde man a priori bare vente at finde 
ett vngre stubbelag (eller uttorkningshorisontl som \idnesbyrd om begge 
disse to avsnit * — hvilket jo ogsaa er tilfældet. — Paa dette punkt er 
korrelationen mellem de marine a\-leiringer og torvmyrene temmelig usikker. 

Nedenstaaende skema viser korrelationen mellem Oyexs nivaaer og 
Blytts perioder: 

A()'/////5-ni\aaet- — arktisk 

Porf/a//dia-n[vaa.et- — subglacial 

Liffon'iia-nW3.a.et — subarktisk (sensu stricto) 

P/io/as-nWaatit — infraboreal 

Macfra-nWaaet — boreal 

7ir7/»^s-nivaaet — atlantisk 

T'nz'/a-nivaaet — 

Os/rflrrr-nivaaet I ) — subboreal 

Osfræa-nWaaet II I — subatlantisk. 

Jeg skal ikke her gaa nærmere ind paa disse detaljer. Sammenlignes 
ØYENS opfatning med de svenske torvmyrforskeres (Serxaxder, vox Post 
etc.), erder for saavidt en vigtig overensstemmelse som Oyex meget be- 
stemt henlægger det p o s t g 1 a c i a 1 c k 1 i m a t o p t i m u m til tiden 
efter Tapes-nivaaet, mens man tidligere antok at den atlantiske tid 
betegnet klimaks^. Likeledes er det meget paafaldende at mens Tapes- 
nivaaet (sensu stricto) utpræger sig som en fugtig akkumulationsperiode, 
svnes Trivia-nivaaets (osf tildels første del av Ostræa-nixaaets) avleiringer 



1 Altsaa en „neoboreal-subboreal" fælleshorisont efter Øyen5 terminologi. 

2 Denne periode er tidligere omtalt under kapitlet „Isavsmeltningen i Giidbrandsdalen". 

3 Cfr. Brøggers redegjorelse hos Serxaxder 1910 1. c. p. 226 — 227. 



KOl.F NOKOIIAGKN. M.-N. Kl. 



at \i(liic (iiii koiiüiiciu.ilc kliiiialloiliold, ^r.m alter synes at slaa om ved 
II meters strainllinjen, (l<i,i; iiten \()l(|soiiiiiie a-ndringer. 

Imidlertid liarmonerei- delte ikke \idei-c ^odt med den nyere svenske 
opl"alniiii4 (Sick.nandek, vos l'osi o. 11.), lixor dcii varme og tørre „sub- 
borealc" tid henlægges til den yngre stenaUler og bronsealderen, den fug- 
tige „suhatlantiske" til o\ergangen mellem bronsealderen og jernalderen. 
Oven niaa nemlig forlægge sit Trivia-nivaa og Ostræa-nivaa I meget længer 
bakovcr i tiden, da Trivia-nivaaets strandlinje ved Kristiania og Trondhjem 
gaar op til mellem 40 og 50 m. o. h., mens strandlinjen i bronse- 
alderen synes at ha indtal omtrent samme stand som i nu- 
tiden (Brøgger 1905 1. cl. Imidlertid kan vi f. eks. i Kristianiafjorden kon- 
statere et andet omslag ved overgangen til nutidsstrandlinjen ; en række syd- 
lige og sydvestlige former utdor, og Myii aroiaria indvandrer (Øvens 
Mya-nivaa = recent tid). Denne er ikke fundet fossil over nuværende hav- 
stand (Oven 1915 1. c. p. 381 1. Den marine faunas utvikling vidner dog 
ikke om nogen voldsomme ændringer. 

Hvorledes disse divergenser skal utjevnes, blir det den fremtidige 
forsknings sak at avgjore. De stratigrafisk-biologiske fænomener som 
ligger til grund for hele denne lære om vekslende klimatiske perioder, er 
i begge tilfælder de samme, og alle ting tyder paa at vi har hat synkrone 
perioder for hele Norden. For saavidt skulde problemet ha utsigt til at bli 
endelig løst. Men dertil kræves nye undersøkeiser og atter nye under- 
søkeiser. 



Blytts teori og Sernanders videre selvstændige utformning av samme 
har i likhet med Oyens opfatning i disse spørsmaal været gjenstand for til- 
dels voldsom kritik og imøtegaaelse. Og jeg kan saa godt forstaa at mange 
forskere finder den anførte inddeling og karakteristik av den postarktiske 
tid altfor indviklet. Det var meget lettere for videnskapen hvis alle 
disse mange perioder kunde undgaaes! 

Men undersokelsene ute i marken viser os atter og atter at forholdene 
har \'æret meget komplicerte. Og læser man med opmerksomhet de nyere 
avhandlinger f eks. av G. Andersson, som kanske mere end nogen anden 
har stillet sig i opposition til klimavekslingsteorier, vil man finde at han ogsaa 
ganske tydelig maa ta sin tilflugt til variationer, ikke bare i temperatur, 
men ogsaa i nedbør, for at kunne forklare stratigrafiske og biologiske for- 
hold. Saaledes skriver Andersson (1909 S. G. U. Arsbok p. 77): „in the 
Baltic bassin after a period that was warmer and drier than the present, 
during the latter part of the Anc}'lus age, there followed gradually under 
the Littorina-age, an equall}^ warm but much wetter period" (senere an- 
fører han en antat aarlig nedbørhøide av ca. 1000 mm.). Samme forfatters 
uttalelser p. 67 og specielt i note 2 paa samme side er ogsaa ganske paa- 



1921. No. 9. 



KALKTUFSTL'DIER I GmiiRANDSDALEX. 



faldende. Man har her tydelig 
ansatser til en klimavekslings- 
teori, selv om periodene ikke 
blir helt synkrone med andre 

forfatteres. 






STBMDCDNAS 
vrŒTATIOM 



UP 



SUB- 
ATLANTISK 



SVÂMSAND 




SUB- 
BOnCAL 



TAT Turr 



ATLANTISK 



HASTIGT BILDAD 
LOS TUFF 



Gaar vi derefter over til 
kalktufs tu diet i norden, 
saa har det ogsaa bragt en 
lang række stratigrafiske og 
plantegeografiske eiendomme- 
ligheter for dagen, som \iser 
at forholdene efter den sidste 
istid ikke har været helt enkle. 

Efter Blytts avhandling 
i 1892, som maa siges at være 
klassisk inden denne forsk- 
ningsgren, er der specielt fra 
svensk side kommet en lang 
række monografiske arbeider 
over svenske tufter (Hulth, 
KuRCK, Sernander, Halle, 
Kjellmark o. fl.). Av disse 
er kanske Skultorptuffen 
i Västergötland den som for- 
tjener störst opmerksomhet, 
da den gir det mest sammen- 
hængende og uforstyrrede 
billede av utviklingshistorien. 
Benestadtuffen i Skaane 
er ogsaa uhyre interessant 
og teoretisk betydningsfuld. 
Om J e m 1 1 a n d s tuffer maa 
man nærmest si at de strati- 
grafiske forhold der er mere 
problematiske og vanskeligere 
at tolke. 

Det av Blytt for Gudbrandsdalens tufter paaviste fænomen, at avsæt- 
ningen ikke har paagaat kontinuerlig, men intermitterende, har senere 
vist sig at ha nærsagt generel betydning. Saaledes konstaterte Hulth for 
Skultorptuften 2 markerte avbrud, karakterisért \ed optræden av humus- 
rænder i tuften eller t3'delige forxitringsfænomener. Hulth fandt 3 tyde- 
lige tutdannende perioder og 2 uttørkningshorisonter, og det laa da nær 



~ai 



SUS- 
ARKTISK 



AMBLVSTECIUM-Turr 

OCH blcke: 






ISSJOBILDNINCAR OCH MORAN 



^o OqO 



Fig. 35. Skultorptuffen. (Sernander 1916). 



ri2 ROLF NOKDHAGF.N. M.-N. Kl. 

at p;iial(lliscrc disse imd de a\ Hi.nti o^ Skknanokk antattc klimatiske 
cpokir (III I. Ill 1899 I.C.I. y\t saagodtsoiii IhIc den jjostarktiske tid var 
rcpi-æscnlcit i Skiiltoi ppmlilct (Hg. 35), frcnigik av floraens gradvise for- 
andringei-; seU- ikke de arktiske elementer manglet i bunden. Av Kl'RCKS 
(1901) og senere Sernanders studier (1916 I.e.) I'remgaar det at ogsaa 
Benestadliifteii har 2 avbrud; men |)aa grund a\- en f)mfattende eksploitering 
av denne kalkforekomsts mre del, tr de yngre lag, som indeholder det 
yngste („subboreale" ) avbrud, fjernet o\'er næsten hele omraadet og der- 
for \anskelig tilgjængelige (efV. dng profilet fra Skvattem<)llan, Sernander 
1916 1. c. p. 160). — Endelig har vi Maeles arbeide over den floristisk 
høist interessante Botarfve-tuf paa Godand (1906 1. c), som foruten en del 
smaa lokale avbrud viser én markert muldstripe gjennem hele tuflen. Da 
findestedet har ligget under hav til langt ut i den postglaciale varmetid, 
burde man her ogsaa a priori \'ente at finde bare ett distinkt avbrud, idet 
nemlig ældi^e lag svarende til o\ennævnte tuffers undre deler, ikke har 
kunnet komme til utxikling paa stedet. Ogsaa fra andre deler av Sverige 
er der beskrevet mindre kalktuffer med humusrænder. 

Av nærværende avhandling \'il det ogsaa fremgaa at Gud brands- 
dale ns tuffer, saaledes som vi nu kjender dem ved Ovexs og Holmes 
samt undertegnedes nye undersøkeiser, taler et meget bestemt sprog og 
stemmer godt overens med de resultater som svenske kalktufforskere er 
kommet til. 

ja, undertegnedes studier over Leinetuften har til og med git det for- 
bløffende resultat, at denne stemmer meget bedre med klimavekslingslæren 
end Blytt selv ante. 

Blytt var selv til en begyndelse meget i t\il om hvorledes Leine- 
tuffen burde tolkes. 1 sin forste beskrivelse (Naturen 1891) tolker han 
f ur u tuffen som synkron med det yngste torv I ag i myrene (sub- 
atlantisk). Dryastuffen blev da subboreal og birketuffen samtidig med det 
næstyngste torvlag, altsaa atlantisk: 

Furutuf — subatlantisk 

Dryastuf og kalkler — subboreal 

Birketuf — atlantisk. 

Imidlertid bragte kalktuffen ved Nedre Dal ham bort fra denne tolkning; 
han ansaa det nemlig høist usandsynlig at en birketuf a\- subalpint præg 
paa saa lavt nivaa som Nedre Dal kunde skrive sig fra en saa ung periode 
som den atlantiske, der tvertom, efter andre undersokelser at dømme, syntes 
at ha været varmere end nutiden. Blytt forkaster derfor den første 
forklaring og hævder at tufiagene maa \-ære ældre (1892). Den nye kor- 
relation blev da: 

Furutuf — atlantisk 
Dryastuf — boreal 
Birketuf — infraboreal. 



I92T. Xo. 9. KALKTL'FSTUDIER I GUDBRAXDSDALEX. IT3 

Imidlertid maa dette siges at \-ære en temmelig dristig tolkning; for 
konsekvensen av det hele blir at der i Gudbrandsdalen ikke har været 
nogen tufdannelse under den sidste av Blvtts fugtige perioder, den sub- 
atlantiske tid. Med andre ord: To r\- m y renes subatlantiske torv- 
lag og subboreale uttorkningshorisont, som jo netop har 
vist sig at være de mest generelle, skulde altsaa ikke ha 
nogen ækvivalent i Gudbrandsdalens tu i' fe r. 

At Blvtt selv har undret sig over forholdet, fremgaar tydelig av det 
eiterte arbeide. Han skriver nemlig (1. c. p. 23): „Man maa jo ogsaa let 
tænke sig at kalkgehalten i grunden kan uttømmes saa at de senere fugtige 
tider kan gaa hen uten tufdannelse. Ja da vi ved Nedre Dal saa at furu- 
tufl'en laa ved siden av birketuffen og ikke over den, er endog den 
mulighet forhaanden, at yngre tuflag kan findes baade ved 
Nedre Dal og Leine^ ihvorvel jeg anser dette for lite rimelig." 

Omtrent 30 aar efter at disse uttalelser fremkom, h'kkedes det imid- 
lertid undertegnede ved Leine virkelig at fremfinde et yngre tuflag \Alim$- 
înffcii\, som hviler diskordant paa furutuffen. Og vi kan nu med be- 
stemthet si at den subboreale uttorkningshorisont og det 
subatlantiske torxTag har sin fuldstændige stratigrafi ske 
mots varig h et inden Leinetuffen. 

Denne kalktuf tvinger os nu simpelthen ind i Blvtts tankegang: 
Alnus-tuf — subatlantisk 
Forvitring og diskordans — subboreal 
Furutuf — atlantisk 

Xerofil Dryasmatte og diskordans — boreal 
Bladtuf og mosetuf — subarktisk li videste forstand). 

Som jeg før har nævnt, bygger Alnus-tuffen paa en udmerket maate bro 
over til nutidsforholden e, hvilket ogsaa befæster ovenstaaende kor- 
relation. 

Jeg var selv oprindelig temmelig skeptisk overfor Blvtts tolkning av 
tufliene, idet jeg nemlig i likhet med mange andre fandt hele dette indvik- 
lede system av perioder en smule usandsynlig. Og jeg fandt ogsaa det 
fænomen, at de to sidste perioder, som andre steder i Skandinavien er de 
mest markerte, skulde ha gaat sporløst hen over dalens tufter, noget 
mystisk. Men naar man saa en vakker dag finder netop det som Blvtt 
savnet, men hvis tilstedeværelse han i grunden har forutsagt, er dette altfor 
paafaldende til at bero paa nogen tilfældighet. 

Naar man videre ser hen til at nøiagtig de samme perioder og av- 
brud synes at komme igjen i Gillebu — Ting\-oldprofilene (Øven 1920I 
tavleiringene fra den sidste periode dog i en noget modificert, men ut- 
præget form I, maa man uvilkaarlig studse. I likhet med Skultorp og Bene- 



' Uthævet her. 

Vid.-Selsk. Skr. L M.-N. Kl. 1921. Xo. 9. 



M4 



KOI.K NORDHAOKN. 



M.-X. Kl. 



.sl:i<IUiH( 11 \ isci' h.'iadc avsætnin^cnc \■^:(\ I.cirif o^ Ciillchii I ing\olri 3 t^'dc- 
lig'c opsxiiliniiini;«!- ;i\- de \iiksoiiiiiic kilder og 2 avhrud i avsætningen. 
Overenssteinnielscn med l<ii\niyrlV)r.sUningen.s resultater og de marine av- 
Iciringcrs vidncshx rrl cv likclcdes paafaldende. 

All i all iiiaa man ogsaa indromme, at klimavekslingsteorien paa en 
iiUimgcn iiiaalc t'orniaar al lose alle disse stratigrafiske problemer. \'i har 
\\)\- liden ingen anden arbeidshypotese som er bedre'. Og akcepterer 
man ikke denne forklaring, blir Gudbrandsdalens kalktuffer en eneste 
lang række av s po rsm aalstegn, en ind\iklet serie av merkelige og 
uforklarlige, men lo\'mæssige vekslinger. 

jeg stiller mig skeptisk overfor mange av de konsek\enser som Blytt 
i sin lid trak av sin teori. Saaledes tror jeg at hans floristiske grupper 
tildels er gro\e skemaliseringer, og al \oi- flora ogsaa gjemmer problemer 
som Bi.YTT ikke var opmerksom paa. I Ians plantevandringslære indeholder 
sikkert ogsaa store overdrivelser og mange ensidige synsmaater. 

Men luad der for undertegnede i denne forbindelse staar som hoved- 
saken, er selve k 1 i ni a \- eksl i ngslæren som kvartærgeologisk arbeids- 
hypotese. Til trods for de mange divergerende opfatninger som her ogsaa 
gjor sig gjældende, maa man dog indromme at denne fremdeles er et 
aktuelt centrum — eller stormcentrum - i nordisk kvartærgeologi^. 



Korrelation. 



BlytTo periodei' 


I.,inr 


(iillebu- ringvold 


Nedre Dal 


Recent tid 


I""or\itring, s\ak 
mosetufdannelse 


Forvitring 


Forvitring 


Subatlantisk 


Alnus-tuf 


Erosion med 
avsætning av 
grus og tufstyk- 
ker paa sekun- 
dært leiested 


Utglidning? 



' Cfr. Sandegrens oversigt over svensk kvartærgeologisk forskning i sidste hefte av 
Geol. Foren. Förh. Stockholm rgai, hvor den Blvtt-Sernanderske teori og dens 
store betydning beljs'ses paa en udmerket maate. 

2 Henrik Printz har i sit netop publicerte store arbeide over „Th3 vegetation of the, 
Siberian — Mongolian Frontiers" (I. c.) git en række interessante oplysninger om klima- 
tiske forandringer i de angjældende trakter av Asien i nutiden (forskvvninger i steppens 
utbredelse paa skogens bekostning som en folge av klimaets utvikling i tor retning). 
Printz synes ogsaa at ville soke forklaringen herpaa i Blvttsk retning, og hans arbeide 
turde faa betydning for diskussionen om den „subboreale" periodes klima og plantevekst. 



1 92 1. No. 9. 



KALKTUKSTLDIER 1 GL"DBRANDSDALEX. 



Hlytts perioder 


l.nm- 


Gillebu -Tingvold 


Xcdre Dal 


Subboreal 


Forvitrings- 
horisont 


Forvitring 


-i 


Atlantisk 


FurutLit" 


Furutuf 


Furutuf 


Boreal 


Drvastut' og 
diskordans 


Jordstripe 


Diskordans 


Subarktisk 


Bladtut" og 
mosetuf 


Bladtuf 


Bladtuf 


(I Gudbrandsdalen 
2 forskjellige av- 
snit, det ældste 
meget kontinentalt) 


Rod 1ère 


For\-itret 
moræne, anrik- 
ning av Jern- 
forbindelser 





Sidste isbræ .... 


Morænelere 
med blokker 


Morænegrus 


Moræne 



IV. Bemerkninger om Gudbrandsdalens plantevekst 
i postarktisk tid. 

I det følgende skal jeg behandle de enkelte avsnit i kalktuflfenes ut- 
viklingshistorie mere indgaaende og omtale de konsekvenser som oven- 
staaende korrelation medforer. 



A. Den subarktiske tid. 

Jeg anvender her Blvtts oprindelige og videste betegnelse, idet jeg 
ikke føler mig helt overbevist om at Blytts „infraboreale" og „subark- 
tiske" (sensu stricto) perioder er av generel natur. Paa dette punkt maa 
man av\ente torvmyrforskningens resultater med hensyn til de „præ- 
boreale" lag. 

Jeg har allerede under gjennemgaaelsen av tufifene omtalt den konti- 
nentale periode umiddelbart efter isens endelige avsmeltning i dalen, som 
udmerket sig \-ed forvitrings- og oksydationsfænomener i overflatelagene. 
\'i har her et forhold som er av stor betydning for diskussionen om 
tuftenes alder og opstaaen. 

Andersson og Birger, som omtaler baade Jemtlands og Gudbrands- 
dalens tuffer i sit store arbeide om Norrlands flora (191 2), uttaler føl- 
gende herom: 



Il6 KOI.I- .NOKDIIAf.KN. M.-X. Kl. 

„Del tonic kiinna starkl itVagasättas om ej kaiktufïenia.s i tiden uppen- 
hai'ligen skarpt lokalisci adc frirckonist, kan ställa.s i samband med den stora- 
mängd gi"iinfl\att< n som tillfördes dessa kalktrakter just i samband med isens 
aismaltning. TiilTbildningen skulle, om ofvan framkastade möj- 

lighet visar sig riktig, ickc statt i samband med nagon „vat" period, som 
fiera forskare fcH-modat, utaii med afrinningen af den sista delen af den i 
inlandsisen under artusenden magasinerade nederbcirden" (I.e. p. 144 — 1451. 
Om jemtlands tuffer uttaler forfatterne senere: „Enligt här häfdade upp- 
fatning skulle de egentliga kalktufferna till alldeles öfvervägande del bildats 
under nàgra arhnndradcn alldeles \id tiden for de sista isresternas defini- 
tiva afsmaltning." Leine og Nedre Dal „ligga i ungefar samma läge till 
dessa trakters isdelare, som de jämtländska tufferna (fig. 20) och ha en i 
alla detal jei- fullstäiidigt mots\arande flora" (1. c. [). 145 1, og forfatterne synes 
at ville utstrække den samme forklaring til ogsaa at gjælde disse norske 
tuiforekomster. 

Av redegiorelsen i nærværende arbeide vil det imidlertid fremgaa a t 
t u f d a n n e 1 s e n i G u d b r a n d s d a 1 e n slet ikke h a i" b e g \' n d t i til- 
slutning til i s a v s m e 1 1 n i n g e n , men forst senere. Der er et 
markert og eiendommelig interval imellem, som antageligvis ikke har været 
helt kortvarig. Den av Andersson og Birger hævdede opfatning maa alt- 
saa forkastes. Desuten er baade Leine- og Gillebu — Tingvold-avleiringene 
saa avvekslende og indviklede, at det bare av den grund er en umulighet 
at anta at altsammen er dannet i løpet av nogen aarhundreder, ved den 
tid da de sidste isrester avsmeltet. Paa forfatternes kartskisse (fig. 20 1. c. 
p. 138), hvor isskillet og kalktuflokalitetene er indlagt, er isskillet over det 
østenfjeldske Norge ogsaa indtegnet feilagtig; det er placert like nord for 
Nedre Dal, passerer altsaa ret over Gillebu i Oier herred. Og Leine er 
anbragt et godt stykke nordvest for isskillet, hvilket altsammen ikke 
stemmer med de faktiske forhold, saaledes som disse for længe siden er 
klarlagt av Rekstad og Oye\. Isskillet laa nemlig flere mil hoiere oppe i 
dalen end Leine, saaledes som jeg for har omtalt. 

Da Gudbrandsdalen i nudden har et utpræget kontinentalt klimat, sam- 
menlignet med andre deler av \ort land, er det forbundet med adskillig 
risiko at ville generalisere den kontinentale periode h\orom tuftenes under- 
lag bærer vidnesbyrd. Fordi om forholdene \ar ekstreme oppe i dalen, 
behøver dette ikke at ha været tilfældet andre steder i landet. 

Da Oyens Littorijia-iiivaa repræsenterer den første tid efter isens 
endelige avsmeltning, og udmerker sig ved svak akkumulation og typisk 
skjælbankedannelse baade i Kristiania- og Trondhjemsfeltet, betragter Oven 
forvitringshorisonten i bunden a\- kalktuftene som et hermed o\-erensstem- 
mende og synkront fænomen, med andre ord: perioden skal ifølge Oven 
allikevel ha været av generel natur (1920 1. c. p. 335). 

Hvorledes Gudbrandsdalens plantexekst var beskaften i den allerførste 
tid efter isens forsvinden, er uvisst. Da kildene begyndte at springe frem 



I92I.X0. 9- KALKTUFSTUDIER I GLDBRANDSDALEX. II7 

som en folge av oket nedbor, fandtes der en subarktisk lovtrævegetation 
baade ved Leine, Gillebu og Nedre Dal. Ved Leine har \istnok enkelte 
f;eldplanter været tilstede. Man faar et le\ende indtryk av at Gudbrands- 
dalen maa ha frembudt et billede som et av vore subalpine dalforer med 
frodige bjerkelier tilblandet asp, heg og Salices (baade gronne og 
graalodne). 

Forekomsten av fur upollen kan bero paa en sporadisk optræden 
av furu paa specielt gunstige lokaliteter; men som for næ\'nt er langflugt 
ikke utelukket ^ Flere ting vidner ialfald om at furuen maa ha været lø\'- 
trærne helt underlegen. 

\"ed Gillebu dækkes tuffen og bækkedalens omgivelser i nutiden av 
en blandet furu- og granskog, som er meget vakker. Terrænget passer 
ogsaa udmerket for furu (utvasket morænegrusl; at granen her som alle 
andre steder har presset sig frem paa furuens bekostning, er en sak tor 
sig. \'i har ogsaa set at under furutuftens tid har findestedet hat typisk 
furuskog. Hvis nu furuen under subarktisk tid hadde været almindelig i 
Oier rundt omkring findestedet, hvorfor skulde den da „tilfældigvis" 
mangle netop paa denne lokalitet, h\or den i den efterfølgende periode (furu- 
tuftens tid) s3-nes at trives fortrinlig, til trods for at jordbundsfor- 
holdene hele tiden er de samme, og likeledes i nutiden? 

Det samme gjælder Leine. Under furutuftens tid har lerbakkene og 
tuf kildens nærmeste omgivelser været rigtig et ønskested for turuen, at 
dømme efter den kolossale masse med avfald som tuft'en indeholder. Skal 
ogsaa her den typiske bladtufbænk, uten makroskopiske fururester, skrives 
paa tilfældighetenes regning? Mangler furuen her bare fordi der „tiliæl- 
digvis" stod en hel del lovtrær „iveien" netop paa tuffindestedet? 

Hvis fænomenet kun hadde været karakteristisk for én av lokalitetene, 
kunde et saadant standpunkt forsvares. Men ogsaa tuffen ved Nedre Dal, 
som ligger paa dalens sydside like i nordvest for Lillehammer, med østlig 
eksposition, viser den samme mægtige bladtuf. Alt dette tyder med be- 
stemthet paa at \-i her i det centrale Norge i subarktisk tid har hat en 
dominerende subalpin løvtrævegetation, en klimatisk betinget bjerk — asp- 
periode, selv om furuen sandsynligvis allerede paa dette tidspunkt var 
indkommet til dalen. 

Spørsmaalet om pollenets langflugt er endda ikke tilfredsstillende 
utredet. At man ikke bør negligere denne feilkilde, fremgaar kanske bedst 
av Hesselmans forsøk paa fyrskibene utenfor Sveriges kyst, hvor pollen- 
regnet paa havet var ganske paafaldende (1919 I.e.). Det omdisputerte 
furupollen paa Novaja-Zemlja bør ogsaa mane til forsigtighet. Jeg har 
selv under en geologisk ekskursion til Finse (september 19 14) indsamlet 
løvblad er av asp, bjerk og or (eller hassel) paa Hardangerjøkelen flere 



' HoLMSKXS opsats om pollen i kalktuf har ikke været mig tilgjængelig under manu- 
skriptets utarbeidelse (trykt i sidste hefte av Norsk Geol. Tidsskrift 192 1). 



I r8 KOLK NORDHAOr.N. M.-.\. Kl. 

mil fra de naiiix sic loickoinstcr l)aa(lc i ost o^ vest', l-'ra Alijcrne er 
lignende luiid (liokeblader) Ira en ra-l<l<e bræer omtalt i Schkötkus „Das 
Pflanzenleben der Alpen" (p. 738). Fx-kjendt er ogsaa „salthaglen" paa 
Gotthardt i 1870; de saltkrystaller som lier taldt til jorden (op til 0,76 gr.l, 
maa mindst ha j^assert 250 à 300 km. s xeilaii^^d»- ItVa (icnna eller Venedig), 
kanske betydelig længer. Ytterst interessant ( r likeledes Kf:r.\krs oplys- 
ning (Pflanzenleben I, 1888 p. 36) om, at han i alle de prøver av „rød 
sne" som han har ii n d c is o k t fra A I p e i" n e s gletschere, har 
fundet pollen a \- fo r s k J c 1 1 i ge naaletræM". -- Fordi om furupolien 
er llindet i bladtiiffen ved Leine, er det forhastet at trække den slutning, 
at træet har \okset i nærheten. Blomsterstøvet kan med opstigende luft- 
strømmer være kommet længer nedenh-a dalbunden, kanske langveisfra^. 

Spørsmaalet om der i Skandinavien har været en almindelig „bjerk — 
asp-periode", saalcdes som Steenstrup oprindelig antok, er i den senere 
tid h\'p|:)ig diskutert. 

Stkenstrups opfatning (1842 1. c.) synes nu mere og mere at maatte op- 
gives. Allerede i 1894 undlot Sernander i sine „Studier öfver den got- 
ländska vegetationens utvecklingshistoria" at medta nogen bjerk -asp-tid 
for denne o, da han ikke fandt beviser herfor i Gotlands myrer. Hoops 
fremhæver i 1905 med bestemthet at hverken i Nord- eller Mellem-Europa 
hai- bjerk og asp alene været dominerende efter isens avsmeltning, men 
furuen har fulgt like efter de nævnte træsorter, ialfald ubetvdelig senere. 
L. VON Post mener paa grundlag av pollenanalyser helt at maatte stryke 
bjerk — asp-perioden i det sydlige Sverige (191 8 1. c.l. Og helt nylig har 
K. Jessen for Sjællands vedkommende hævdet at „Rene Birke — Bævreasp- 
skove i den Forstand, at Skovfyr manglede i dem, synes ikke at have 
været eneherskende i Nordostsjælland. Baade makroskopiske Rester og 
anseelige Mængder af Pollen af Fyr er fundne saa d3'bt i den alluviale 
Gyttje som til nogle faa cm. fra Dryaslerets Overkant" (I. c. p. 221 1. Holm- 
SENS nye undersøkeiser i Norge synes ogsaa, saavidt jeg har kunnet erfare 
av de foredrag han har holdt, tildels at gaa i samme retning. Baade Holm- 
boe (1903) og Wille (1915) opforer imidlertid en „Bjerkeperiode" for 
Norge. 

Da det Steenstruplanske skema ogsaa i andre henseender ikke 
synes at svare til de faktiske forhold (von Post i 9 18 1. c), er der ingen 
grund til at holde paa det. Men jeg tror dog, at Gudbrandsdalens 
tuffer paa dette punkt taler et sprog som ikke helt kan negligeres. Der 
synes i det centrale Norge virkelig at ha været en bjerk — asp-periode. 



' Konservator J. Lid, Kristiania, har meddelt mig, at han flere ganger har konstatert at 
lovblader blaaser over fra Hardanger til Voss. 

2 En pollenundersokelse av den slaggagtige mosetuf (allerunderste horisont ved Leine), 
bragte uhyre sparsomt bjerkepoUen for dagen, og i de 10 præparater (dækglas 18 X 
t8 mm.) som blev gjennemset, fandtes kun et eneste pollenkorn av furu. Dette kan 
skyldes forurensning og kan ialfald ikke tillægges nogen avgjorende betydning. 



ÏÇST. Xo. 9- KALKTLTSTUDIER I GUDBRAXDSDALEN. II9 

om end ikke saa decidert som palæobotanikerne i sin tid antok ^ Det er 
ingen grund til at anta at forholdene i Gudbrandsdalen har stemt fuld- 
stændig overens med forholdene paa Sjælland eller i Mellem-Sverige; i 
nutiden er jo forskjellen betydelig mellem disse omraader. 

Blvtt har i sin avhandling trukket en hel del konsekvenser av den 
subarktiske bladtufs forekomst i bunden av tuffene. Han har utvilsomt ret 
i at klimatet dengang \ar koldere end nu (1892 p. 27I; ti alle de iagttatte 
planter og snegler er subalpine, tildels alpine; ingen sydlig art er fundet. 
Dette er saa meget mere paafaldende som baade Leinebakkene, Gillebuskraa- 
ningen og lien ved Nedre Dal i nutiden opviser en mængde sydlige, varme- 
kjære arter. Men naar Blytt desuten paa grundlag av de manglende makro- 
skopiske fururester hævder at hele Gudbrandsdalen fra Mjøsen av har ligget 
over furuens grænse i denne forste tid, kan neppe dette ræsonnement be- 
tragtes som overbe\'isende, selv om det er ilildt logisk konsekvent. Furu- 
pollenets forekomst staar her iveien og maner til forsigtighet. Jeg skulde anta 
at Leine har ligget over furugrænsen, og at denne har ligget betydelig lavere 
end i nutiden; men opfatningen paa dette punkt maa nodxendigvis bli 
noget subjektiv. 

Den subarktiske bladtufbænk er meget vigtig, idet den viser os at 
allerede i denne periode hadde bræene i det centrale Norge 
trukket sig til bake til hoifjeldet. En vedvarende ispolse over 
midtre del av Gudbrandsdalen helt til boreal tid er saaledes en levende 
umulighet; vi har her frodig lovskog hele veien fra Kvam til Faaberg. 
Under omtalen a\' Jemtlands tuffer kommer jeg tilbake til dette vig- 
tige punkt. 

En speciel interesse knytter sig til forekomsten av Hippopliars rltaiii- 
Jioidcs ved Gillebu og den undre Z)rv(7s-horisont ved Leine. Hippophacs 
og dens indvandring \il bli nærmere behandlet i det specielle avsnit 
„Hippophaës-problemet". Hvad vi her særlig skal merke os, er at denne 
busk er en av de planter som har lettest for at tape i konkurrencen med 
andre trær og busker. Av hele dens forekomstmaate i nutiden har en 
række botanikere trukket den bestemte slutning, at Hippophacs i hoi grad 
er 1 y s e 1 s k e n d e, og at dens værste fiende er tætte bestander av skog- 
dannende trær. Derfor formaar den i nutiden kun at hævde sig dels paa 
strandkanter langs Nordeuropas kyster, hvor andre planter ikke kan 
gjore den rangen stridig (flyvesand, storstenet strand, overalt lokaliteter 
hvor konkurrenceforholdene er lette), desuten langs flodbredder og 
gruset-stenete bækkedaler i Mellemeuropa og Centralasien, h\or likeledes 
konkurrenceforholdene er lette paa grund av den urolige jordbund. I kli- 
matologisk henseende er den indifferent i hoi grad, idet den 
kan trives saavel langs Norges og Englands nedborrike kyst som i Asiens 



' Allerode ovenfor (Generel del, avsnit II) har jeg berørt dette problem under omtalen 
av plantevandringer fra Norge og østover ind i det centrale Sverige. 



uoi.i- N()i<i)ii.\(,K.\. M.-N. Kl. 



orkcntraklcr (Kim-pkn 1. c. p. 644I. I*<ii gaar i Himalaya 5000 in. o. h., i 
Alperne 1900 m. <>. h. (ifolj^c nicfklcleise tV;! I )r. II. G.wis, München). 

F'orekonisUn av I lippoplinr^ i det centrale \orge i subarktisk tid vificr 
at forholdene her hai- været ganske eiendommelige, og taler til gunst for 
den opfatning, at lysaapne bjerkeskoger har været fremherskende. 

Det er likeledes hoist interessant at 1 lippopluu's blir mere sporadisk 
opo\er i bladtulTen, og at den forsvinder med denne, altsaa for 
lui-utiiffens dannelse begynder. Da jordstripen \ed Gillebu maa tolke.s .som 
boreal, kan man med god grund paastaa, at llippophaës i Gud brands- 
dal en utdor under boreal tid eftersom fu ruskogen tykner 
til, og at \'\ netop i fur uens sukcessive dominans har aar- 
s a k e n til artens u t d o e n ( beskygning) ' . I det hele tat kaster Hippo- 
pliaës et eiendommelig lys over denne subarktiske periode. Livskaarene 
maa for plantene tildels ha været ganske ekstraordinære, og dette tidsrum 
gjemmer utvilsomt paa mange botaniske henuneligheter som nok kunde 
gi os noklen til forstaaelsen av visse arters eiendommelige utbredelse i 
nutiden. 

Den undre Dryashorisont ved Leine supplerer Plippnphacs paa en ud- 
merket maate. Rigtignok er Dryas octopetala økologisk set en noget anden 
t3^pe end Hippopliai's, men begge er lyselskende, og det er ganske 
paafaldende, at det \'ed Leine er furuen som i slutten av 
boreal (senboreal) tid gj o r ende paa fjeldplantene i likhet 
med Hippophaës ved Gillebu (ofr. den lokale „klidagtige" förna-tuf ved 
Leine med de sidste Z)r>'as-blader). I furutuffen finder man hverken Dryas, 
Salix reticulata eller Hippophaës. Der er her en loxmæssig paralellitet til- 
stede mellem furuens tiltagende kvantitet (fra subarktisk til senboreal tid^ 
og ovennævnte fotofile planters avtagende frekvens og utdøen, som ikke 
beror paa tilfældigheter. Og disse rent biologiske forhold bekræfter 
efter min mening den opfatning som jeg ovenfor har gjort gjældende med 
hensyn til Gudbrandsdalens plantevekst i subarktisk tid. Disse fænomener 
er ogsaa av den storste betydning naar vi senere skal prove at sammenbinde 
Jemtlands og Gudbrandsdalens kalktuffer. 

B. Den boréale tid. 

Under boreal tid har tufavsætningen stadig avtat og snart ophørt saa- 
godtsom fuldstændig, og isteden indtrær forvitring baade ved Leine og 
og Gillebu. Samtidig med denne avtagen i fugtighetsforholdene ændres 
vegetationens karakter. En xerofil Dryasmatte utbrer sig ved Leine paa 
den fugtige bjerkeskogs bekostning eftersom lerbakkene tørres mer og mer 
ut, og til slut finder vi her en meget aapen og spredt bjerk-furuskog, som 
hai tillatt en bundvegetation av fotofile arter. Furuen gior nu sin indtræ- 



' Cfr. SiEGRiST, R. : Die Auenwälder der Aare 1. c. Her findes ypperlige eksempler paa 
hvorledes furuen langs llodbredder i Schweiz dræper Hippo/iat^s-hest&ndene lp. 156— 157I. 



I92I. Xo. 9- KALKTUFSTL'DIER I GUDBRAXDSDALEX. 121 

delse paa findestedet, utvilsomt som en følge a\- gunstigere klimatiske for- 
hold (høiere sommertemperaturi, og fordi den subalpine løvskogs magt nu 
er brutt. Ved Leine er vi videre i den heldige situation, at vi kan kon- 
statere hvorledes furuen utvider sig jevnt og sikkert og til slut bringer 
Dryas og de andre t^eldplanter til undergang paa grund av den tiltagende 
besk3-gning li senboreal tid). 

Det samme har utvilsomt været tilfældet ved Gillebu, hvor Hippoplincs 
møter samme skjæbne. Imidlertid fortæller den jordstripe som her ækviva- 
lerer Dryastuften, intet om periodens plantevekst. 

Blvtt betragtet Dryasmatten ved Leine som et lokalfænomen, beroende 
paa lokalitetens store høide over havet (1. c. p. 1 1 ). Og mine undersøkeiser 
har vist at vi ikke noget andet sted i Gudbrandsdalen har antydning til 
nogen Dryastuf Derimot finder vi stratigrafisk og genetisk likeværdige 
horisonter i form av jordstriper eller diskordanser. Dr3-astuffen ved Leine 
er imidlertid betydningsfuld i flere henseender. Jeg har tidligere nævnt 
at lerbakkene i boreal tid maa ha frembudt et enestaaende syn. Specielt 
er denne masseoptræden av Drvas interessant, naar man tar i betragtning 
at planten i nutiden hverken findes i lerbakkene eller paa skiferklippene 
høiere oppe, eller i skraaningen under Tunsbergfjeldet. 

Paa den anden side er det utænkelig at Drvas, dengang den vokste 
paa tuffindestedet, ikke ogsaa fandtes paa skiferklippene. Disse h a r i 
boreal tid sikkert været dækket av en Dr y as matte. Følgende 
plantearter, som allesammen vokser paa klippene i nutiden (cfr. fortegnelsen 
i den specielle del p. 1 1 1, indgaar i Dryadeta octopetalae som jeg har studert 
baade ved Finse og paa Svlene: 

A)it('iiiiari(i alp'nia Gcntiaiia tcuclla 

Astragalus alpiiiiis Jinicns trifidiis 

Caj'i'x capillaris Poa alpina 

Carcx sparsißora Poa cæsia 

Ccrastiiiiii alpiiiiim Pariiassia palustris 

Cetraria nivalis Polygonum viviparuni 

Draba hirta Potent ilia verna 

Draha iiwana Sclaginella spinulosa 

Gentiana nivalis J^'ronica saxatilis 

Flere av disse er til og med typiske indikatorer paa denslags los, smuldrende 
skiferbund som Dryas netop ynder. Antageligvis er flere av disse arter at 
betragte som association s -relikter fra den tid da Dryadeta i boreal 
tid dækket skraaningene. I motsætning til Dryas selv har de klart sig 
under varmetiden med dens hævning og ekspansion av furuskogen i disse 
trakter. Dog er det mulig, at enkelte arters forekomst i dette nivaa ogsaa 
skyldes senere nedvandring, i tiden efter det klimatomslag som gjorde ende 
paa varmetiden og rykket skøgen nedover igjen. En ting er ialfald sikker: 
Drvas selv dode her ut under furuskosrens forste tid og har endda ikke 



ROLF NORUHAfiFCN. M.-N. Kl, 



(oiiii.iadd ;it spi'c si^ jiaany til disse skiferbakker eller længer ned- 
over lien. 

Naar siliialioncn tor Leines \edkr)inniende faktisk var den, at Dryas, 
Salix rcliciildhi etc i hnreal tid liadde en meget større utbredelse end 1 
nutiden, iiiaa man lia li>\ til at generalisere dette fænomen ogsaa for andre 
hoiereliggende strækninger i dalen. Den boréale tid synes at ha 
været gunstigere for f J el d pi an ten e end nutiden; antagelig- 
\is 11 a 1- regio alpina dengang været større end den nu- 
værende. y\lting tyder paa at isen under denne tid var avsmeltet 
omtrent til sin nu\-ærende lokale utbredelse i norsk hoifjeld, og fjeldfloraen 
synes i denne tid for skoggrænsen begyndte at bevæge sig opad, at ha 
hat glimrende betingelser. Dryastuffen \ed Leine kan ikke tolkes paa nogen 
anden maate. At furuen her kommer til sammen med Dryas, viser at vi 
her saa at si befinder os i den undre grænse for fjeldplantenes domæne. 
Ned til Ciillebu og Nedre Dal har de ikke evnet at komme. 

Spørsmaalet om „regio subalpinas" og „regio alpinas" tidligere ut- 
s trækning under de postarktiske perioder sammenlignet med nutidsforhol- 
dene, er blit mere og mere aktuelt i de senere aar. Da furuen som bekjendt 
under varmetiden (specielt under optimum) gik 150 — 300 m. hoiere tilfjelds 
end nu, antok man oprindelig at ogsaa bj er kens ovre grænser var for- 
skjovet opad i tilsvarende grad, til trods for at de palæontologiske fakta 
ingenlunde kunde siges at bevise at saa hadde været tilfældet. Fund av 
fossile bjerkerester ovenfor grænsen for fossil furu er nemlig meget sjeldne 
(Birger 1908, Smith 191 i. Fries 1913, Smith 1920), og hele spørsmaalet er 
saagodtsom ikke utredet (cfr. Tengwall 1920 1. c. p. 285). Desuten hviler den 
tankegang som ligger til grund for denne analogi-anskuelse, paa et postulat, 
nemlig at bjerkens og furuens avhængighet av sommertemperaturen ikke 
er væsensforskjellig, kun kvantitati\-t forskjellig (furuen synes at kræve 
betydelig hoiere sommertemperatur end bjerken). 

Imidlertid er svenske botanikere ved sine nyere undersøkeiser kommet 
til det resultat, at skoggrænseproblemet er vanskeligere at løse end man 
oprindelig tænkte sig (Fries 1913, 1918, Tengwall 1920, Smith 1920 1. c.I. 
Det ser nemlig ut som om ikke alene sommertemperaturen for bjerkens 
vedkommende bør være over et visst minimum; bjerken synes desuten at 
kræve en Vegetationsperiode av en viss længde, som ikke maa 
avknappes (o: bjerken kræver for at kunne leve en minimumstid i hvilken 
den kan utfore alle sine normale livsfunktioner, hvilket selvfølgelig forut- 
sætter at ogsaa temperaturen den hele tid befinder sig over et visst lav- 
maal). Man har nu \idere tænkt sig, at kanske varmetidens klima (ialfald 
under optimum) \ar av den natur at vistnok den gjennemsnitlige sommer- 
temperatur har været hoiere end nu, men selve Vegetationsperioden (tiden 
mellem lovspræt og lovfald) behover ikke at ha faat en tilsvarende for- 
længelse, med andre ord: furugrænsen kan ha bevæget sig opad 150 — 
300 m. hoiere end i nutiden (hvilket er konstatert), men vi har ingen ret 



I92I. No. 9- KALKTLFSrrniER I GL'DBRAXDSDALEX. I23 

til at slutte det samme om bjerken. Bjerkegrænsen kan tor den saks skyld 
gjerne ha ligget relativt stille eller kun undergaat svingninger i mindre 
skala. Ja, man maa regne med den mulighet, at i postglacial tid har kanske 
furiien, altsaa et naaletræ, dannet baade skoggrænsen og trægrænsen i de 
skandinaviske tjeldtrakter i likhet med forholdene i Alperne, Sibirien o. s. v. 
i nutiden. Det nuværende bjerkebelte eller „regio subalpina" skulde da i 
overensstemmelse hermed \-ære et forholdsvis recent fænomen, opstaat efter 
varmetidens slut (furuen har rykket ned, mens bjerken er blit staaende, 
eller har ialfald ikke rykket saa langt ned som den forste; de har opført 
sig forskjellig og ua\hængig av hinanden). 

Dette ræsonnement, som i første række skyldes Dr. Th. C. E. Fries' 
epokegjørende studier i Torne lappmark, er ialfald teoretisk set meget 
vigtig og skjærper problemstillingen i hoi grad, likesom det angir retnings- 
linjene for den fremtidige forskning. 

Selv om Gudbrandsdalens tuffer, paa grund av furuens noget tvilsomme 
stilling i den første tid, ikke taler et helt avgjorende sprog paa dette punkt, 
synes det mig dog txingende nødvendig at anta at vi i subarktisk (\istnok 
ogsaa i boreal) tid, altsaa for den egentlige varmetid, har hat en subalpin 
bjerkeskogsregion i det centrale Norge likesom i nutiden, fremkaldt av et 
subalpint klima, men med andre grænser end i nutiden. 

Om regio alp in a, hvis utstrækning hænger paa det noieste sammen 
med regio subalpinas, uttaler Serxander som sin mening (Herjedalen 1910 
1. c. p. 208) at „regio alpina icke en gang omedelbart efter isens afsmältande 
kunnat vara mera utstrcäckt an i nutiden, utom tvärtom mycket snart maste 
genom skøgens uppryckning ha reducerats i sådan skala at troligen mànga 
af de forst inkomna Ijällväxterna utdött". Dryastufifen ved Leine og det 
ræsonnement som Jeg har fremfort i anledning av de merkelige skifer- 
klippers vegetation, fører tiog til det resultat, at fjeldplantene i det mindste 
i begyndelsen av boreal tid har hat en større utbredelse end i nutiden. 
Jemtlands kalktufter synes at vise det samme. Ialfald har y\ her et inter- 
essant fænomen som vi bør stoppe op for og ikke negligere. Vi bor være 
taknemmelige for ethvert lite indblik vi faar i disse gamle perioders plante- 
vek?t og klimatiske forhold. For endda er de i stor utstrækning indhyllet 
i mørke. 

For det sydlige Skandinavien antar man med rette at den boréale tid 
indwarsler de varmekjære lovtrærs æra i norden og i det hele tat ind- 
vandringen av sydlige og sydostlige typer, som saa i løpet av warmetiden 
spredte sig vidt utox^er landet baade nordo\er og vesto\'er (Blvtt, Haxsex, 
Serxaxder 1. c.). 

El/ sa!ll/^lCl//ig/n'//^^ nicllnn Giidbraiidsdalciis og JciiitlaiKh kalktiiff'cr. 

En korrelation mellem Gudbrandsdalens og Jemtlands tuffer er tildels 
vanskelig at gjennemføre. Ovex behandler dette problem i sit sidste arbeide, 
og angriper her baade Sernanders og Halles tolkning av de jemtlandske 



124 KOLK NOUIJIIAGKN. M.-N. Kl. 

rorckomstcr. Men nogen liltVedsslillende forklaring (aar man clog ikke; 
i grunden l.ii- ()^'i:.\ sporsniaalel staa liildstændig aapr-nt. 

Imidlertid aapner ll.\ii.i;s interessante under.sokel.ser nye muligheter 
l(ir en k()ii-elati(in mellem disse iK^rske og svenske tuffer. Hallk har 
trukket frem en række nye stratigrafiske og biologiske momenter (19J5 
I. c. I, som er :\v den allerstørste værdi, og som i al korthet kan resumeres 
i folgende punkter: 

1. I hunden a\' jenitlands tuffer er dei' i de ældste tuffer antydning til 
en smal undre sone uten fururester; ialfald er furuen her sparsommere 
end ellers. 

2. Drvas octopetala og Hippophäcs optrær i denne undre sone i stor mængde 
og karakteriserer den. De fors\inder opad eftersom furuen ©ker, 

3. I loiere oppe i lagrækken dominerer furuen. Granen mangler. 

I'Virskjellen mellem disse tuffer og de norske ligger i forekomsten av 
blad tufbænken i Gudbrandsdalen, hvilket baade Andersson og Birger 
(1912 1. c. p. T45) og Halle (191 5 p. 38) finder paafaldende og vanskelig 
at forklare. Saaledes stiller altsaa saken sig set med svenske eine; for 
os, som har arbeidet med de norske tuffer, er netop den manglende 
bladtuf i Jemtland vanskelig at forstaa. 

Men bortset fra denne undre tufbænk, er likheten meget fremtrædende. 
L i k e s o m i Jemtland f o 1 g e s D r 3' a s og H i p p o p h a ë s ad o g s a a i 
de norske tuffer og forsvinder samtidig. Rigtignok er de her 
endda ikke fundet sammen, men dette betyr ikke saa meget da stratigrafien 
ved Leine og Gillebu stemmer fuldstændig overens. — I begge til- 
fælder er det en ökning i furuens frekvens som er utslag- 
gi vende og skjæbnes vanger. 

Seknander har senere i overensstemmelse med Halles fremstilling 
tolket den Z^rvas-matte som optrær i bunden av visse jemtlandske tuffer, 
som en „senboreal hede" (1915 1. c. p. 540), og det ligger i det hele tat 
snublende nær at sammenstille denne med Dryastuffen ved Leine, som vi 
ikke finder anden plads for end netop i den boréale tid. 

Men hvorfor er der da ikke ogsaa utviklet en bladtuf i Jemtland? 
Dette kan hovedsakelig bero paa to ting: enten har de klimatiske forhold 
her lagt hindringer iveien for tufdannelsen i den første tid efter isens av- 
smeltning, eller rester a\' indlandsisen har persistert længer 
i denne del av S k a n d i n a \- i e n end i det centrale Norge. 

Ved studiet av Ragundasjoens uttapning har svenske geologer (De 
Geer 1915 1. c. p. 191) forsokt at fiksere tidspunktet for den ostlige isrests 
tvedeling (bipartition) i en nordostlig og en sj-dvestlig del, og man er 
nu tilboielig til at ville henlægge denne begivenhet til boreal tid (Sernander 
19 1 6 p. 539). At der i det centrale Norge umulig kan ha ligget igjen nogen 
isrest i boreal tid, har jeg allerede ovenfor omtalt. Det store sporsmaal 



ig2I.Xo. 9- KALKTUFSTUDIER I GUDBRANDSDALEX. I25 

blir da h\'oiiedes disse opfatninger skal kombineres sammen. Hvis \irkelig 
isrestens bipartition forst skedde saa sent som i boreal tid, hvor stor ut- 
bredelse hadde da den tilbakeblivende sydvestlige del av isen? De Geer 
synes at være fuldt opmerksom paa at der her er \anskeligheter tilstede, 
idet han paa sit kart (19 15 1. c.) over isens bipartition ikke bringer den 
sondre islobe til nogen naturlig avslutning vestover i Norge. 

Jeg skal her ikke gaa nærmere ind paa dette problem, som er meget 
vanskelig. Svenske og norske opfatninger divergerer paa dette punkt ganske 
sterkt. M3'tilus-nivaaet og Pordandia-nivaaet spiller her ind paa norsk side, 
likeledes tidfæstelsen av de brædæmte sjeer (cfr. avsnit I) o. s. v. 

Men at der efter Halles undersokelser er banet vei for en korrelation 
mellem de norske og svenske forekomster, anser jeg for gi\et. Mange 
gode grunder taler for at Drvas-Hippop/iars-sonen i Jemtlands tutfer skriver 
sig fra boreal tid, og dette bekræfter den mening som jeg ovenfor har 
hævdet, nemlig at denne periode maa ha været specielt gunstig for fjeld- 
plantene. Aarsakene hertil kan dog ha været forskjellige. 

Furutuflfen i jemtlands-forekomstene tolkes av Sernaxder som adantisk, 
nærmere betegnet som gammelatlantisk (1. c. p. 540). Forfatteren tænker 
sig at tufavsætningen her er blit avsluttet forholdsvis tidlig „genom det 
•öfversilande vattnets sjäfdränering i distinkta erosionsfàror". Imidlertid er 
det al grund til i fremtiden at undersoke det muld- eller humuslag noiagtig 
som pleier at overdække furutuften; for efter den erfaring som undertegnede 
har gjort baade ved Leine og Gillebu, er det meget let ved en hastig gra\-- 
ning at overse eventuelle forvitrede 3'ngre tufrester i jordlaget. Disse er 
ofte forsvindende smaa, men kan allikevel gi værdifulde oplysninger (cfr. 
Alnus-tuffen ved Leine og de ubetydelige furutufrester i visse profiler ved 
Gillebu). 

Hippop}iai-s-problci)ici. 

I den foregaaende fremstilling er Hippophars rliaiiiiioidrs nævnt flere 
ganger. Denne arts merkelige utbredelse i nutiden og dens fossile fore- 
komst paa Gotland, i Jemtland og Asele lappmark har siden lang tid til- 
bake sat plantegeografenes fantasi i bevægelse. Meningene om dens ind- 
vandringsveier til Skandinavien har \æret noget delte, og flere teorier har 
været opstillet i sakens anledning. 

Paa kartet fig. 36 er alle kjendte norske og svenske lokaliteter for 
nulevende og fossil Hippophars indtegnet. For F"inlands og Danmarks ved- 
kommende er fremstillingen noget ufuldstændig, men hovedlinjene i plantens 
utbredelse trær dog tydelig frem. Langs Tysklands Ostersjok3'st er bare 
nogen faa lokaliteter indlagt. Hippophars forekomst i Kurland er tvilsom 
(ældre angivelser hos Koppex I. c, men ikke gjenfundet i nyere tid). 

Av de fossile findesteder er F ^ Fröjel paa Gotland (G. Anderssox 
1895 1. c. p. 45), G =^ Gillebu i Oier (O vex 191 7). De jemdandske lokali- 



120 Kr)l.r NOKDHAfiKN. M.-X. KI. 

IcUr ry ifitli^c ll.M.i.i; (1915) 1' »Iwciulc : Kaflkälcn (Xaiii'iksi 1885 I.e.), 
Digcniäs (Si:rn.\.\I)i:u 1899 I.e.), Sikaskälcii (Carlson ifolge IIai.ij-: 1915 
|x 39), Gaxsjö (Carlson), I'ilsta ((1. Andkrsscn), Tracksta i Hallen (Kjp:ll- 
makk), Gulastjärn (Hallk I. c.l. Ihn nordligste lokalitet i Sverige er Làng- 
scle i Dorotea sogn, Asele lappmark (Nathorst 1885, planche 18). 
/ /i/ipo/i/idrs-poWcn er paavist i sandiag med arktiske planterester paa Snasa- 
högarna og V. ICnadalshöJden av II. Smith (1920 1. c. p. J38); disse lokali- 
teter er ikke indtegnet paa kartet, da der vel kan disputeres om hvor stor 
\egt man skal tillægge forekomsten av pollen av en x'indbestever. Forövrig 
slutter disse forekomster sig meget smukt til de ox'rige. 

U t b r e d e 1 s e i S \- e r i g e i nutiden. 

Norrholh'ii : Neder-Kalix skärgard; Luleå skärgard: Junkön, Smaskaren 

0. s. V. (Backman og Holm p. 247). Seskaroen ( 7 1899 E. Hammarkn 
Hb. Ups.). Haparanda skärgard (J. A. Z. Brundin ^ 7 1900 Hb. Ups.). 
Pitea skärgard: Hunden ^6 1892 E. Lundberg Hb. Ups.), Buskin (1870 
A. N. Lundström Hb. Ups.), Mellerstön, Kluntarne, Rebben, Svartnäs, 
Trundön; Neder-Lulea: Alhamn (1917 E. Markllnd 1. c. p. 796). Kalix 
skärgard: Granön (^^6 1906 F. E. A. Block Hb. Ups.). 

Västcrhottcii : Umea: 'Latla (1887 N. L. Andersson Hb. Ups.). 

Mcdclpad : Brämön ; Njurunda: GaltstrtHii mellem P'uruskär og Sathamn,. 
Björköbyn, Björköviken; Alnön: ej langt fran Spikarna, Rödön, Lillkalven, 
Granön; Tynderö: Skilsakersmalen, Astön söder om Kalvhällberget (Col- 
LiNDER p. 118). 

Hälsingland: Gnarp: Ragvaldsnäs fram tili Medelpadsgränsen ; Jätten- 
dalkredsands allmänning. — Fleresteds dyrket „såsom vid Ljusdal, Hudiks- 
vall etc." (Wiström). 

Gnstriklaiid : Gevle: Brvnäs ( 7 1823 herb. Wahlenbcrg in Hb. Lps.l, 
Brobanken C*/6 1852 R. F. Fristedt cV F. J. Björnström Hb. Ups.), Iggön 
( il 1895 Torsten Arnell Hb. L'ps.), Granön ('^7 1894 T. Arnell Hb. 
Ups.), Miramar (/6 1885 C. O. Schlyter Hb. Ups.). 

Upplaiid: om HippopJiacs utbredelse i Uppland cfr. E. Almouist 192 i 

1. c.'. Den er her almindelig ned til Grisslehamn; sondenfor er den betydelig" 
mere sjelden. De sydligste lokaliteter er Niekö i Ljusterö og Svartlöga i 
Blidö, beliggende ved ca. 59*^ 35' (E. Almouist in litt.). Lokalitetene paa 
kartet er indtegnet efter et kartkoncept utfærdiget av Almouist. I Uppland 
træffes den mange steder paa enger, langs veikanter o. s. v. i kystens nær- 
het, undertiden flere km. fra nu\ærende strand (kulturpaavirkede stederL 

Boliii:>hïii : Oxe\ik i Dragsmark sn. ( 7 1884 G. \Vailix Hb. Ups.); 
Dragsmark (7 1889 A. U. Jonsson Hb. Ups.); Tanum: 0,5 km. sydvest 
for Tanums jernbanestation (Magnusson Sv. Bot. Tidskr. 1918 1. c. p. 472); 
Svenneby: Spanslätt ('-S 1869 G. M. L. Gerke Hb. Göteb., ifolge Magnus- 



' I>onnc avhandling vil bli publicert i Svensk Bot. Tidskrift 1921. 



I92I. No. 9- KAI.KTL'FSTL'DIER I GUDBRANDSDALEN. 



127 




Fig;. 36. Kart over Hippophais iluuniioidcA i Fennoskandina\-ien. # = voksesteder 
i nutiden. O = fossilforekoinster. G. = Gillebu, F. = Frojel paa Gotland. 



128 ROLF NORDHAGE.N. M.-N. Kl. 

.SON I.e. vistnok spontan); Moilanda: I IcniianfJn (I'.m.mkk .S\'. liot. 'lid.skr. 
1920 p. 8g)'. Denne oversigt er sammenstillet av l'il. ma<^. K. .Stf:k.\er, 
L^ppsala, som elskvæixlij^st har stillet den til min disposition. 

Uth redelse i l'inland. 
1 l'"inlan(l < r / lif^/xipluu's noie kn\-ttet til den iiottniske Bugt, hvor den 
optrær paa sti'andkantene Ira Tornea i nord til Alandsarkipelet i .syd. Syd- 
grænsen paa fastlandet er Ny.stad, i skjærgaarden (Aland fraregnet) een 
Enskär i Gustafs sogn i Abo-skjærgaarden („regio aboënsis" efter Herbarium 
Musei Fennici's inddeling 1889I. Lokalitetene er indlagt efter oplysningene 
i Palmgkens monografi (1912), Inor al litteratur vedrørende HippnpJiavs i 
Finland citeres. 

U tb red else i Danmark. 

I Lange: Haandbog i den danske Flora (j886 — 88) anføres følgende 
om plantens forekomst: .Strandklinter (paa 1er og kalk) og sandklitter; i den 
nordlige del av J3iland aim., sjeldnere mot syd (Ris skov ved Aarhus) og 
paa øene. Fyen: Vedelsborg. Sjælland: Refsnæs, Lerchenborg, Stevns 
klint, Moens klint i stor mængde. Falster: Mellem Korselitze og Grønsund 
og mellem Tromnæs og Boto. Laalland: Oreby strand, Aalholm strandklint. 
Bornholm. Warming (1907 — 1909) har en mængde oplysninger om Hippo- 
pliacs utbredelse og økologi. Folgende findesteder omtales her: Almindelig 
i klitterræng nord for Limfjorden, ogsaa ved dennes kyster, f. eks. ved Kaas, 
og inde i landet, f eks. omegnen av Vildmosen, ved Tolne og Vogn bakker, 
Lerup kirke. Mot syd paa vestkysten mere spredt og .sjelden, bl. a. i Xissum- 
Husby klitter, Holmslands klit, Nymindegab. Han omtaler den videre fra 
Skagens nordstrand, Uggerby og Tversted aaer og Rubjerg Knude vest 
for Hjørring. X'idere: i mængde f. eks. paa Møens Klint, paa Refsnæs og 
paa nordøstkysten av Falster. Paa kartet er de av disse lokaliteter ind- 
lagt som jeg har kunnet fiksere nøiagtig. 

Utbredelse i Norge. 
S. Trøndclag og N. Trondclag fylke: Inderøen: \-ed Borgenljorden 
nær Sund ('/v 1918 I. Jørstad). Schübeler anfører følgende: Paa Inder- 
øen 5 m. hoie trær. 2 stammetversnit maalte 13 cm. med 54 aarringer og 
20 cm. med 44 aarringer (uten bark). — Værdalen: Helgeaaens bredder 
nær Hagagaard ved Nes (J. W. Zett.). — Skogn: paa stranden mellem Tynes 
og Levanger i største mængde (j. W. Zett.); Holme (A. Bl.). — Skatval: 
i hammeren mellem X'elvang og Olderen ved kalktuften (juni 19 15 Rolf 
Nordh.agen). — Stjordalen: Sutteroen (Hoffstad). — Prosta: Tautra m. 



1 Litteratur over Hippophah' utbredelse i Sverige: Backmax, C. J., og Hol.m, V. Y.: 
Elementarflora öfver Västerbotten och Lapplands fanerogamer och bräkenartade växter. 
Upsala 1878. Marklund, E., .Svensk Bot. Tidskrift 191 7. Collinder, E., Medelpads 
Flora. Norrländskt handbibliotck III. Upsala 1919. Wiström, P. W.: Förteckning 
öfver Helsinglands fanerogamer och pteridofvter. Wimmerby 1898. 



192 I. No. 9. 



KALKTUFSTL'DIER I GUDBRANDSDALEX. 



129 




aim. paa strandkanter (A. Bl.I; Prosta (•' s 1875 A. Bl.); Holmberget (-'/ô 
19 1 5 I. Jørstad). — Strinda: Dævlehavnen og østover (1869 Storm); 
Saksvik (1876 Ove Dahl); Tinveden (Bl. 1897); Strinden (Oxaal); Trond, 
hjem IM. N. Blytt). — Skjern: Eid (-7 19 15 I. JorstadI. — Ørlandet: 
mellem Fladnes og Hovde rikelig over den ilate utmark (Norman 1883); 
Beian (Boeck & Korex. Herb. Kristiania); Berg og Ostraat (1900 Hoff- 
stad); Ostraat (Korex). — Bjugn: nær grænsen til Orlandet (Normax- 
Storm). — Aafjorden; 
Melem fleresteds; Kven- 
gjerdet; Valdersund (Nor- 
man, Blvtt); Hyldenesset 

(storvoksen) (Norman); 
Valso (Norman); Lovø og 
Oian (1900 Hoffstad); 
Lysø (Angström). — Stok- 
sund: Hosen paa fl^^vesand 
(1900 Hoffstad). — Nærø: 
prestegaardstrakten (Nor- 
man 1883). 

Nordland fylke: Al- 
sten : Strandaasen nær 
Sandnessjoen (Ove Dahl 
191 i); Horvnesodden. — 
Tjøtta: Løvønesset, et helt 
litet krat av indtil mands- 
høie busker (Ove Dahl 
191 1 ). — Hero: Nordviken 
paa Donna (Ove Dahl 
191 i). — Meisfjorden: Lei- 
nes (Ove Dahl 191 i). — 
Salten : Fiskvaagflauget 

(Sommerfelt; A. Land- 
mark har senere fundet den her omtrent 100 m. o. h. ctr. Blytt 1892 1. c); 
Kvitberget øverst i Jordbrudalen ca. 600 m. o. h. (Ove Dahl og R. Nord- 
HAGEx ^/7 1920); Junkersdalsura (Fridtz, Bohlin, Dvrlng (1900), ipse!). — 
Bodo: nedenfor landskirken ved soen paa flåten dominerende busk, steril i 
1876, blad 6,5 cm. langt (Sommerfelt, Norman). — Steigen: Engeløen 
ved Laskestad paa Prestkontindens fot i nøken grusbakke m m. o. h., 137 m. 
o. h. og 20 m. nedenfor birkefeltets midtlinje, undtagelsesvis i blomst ^6 
1867, steril i 1881 (Haukland, Norman, Notø ^^7 1912I. 

Hippophacs' forekomst i Salten, det eneste omraade i Skandinavien 
hvor arten nu findes i indlandet og ikke paa strandkanter, er meget inter- 
essant og lærerik. I Junkersdalsuren blev den opdaget i 1889 av 

Vid.-Selsk. Skr. L M.-X. KL 1921. Xo. 9. 9 




Fig. 37. Fra Junkersdalsuren. E. H.\vréx fot. 



I30 



ROLF NORDMAOEN. 



M.-X. Kl. 



R. I'KiDi/ (l)^l<l^(. 1900 1. c). „juiil<<r.s(lal.sura" er cn storslagen kanjon- 
dannclse, noget over en li.iK mil lang, hvis nordside falder av mot elven 
i lodrette stup, som naar op til 500 m. o. h. Under disse findes der væl- 
dige urdannelser, som paa en række steder antar formen av imponerende, 
instabile rasmarker, hvor der vaar og høst gaar svære skred ut i elven. Paa 
disse steder, som befolkningen kalder „flaug" (fig. 37), mangler vegetationen 
oftest helt, og de alternerer med mere stabile urpartier, hvor det kalkrike 
jordsmon og den heldige eksposition fremkalder en frodig og meget inter- 




Fig. 38. Hippophii'es paa en hylde i Junksrsdalsuren. Utsigt mot vest. Arctostaphylos 
uva iirsi \ forgrunden. 5te juli 1920. Nordhagen tot. 



essant plantevekst. Bjerk (baade Bctiila odorata og verrucosa), iuru, heg, 
graaor [Aluns iiicana), rogn, selje o. ti. optrær skogdannende, og i bunden 
findes en lang række græs og urter, hvoriblandt mange sydlige relikter, 
som har gjort denne lokalitet beromt (Andersson og Birger 1912 p. 120). 
Hippophacs derimot optrær i ganske stort indi\idantal netop paa et av de 
midtre flaug oppe i en svær rasmark, h\or andre trær og busker ikke kan 
trives. Voksestedet er for saavidt aldeles topisk: det er de lette kon- 
kurrence v i 1 k a a r som b e \- i r k e r at planten her kan hævde sig. 
Selv formaar den trods det instabile substrat at holde sig, takket være 
den Sterke vegetative formering som udmerker arten (rik skuddannelse paa 
røttene). Desuten findes Hippophacs ogsaa i en ganske liten koloni noget 
længer vest i uren paa en hylde alleroverst under selve den lodrette berg- 
væg, paa et næsten ubestigelig sted (fig. 38). Nedenunder strækker der sig 



1921. No.9- 



KALKTCFSTUDIER I GUDBRANDSDALEX. 



13^ 



en flere hundrede meter lang rasmark, kanske den iarlit^ste i hele uren. 
Den danner her et litet krat paa en 45 ^ skraanende hylde i 4 — 500 m. 
høide. Sammen med den findes et par buskformede individer av Betiila 
odorata, desuten Dryas octopetala, Salix reticulata, Arctostapliylos uva itrsi 
(se billedet), Riibiis saxatilis og Canipaiiiila rotimdifolia , men intet sammen- 
hængende dække. I juli 1920 var den her aldeles steril og ca. 0,5 m. hoi. 

Sammen med konservator Ove Dahl foretok jeg ogsaa samme sommer en 
ekskursion til den saakaldte Jordbrudal i Salten \-est for Russaanes, hvor Russ- 
aaen over en længere strækning 
har underjordisk lop (kalksten). 
W besteg da det merkelige 
tempelformige „Kvitberget" (tig. 
39), som paa sin sydside har 
svære urdannelser av nedraset 
forvitringsmateriale. Paa denne 
ytterst gunstige lokalitet meter 
man tildels de samme arter 
som i Junkersdalsuren [Cypri- 
pcdiuni, Epipactis, Braya gla- 
bella etc.), og det lykkedes mig 
ogsaa at opdage et litet bestand 
av Hippophacs øverst i den stor- 
stenete ur like ovenfor skog- 
grænsen (i 600 meters høide). 
Den var her ikke mere end 
0,3 m. hoi og steril. 

Hippophaës maa utvilsomt 
tolkes som relikt paa disse 
forekomster (cfr. nedenfor). 
Specielt i Junkersdalsuren er 

skredene saa at si fast institution, som ikke viser tegn til ophor, og baade 
konkurrenceforholdene, den kalkrike bund ' og ekspositionen er her gunstig 
for planten og letter dens kamp for tilværelsen. Paa Kvitberget forer den 
en mere hensyknende tilværelse. 

X'oksestedene paa Helgelandskysten er utelukkende kalkforekomster 
nær stranden, hvor vegetationen oftest er noget aapen, men artsrik. Den 
vokser fleresteds sammen med Dryas og andre fjeldplanter, som her op- 
trær i havets nivaa paa kalk og dolomit (cfr. Ove Dahl op. cit.). 

Forekomstene i Trøndelagen stemmer gjennemgaaende med danske og 
svenske lokaliteter. Interessant er findestedet Hosen i Stoksund, hvor Hoff- 
stad fandt Hippophaës paa flyvesand saaledes som i Danmark. „Her 
dannede H. et krat af omkring 0,5 (op til i) meters høide med aapne mellem- 




Fig. 39. Kvitberget i Salten. //;y>/>o/)/(rtis--lokaliteten 

længst tilvenstre, overst i uren. 7de juli 1920. 

Nordhagen fot. 



1 Om plantens forkjærlighet for kalk cfr. Pai.mgrf.n 1. c. 



132 KOLF NORDHAGEN. M,-N. Kl, 

niiii. Inde i krattet lanfltcs st<rilc skud ai' (Ja/iiiiii borcalc ik verum rikclig, dcs- 
utiii spredt Coiivnlldiiii nui julis og Vicia craccn, hvilken sidstnævnte under- 
ti(l( 11 iiaaedc o[) ovei" 11. buskene. C)]) gjennem krattet stak desuden Festuca 
oviiKj, men især Cciitaurca Scahiosa, enkeltvis ogsaa Knaiitia arvciisis og 
Elynnis arenariiis.'' (IIoffstad 1900 1. c. p. 20.) Ved seK'e Trrtndhjems- 
fjorden fra Orlandet og ind til Indeioen danner llijt/xiplinrs mere eller 
mindre aapnc krat eller smaa besiander paa strandkantene. Paa Skatval- 
halvoen har jeg set den i en brat hammer ca. 150 m. o. h. ved Velvang 
(kalkholdig underlag). 

1 F>lvtt: Norges Flora anføres Hippophaës for „Indviken i Nordfjord 
paa strandbredder (ifølge Krogh)" ; men den er senere aldrig omtalt herfra 
og ingen herbarieeksemplarer foreligger, saa forekomsten er meget tvilsom, 
men slet ikke usands^'nlig. 

Palmgren har i sin store monografi over Hippophaës paa Alandsøene 
git en detaljert skildring av artens utbredelse baade i Europa og Asien, 
samt diskutert plantens livskrav (forhold til lys, temperatur, nedbør, 
biotiske faktorer, specielt konkurrencen med andre arter o. s. v.), og plantens 
økologi maa nu i det hele tat siges at være tilfredsstillende utredet. 

Om utbredelsen skriver Palmgren følgende (1. c. p. 27 — 28): „Der 
Seedorn wird, an Flüssen und Bächen in Gebirgsgegenden sowie an Seen 
und am Meer auftretend, von der Mongolei im Osten bis nach England, 
Frankreich und Spanien im Westen angetroffen. Die Art ist am nörd- 
lichsten in Europa in Norwegen unter 67^ 55' nördlicher Breite, in Asien 
im südlichen Sibirien gefunden worden, während sie am südlichsten in 
Europa am Mittelmeer und in Asien am Himalaya unter etwa 30^ nörd- 
licher Breite vorkommt. In Europa steigt der Seedorn bis in eine Höhe 
von 2 000 m., in Asien sogar bis 5 000 m. über dem Meeresspiegel. In 
Europa umfalàt die Verbreitung des Seedorns ein südliches Gebiet, welches 
einen bedeutenden Teil der Gebirgsgegenden Mittel-Europas sowie Teile 
von Spanien, Italien, und der Balkanhalbinsel in sich schließt, und ein 
nördliches Gebiet, welches einen Teil der Küstenstrecken des Englischen 
Kanals, der Nordsee, der Ostsee und der Westküste von Norwegen um- 
spannt. Innerhalb dieses nördlichen Verbreitungsareals ist der Seedorn 
eine ausgeprägte Küstenpflanze — — — — ." 

Om Hippophaës' merkelige utbredelse skriver Koppen (1. c. 1888 — 89 
p. 644): 

„Die merkwürdige Verbreitung des Sanddornes läfat 
sich auf klimatische Ursachen durchaus nicht zurück fü h renV; 
denn er findet sich unter sehr verschiedenen Temperatur-Bedingungen und 
anscheinend auch unter sehr differenten Feuchtigkeits-Verhältnissen; so 
dürften die "natürlichen Bedingungen unter denen er einerseits in Norwegen 
bis zum 68^ n. Br. und andererseits auf der der Hami-Wüste zugekehrten 

' Uthævet her. 



ig2I.No.9- KALKTUFSTUDIER I GUDBRAXDSDALEX. 133 

vorderen Terasse des Nan-schan (unter dem 40*^ n. Br.) wächst, sich außer- 
ordenthch von einander unterscheiden; dort findet er sich in der ausge- 
sprochensten maritimen, hier dagegen in der kontinentalsten Lage. Es scheint 
daß der Standort, den der Sanddorn bevorzugt, am ehesten seine eigen^ 
tümliche \'erbreitung erklären könnte. Ich habe Eingangs bemerkt, dafa es 
die Meeresküsten und die Ufer der Gebirgsbäche hauptsächlich sind an 
denen dieser Strauch sich ansiedelt. Die geringe Entwicklung der Meeres- 
küsten im europäischen Rufaland (abstrahirt von dem klimatisch für den 
Sanddorn unzugänglichen Küsten des Eismeeres, desgl. des Weissen 
Meeres), so wie die fast vollständige Abwesenheit von Gebirgen auf dem 
kolossalen Räume, den die russische Tiefebene einnimmt, — diese beiden 
Factoren dürften hauptsächlich das Fehlen des Sanddornes auf der letzteren 
bedingen." Pal:^igrexs undersokelser paa Aland bekræfter fuldt ut hvilken 
betvdning voksestedet og specielt konkurrenceforholdene har for 
Hippophacs^. Den taaler til en vi.ss grad saltbund bedre end andre trær 
og busker, og langs den Bottniske Bugt, hvor der paagaar landhævning i 
nutiden og nyt land stadig dukker op, har den udmerkede betingelser, idet 
ingen andre vedplanter formaar at utkonkurrere den. Paa de længst fra 
strandlinjen liggende lokaliteter (de ældste) dor den efterhaanden ut og for- 
trænges av andre planter. I Uppland indtar den ogsaa et smalt spatium 
paa strandkantene, mellem den uroligste del av stranden (med koloniartet 
vegetation) og strandkrattene av or længer inde (Serxander 1905 1. cl. 
I det hele tat er der nu gjort saa mange iagttagelser i sakens anledning 
at man kan slaa fast folgende: 

1. Hippophaës er indenfor temmelig vide grænser klimatologisk indifferent 
(baade hvad temperatur og nedbor angaar). 

2. Planten optrær nu kun paa lysaapne voksesteder, hvis natur er saadan 
at en sluttet vegetation specielt av trær og busker er utelukket (stenet 
strand, flyvesand o. 1. ved kystene; rasmarker. sandige elvebredder etc. 
i indlandet). 

3. Desuten er Hippopliars noget kalkyndende. 

ØvEx har i flere av sine nyere skrifter omtalt Hippophaës^ og polemi- 
serer her sterkt mot en række svenske forfattere som har ytret sig i an- 
ledning av Hippop/iaës-pToh\emet. Han konkluderer selv med den bestemte 
mening, at plantens utbredelse, fremrykning og tilbakegang styres av kli- 
matologiske lover, og henforer dens indvandringstid til sit /'/?o/(fr5-nivaa. 



• Cfr. Nathorsts uttalelser 1886 1. c. I september 1921 har jeg hat anledning til at 
iagtta Hippophaës paa utallige steder i Alperne, og mine erfaringer bekræfter fuld- 
stændig ovennævnte tolkning. — Servettaz S3'nes i sin monografi over Elaeagiiaceae 
(1909 1. cl tildels at ha misforstaat Hippophaës' økologi; han hævder at naar planten 
formaar at spre sig til lokaliteter utenfor fugtige elvebredder, saa er det altid til 
kalkfattige voksesteder. Dette er fuldstændig misvisende, likeledes paastanden om at 
planten foretrækker fugtig bund (cfr. Siegrist 1. c. p. 122). 

2 Videnskapsselskapets Skrifter og Svensk Bot. Tidskrift op. cit. 



134 



UrjI.K NOKI>MA(,KN. M.-N. Kl. 



Da dette faunistisk set er av utpræget sydvestlig karakter, tolker han ogsaa 
mppoplinfs som sydvestlig. „Den synes derfor at \'ære paa det noieste 
samnnnknvttct med den lici' oinliaiidlfde periodes he-le klimatpræg, der var 
a\- utpræget sydvestlig karakter" (1920 1. c. p. 322). 

OvK.N indtar altsaa et standpunkt som er diametralt motsat KOh'ENS, 
Pai.m(;rkns og andre botanikeres. Imidlertid synes Øven ikke at kjende 
til disse hovedverkt r inden //.'/'/)ry>//''//',s-litteraturen ; lian citerer dem ialfald 
ikke. Han berører ikke plantens biologi i nutiden, og nogen 
diskussion av forskjellige forklaringsmuligheter finder man ikke. Da jeg 
ovenfor tilstrækkelig har belyst sporsmaalet om Hippopliacs' forhold til 
klimatiske faktorer, skal jeg her kun omtale plantens relation til Pholas- 
nivaaet. 

Da J lippophars ved Gilkhu ei" knytttt til bladtufbænken og denne av 
Oyen tolkes som „infraboreal" og synkron med hans Pliolas-iiivaa (cfr. 
oversigten i avsnit 111), er Oyens standpunkt forstaaelig. 

Men ved Gillebu moter vi det merkelige fænomen, at Ilippophacs 
fa k t i s k h a r sit m a k s i m u m allerede i b 1 a d t u f b æ n k e n s aller- 
underste del (tuffens underside). Den har været tilstede i store mængder 
netop da tufdannelsen begyndte; blokkenes underside viser os et 
avtryk av markens overflate paa dette tidspunkt. Man maa derfor anta at 
denne art har været tilstede eller indvandret allerede under den konti- 
nentale periode umiddelbart efter isens av smeltning, som 
gik forut for tufdannelsens begyndelse. Ialfald er der ikke levert 
noget bevis for at P/ippop/iars forst er indvandret under den periode som 
udmarker sig ved bladtufdannelse. Heller ikke paa dette punkt virker 
ØYENS fremstilling overbevisende. 

Sammenholder man Hippophacs' fossile forekomst paa Gotland (G. An- 
dersson 1895 1. c), i Jemtlands kalktuffer, i Làngsele sydligst i Lapp- 
marken og ved Gillebu i Norge, kommer man til det resultat, at denne 
art maa ha hat en \id og vistnok temmelig sammenhængende 
utb redelse o\'er hele den skandina\iske hal\-o i subarktisk 
og utover i boreal tid for fu ruskogene (i syd skoger av ædle 
løvtrær) begyndte at fa a o ve r take t'. Dog er det mulig at den i 
indlandet har foretrukket kalktrakter. Allerede i boreal tid begyndte ned- 
gangen, og litt efter litt dode den ut baade i det centrale Norge og i 
Jemtland— Lappmarken. Kun langs kystene og paa et par specielt gun- 
stige indlandslokaliteter i Nordlands fylke har den evnet at holde sig til 
nutiden. Ved den Bottniske Bugt synes den at like sig bedst, takket være 
landhævningen i disse trakter. Det er ingen grund til at anta at 
denne kystutbredelse i og for sig er et sekundært fænomen. 
G. Anderssons fund av fossil Hippophacs paa Alnön i Medelpad i „littcrina- 
gyttja" viser, som forfatteren selv gjør opmerksom paa (1895 p. 45), at den 



' CtV. G. Andersson 1896 p. 455. 



igai.Xo. 9- kalktl'fstudier i gudbrandsdai.en. 135 

i lang tid har været kystplante. At sænkninger av landplaten langs andre 
kyststrækninger i postglacial tid maa ha været skjæbnesvanger for Hippo- 
pliacs og utryddet den over \isse strækninger, har Halle tidligere gjort 
opmerksom paa (1. c. p. 43). 

Da Hippopliacs i klimatologisk henseende er indifferent, er det bare 
naturlig at den i nutiden forandrer og utvider sit utbredelsesfelt og an- 
lægger nye kolonier ^ Hvis forekomsten i Bohusian er av geologisk talt 
ung alder og staar i forbindelse med det danske utbredelsesfelt (cfr. kartet), 
hvilket enkelte forfattere har hævdet (Arnell 1912 p. 232), kan det godt 
hænde at Hippopliacs i fremtiden vil faa en renæssance langs Kattegat- 
Skagerakk vsten . 

Selv om problemet Hippopliacs nu kan betragtes som løst i hoved- 
saken, er det faktiske billede vi for tiden kan gi av dens tidligere ut- 
bredelse, mangelfuldt. Men fremtidige fund vil nok utfylde hullene. Spors- 
maalet om arten er indvandret til Jemtland fra Norge eller fra ost, to 
alternativer som Halle diskuterer, staar fremdeles aapent. Halle synes 
at helde til den forste antagelse, lalfald er det nu en kjendsgjerning at 
planten har vokset i det centrale Norge i subarktisk tid^, og naar saa 
mange andre planter har vandret fra Norge ind i Sverige (cfr. Serxander 
19 1 o), kan dette ogsaa meget vel ha været tilfældet med Hippopliacs. 

Om denne art oprindelig fulgte efter det avsmeltende isdække fra syd 
eller ost eller fra flere kanter samtidig, er omdisputert^. Schwellexgrebel 
betegner Hippopliacs som en asiatisk steppeplante av østlig type (1905 1. c. 
p. 191 1. Warmlxg skriver (Dansk plantevekst II Klitterne 1907 p. 150) at 
den kunde „maaske være indvandret til Sverrig nord om den Botniske 
Bugt fra Stepper i Vestasien og Østrusland, mens den vel er kommet til 
Østersøens andre Kyster og til \'esterhavets fra Mellem-Europa, i hvis 
Bjerge og langs hvis Floder den \-okser. " Da denne sak er meget van- 
skelig at avgjøre, skal jeg her ikke diskutere den nærmere. 

Alt i alt er Hippopliacs' fossile forekomst meget interessant, idet den 
viser os at den akt u alistiske tolkning av denne art som en plante for 
hvem konkurrenceforholdene og derigjennem de edafiske forhold er de ut- 
slaggivende, er fuldstændig korrekt. Fortiden supplerer her nutiden og 
omvendt. Og naar Øyen i Hippopliacs mener at ha fundet et typisk 



' BucHENAV har fra de ostfrisiske oer omtalt hvorledes Hippophæ's i slutten av det 19de 
aarh. har bredt sig østover; den har endda ikke naadd øene Spiekeroog og Wangeroog 
lAbh. natur \v. Vereins Bremen XVII 1903). 

2 Smith (1920 1. c. p. 139» antyder ogsaa en vestlig indvandringsvei for Hippophacs til 
Jemtland (cfr. denne forfatters pollenfund I. 

3 Smith (1920 1. c.) fremsætter den antagelse, at Hippopha-s har overlevet den sidste 
istid paa Norges vestkyst. Dette er dog usandsynlig. Den hoiarktiske fauna som er 
fundet paa Ørlandet, viser at klimatet paa nunatakker og isfrie omraader langs vest- 
kysten maa ha været meget strengt. — Desuten viser 'forekomsten paa Gotland at 
Hippophtt's her har fulgt efter isranden fra syd eller ost. Til Gillebu er den utvil- 
somt ogsaa kommet sondenfra. 



136 ROLF NORDHAGEN. M.-N, Kl. 

eksciiipcl paa luoilcdcs plantcartcncs iilhiv dclsc rent generelt styres av 
klimatologiske lo\(r (I. c. ]). 323I, saa maa dette eksempel siges at være 
meget litc lykkelig \algt. 

i 1\ is ikke I llp/^apliars hadde været kjendt i fossil tilstand, vilde menin- 
gene om dens iiid\ andiingstid og indvandringsveier utvilsomt ha været endda 
mere divergerende. Med utgangspunkt i h\ad vi nu vet om denne art, 
kunde det sporsmaal reises, om vi ikke inden vor flora ogsaa har andre 
arter som endda ikke er fundet fossile, men Inis historie har faldt sammen 
med I lipptipliai's' . I (iiidhrandsdalen maa man i denne forbindelse specielt 
ta^nke paa „klaariset", Myricaria ^cniiaiiica. Den er meget sterkt av- 
hiengig av de edafiske forhold og utkonkurreres meget let av andre trær 
og busker hvor bunden er stabil; kun paa sandige og stenete elvebredder 
som oversvømmes i flomtiden, danner klaariset egne krat. Planten har i 
Skandinavien et nordlig utbredelsesfelt i Tromso — Finmiark fylke og et 
centralskandinavisk fra Honefoss til Namdalen (med forgreninger over til 
Sogn og Nordfjord). Fra det trondhjemske har dette felt en utloper ind i 
Sverige (ialfald tilsynelatende), hvor Myricaria langs Indalselven naar frem 
til den Bottniske Bugt. Man maa anta at denne art i tiden efter isens av- 
smeltning og for furuskogenes tid har hat specielt gunstige livsbetingelser og 
rike spredningsmuligheter. Men desværre vet vi endda intet om dens ind- 
vandringstid ^ I Central-Asien optrær den ofte sammen med Hippophaës^; 
saaledes skriver Drude: „Hippophacs rhamnoidcs, noch am Kuku-nor bis 
3600 m. Höhe ansteigend, wird bis 20 Fuß hoch; oft begleitet ihn die in 
Hochtibet am höchsten steigende Myricaria i^rnnauica'''' (Handbuch der 
Pflanzengeographie 1890 p. 409). 

En anden merkelig art av subarktisk præg er Astcr sithintcgcrrinms 
(Resv. & Ostenfeld), som i nutiden har et meget isolert utbredelsesfelt 
ved Aursunden nord for Roros. Ifolge dr. Resvoll er planten sterkt av- 
hængig av de edafiske faktorer; den er knyttet til et parti av den gruset- 
sandige strand hvor skogen ikke gaar ned til sjoen^. Netop av denne 
grund frygter man for at reguleringen av Aursunden kommer til at bety 
plantens dod; den magter nemlig ikke at trænge sig ind i sluttede natur- 
lige samfund. — Hvis Hippophacs ikke hadde eiet sin merkelige klimato- 
logiske plasticitet og sin evne til at taale en god portion salt i substratet, 
vilde den i nutiden utvilsomt ha fristet en 3-tterst kummerlig tilværelse. 
Den er utdød over milevide strækninger fra Oier og helt til Lappmarken. 



1 Cfr. Andersson & Birger 19 i 2 p. 167. 

2 Dette er vistnok ogsaa tilfældet et par steder ved Trondhjemsljorden og i nærheten 
av Sundsvall. For Tysklands vedkommende cfr. Graebner : Die Pflanzenwelt Deutsch- 
lands 1909. I Schweiz optrær disse to arter ofte sammen, f. eks. i Tessin, hvor jeg 
mellem Mesocco og Bellinzona har set vakre blandede krat paa elvenes grusavleiringer. 

3 Cfr. Th. Resvoll & Ostenfeld : Den ved Aursunden fundne Aster. Nyt Magazin f. 
Naturv. B. 54. 19 16 p. 6. 



192 1. No. 9. 



KALKTL'FSTUDIER I GUDBRANDSDALEN. 



137 



Hvem vet oni ikke Asfcr sithiiitrgcrriiiiits, som synes at \'ære daarlig rustet 
i kampen for tilværelsen (ialfald i nutiden), tidligere har hat en noget mere 
sammenhængende utbredelse? (Cfr. Resvoll & Ostenfeld 1. c. p. 6, hvor 
ogsaa denne mening fremsættes) '. Det nærmeste voksested ligger nu i det 
nordostlige Finland. 

I denne forbindelse vil jeg ogsaa fæste opmerksomheten ved den 
merkelige ansamling av "østlige planter som vi nu for tiden har i selve 




Fig. 40. Aster subiiitcgei-riiitiis paa stenet sandstrand ved Aursunden. Desuluii iccs Astra- 
galus alpiniis. Th. R. Resvoll tot. 

Gudbrandsdalen. Afliyriitiii cirnatimi, Cystoptcris siKÙtica og Afragene 
sihirica har her hver for sig et isolert skandinavisk utbredelsesfelt (fig. 41). 
Eiendommelig nok „hopper" alle tre fra den finsk- russiske 
grænse og til Gudbrandsdalen. 

I. Athyriitm crenatitin: Utbredelse i Finland — Rusland : fra Ponoj- og Imandra-Lappmarken 
i nord til Tavastehus i s\'dvest; ostover gjennem Ladoga-Karelen, den kareliske land- 
stripe mellem Ladoga og Onega og Nowgorod (Her.m.\xn : Flora 1912). Sibirien, Mon- 
goliet, Sakhalin. Utbredelse i Norge: Sel: Rosten ovenfor Laurgaard (Moe). Nord 



' En anden art som i biologisk-okologisk henseende minder om Aster subintegcrriinus, er 
Car ex bicolor, som foruten et nordlig utbredelsesfelt i Skandinavien har 2 meget isolerte 
forekomster i Centralskandinavien, nemlig i Foldalen i Norge (sandige elvebredder) og 
i Härjedalen. 



138 ROLF NOKDIIAGKN. M.-N. Kl. 

for Klingen langs vcicn (M. N. H1..I, Kvam: Slfirurcri (Nok.man HctIj. Kri.st,). Ringcbu: 
ved Laag(;n nedenfor Randkiev bro (lercslcds i mængde (Kaalaas); Klstad paa den anden 
side av elven like overfor gaarden (Ørsanden if. A. Hlvttj. Den vokser lier „ved foten av 
l)ergenc paa skyggefulde miildete steder blandt raatne omstyrtede stammer ogsaa paa 
græsbund i sclskap med subarktiske planter sansom Palypodhim Phegopteris, Stellaria 
ncinontiii, Acoiiititiii, Cnaiiiitin siliuitiriiiii, liniuiiictihi-i <ic<;i- <.ti-." (haandskrcvne notiser 
av A. Bl. i HI. N. FL). — Mangler i det ovrige Europa. 

2. Cystopteris siidcticn : Rusland: Nowgorod (Hermann 1. c). Desuten i Karpaterne. Norge: 
\r(l V'instraolven i CaKlbrandsdahn rit o\-,rf()i- l\i)ngsli, i fiype vanskelig tilgjængelige 
klofter i selskap med ('vs/n/i/cn's iiion/niia og Ciiiiia pendula (Kaalaas 1897). 

3. jUrngi'>ir Sibirien: Kiiiiaiid Rusland; Onega-Karelen og Nowgorod og i lerkeskog i 
Onegadalen (Hermann 1. c, Wille 1. c. p. 251). Nord- og Mellemrusland, Ural, Sibirien, 
Mongoliet, Turkestan. Paa begge sider av sjoen Losna i Gudbrandsdalen (ved grænsen 
mellem Øicr og Ringebu herreder), jiaa østsiden mellem gaardene Enge og Vedum, 
paa vestsiden ved Huglicn og vod Rugakersætrene (op til over 900 m. o. h. Wille 
1. c. p. 243). 

Alle tre arter er av subarkti.sk type og viser en ganske iVappant over- 
en.ssternmelse hvad utbredelsen angaar. \'i har her kanske en av de 
største gaader som vor flora gjemmer. Da alle tre er urteagtige 
eller ialfald visner ned som urter, er chancene for at finde fo.ssile rester av 
dem uhyre smaa. Men et eneste fossilfund kan her si det avgjorende ord. 
For Atrageiic siltiriai's vedkommende har Wille (191 7 1. c.) fremsat den 
anskuelse, at den er indkommet til Gudbrandsdalen ved tilfældig spredning 
i nyere tid, hvilket er meget \'el tænkelig, hnidlertid mener jeg ogsaa de 
to andre arter maa tages med i ræsonnementet, og med henblik paa Hip- 
pQphMS og dens utdoen over hele Centralskandinavien, maa man indromme, 
at der her foreligger muligheter for at ovennæ\'nte arter, som nærmest er 
av subarktisk type, ogsaa kan ha hat en kompliceret historie i det lange 
tidsrum som ligger mellem isens foi'svinden og nutiden. Men opfatningen 
paa dette punkt maa selvfølgelig bli subjektiv; noget videnskabelig bevis 
kan man jo ikke levere. 

C. Den atlantiske tid. 

Denne periode har i Gudbrandsdalen udmerket sig \ed tætte furu- 
skoger. Antageligvis har sydlige lovtrær spillet en viss rolle ialfald i 
dalens nedre del paa gunstige lokaliteter (hassel, aim, muligens lindM. Av 
hasselnøttfund i torvmyrer vet vi at Coryhls Avellaiia har hat en noget større 
utbredelse under den postglaciale varmetid i dalens sydlige del (cfr. Holm- 
boe 1903). ^ 7/;///s-poUen, som det har lykkedes mig at paavise i smaa mængder 
i furutuffen ved Gillebu, er meget interessant, men gehalten er altfor ube- 
tydelig til at kunne tillægges nogen større betydning. 

Fundet av Tofxeldia palustris i den atlantiske tuf \ed Leine \iser at 
enkelte fjeldplanter ogsaa und(^r denne periode har hat tilhold i Leine- 



' Uliniis iiioiitana har nu sin nordgrænse i Faaberg. Gaardsnavnet Ahne i S. Fron viser 
ifolge A. M. Hansen i 19 i 4 1. c.) at den er i tilbakegang. Acer plafmtoides gaar til 
Faavang, Corvliis og Tilici til Ringebu, men er meget sparsomme i sin optræden. 



T92I. No. 9. 



KALKTÜFSTUDIER I GUDBRANDSDALEX. 



139 



hakkene. Muligens har dette hare været tilfældet med alpine eller subalpine 
mvr- og kildeplanter. Drvas, Sn/ix reticulata og de øvrige som dominerte 
i horeal tid, er ialfald nu forsvundet. — Forekomsten av fjeldplanter i 
Jemtlands kalktufter har Sernaxder oprindelig villet forklare som atlantiske 
nedvandringer (1899 1. c, 1916 1. cl. Imidlertid tyder Halles undersokelser, 
specielt paavisningen av den . småle Drvas-Hippophacs-sone i hunden, paa 
at den mere sporadiske optræden av fjeldplanter hoiere oppe i furutuffen 
paa de jemtlandske forekomster snarere er en reminiscens fra horeal tid. 

\'ed Leine rober Fragå- . 

ria vcsca og Bctula verrucosa 
at klimatet har været gunstig, 
ialfald ikke daarligere end i 
nutiden. 

Haxsex har med styrke 
fremhævet hvilken stor betyd- 
ning den postglaciale varme- 
tid med dens hævning av 
skoggrænsene har hat for 
fjeldvegetationen (1904 1. cl. 
Under denne tid er mangt 
et kontinuerlig utbredelsesfelt 
blit splittet ad og mange ark- 
tisk-alpine arter bukket under 
hvor fjeldene ikke var hoie 
nok. Tuffene vidner tydelig 
om denne utryddelseskrig, 
specielt av fotofile arter under 

^4-i^v,^,vi. 4.:^ i^f,- n.-,,^^ ^^ Fia;. 41. Kart over Gudbrandsdalen med indlagte vokse- 

atlantisk tid (cir. JJrvas og , , . ,,, a- • /^^ 4// ■ 4 ,m, 

^ steder tor A/ragene stotrtca ( X j, Athyriiini crctia/iiiii i • 1 
HIppOphacs). og Cystopteris siidetica {"»,). 




D. Den subboreale tid. 

Om dette interessante tidsrum fortæller kalktuflfene i dalen ikke andet 
end at kildene torket ind, og at der indtraadte en forvitring av den tid- 
ligere dannede tuf. 

Som ovenfor omtalt er torvmyrforskerne kommet til det resultat, at det 
„postglaciale klima-optimum", altsaa selve varmetidens klimaks, har indtraadt 
efter atlantisk tid, nemlig under den terre og varme subboreale periode. 
De marine avleiringer baade i Kristiania- og Trondhjemsfeltet viser ogsaa, 
som Oven har gjort opmerksom paa, at vi ingen grund har til at sætte 
Tapesnivaaet (sensu strictol som kulminationspunkt; Trivia-nivaaets og del- 
vis Ostræa-nivaaets fossilrike avleiringer vidner nemlig om ytterst gunstige 
klimatiske forhold, ogsaa efter Tapes-nivaaet. I avleiringer paa Frooene, 
utenfor Trondhjemsfjorden, fra Trivia-ni\aaet fandt jeg i 1915 foruten 



140 



KOLK NORIJIIAGEN, 



M.-X. Kl. 



Trivia cnropœa og en mænj^dc andre sydlige typer ogsaa Solecurtus can- 
didits Rkn. Dr.sir., en niiisling sdhi nii for tid(.-n ikke findes nordenfor den 
Irske Sjo (N<)ki)1i.\(,i'.n 1917 I. c). Dette og andre tidligere fund i de norske 
marine avsætninger viser hvilke voldsomme forandringer og forskyvninger 
vi har at regne med i havets dyreverden i postglacial tid. Og landfloraen 
har gjennemgaat tilsxarende, om ikke større ændringer (cfr. Andersso.n 
1902 1. c., Sernandek 1908 1. c, Mai.mstköm 1920 1. cl. 

En række med plantearter 
av sydlig eller sydostlig, varme- 
kjær og kontinental type, som 
nu har sin nordgrænse i Gud- 
brandsdalen paa gunstige loka- 
liteter (sydskraaninger med 
kalkholdig og ter bund) er 
utvilsomt at opfatte som relikter 
fra subboreal tid. Dracocepha- 
liiiii Riiyscliiaiia og Brachy- 
podium piiniafiini (fig. 42 og 
43), to arter som Sterner 
(i 92 1 1. c.) specielt har studert, 
horer til denne gruppe (cfr. den 
specielle del p. 9). Den sidste 
sætter ikke moden frugt ved 
Leine, saavidt jeg har kunnet 
konstatere; men den har rik 
vegetativ formering. Da Leine- 
bakkene i nutiden huser saa 
mange rariteter, maa man anta 
at de i subboreal tid hadde en 
meget interessant vegetation. 
Fig. 42. Dracoccphaliuu Riiyscliioiia i Skandinavien. Desværre er der ikke opbe- 

Efter Sterner (192 i). For Norges vedkommende er y året hverken tuf eller andre 
voksestedene sammenstillet av R. Nordhagen. Dette _ . 

gjælder ogsaa Brachypodiir.u piinmtum. reStCr fra dette tidsrum'. 




E. Den subatlantiske tid. 

Den merkelige Alnus-tuf og gruskeglen ved Gillebu vidner om store 
forandringer efter den subboreale tid. Fugtige orekrat lik de nuværende 
har udmerket Leinebakkene, men furuen har dog fremdeles været tilstede. 
Gillebutuffen fortæller intet om vegetationens karakter i Oier under denne 
periode. 



' Jeg har tidligere omtalt de divergenser og vanskeligheter som knytter sig til beteg- 
nelsen „subboreal". Noget bidrag til denne periodes tidfæstelse kan kalktuffene i Gud- 
brandsdalen selvfolgelig ikke levere. 



1921. No. 9- 



KALKTUFSTUDIER I GUDBRANDSDALEN. 



141 



Man antar nu at granen indvandret til Norge i subboreal tid (Holm- 
sen 1 919 og 1920 1. c), og det er Ibr saavidt merkelig at rester av denne 
art mangler i den subatlantiske tuf" ved Leine. I nutiden er dog granen 
sjelden i de overste b3'gder i Gudbrandsdalen (f. eks. i Lesje, hvor den 
betragtes som en raritet og bindes til kranser, omtrent som barlind andre 
steder i landet). Den holder sig her mest til bakliene (Hellaxd 1. c). 
Dette tvder paa en relativt sen fremtrængen i dalens nordre del, selv 
om kanske klimatiske forhold 
i nutiden ogsaa spiller ind. - 
Jeg har forovrig tidligere præ- 
cisert at den negative kjends-' 
gjerning at granrester mangler 
ved Leine, ikke behover at bety 
saa meget. Da Alnus-tuften er 
sterkt for\-itret, ligger det nær 
at tænke sig at tuidannelsen i 
subatlantisk tid væsentlig har 
paagaat i periodens forste del. 
^'ngre lag er muligens ogsaa 
helt forvitret. 

I subatlantisk tid rykket 
skoggrænsene paa fjeldene atter 
ned fra det hoitliggende nivaa 
og til det nuværende. Regio 
alpina blev herigjennem sterkt 
utvidet, og nye sprednings- 
muligheter aapnet sig for Ijeld- 
plantene. De tjeldarter som vi 
finder i lerbakkene i Kvam i 
nutiden, er vistnok subatlantiske 
og recente nedvandrere. Ler- 
skredene i nyere tid maa saa- 
ledes som tidligere omtalt, ha 

skapt gode vilkaar for fjeldplantene ved at blotlægge nye arealer og skape 
lette konkurrencevilkaar og rik lystilgang'. En art som Dryas odopctala 
har dog endda ikke fbrmaadd at gjenerobre det tapte terræng. 

Omvendt har sydlige typer under subatlantisk tid undergaat store for- 
sky vninger. Naar dette gjælder arter som hassel, Trapa .iiataiis etc., saa 
maa man ha lov til at generalisere forholdet. Dog er det mulig at for- 
skyvningene i den varme og lune Gudbrandsdal har foregaat i mindre skala 
end manse andre steder. 




Kig. 43. Brachvpodiitiii piiuiaium i Skandinavien. 
Efter Sterner (i 92 i). 



' Denne aapne plantevekst i skraaningene er ogsaa hensigtsmæssig for de sydlige relikte 
typer, som her ikke har saa let for at bukke under i konkurrencen med andre arter. 



142 



KOKK NOUOMAfiKN. M.-N. Kl. 



' I dalens vegetationshistoi-ic har den subatlanti.ske tid været en av 
de allervigtigste, idet granskogene litt efter litt har presset sig frem, 
specielt paa furuens bekostning. Dette har været mere fremtrædende i 
dalens nedre del end i de øvre trakter, hvor granskogen væsentlig er 
knyttet til de fugtigere og skyggefulde baklier med nordlig og østlig 
eksposition. 

Endelig har mennesket i historisk tid paavirket vegetationen paa utal- 
lige maater, likeledes vandflommene og lerskredene. 



Alt i all kan \ i for (Judbrandsdalens vedkommende adskille tre epoker 
i s k o g e n e s h i s t o i- i e : 

1. Den første subarktiske bjerk-aspeskogs tid. 

2. Furuskogenes tid. 

3. Granskogenes tid. 

Under den postarktiske varmetid har antageligvis ogsaa ædle løvtrær 
spillet en viss rolle paa gunstige lokaliteter, saaledes som ovenfor antydet. 

Denne inddeling sigter kun til de dominerende træslag; det er saa 
at si en kvantitativ inddeling, og den falder ikke sammen med den over- 
sigt som andre forskere opstiller paa grundlag av undersøkeiser i Danmark, 
Syd- eller Mellemsverige og det sydlige Norge. Her har forholdene været 
anderledes, ikke alene kvantitativt, men ogsaa kvalitativt. I det hele tat 
er det umulig at fastsætte en viss postarktisk „norm" for 
hele Skandinavien grundet paa palæobotaniske principper, 
saaledes som ældre plantegeografer antok. Forholdene har variert be- 
tydelig i de forskjellige deler av Skandinavien, om end visse større geo- 
grafiske provinser med et visst fællespræg i vegetationshistorien kan ut- 
skilles. For Norges vedkommende er disse spørsmaal endda lite studert. 



Fortegnelse over de fundne plantearter og snegler. 

Karplanter. 

Eqitisetiiin varicgaiitiii All. Uhyre aim. i mosetufbænken ved Leine 
(Blytt i 891); ogsaa i den undre Dryashorisont og i Dryastufifen. 

Equisetitiii hiemale L. Furutuflfen ved Leine, men især aim. i Aliuts- 
tuften. .Nedre Dal i mængde i bladtuffen (Blytt i 892 1, muligens ogsaa i 
furutuffen. 

Piiius silvestris L. I furutuffen ved Leine, Gillebu og Nedre Dal i 
uhyre masser (naaler, bark, grener, kongler); i Dryastufifen ved Leine spar- 
somme naaler, nederst kortere og smalere, men allerede i Dryastut kom- 
pleksets midtre del lange og brede (cfr. Blytt 1892). Sparsomme naaler 
i Alnus-tufifen ved Leine. Pollen i bladtuflfen ved Leine og Gillebu og i 
bladtufblokkene ved Onset. 



I92I.X0. 9- KALKTUFSTLDIER I r.rnBRANDSDALEX. 1 43 

Picea cxcclsa (Lam.) Link. Nogen naaleformige avtr3^k i et par smaa 
stvkker fra Onset (Blytts samling) og i undertegnedes samling av lose 
tufbiter i jorden sammesteds tilhorer vistnok denne art. — Mangler i Gud- 
brandsdalens tuffer. Kun ved analyse av forvitret furutuf fra Gillebutuffens 
Gverste del har jeg set et par pollenkorn av gran, hvilket utvilsomt skyldes 
recent tilforsel med nedsivende vand. 

Popitlits trcmiila L. Bladtuften ved Leine, Gillebu og Nedre Dal 
(Blvtt 1892); Dryastuffen, furutuften og Alniis-iu'^ew ved Leine; furutuffen 
ved Gillebu. Onset i bladtuf blokkene (O yen 1920). 

Salix capnva L. Ved Leine, Gillebu og Nedre Dal aim. i bladtuften 
(Blvtt 1892 1. \'ed Leine sparsomt i Dryastuften, desuten i furutuften og 
Alnustuften sammesteds. 

Sa/ix nigricans Sm. 1 bladtuften \ed Nedre Dal (Blytt 1892); \ist- 
nok ogsaa i bladtuften ved Leine og Gillebu; i bladtufblokkene ved Onset 

(O YEN 1920). 

Salix ofr. phylicifolia Sm. I bladtuften \ed Leine og i Dryastuften 
samt i bladtuften ved Gillebu forekommer en del blader som maa henføres 
til denne art; men bestemmelsen av fossile Saliccs er meget vanskelig. 

Salix arbuscula L. \>d Leine i den undre Dr3-ashorisont, men især 
i Dryastuften almindelig (Blytt 1892). I et profil er den iagttat nederst i 
furutuft'en. — Planche I, fig. 8. Foruten blader har jeg set småle Ç-rakler 
av denne art (planche II, fig. 2). 

Salix hastata L. Et par blader fra bladtuften ved Leine (Blytt 1892'. 
Denne bestemmelse er utvilsomt rigtig. 

Salix glauca L. Bladtuffen ved Leine temmelig aim. (Blytt 1892) 
Kollodiuma\tryk av et vakkert opbevaret avtryk viser at epidermis har 
været sterkt haaret. Bestemmelsen er ganske sikker (planche I, fig. 6). 

Salix cfr. laiiata L. Et par mindre avtryk fra bladtuft"en ved Leine 
tilhører muligens denne art, men bestemmelsen er meget usikker. 

Salix reticulata L. DryastufFen ved Leine (Blytt 1892). Blytt fandt 
bare to blader; men da jeg selv har fundet 13 avtryk, er arten neppe 
sjelden i dette lag (planche III, fig. 2I. 

Salix herbacea L. Av denne art har jeg fundet et fragmentarisk, men 
typisk bladavtryk (planche I, fig. 7) i Dryastuft"en ved Leine. Karakteristisk 
for denne art er at bladets hovednerve (midtnerve) ikke loper distinkt ut 
til bladranden, men svækkes og blir mere utydelig et stykke fra randen. 
Den fine tanding og bladets smaa dimensioner stemmer likeledes med denne art. 

Foruten de anførte arter har jeg fundet avtryk av andre Salices; men 
det er ikke mulig at bestemme dem med sikkerhet. 

Bctiila odorata Bechst. Alm. i bladtuff"en (og mosetuffen) ved Leine, 
Gillebu, Nedre Dal (Blytt 1892). Ogsaa i Dryastuften, og i furutuften paa 
alle tre forekomster, desuten i ^/////5-tuften ved Leine. Onset i bladtuf- 
blokkene. — Foruten blader og bladfragmenter i store kvantiteter har jeg 
ogsaa fundet forkalkete Ç-r'akler. 



'44 



KOI. F NORIJIIAGEN. M.-N. Kl. 



lirliild vcrniiosii Va\\<\\. lii .^ii .inrorcr et bladfragmcnt fra furutuflTen 
ved l.cinc (ilolilx lltandit, na-stdi l:i|)|)< I rand), (\()\s, utidir t\il, lil<(dcdcs et 
lani^stilkcl lilad tia Ini iiliifTcii \-ed Nedre l)al (usikkert). — Jeg har selv i 
riii-iiliirfcii \cd l.ciiii fundet et hjerkeblad med langt uttrukket .spid.s og 
dohlxlttandct rand, som niaa heni'ores til /j. verrucosa. 

lirhild cli". iiitcniirdid liioM. „1 )ryastutr(n \'efl Leine, meget usikker" 

(lîl.^TT l891>l. 

Betitln c'iv. iidiKi L. „Va hi-ndstykkc av et blad, meget usikkert fra 
Dryastuften \ed Lcinc" (Iîi.vtï 1892). 

Belula ciV. a/pes/ris Fr. „Leine, den nederste del av birketuiTen. Et 
eneste blad uten top, hvorfor bestemmelsen er usikker" (Blytt 1892). — 
Jeg har selv hat anledning til at se de angjældende a\tryk av B. nana og 
iiiferniedia og finder dem høist usikre. I mine egne samlinger har jeg heller 
ikke fundet spor av disse former. 

Aluns iticniin DC. Blytt anibrer et par fragmentariske bladavtr^yk fra 
birkrtuffen \-ed Leine som „cfr. Alnus incana" ; men bestemmelsen er meget 
usikker da randen mangler. I ^/////s-tuffen ved Leine uhyre aim., ogsaa 
en hel del Q-rakler. Bladstorrelsen varierer adskillig (planche V). — Alm. 
i bladtuf blokkene ved Onset (Blytts samling. Øven i 920 1. 

Corylus Avellaua L. Avtryk og hulhet efter en nott i et kjst tufstN'kke 
i jorden ved Onset. Alder ubestemt. 

Ribes cfr. ruhruni L. Blytt anforer et haandnerxet blad fra birke- 
tuffen ved Leine. Noget usikker artsbestemmelse. 

Sorbits Aucupariah. I furutuffen ved Leine et bladfragment (planche II, fig-S). 

Cotoncastcr hitcgcrrinia Medic. Blytt opforer et blad fra Dryastuffen 
ved Leine; men bestemmelsen er usikker. 

Prit lins Padus L. „Den nedre del av et blad 0,04 m. bredt paa 
midten, fra birketuffen ved Dal" (Blytt 1892). 

Dryas octopetala L. Blader og hele stammer med paasittende blad- 
rester i store masser i Dryastuffen ved Leine (Blytt 1892). Desuten i den 
undre Dryashorisont. Et enkelt blad i bladtuften. Forskjellige eiendomme- 
lige kalkkjerner og avtryk (blomster i begyndende frugtstadium (planche I, 
fig. i), blomsterbund med nierker efter smaafrugtene (planche I, hg. 2, 3, 
4, 5)) skriver sig ogsaa fra denne art. 

Fragaria tiesca L. 1 furutuffen \ed Leine fandtes et mindre, trekoblet 
blad, som i ett og alt stemmer med jordbærblader. Desværre var bladet 
saa skjort at det under transporten til Kristiania blev noget odelagt. Av 
planche III, hg. i ser man at de tre deler er fossilificert i forskjellige planer. 
{Ulniaria pentapctala, Ribes riibritni o^^ Ritbus saxatilis er utelukket, h\'ilket 
den direkte sammenligning paa findestedet dengang avtrykket ble\" frem- 
fundet, viste.) 

Hippophacs rliaiiiiioides (L.) Aschers. Bladtuften ved Gillebu, i mængde- 
vis i tuf bænkens bund og paa dens underflate (Oyen 191 7, det. Nordhagen), 
sparsommere opad. 




I92I. No. 9- KALKTUFSTLDIER I GUDBRANDSDALEX. I45 

Pyrola cfr. luitior L. I Dryastuffen og furutuffen ved Leine flere 
blader og en hel forkalket liten plante av en Pyrola, som stemmer godt 
med P. minor; men artsbestemmelsen er ikke helt sikker. 

]\iccin{nin vitis idcea L. I furutuften ved Leine m. aim. og Nedre Dal 
aim. (Blytt 1892); ved Gillebu sparsom baade i bladtuften og furutuften. 
Blvtt anfører den ogsaa, om end noget usikkert, fra Dryastuff"en ved Leine. 

Vaccinium uliguiosiun L. I bladtuften ved Nedre Dal et eneste blad 
(Blytt 1892). Furutuften ved Leine flere blader. 

Cfr. Arctostaphylos iiva iirsi (L.l Spreng. Blytt anforer et usikkert 
brudstykke (den nedre halvdell av et blad fra Dryastuften ved Leine. — 
Bestemmelsen er hoist usikker; jeg har 
&Av fundet smaa glatte bladavtryk i Dryas- 
t :ffen som ligner baade Cotoncastcr og 
z idre arter med helrandete blader, men 
s >m vistnok bare er ax'tryk av Salix 
arbuscula, hvor det øverste kalklag (med 
r "vaturavtrj^kket) er avspaltet eller ut- 
v.?'<et. 

,• T T- 1 1 7 • r ^i»- 44- Cellestruktur av Pellia sp., 

Ltnnæa borealis L. Lt blad 1 turu- ^ . ,-. 1 n j- .1 i iw .- 

tegnet etter kollodiumavtryk av kalktut. 

tuft'^n ved Nedre Dal (Blytt 1892). 

Cirsium heterophyllinn (L.l All. Av denne art fandtes flere store kurver 
og -' n del bladfragmenter i furutuften ved Leine; desværre er de største 
blit ^nust under transporten. En mindre kurv er avbildet paa planche II 
(fig. I. Bestemmelsen er helt sikker. 

Carex sp. En liten forkalket plante fandtes i Dryastuften ved Leine. 
I ^/a 75-tuft'en grove blader som vistnok skriver sig fra Cariccs. 

Gj-amiua. Ubestemmelige bladfragmenter og hulheter efter græs er 
aim. i tuffene. I Alnustuff'en ved Leine flere grove græsblader. 

J ofieldia palustris Huds. I furutuffen ved Leine i den sedentære sone 
fandtes flere vakre bladvifter (planche II, fig. 4I. Kollodiumavtryk av epi- 
dermis viste at spalteaapningenes anordning og størrelse var som paa 
levende materiale. De eiendommelige småle og „ridende" blader kan ikke 
forveksles med andre planters end Narthcciiini's, som er meget bredere og 
grovere. 

Moser. 

Blytt anfører AmblystegiuDi falcatnm Brid. og Nbiiiini piuictatinn 
Hedw. fra Leine samt Hypna, som ikke kan bestemmes med sikkerhet. — 
Baade ved Leine og Gillebu findes vakre mosetuft'er, men artsbestemmelsen 
er meget vanskelig. Av Icvcrnioser har jeg kunnet identificere Marchantia 
polymorpha L. ved hjælp av kollodiumavtryk (cellestruktur paa oversiden og 
undersiden, „aandehuler", bukskjæl. Cfr. planche I, fig. 9), desuten Pellia sp. 
(fig. 44). Den første fandtes i mosetufbænken ved Leine, den sidste i den 
undre Dryashorisont. 

Vid.-Selsk. Skrifter. L M.-N. Kl. 1921. No. 9. 10 



146 ROLF NORDHAGKN. M.X. Kl. 

Lavarier. 

I fiinitiirfcn \((1 Leine landtes et eiendoninielig avtryk i den sedentære 
sone (phinelie II, Wy;. 5). \'ed kollodiiimavtryk kunde jeg konstatere at det 
ikke (1" nogen levermos. Det er utvilsomt en lavart, antageligvis I'an)i(lia 
physodcs (L.l AcH., som optrær i enhver furuskog i store mængder. Stør- 
relsen, lobernes form og avslutning stemmer godt med denne art. Ogsaa et 
par andre, men mere utydelige avtryk fra Leine skriver sig vistnok fra lav- 
aiter. Hi .^■l r landt vakre avtryk av Prltli^rra cnniiin \\ ..] Tii. Fr. sparsomt 
i furutuffen ved Leine. Fossile lavarter er ellers meget sjeldne (cfr. Ser- 
N ander: Subfossile Flechten. Festskrift til Stahl 1918 1. c.j. 

Characéer. 

I mosetuffen i bunden av Leinetuffen optrær der en mængde fine smaa 
kalkror, som tildels er kantet og forgrenet, og som utvilsomt skriver sig 
fra Cliaracr'cr. Restene har en paafaldende likhet med Characékalk fra 
sjøer paa Gotland (Vaxtbiologiska Institutionens samling. Uppsala). Ogsaa 
grovere skud forekommer. En del smaa rer er avbildet paa planche 1 
(fig. 10). Muligens Nitclla opaca, men usikkert. 

Snegler. 

Bestemt av docent cand. real. F. Oklaxd, Aas landbrukshøiskole. 
Blytts samling blev bestemt av avdode frk. B. Esmark. 

Claiisilia nigricans (Strom). Bladtuf blokker ved Onset (det. B. Es- 
mark, OvE.N 1920). 

Claiisilia plicatula Drap. Sammen med foregaaende. 

Coclilicopa liihrica Müll. Dryastuffen ved Leine (Blytt 1892); ogsaa 
i Alnustuffen sammesteds. 

Coniiliis fiilviis Müll. I morænelerens øvre del, mosetuffen, Dryas- 
tuffen og furutuffen (Blytt 1892) ved Leine. Onset i bladtufblokkene 
(det. Esmark, Oyen 1920). 

Helix arhiisfonini Less. I furutuffen og bladtuffen ved Nedre Dal 
(Blytt 1892). 

Hyaliiiia radiaUtla Alder (incl. var. pctroiiclla Charp.). 1 Dryastuffen,. 
furutuffen og Alnustuffen ved Leine (Blytt 1892). Bladtufblokkene ved 
Onset (det. Esmark, Oyen 1920). 

Pupa iiiitscoruiii Müll. Bladtuffen ved Nedre Dal (Blytt 1892). 

Pyranndula ntdcrata Studer. Furutuffen ved Dal (Blytt 1892)^ 
Dryastuffen og Alnustuffen ved Leine. Bladtuffen ved Gillebu (Oyen 1920)- 

SphyradiiDJi vdcntiihtui Drap. Mosetuffen ved Leine. 

Siicciiica sp. Bladtuffen ved Gillebu (Oyen 191 7, ipse!). 



1 92 1. No. 9. KALKTUFSTUDIER I GUDBRAXDSDALEX. 1 47 

î'a/oiiia pnkliclla MCll. Bladtufblokkene ved Onset (det. B. Esmark, 
Oyex 1920I. 

Vertigo alpestris Alder. Alnustuffen ved Leine. 

Vitriiia pcUncida Müll. Morænelerens overste del, samt i furutuften 
ved Leine (Blytt 1892). Bladtuffen ved Nedre Dal (Blytt i 892 1. 

Liliiiiaa tnmcatitla Müll. Morænelerens ovre del, Leine (det. Dr. 
N. Odhner, Stockholm). 

Hxdrohia Stcuii v. Mart. Alnustuffen ved Leine. 



148 RfM.K NDKDIIACKN. M.-N. Kl. 



LITTERATURFORTEGNELSE. 

Andersson, Gunnar, Om nagra växtfossil frân (iotland. Gcol. Foren. Förh. Stockholm 
1895. Bd. 17. 

— Die Geschichte der Vegetation Schwedens. Botan. Jahrb. Leipzig 1896. Bd.- 22. 

— Hasseln i Sverige fordom och nu. Sveriges Geol. Unders. 1902 Ser. Ca, No. 3. 

— Die Entwicklungsgeschichte der Skandinavischen Flora. — Resultats scientifiques du 
Congrès international de Botanique. Wien 1905. Jena 1906. 

— The climate of Sweden in the late-quaternary period. Facts and theories. — Sve- 
riges Geol. Unders. 1909 Ser. C, No. 218. 

— och Birger, Selim, Den norrländska florans geografiska fördelning och invandrings- 
historia. Norrländskt Handbibliotek V. Uppsala och Stockholm 1912. 

Arnell, H., Hippophae rhamnoides och Thymus serpyllum. Svensk Bot. Tidskrift. 1912 Bd. 6. 
Birger, Selim, Om Härjedalens vegetation. — Arkiv f. botanik. Upsala 1908. 
BjøRLYKKE, K. O., Det centrale Norges fjeldbygning. N. G. U. 1905. No. 39. 

— En „hardpan"-dannelse i Norge i arid khma. — Norsk Geol. Tidsskrift. 1911. 

— Norges kvartærgeologi. — N. G. U. 1913. No. 65. 

— Norges jordbundsprovinser og klimatiske hovedstrok. — Foredrag ved de Skandi- 
naviske naturforskeres 16. mote i Kristiania 19 16. Kristiania 19 18. 

Blytt, Axel, Essay on the immigration of the Norwegian flora during alternating rainy and 
dry periods. Kristiania 1876. 

— Iagttagelser over det sydostlige Norges torvmyre. Christiania Vidensk. Selsk. Forh. 
1882, No. 6. 

— Om de fytogeografiske og fytopalæontologiske grunde for at antage klimatvexlinger 
under kvartærtiden. Ibidem 1893, No. 5. 

— Kalktuffen ved Leine. Naturen 1891. 

— Om to kalktufdannelser i Gudbrandsdalen med bemærkninger om vore fjelddales post- 
glaciale geologi. Chri.a Vidensk. Selsk. Forh. 1892. 

— Zur Geschichte der Nordeuropäischen, besonders der Norwegischen Flora. — Englers 
Jahrbücher. B. 17. Beiblatt 41. 1893. 

— Nye bidrag til kundskaben om karplanternes udbredelse i Norge. Chri.a Vidensk. 
Selsk. Forh. 1882, 1886, 1892 og 1897. 

— Haandbog i Norges Flora. Udgivet ved Ove Dahl. Kristiania 1906. 

— Theorien om den norske floras indvandring under vekslende torre og fugtige perioder. 
Et efterlatt manuskript. Bergens Museums Aarbok 1909, No. 8. 

Blytt, M. N., Norges Flora I — III. Kristiania 1861 — 1876. 

Brøgger, W. C, Om de senglaciale og postglaciale nivaaforandringer i Kristianiafeitet. N. G. U. 
No. 31. Kristiania 1900— 1 901. 

— Strandlinjens beliggenhed under stenalderen i det sydostlige Norge. N.G.U. 1905. No. 41. 

— Norges geologi i „Norge 1814— 1914". Kristiania 191 4. 
Clements, F. E., Plant succession. Washington 1916. 

Cleve, Astrid, Zum Pflanzenleben in nordschwedischen Hochgebirgen. Bih. K. Svenska Vet. 
Akad. Handl. 1901. Bd. 26. 



I92I.N0. 9- KALKTUFSTL'DIER I GL'DBRAXDSDALEN. I49 

Dahl, Ove, Botaniske undersokelser i Helgeland I. Kri.a Vid.-Selsk. Skrifter 191 1. 
Danielsen, Daniel, Glacialgeologiske undersokelser omkring Kristiansand. Nyt Mag. f. Xaturv. 
1908. B. 47. 

— Bidrag til Sorlandets kvartærgeologi. N. G. U. 1910 No. 55. 

— Nyt fund av glaciale plantefossiler. Naturen 19 10. 

— Kvartærgeologiske streiftog paa Sorlandet. Nyt Mag. 1912. Bd. 50. 

De Geer, Gerhard, On late-quaternary time and climate. Geol. Foren. Forh. Stockholm 
1908. Bd. 30. 

— Om finiglaciala 1 oW/Vi-relikter. — Geol. Foren. Förh. Stockholm 1913. Bd. 35. 

— Om naturhistoriska kartor ôfver den baltiska dalen. Populär naturv. Revy. 1914- 
Stockholm 19 15. 

— Om internationell användning af den svenska kvartærkronologien. Geol. Foren. Förh. 
Stockholm 1916. 

— Om tidpunkten for Allerôdoscillationen. Ibidem 1916. 

Dyring, Joh., Junkersdalen og dens flora. — Nyt Mag. f Natur\-. Kristiania 1900. Bd. 37. 
Enouist, Fr., Cher die jetzigen und ehemaligen lokalen Gletscher in den Gebirgen von Jämt- 
land und Härjedalen. Sveriges Geol. Unders. 1910. Ser. Ca, No. 5. 

— Die glaziale Entwicklungsgeschichte Nordwest-Skandinaviens. Sveriges Geol. Unders- 
Àrsbok 1918 Ser. C, No. 285. 

Five, Ixgbr., Om saltbitterjorden i Nordre Gudbrandsdalen. Jordbundsbeskrivelse nr. 5. 

Kristiania 1 9 1 1 . 
Fries, Th. C. E., Botanische Untersuchungen im nördlichsten Schweden. Uppsala 1913. 

Akad. avhand. 

— Über die regionale Gliederung der alpinen Vegetation der fennoskandischen Hoch- 
gebirgen. Vet. och prakt. unders, i Lappland anordn, af Luossavaara-Kirunavaara 
A.-B. Uppsala 191 7. 

— Nàgra kritiska synpunkter pà skogsgränsproblemet. — Sv. Bot. Tidskr. Stockholm 
1918. Bd. 12. 

Graebner, p.. Die Pflanzenwelt Deutschlands. Leipzig 1909. 

Halle, Thore G:son, En fossUförande kalktutf vid Botarvfe i Fröjels socken pä Gotland. 
Geol. För. Förh. Bd. 28. Stockholm 1906. 

— Nàgra jämtländska kalktuffer och deras flora. Sveriges Geol. Unders. Arsbok 8 I1914). 
Ser. C, No. 260. Stockholm 1915. 

H.\nsen, Andr. M., Hvorledes har Norge faat sit plantedække? Naturen 1904. 

^ Landnàm i Norge. Kristiania 1904. 
Hartz, N., og Milthers, V., Det senglaciale 1er i Allerod teglverksgrav. — Meddel, fra 
dansk geol. forening. 1901. 

— Bidrag til Danmarks senglaciale flora og fauna. Danmarks Geol. Unders. 1903. 
Ser. II, No. 11. 

Havren, Ernst, Fràn Bodo till Junkerdalen. — Finlandias Arsbok 19 19. 

Helland, Aml'^d, Norges land og folk. Kristians amt I — III. Kristiania 1913. 

Hermann, F., Flora von Deutschland und Fennoskandina\-ien sowie von Island und Spitz- 
bergen. Leipzig 1912. 

Hesselm-AN, Henrik, Studier over salpeterbildningen i naturliga jordmåner. Meddel, fràn 
Statens Skogsförsöksanstalt. Stockholm 191 7. 

— Om pollenregn på havet och fjärrtransport af barrträdspollen. Foredrag. Geol. Foren. 
Förh. Stockholm 19 19. 

Hoel, Adolf, Referat av Øyens „ Kvartærstudier i Trondhjemsfeltet IH". Geol. Foren Förh. 

Stockholm 19 16. 
Hoffstad, O. A., Vegetationen og floraen paa kysten af Trondhjems stift nordenfor Trond- 

hjemstjorden. Nyt Mag. f. Naturv. 1900. Bd. 37. 
Holmboe, J., Planterester i norske torvmyrer. Kri.a Vidensk. Selsk. Skr. 190.3. 

— En undersjoisk tor\Tnyr ved Nordhassel paa Lister. Naturen. 1909. 



I50 



ROLF NORDHAGEN. M.-N. Kl, 



HoLMSF.N, Gunnar, Hrædæmte sjøer i Nordre Østerdalen. Norges Geol. Unders. 1915. No. 73. 

— Om sporene efter en lokal brædannelse yngre end storbræens avsmeltningsperiode i 
den Skandinaviske halvøs bræsjøomraadc. — Norsk Geol. Tidsskrift Bd. III. 1915. 

— Sulitelmatrakten. Norges Geol. Unders. Aarbok 191 7. 

— forskyvninger i snelinjens høide under avsmeltningsperioden. Norsk Geol. Tidsskrift 
Bd. IV. Kristiania 19 18. 

— Litt Gin grangrænsen i F'æmundstrakten. Tidsskrift for Skogbruk. Kristiania 1919. 

— Naar indvandret granskogen til Kristianiatrakten? Ibidem 1920. 

— Notis om pollen i kalktiif. Norsk Cieol. Tidsskrift 1921. 

Hoops, Johannes, Waldbäume und Kulturpflanzen im germanischen Altertum. Strassburg 1905. 

Hult, Ragnar, Försök till en analytisk behandling af växtformationcrna. Akad. Afh. Hel- 
singfors i88r. 

HuLTH, J. M., Über einige Kalktuft'e aus Westergötland. Bull, of the Geol. Instit. of 

Upsala. No. 7. 1899. 

Jessen, Knud, Moseundersogelser i det nordostlige Sjælland. — Danmarks geol. Unders. 
Nr. 34. Kobenhavn 1920. 

Johansen, A. C., Om den fossile kvartære molluskfauna i Danmark og dens relationer til 
forandringer i klimaet. — Diss. Kobenhavn 1904. 

Kaldhol, H., Glacialgeologiske iagttagelser paa Sondmore. — Norges landbrukshøiskoles 
festskrift 1859 — 1909. Kristiania 1909. 

— Flytblokker fra Kristianiatrakten og Danmark paa (ijernnmdnes i Romsdalen. Trond- 
hjems Vidensk. Selsk Skr. 191 1. 

Kerner von Marilaun, A., Pflanzenleben. Leipzig 1888. 

Kjellmark, Knut, Om nâgra jämtländska kalktuffer och blekeförekomster. (ieol. Foren. 

Förh. Stockholm 1904. 
KoLDERUP, C. F., and Monckton, The Geology of the Bergen District, Norway. 191 1. 
KuRCK, C., Om kalktuften vid Benestad. — Kungl. Vet. Akad. Bih. Bd. 26, Afd. II, 1901. 

— .Studier öfver några skånska kalktuffer. Arkiv for kemi, mineral, och geol. Vol. i. 1904. 
Koppen, Fr. Th., Geographische Verbreitung der Holzgevvächse des europäischen Russlands 

und des Kaukasus. Erster Teil. St. Petersburg 1888. Zweiter Teil 1889. 

Malmström, C., Trapa natans L. i Sverige. — Svensk Bot. Tidskr. 1920. 

Meteorologisk Institut, Nedboriagttagelser i Norge. Tillægshefte til aargang XVIII. Kri- 
stiania 19 1 2. 

Nathorst, A. G., Forberedande meddelande om floran i några norrländska kalktufter. Geol. 
Foren. Förh. Bd. 7. 1885. 

— Ytterligare om floran i kalktuffen vid Långsele i Dorotea socken. Ibidem Bd. 8. 1886. 

— Om några arktiska växtlämningar i en sötvattenslera vid Alnarp i .Skåne. Lunds 
Universitets Arsskr. Bd. 7. 1870. 

— Über den gegenwärtigen Standpunkt unserer Kenntnis von dem \'orkommen fossiler 
Glacialpflanzen. Stockholm, Vet. Akad. Bih. Bd. 17. 1891. 

NoRDH.\GEN, Rolf, Planteveksten paa Frooene og nærliggende oer. Trondhjems Videnskaps- 

selsk. Skr. 1917- 
NoRDMANN, V., Danmarks pattedyr i fortiden. — Danmarks Geol. Unders. III. Række. 1905. 

— Meddel. Dansk geol. forening. Bd. 4. 1912 p. 94 — 96. 

— Om inddeling av Nordeuropas kvartær paa grundlag av iagttagelser i Danmark. Fore- 
drag ved 16. skandinaviske naturforskeres mote i Kristiania 191 6. Kristiania 1918. 

NoRM.\N, J. M., Botanisk reise i et strog af kysten mellem Stavanger og Bergen. — Nyt 
Mag. for Naturv. Bd. 8. 1855. 

— Voksesteder for nogle af den norske floras karplanter sondenfor polarkredsen. — Archiv 
f. math, og naturv. Bd. 5. Kristiania 1881. — Yderligere bidrag etc. Ibidem Bd. 8. 1883. 

— Norges Arktiske Flora I — II. Kristiania 1894 — 1901. 

Palmgren, A., Hippophaës rhamnoides auf Aland. — Acta Soc. pro Fauna et Flora Fennica. 
36. No. 3. Helsingfors 191 2. 



1921. No. 9- KALKTUFSTUDIER I GUDBRAXDSDALEX. I5I 

Post, L. vos, Strati graphische Studien über einige Torfmoore in Närke. Geol. Foren. Förh. 
1909. Bd. 31. Stockholm 1910. 

— Norrländska tortmossestudier. I. Drag ur myrarnas utvecklingshistoria inom „lidernas 
region". — Geol. Foren. Förh. Bd. 28. Stockholm 1906. 

— Einige südschwedischen Ouellmoore. Bull, of the Geol. Inst. Upsala 1916. 

— SkogsträdpoUen i s\'dsvenska torvmosselagerföljder. — Foredrag ved 16. skand. natur- 
forskeres mote i Kristiania 1916. Kristiania 191 8. 

— Diskussionsindlæg i anledning av Hesselmans foredrag om pollenregn paa havet. Geol. 
Foren. Förh. 1919 p. 89 ft". 

Printz, Henrik, The vegetation of the Siberian-Mongolian Frontiers. Videnskapsselsk. i 

Trondhjem. 1921. 
Rabexhorst, L., Kryptogamenflora von Deutschland, Österreich und der Schweiz. II. Auflage. 

Leipzig 1906 — 1911. 
Ramann, E., Bodenkunde. III. Auflage. Berlin 191 1. 
Resvoll-Holmsen, Hanna, Gm fjeldvegetationen i det ostenfjeldske Norge. — Archiv tor 

math, og naturv. Bd. 37. Kristiania 1920. 
Resvoll, Thekl.a. R., og Ostenfeld, C. H., Om den ved Aursunden fundne Aster (Aster 

subintegerrimus). — Nyt Mag. for Naturv. Bd. 54. Kristiania igi6. 
Rekstad, J., Bræbevægelsen i Gudbrandsdalen mot slutningen av istiden. — Archiv I. math. 

og naturv. Bd. 17. 1895. 

— Merker efter istiden i det nordlige av Gudbrandsdalen. Archiv f. math, og naturv. 
Bd. 18. Kristiania 1896. 

— Merker efter istiden i Gudbrandsdalen II. Ibidem Bd. 20. 1898. 

— .Skoggrænsens og snelinjens storre hoide tidligere i det sydlige Norge. — Norges 
Geol. Unders. Aarbok 1903. 

— Fjeldstroket Fauske— Junkerdalen. — N. G. U. Aarbok 191 7. 
Reusch, H., Norges geologi. — N. G. U.s skrifter No. 50. 1910. 

— Nogen bemerkninger i anledning av seterne i Østerdalen. — N. G. U.s Aarbok 191 7. 
Samuelsson, g., Studien über die Vegetation bei Finse im inneren Hardanger. — Nyt Mag. 

f. Naturv. Bd. 55. Kristiania 1916. 
Sandegren, R., Den kvartara forskningen i Sverige under de senaste 25 åren. Geol. Foren. 

Förh. Stockholm 1921. 
.ScHETELiG, Jakob, Foredrag om Atnebræen og Jutulhugget. Referert i Norsk Geol Tidsskr. 

Bd. I. No. 13. 1910. 
Schröter, C., Das Pflanzenleben der Alpen. Zürich 1908. 
Schwellengrebel, N., Über Niederländische Dünenpflanzen. — Bot. Centralblatt. Beiheft 

Bd. 18. 1905. 
Schübeler, f. C, Viridarium norvegicum. Bd. I. Christiania 1886. 
Sernander, R., Studier öfver den gotländska vegetationens utvecklingshistoria. — Akademisk 

afhandling. Upsala 1894. 

— Några ord med anledning af Gunnar Andersson, Svenska växtvärldens historia. Bot. 
Notiser 1896. 

— Studier öfver vegetationen i mellersta Skandinaviens fjälltrakter. 2. Fjällväxter i barr- 
skogsregionen. Bih. K. Sv. Vet. Akad. Handl. Bd. 23. 1899. 

— Växtvärlden (Upplandi, Skildring af land och folk, utgifven af Kungl. Humanistiska 
Vetenskapssamfundet i Uppsala genom Axel Erdman och Karl Hildebrand. Bd. I. 
Stockholm 1905. 

— Stipa pennata i Västergötland, en studie öfver den subboreala periodens inflytande på 
den nordiska vegetationens utvecklingshistoria. — Svensk botan. tidskr. Bd. 2. 1908. 

— Pinguicula alpina och P. villosa i Härjedalen. Några synpunkter på den central- 
skandinaviska tjällflorans vandringshistoria. — Sv. bot. Tidskrift. Bd. 4. 19 10. 

— Die Schwedischen Torfmoore als Zeugen postglazialer Klimaschwankungen. — Post- 
glaziale Klimaveränderungen. Stockholm 191 o. 



152 



KOLK NOKDHAGKN. M.-N. Kl, 



Sernanuku, R., Svcnska kalktuircr. (icol. Vnri:n. I'ijrli. Stocklifilm rgis - 1916. 

— Förna ocli Alja. ficol. l'ôv. Förh. rgiS. 

— Siibfossile Flechten. Festskrift til K. Stahl. Flora 1918. 

Servettaz, Camille, Monographic des Elæagnacées. Dresden 1909. Beihlatt Hot. Ctntralblatt. 
.SiEGRiST, R., Die Auenwälder der Aare. Jahrb. Aargaii. Nat. Ges. 1913. 

Smith, Harry, Postglaciala regionförskjiitningar i norra Härjcdalcns och södra Jämtlands 
rjälitrakter. — Geol. Foren. Förh. Bd. 33. Stockholm 191 1. 

— Till kännedomen cm de central-svenska fjällens första flora efter istiden. — .Skogs- 
vardsfören. Tidskrift. Stockholm 191 7. Bil. i. 

— Vegetationen och dess utvecklingshistoria i det centralsvenska högfjällsomradet. — 
Norrländskt Handbibliotek IX. Uppsala 1920. 

SiEENSTRUP, Jai'kius, Gcognostisk-geologisk Undersøgelse af Skovmoserne Vidnesdam og 
Lilleinose i det nordlige Sjelland. — Vid. Selsk. naturv. og math. Afd. IX. Del. 
København 1842. 

Sterner, Rikaru. Avhandling om sydostlige plantearter i .Skandinavien. Under tryk- 
ning. Uppsala 1921. 

Storm, V., Vciledning i Trondhjems omegns flora. Trondhjem 1869. 

Tamm, Olof, Markstudier i det norrsvenska barrskogsområdet. Akad. Avh. Stockholm 1920. 

Tengwall, T. A., De sydliga skandinaviska fjällväxterna och dcras invandringshistoria. — 
Svensk Bot. Tidskr. 1913. Bd. 7. 

— Die Vegetation des Sarekgebietes. — Naturwissenschft. Untersuch, des Sarekgebietes 
in Schwedisch-Lappland. Geleitet von Dr. A. Hamberg. Bd. III. Botanik. Stock- 
holm 1920. 

V.\hl, Martin, De kvartære stepper i Mellemeuropa. — Geografisk Tidsskrift. Koben- 
havn 1902. 

Warming, E., Den danske planteverdens historie efter istiden. Universitetsprogram. Koben- 
havn 1904. 

— Dansk plantevekst II. Klitterne. Kobenhavn 1907— 1909. 

Weber, C. A., Was lehrt der Aufbau der Moore Norddeutschlands über den Wechsel des 
Klimas in postglacialer Zeit? Zeitschr. d. deutsch. Geolog. Gesellsch. Abhandl. Bd. 62. 
Berlin 19 10. 

Werenskiold, Werner, Sondre Fron. — Norges Geol. Unders. No. 60. Kristiania 191 1. 

Wille, N., Norges Planteverden. ~ Salmonsens Konversationsleksikon Bd. X. 

— Om Indvandringen af det arktiske Floraelement til Norge. — N\-t Mag. f. Naturv. 
Bd. 43. 1905. 

— The Flora of Norway and its immigration. — Annals of the Missouri Botanical 
Garden. 1915. Vol. 2. 

— Atragene sibirica L. vildtvoxende i Norge. — Botaniska Notiser. Lund 191 7. 
ØYEN, P. A., Kontinentalglaciation og lokalnedisning. Archiv for math, og naturv. Bd. 21. 

Kristiania 1899. 

— Porilandia nrctica Gray og dens forekomst i vort land under ratiden og indsjø- 
perioden. — Kri.a Vid. -Selsk. Forh. 1903. 

— Nogle bemerkningcr om klimatforandringer. Ibidem 1904. 

— Dryas octopetala L. og Salix reticulata L. i vort land for indsjoperioden. Ibidem 1904. 

— Skjælbanker i Kristianiatrakten. Nyt Mag. f. Naturv. Bd. 44. 1906. 

— Det sydlige Norges „boréale" strandlinje. Kri.a Vid. -Selsk. Forh. 1906. 

— Nye bidrag til bestemmelse av PAo/as-nivaaet. Ibidem 1907 A. 

— Skjælbankestudier i Kristiania omegn. Nyt Mag. f. Naturv. 1907 B. 

— Kvartærgeologiske streiftog omkring den indre del af Bundefjorden. ~ Archiv f. 
math, og naturv. Bd. 30. 1909. 

— Portlandianivaaet ved Skaadalen station. — Kri.a Vid. -Selsk. Forh. 1909. 

— Trivianivaaet ved Svelvik. — Archiv, f. math, og naturv. Bd. 30. 1909. 



1921. No. 9- KALKTUFSTUDIER I GUDBRANDSDALEN. I53 

0YEN, P. A., A brief Summary of the evidence furnished by glacial phenomena and fossiliferous 
deposits in Norway as to latequaternary climate. — Geologkongressen i Stock- 
holm 19 IQ. 

— Nogle bemerkninger om ra-perioden i Norge. — Norsk Geologisk Tidsskrift Bd. II. 191 1. 

— Kvartærstudier i Trondhjemsfeltet I og II. Trondhjems Videnskapsselsk. Skr. 1909 
no. 4 og 1910 no. 9. 

— Meddelelse i Norsk Geol. Tidsskrift 1913. 

— Kvartærstudier i Trondhjemsfeltet III. — Trondhjems Videnskapsselsk. Skr. 1914 Bd. 2. 
Trondhjem 19 15. 

— Istiden. Naturen 1916. 

— Norske kalktuft'orekomster. — Naturen 19 19. 

— Hippophaës rhamnoides L. fra en kalktuf i det svdlige Norge. — .Svensk bot. tidskr. 
1919 Bd. 13. 

— Kalktuf i Norge. — Norsk Geol. Tidsskrift. Bd. V. Kristiania 1920. 



154 



ROI.F NOkDIIA'.r.N. M.-N. Kl. 



TEKST TIL PLANCHENE. 

I. 

Blomst av Dryas octopelala i begyndende frugtstadium. (3 x ). 

og 3. Avtryk og kalkkjerner, vistnok blomsterbiind og basaldelen av Dryasblomster 

(set nedenfra). (4 x ). 

4. Blomsterbund av Dryas med nischer efter smaafriigtene. (3 X ). 

5. Hulrum efter smaafrugter av Dryas. (3 x ). 

6. Blad av Salix glauca. C/l). 

7. Bladfragment av Salix herhacea. (2V2 x ). 
Blad av Salix arhtiscula. C/l). 

Avtryk av undersiden av Marcliaiilia polyiiiorpha (merke efter bukskjællene). (2 x ). 
Kalknirfragmenter efter Characéer. (3 x ). 

II. 

Kurv av Cirsutiu lictcioplivllitiii (med undersiden op og den avbnikne stilk tilhoire). (a X). 
Rakle ($) av Salix arbtiscula med aapncde kapsler. (3 x). 

3. Bladfinne av Sorbits Attciiparia. Cl). 

4. Bladvifte av Tofieldia palustris. (2 x |. 

5. Avtryk av en lavart, vistnok Paruielia pliysodcs. (41/2 x ). 

III. 

Blad av Fragår ia vesca, Leine. (1I/2 x). 

Dryas octopelala og Salix reticulata, Leine. (1V2 x ). 

3. En fuglefjær, Leine. (i''2 X ). 

4. Hippopha'és, Salix caprœa, Betula odorata, Gillebu. C/l). 

5. Forgrenet dvergstamme av Dryas med bladrester, Leine. ('2). 

IV. 

1. Forvitret furutufstykke fra profil XXI, Leine, med avtr\'k efter tre kongler av Pinus 
silvestris. C /2). 

2. Bladtuf med anrikning av blader av Salix efr. nigricans, Leine. (1/2). 



V. 

Alnus-tuf fra Leine. Bladavtryk og kvister av Aluns iiicaiia. Det overste st\-kke '/2, 
de nederste '/l. 



T92I.No. g. KALKTUFSTLDIER I GUDBRAXDSDALEN. 



RETTELSER OG TILFØIELSER. 

Side 46, linje 8 ovenfra : Schroeter, læs Schröter. 
„ 84, „ 22 ovenfra : M. tciicrriiua iitgaar. 
„ 84, „ 24 ovenfra ; H. Bambergi, læs H. Baiitbergcri. 
„ 84, note I : Olivcorona, læs Olivecrona. 
„ 96, linje 3 ovenfra : leret, læs leren. 

„ 84. Da forfatteren har læst korrektur paa avhandlingen under en reise i Mellem- 
europa og for længere tid ad gangen har været avskaaret fra videnskabelig litteratur, er 
desværre nomenklaturen for de i avhandlingen omtalte moser helt negligeret. Da Dr. Hj. 
Möller, som har bestemt mine prover, tildels anvender en nomenklatur som avviker meget 
fra den mellemeuropæiske, anføres nedenfor autorsnavn og s\'nonymer for de anforte mosarter :. 
Ainblystegiuiit filiciiiitiit De Not. = Cratoiicuroii filiciiutm (L.) Roth. 

— proteiisitiii LiNDB. = Caiiipyliitiii frotciisitm (Brid.i Kixdb. 

Dilricliiiiii fle.vicaule (Schleich.) H.^.mpe. 
Leucodon sciitroidcs (L.) Schwgr. 
Mollia æntgiiiosa Lindb. = Gyiiiiios/oiuuiii ntpestre Schleich. 

— tortnosa .ScHR.\NK = Tortella tortitosa iL.I Li.mpr. 
Mviirelld npiciilata ( Hüben 1 Br. eur. 

— jitlacea iVii.L.I Br. eur. 

Hypiiiiiii Bainhergeri Schpr. = Stereodon Bambergeri Lindb. 

— ser ia II ni L. = Honialothecitini sericeiint (L.) Br. eur. 
Sii'artsia niontana Lindb. — Disticliiiini capillacctiin (.Sw.i Br. eur. < 
Leersia cnntorta Lindb. = Encalvpta contortn iWulf. I Lindb. 



Trykt den 9. november 1921. 



Vid.-Selsk. Skr. I. M.-\. Kl. 192 1. No. 9. 



Planche I. 



I 



•v^ 



î^ 



^ 




■if 

,* 

M 



A- 






KÆ 
%^^ 



y 



» 



// 



il 



et 



10 



R. Xordlictgiti (hliii. 



Vid.-Selsk. Skr. I. M.-N. Kl. 1921. No. 9. 



Planche II. 



■% 




% 



^.-r 






-C^'- 



\ ,.</ 



A'. Xordliaircii ilcliii. 



Vid.-Selsk. Skr. I. M.-X. KI. 1921. No. 9. 



Planche III. 







cT 



B. Larssen /of. 



Vid.-Selsk. Skr. I. M.-N. Kl. 1921. No. 9. 



Planche I\' 



^<^r^S >■V■ 





B. Larssen fot. 



Vid.-Selsk. Skr. I. M.-X. Kl. 1921.X0.9. 



Planche V 








ÄOL'IVALEXTE DER l'XTERSILURISCHEN 
EULOMA-NIOBEFAUNA BEI PLZENEC 

IN BÖHMEN 



VON 
J. V. ZELIZKO 



MIT 5 TAFELN, I KARTENSKIZZE UND I TEXTABBILDUNG 



(ViDENSKAPSSELSKAPETS SkRIFTER. I. MaT.-NATURV. KlASSE. 192t. No. lo) 



UTGIT FOR FRIDTJOF NANSENS FOND 



KRISTIANIA 

IN KOMMISSION BEI JACOB DYBWAD 
1921 



Fremlagt i den mat.-naturv. klasses mote den sdje juni 1921 av prof. Johan Kiær. 



A.W.BRØGGERS BOKTRYKKERI A /S 



Vorwort. 



A, 



ils mir im Jahre 1908 Herr Prof. C. v. Purkyne eine größere, be- 
kanntlich von ihm für die geologisch-paläontologischen Sammlungen des 
städtischen historischen Museum in Pilsen erworbene Menge untersiluri- 
scher Fossilien der Stufe d i ;' aus der sogen. Cenid strdii (= Schwarze 
Lehne) oder Harka'^ an dem rechten Ufer des Uslavaflusses bei Plzenec 
(Pilsenetz), zur wissenschafdichen Bearbeitung anvertraute, ahnte ich nicht, 
dafe die betreffende Gegend später ein willkommener Gegenstand meiner 
einige Jahre dauernden Forschungen sein wird. 

Im Jahre 1910 besuchte ich Hurka zum ersten Male, wo ich teilweise 
für das historische Museum in Pilsen, teilweise für die geologische Reichs- 
anstalt weitere, gröfatenteils neue Fossilarten aufsammelte, dann setzte ich 
auch im Jahre 191 1 meine Forschung fort, sodafa die Anzahl der Fossilien 
wiederum stieg, wie auch aus dem tabellarischen Verzeichnis Purkyne's-, 
wo einige von mir nachträglich festgestellten Arten angeführt sind, ersicht- 
hch ist. 

Außerdem bleibt noch eine Reihe zur Bearbeitung vorbereiteter Ver- 
steinerungen wie: Orthoceren, Pteropoden, Gastropoden und Lamellibran- 
cliiatoi 

Schliefalich veröffentlichte auch C. Kloucek einen Beitrag zur Kenntnis 
der Fauna der Stufe d i ^' von Hurka ^. 

Aus dem oben angeführten geht am besten hervor, dafa diese meistens 
schwer zugängliche steile Lehne viel besser durchgeforscht war als man 
eigentlich annimmt"*. 



* I. V. Zeli'zko. Faunistische Verhältnisse der unter sihirischen Schichten bei Pilsenetz in 
Böhmen. Verhandl. d. k. k. geol. Reichsanstalt. Nr. 3. Wien 1909. Zpräva o zvirene 
spodniho sihiru od Plzence. Sbornik historického musea v Plzni. Jg. I. Pilsen 1909. — 
Zur Frage über die Stellung der Hyolithen in der Paläontologie. Centralblatt für Min., 
Geol. u. Pal. N. 12. Stuttgart 1908. — Zwei neue Conularien aus dem älteren Paläo- 
soicuni von Böhmen. Ibid. 1913. 

■' C. Rytjr Purkyne. Geologie okresu Plsenského. S. 53 — 57. Pilsen 1913. 

C. Kloucek. O geologickém horisoyttu rudnilio lo:iska na Karyzku. Rozpravy Ceské 
Akademie. Ig. XXII. N. 9. Prag 1913. 

* Derselbe. O vrstvàch D i y, jich trilobitech a nalesistich. Ibid. Ig. XXV\ N. 39. 1916. 



J. V. ZKI.IZKO. 



M.-N. Kl. 



Vor Pltrkyne war die Lokalität faunistisch sehr wenig durchgeforscht, 
denn das Museum der k. k. geologischen Reichsanstalt hat von da nur drei 
von LiPOLD mitgebrachten Fossilarten \Stroplioincnn primula, Palcru'a hohe- 
inica, und Lingnla sp). 




,Sr(lIco 



Sciacfer 



.ciscliiefer J -S 



■^yj^^^z^ ^ahfûaner Sckchlm/ f Oseh-lfvmr Sanerer) 



Ur3fiver Qnarzite dz 

fossfienïnJirenae Sân:ef(X djß 
ØeciiwaMte dirldoma, -JfaMmsJ, 
iüJu^Mlm^'se cUsjirajelrmibriscJi' oder 
mkrsilmsch/verzeiclmeP, tcüweise 
onl/^canTit. 






^ Imjofimdtntäerlossib'en/ 
* Innioit derlTocliiojioim/ 

Lierai se 



lull ÆleriTose 



) ^ 



ßehäjißehmju. Sand 

Sand, Sdiotto' nJéhn aerûeffmoart 



Geologische Kartenskizze der Gegend von Plzencc. 

I : 30 000. Nach C. Piirkyne, ergänzt v. J. V. Zelizko. 

I. Krejci und K. FeisTiMantel^ die sich auch mit den geologischen 
Verhältnissen der Umgebung von PIzenec befafsten, sprechen nirgends von 
Versteinerungen. 



^ Orograpliisch-geoteklonische Übersicht des sihtrischen Gebietes im mittleren Bö/iiiieii. Archiv 
der naturwiss. Durchforschung von Böhmen. Bd. V. No. 5. Prag 1890. 



1921. No. I o. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. 5 

Als ich nach Beendigung meiner Forschung auf der Hurka im Jahre 
191 2 bei Plzenec zufäUig Schiefer der Komorauer Stufe (d i ß) mit über- 
raschender, meistens bisher unbekannter Fauna \ welche der Hauptgegen- 
stand der vorliegenden Arbeit ist, entdeckte, besuchte ich anläßlich der 
Durchforschung neuer Lokalitäten Plzenec noch im Jahre 1913 und 1915. 

Dafà ich die Forschung mit solchem Erfolg zustande brachte, wie 
später gezeigt wird, dafür bin ich in erster Reihe dem löbl. Verwaltungs- 
ausscliusse des Museums des Königreichs Böhmen, welcher mir zu diesem 
Zwecke im Jahre 1913 eine Subvention aus dem Barrande'schen Fonde 
gewährte, v^erpflichtet. 

Die Schloenbachstiftung der k. k. geol. RcichscDistall ermöglichte mir 
dann im Jahre 1914 eine Studienreise nach Skandinavien zur Besichtigung 
dortiger altpaläozoischer Ablagerungen, sowie der diesbezüglichen, zum 
Vergleich der von mir bei Plzenec festgestellten Fauna wichtigen Samm- 
lungen. 

Während meiner Anwesenheit in Plzenec war Herr Ing. Rud. Cisar 
stets ein opferwilliger Teilnehmer meiner Arbeiten, wofür ich ich ihm an 
dieser Stelle herzlich danke. 

Gleichfalls danke ich dem p. t. Goncindcamte in Plzenec für die Erlaubnis 
zur Durchforschuug der der Gemeinde gehörenden Lokalität »U Blazeje«. 

Ferner ist es mir eine angenehme Pflicht, Herrn Prof. C. Ritter von 
PuRKY.N'E für die allseitige, meinen Forschungen im westlichen Böhmen 
jahrelang gewidmete Unterstützung zu danken. 

Schliefslich bin ich Herrn Prof. Joh. Kiær in Kristiania, welcher sich 
meiner Publikation freundlichst annahm und dieselbe zur Veröffentlichung 
in »V'idenskapsselskapets Skrifter« empfahl, besonders zum innigsten Dank 
verpflichtet. 



^ I. V. Zeli'zko. Neuer Beitrag sur Geologie der Gegend von Pilseneis in Böhmen. Ver- 
handlung der k. k. geolog^. Reichsanstalt. No. 5. 1913. 



J. V. ZKi.izKo. M.-\. Kl. 



Schiefer der Bande D — d 1 /^ 

deren Liegendes und Hangendes. 

a) Die Schiefer >U Blazeje«. 

Westlich von Plzenec, an dem linken Ufer der Uslava, zieht sich eine 
aus der Ferne gut sichtbare Lehne, genannt »U Blazeje«, welche mit einer 
ziemlich hohen felsigen, nordwestlich streichenden Böschung endet. 

In einer Schlucht der betreffenden Lehne, an dem rechten Ufer eines 
unweit des Eisenbahnwächterhauses No. 269 flieisenden Bächleins treten 
dünngeschichtete, verwitterte Schiefer zu Tage, die bei flüchtiger Besich- 
tigung an die Schiefer der Osek-Kväner Stufe (d i y) erinnern, welche be- 
kanntlich auf der gegenüberliegenden Hurka vertreten sind und scheinbar 
auch auf das linke Uslavaufer übergehen. 

Diese Schichten, welche teilweise von Diluvialablagerungen bedeckt 
sind, hat Herr Prof. Purkyne in seiner Karte ^ ursprünglich als der Osek- 
Kväner Stufe angehörend verzeichnet, später jedoch ^ die Ansicht ausge- 
sprochen, daf3 dieselben wahrscheinlich dem Eulomahorizonte der Stufe 
d I /j zuzurechnen sind. 

Bei näherer Untersuchung der besprochenen Schiefer fand ich, dafs 
der petrographische Charakter derselben ein anderer ist, als die typischen 
schwarzen Schiefer von Hurka aufweisen. 

Das Gestein besteht im frischen Zustande aus graugrünem, feinkörni- 
gem, .glimmerigem und festem Schiefer, welcher verwittert eine lichte, 
graue oder gelblichbraune Farbe aufweist. Die Flächen der Schieferplatten 
sind häufig von Eisenhydrooxyd verschiedenartig gefärbt. 

Genannter Schiefer erinnerte mich schon auf den ersten Blick an einen 
ähnlichen, den ich im J. 1906 bei Klabava, zwischen Pilsen und Rokycan 
sah, von wo ein Jahr früher Herr Prof. B. Horak einige neue Versteine- 
rungen für das städtische Museum in Rokycan erworben hat, und welche 
mir zur späteren Bearbeitung zugesandt waren. Als dieses Material bald 



^ Geologickä mapa zastupitelského okresit Plsenskeho, r : 30 000. Pilsen 1910. 
2 Geologie okresu Plsenskcho. S. 52. Pilsen 1913. 



I92I. No. 10. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. 7 

darauf durch die Aufsammlungen Holubs bedeutend vermehrt wurde, habe 
ich auf Ansuchen des Herrn Holub. welcher die Zugehörigkeit der Funde 
zu einem für Böhmen neuen Eulomahorizonte nachgewiesen hat^, ihm den 
betreffenden Teil zur Bearbeitung überlassen. 

An der Exkursion nach Klabava im J. 1906 habe ich mit Herrn 
Prof HoRAK und Purkyne ausschliefelich zur Untersuchung der geologischen 
Verhältnisse der Lokalität teilgenommen, denn zum Suchen nach Fossilien 
war damals die nötige Zeit nicht vorhanden, infolgedessen konnte man 
natürlich von diesem Ausflug keinen besseren Erfolg erwarten ^. 

Die auffallende petrographische Übereinstimmung des Schiefers von 
Klabava mit dem des Fundortes »U B!azeje'< bei Plzenec veranlafate mich 
zu emsigerem Suchen nach \'ersteinerungen, was tatsächlich nicht ohne 
Erfolg blieb. Auf diese Weise wurde neuerdings bestätigt, dafe es nicht 
immer richtig ist, den petrographischen Charakter eines Gesteines zu unter- 
schätzen, welches nach dem oberflächlichen Ansehen nicht fossilführend zu 
sein scheint. 

Nach längerem geduldigem Suchen im neuen Fundorte gelang es mir 
endlich überzeugende Beweise zu gewinnen, dafe die hier zutage tretenden 
Schichten auch faunistisch mit jenen von Klabava übereinstimmen, d. h. 
daß sie eine Fazies der Stufe diß bildeji. 

In teilweise festem, teilweise halbverwittertem, tonigem Material wurden 
neben undeutlichen Graptolithen auch mehrere Stücke von Brachiopoden, 
die den bereits von Barrande angeführten Lingidide)i, und zwar der Lin- 
gula sulcata^ und der Lingula rugosa^ und anderen verwandten Formen 
angehören. 

Barrande bezeichnete als Fundort beider Arten Klabava (»Vallon de 
la Klabava — d i — au nord de Rokitzan«). 

Außer diesen Versteinerungen kommen im neuen Fundorte bei Plzenec 
auch bekannte enge, den Orthoceren ähnlich gegliederte Bildungen, und 
undeutliche Trilobitenreste vor. 

Noch besser aufgeschlossen findet man fossilienführende Schiefer einige 
Schritie nordwesdich von der obenerwähnten Stelle, nämlich hinter dem 
Kreuz »U Bl^zeje«, wo die Lehne in einer ziemlich steilen Böschung ab- 
lauft. Da an dieser Stelle das Gestein öfters zum Schottern ausgebrochen 
wurde, sind die besprochenen Schichten überall gut zugänglich. 



1 K Holub. Nova fauna spodniho silurii v okoli Rokycan. Rozpravy Ceské Akademie. 

Jg. XX. Nr. 15. Prag 191 1. — Dophiky ku faune eitlomového horisontu v okoli Rokycan. 

Ibid., Ig. XXI, Nr. 33, 1912. 
^ K. HoLTjB. Nova fauna etc.. pag. i. 
^ Systeme Silurien. Vol. V, PI. 106, Fig. HI (-2, 31. 
* Ibid. PI. 152, Fig. V (2, 31. 



8 J. V. zKi.izKo. M.-N. Kl. 

Dieser Teil der Lehne wurde vom Herrn Prof. Purkyne als präkam- 
brische Schiefer aufgenommen. 

Auf der alten, handkolorierten Karte der k k. geol. Reichsanstalt 
(Z. 7 C. IX. Pilsen und BUjwitz) ist drs an dem linken Ufer des üslava- 
flusses liegende Schichtenkomplex überhaupt als präkambrischer (Pribramer) 
Schiefer B bezeichnet. 

Das Gestein in dem nordwestlichen Zipfel der Lehne »U Blazeje« 
bildet nordwestlich zwischen 30—40" einfallende, feste und unregelmäöig 
mächtige Bänke, die besonders gegen NW allmählich in ganz dünne blätt- 
rige Schichtchen übergehen. 

Auch an dieser Stelle glückte es mir nach längerem Suchen einige 
schön erhaltene Versteinerungen im gelbbraunen halbverwitterten, sowie 
im festen, graugrünen Schiefer zu gewinnen. Während meiner dreijährigen 
Forschung fand ich hier zahlreiche Bradnopodoi, einige Orthoceren, Ptero- 
poden, namentlich aber schön erhaltene GraptoliÜien in Form von bis über 
40 cm. langen Büscheln und Ruten (Taf V), mit deren Bearbeitung sich 
soeben Herr Dr. Perner beschäftigt. Am besten erhaltene Exemplare 
haben grofee graugrüne Schieferplatten geliefert. Schliefàlich sollen noch 
einige neue Trilobiten als Seltenheit erwähnt werden. 

Die Schichten der Stufe d i /!? an dem linken Uslavaufer sind von 
Diluvialablagerungen der III. Terrasse Purkyne's bedeckt. Dieselben kann 
man von dem Hauptfundorte »U Blazeje« auch noch gegen WSW, sowie 
gegen O, wo sie quer über den Fahrweg etwa in halber Entfernung 
zwischen der Stelle »U Blazeje« und Plzenec an einigen Punkten zum Vor- 
schein kommen, verfolgen. 

b) Die Schiefer in Plzenec. 

Dieselben untersilurischen Schiefer der Stufe d i /î? wie in der Lehne 
»U Blazeje« kommen auch zn einigen Stellen in Plzenec, wie z. B. hinter 
dem Brauhaus und südöstlich von diesem auf der Stelle »U krizku« ( -= beim 
Kreuzchen) noch an dem linken Uslavaufer vor. Das Gestein ist aber 
größtenteils verwittert und an der Fläche rostig gefärbt. 

Das Liegende der oberwähnten untersilurischen Schiefer des linken 
Uslavaufers bilden die präkambrischen Schiefer, welche überhaupt südöstlich 
von Plzenec einen ausgedehnten, häufig von Lydit- und Spilitstreifen und 
Inseln durchtretenden Komplex einnehmen. 

Sie sind entweder von Diluvial- oder Alluvialablagerungen bedeckt, 
und da sie leicht verwittern, verwandeln sie sich oft in mächtige eluviale 
Tonschichten. Ein solches Beispiel liefert am besten die Umgegend von 
Spâlené Porici (Brennporitschen) südöstlich von Plzenec, wo man Schiefer 



I92I.N0. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. 9 

in eine beträchtliche Tiefe so stark zersetzt fand, sodafâ es oft schwer ist 
das ursprüngliche Gestein durch einige feste Bruchstücke nachzuweisen^. 
Das eigentümhche Aussehen dieser tonigen Schichten lockte vor einigen 
Jahren verschiedene Privatunternehmer zur Schürfung nach Kohle in un- 
mittelbarer Nähe von Spålené Porici, was, wie sich denken läßt, erfolglos 
blieb. Die mir damals zugesandten ca. 26 m. in die Tiefe reichenden 
Bohrproben zeigen wenigstens in wie weit der Zersetzungsprozefà der 
Schiefer vorgeschritten ist, wie folgt: 

1. Probe. Vollkommen zersetzter, in grauen, sandigen Ton verwandelter 

Schiefer bis 15 m. 

2. — Derselbe von einer dunkleren Farbe » 16 » 

3. — Derselbe von schwarzgrauer Farbe » 17 » 

4. — Grauer, sandiger Schiefer » 18 » 

5. — Schuppen von schwarzen glänzenden Schiefer mit 

Quarzkörnern » 19 » 

6. — Dunklere Schiefermasse mit größeren Schieferbruch- 

stücken » 20 > 

7. — Dieselbe mit kleinen Quarzkörnern » 21 » 

8. — Gemisch aus Bruchstücken des schwarzen Schiefers, 

Grünsteines und Quarzes » 22 > 

9. — Schwarzer Schiefer mit Quarz und vielem Pyrit . » 23 » 

10. — Bruchstücke des festen schwarzen Schiefers ... » 24 » 

11. — Grauschwarzer Schiefer » 25 » 

12. — Fester, schwarzer, glänzender Schiefer .... » 26 » 

c) Die Schiefer bei Sedlec. 

Dafe man Schiefer der Stufe d i rj auch an dem rechten Uslavaufer 
verfolgen kann, davon kann man sich bei Sedlec, einem ca. 1.5 km. östlich 
von Plzenec entfernten Dorfe überzeugen. 

Nördlich von dem genannten Dorfe, beim Kreuz auf der linken Seite 
der nach Timâkov führenden Strafae, sowie an einigen anderen Punkten 
in Sedlec selbst, kommen meistens verwitterte, dünngeschichtete gelbbraune, 
nw. bis nww. unter einem 35—40*^ einfallende Schiefer zum Vorschein. 
Da dieselben in einer ziemlichen Tiefe zersetzt sind, konnte ich darin bloè 
undeutliche Graptolitlienspuren und Bruchstücke von einer Lingula, und 
einigen länglichen Konkretionen finden. 

Die besprochenen Schiefer der Stufe d i ß bilden die Unterlage der 
konkordant liegenden schwarzen Schiefer der Stufe d i y von Hurka. 



1 F. Katzer. Geologie von Böhmen, pag. 636 — 637. 



lo J. V. zEiJzKo. M.-N, Kl. 

Die genaue Grenze zwischen beiden Stufen festzustellen, verhindern die 
die Schiefer bedeckenden Diluvialablagerungen. 

Ks scheint, dafà die in den Schluchten des südöstlichen Teiles von 
Hurka, näher gegen Sedlec, später gefundenen Versteinerungen wahr- 
scheinlich jenen Schichten angehören, auf welchen unmittelbar Schiefer der 
Stufe d I ji? ruhen. Die fossilienführenden Schiefer des südöstlichen Teiles 
von Hurka sind stark verwittert und zerfallen in dünne, durch Eisenoxyd 
rostig gefärbte Blätter und Plättchen. Die Farbe derselben ist meistens 
schmutziggrau, im Gegensatz zu den dunklen und festen Schiefern aus der 
höheren Lage der Hurka. 

Wie es der Erhaltungszustand der Versteinerungen zuläfat, habe ich 
folgende Arten bestimmt: 

I. Trilohiten. 
Placoparia Zippci Boeck. sp. 

Einige kleine, vollkommen erhaltene Exemplare, isolierte Kopfschilder 
und häufige Thoraxteile. 

Das Vorkommen dieses Trilobites ist um so merkwürdiger, da derselbe 
in der zwar reichen Fauna von Plzenec^ in keiner der höheren Lagen 
dl;' bisher nachgewiesen wurde. 

Aeglina rcdiviva Barr. 
Ein größeres und ein kleineres Pygidium. 

Aeglina cf. sulcata Barr. 
Ein Abdruck und der Gegenabdruck eines kleineren Pygidiums. 

II. Cephalopoden. 

Orthoceras expcctans Barr. 
Einige Bruchstücke. 

III. Brachiopoden. 

Strophomena primula Barr. 
Einige Exemplare. 



PateruJa bohcuiica Barr. 



Zwei Exemplare. 

1 1. c. I, !. c. 2 pag. 3. 



I92T.No. lO. ÄQUIVALENTE DER UNTERSILURiSCHEN EULOMA-NIOBEFAUNA. II 

IV. Gastropoden. 

Plenrotoniaria (LopliospiraJ viator Barr. 
Einige Exemplare. 

Plcurotomaria (Lophospua) nov. sp. 

Ein negativer Abdrucl^ eines etwas gröfseren Exemplares als ich im 
J, 1909 anführte {Faiiinstische Vcrliälttüsse der witersiliirisclicii Schichlcn 
bei Pilsenetz in Böhmen. Verhandl. d. k. k. geolog. Reichsanst. No. 3). 

Hclicotonia ? nov. sp. 

Ein Abdruck und der Gegenabdruck mit erhaltener feiner Schalen- 
skulptur. 

Aufserdem noch einige Reste wahrscheinlich verschiedenen Pleuroto- 
marien angehörend. 

V. Pteropoden. 

HyolitJnis sp. 

Ein Abdruck und Gegenabdruck eines seitlich geprefsten, längs der 
Schale mit feinen Rippen verzierten Exemplares. 

Coniilaria exqnisita Barr. 

Einige kleine Bruchstücke und ein Fragment eines gröfaeren Exemplars 
mit gut erhaltener Skulptur. 

VI. La mein bra nchiaten. 

Leda boheniica Barr. 

Ein Bruchstück einer Doppelschale und ein kleines, vollkommen erhal- 
tenes Exemplar. 

VII. Crinoiden. 
Einige undeutliche Sten^elreste. 



Die oben beschriebene Fauna liefernden Schichten gehören wahr- 
scheinlich ein und demselben Horizonte d i / der nördlich liegenden Lokali- 
täten bei Ejpovic ^ an, wo v,-ie bekannt Placoparia Zippei als häufigstes 
Fossil vorkommt. 



' I. V. Zelizko. Weitere neue Beiträge sur Keim/nis der Faima des böhmischen Untersilurs. 
Verhandl. d. k. k. geolog. Reichsanst. No. 2. 1902. — Zur Paläontologie der unter- 
silurischen Schichten in der Gegend sivischen Pilsen und Rokycen. Ibid. No. 16. 1908. 



12 J. V. zEi izKo. M.-N. Kl. 

Das Liegende der Schiefer der Siufe d r ß bilden bei Sedlec die auf 
den präkambrischen Schiefern ruhenden Krusnähoraschichten (dia). 

Zum Schlufi möchte ich noch bemerken, dafj die Verbreitung der 
Schiefer d i ß in der Gegend von Pizenec viel grö6er ist als in dieser 
Publikation behandelt wird, dafür sprechen meine neuerlichen Forschungen, 
über welche ich anderorts berichten werde. 



Paläontologie der Bande D — d 1 ß. 

Tri /oh /ten. 

Asaplicllus bohcmicus Zel. 
(Taf. I, Fig. I, Taf II. Fig. i, 2.) 

Es liegen zwei große, fast vollständig erhaltene Exemplare vor, wo- 
runter ein im positiven sowie im negativen Abdruck vorhanden ist. 

Zu derselben Art gehört auch ein Gegenabdruck eines Rumpfes mit 
Pygidium und ohne Kopfschild. Alle Stücke sind durch Schichtendruck 
einigermafaen geprefat, sodafa manche Merkmale schwer nachweisbar sind, 
dessen ungeachtet läfat sich doch ihre Gattungszugehörigkeit bestimmen. 

Das erste vollständige Exemplar (Taf. 1) ist 12 cm. lang. Das halb- 
kreisförmige, mit kurzen Wangenstacheln endende Kopfschild ist mit 
einem sehr breiten, flachen und glatten Randsaum umgeben, und fast 
doppelt so breit wie hoch, denn seine Breite beträgt 83.5 mm. und die 
Höhe 42 mm. Die Grenze zwischen den Wangen und Glabella ist schwach 
angedeutet. Die halbmondförmigen, ursprünglich erhobenen, ziemlich grofsen 
Augen sind abgewetzt. 

Die Glabella ist leider gröfatenteils abgeschält, sodafa sich ihre gleich- 
mäfaig breite, oben abgerundete Form nur nach dem Umrifa feststellen läfat. 
Der Abstand zwischen der Stirn und dem Oberrand ist ziemlich beträchtlich. 

Der aus acht Segmenten bestehende Thorax samt Pleuren stimmen in 
ihrer Form mit denen bei verschiedenen Asaphiden-Gattungen überein. 

Das Pygidium ist halbelliptisch, mit sehr breitem und glattem Rand- 
saum umgeben, 69 mm. breit und 43 mm. hoch. Die beschädigte Axe ist 
lang, ziemlich schmal und erstreckt sich kaum in zwei Drittel des Pygi- 
diums. Die ursprüngliche Segmentierung der Axe und der Seitenlappen 
ist verwischt, nur bei dem zweiten, fast genau so grofaem Exemplare (Taf II) 
ist dieselbe sehr schwach angedeutet. Bei diesem aus einem Abdruck und 
einem Gegenabdruck bestehenden Asaphellus nimmt die lange, schmale 
Axe mehr als ein Fünftel der Gesamtbreite des Pygidiums ein. 



I92I. No. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. I3 

Das im negativen Abdrucl: erhaltene PN-gidium samt den aus fünf 
Segmenten und Pleuren bestehenden Thoraxteilen des bereits oben ange- 
führten dritten Exemplares, weist gieichfalls dieselbe Form und Gröfee auf, 
wie die zwei vorher beschriebenen Stücke. 

Von dem Kopfschild der beiden Abdrücke des zweiten Exemplares 
hat sich bei dem negativen Abdruck (Taf. II, Fig. i| nur ein Teil der linken 
Wange mit dem Stachel und ein undeutlicher Abdruck der Innenseite vom 
HN'postom erhalten, die übrigen Teile waren abgelöst. 

Bei dem positiven Abdruck (Taf. II, Fig. 2) sind beide Wangen und 
ein fast tadelloses Hypostom vorhanden, bekanntlich eines der wichtigsten 
Merkmale zur genauen Bestimmung dieses Trilobites. Das in situ befind- 
liche Hypostom ist 25.5 mm. lang, hat einen eiförmigen, gegen den Ober- 
rand verschmälerten Umri6 und ziemlich breite Randsäume. 

Die eigentliche Einbuchtung in der Mitte des Hinterrandes scheint 
vorhanden zu sein, jedoch ist die genannte Stelle zufällig so mangelhaft 
erhalten, dafs eine präcise Feststellung schwer durchführbar ist. Nach 
allem aber war die Einbuchtung nicht zu groß. Die übrigen Merkmale 
lassen sich aus unserem Bilde, wo der L'mriB mit weiter Farbe ange- 
deutet ist, am besten erkennen. 

Der Form nach erinnert unser Hypostom am meisten an den von 
Asaphcllus Homfray var. aus dem Untersilur (Tremadoc) von Cape Breton 
in Britisch- Amerika, wo derselbe nach Matthew ' in den sogen. Bretonian 
{c2-Asapliel!iis-Zor\e) vorkommt. 

Das mehr als um ein Drittel kleinere Hypostom der amerikanischen 
Art weist einen fast kreisförmigen Umrifs und ziemlich breite Randsäume 
auf, die in der Mitte des Unterrandes eine mäfeige Einbuchtung bilden. 
Der von ^Matthew abgebildete voUständie Asaphellus Homfray var. (1. c. 
1902 und 1903, PI. X\'III. Fig. 10 a), welcher mehr als um eine Hälfte 
kleiner ist als der von Plzenec, ähnelt diesem insofern, dafä sein Kopf- 
schild und das Pygidium die gleiche Form besitzen, während die Rand- 
säume beider Körperteile schmäler sind, infolgedessen die Stirn bedeutend 
mehr zum Oberrand des Kopfschildes und die schwach segmentierte Axe 
mehr zum Hinterrand des Pygidiums geschoben sind. 

Die Glabella ist ebenfalls fast gleichmäßig breit, hat eine abgerundete 
Stirn und das Kopfschild mäfsig lange Wangenstacheln. 

Das Hypostom des Asaphcllus Homfray aus den sogen. ■!>Shineton shales«- 
von Shropshire in IValcs, dessen oberer Teil faunistisch dem norwegischen 



^ G. F. Matthew. Additional notes on the Cambrian of Cape Breton. Bulletin of the 
Natural History Society of New Brunswick, Canada, No. XX, Vol. iv, Part v. —Report 
on the Cambrian Rocks of Cape Breton. Geological Survey of Canada, 1903. 



14 J V /jA.i/.Ko. M.-N. Kl. 

Bu/oma-Nto/u'-l luv i/jjidc ciiispriclii, hat dieselbe Form wie der gleich- 
namige amerikanische AsaplicllnsK 

Als ßegleitfauna des amerikanischen Asaphellus ist unter anderen 
Paral)oli)u'lla, Triartlirns, Bcllcrophoii angeführt. 

Der von Salter beschriebene typische Asaphellus Homfray Salt,^ 
aus dem englischen Untersilur (Upper Tremadoc) von Wales stimmt nur 
teilweise mit dem von Flzenec überein. 

Das Hypostom des Asaphellus bohemicus erinnert auch an einige von 
Brøgger bei den skandinavischen Megalaspiden beobachteten Ilypostome^, 
welche durch ihre Form den Asaphellus-W^'ç>o'=A.ovazx\ ähneln, eine zur Be- 
stimmung der Untergattung des Asaphellus mafagebende Charakteristik, wie 
auch beim Asaphellus bohemicus neuerdings bestätigt wurde. 

Was die Stellung der Asaphiden [Asaphus, Ogygia, Niobe) anbelangt, 
soll hier vor allem auf die Studie Brøggers -»Über die Verbreitung der 
Euloma- Niobe- Fauna {der Ceratopygenkalkfauna) in Europa<i^ hingewiesen 
werden. Die neueste amerikanische Literatur Raymonds zur genaueren 
Definition der Asaphiden ist mir leider infolge des Weltkrieges nicht ein- 
gelangt. 

An unseren Asaphellus hohemicus erinnert ferner einigermaßen der 
mit breitwangigem Kopfschild versehene riesige Asaphellus glabratus Salt., 
welchen Frech aus dem ^5<7/)////5-Schiefer (Mittleres Untersilur) von Cabri'eres 
in Frankreich abbildete''', der aber einen ausgesprochenen Asaphellus- 
Habitus aufweist. Da in demselben Schiefer (vom Alter des Llandeilo- 
Flags) auch Placoparia Tourtwmijiei Rouault vorkommt "^, welche der 
böhmischen Placoparia Zippei nahe steht, entspricht der betreffende Horizont 
der böhmigen Stufe d i ^. 

Der von Frech aus zwei verschiedenen Exemplaren kombinierte Asaphus 
glabratus ist durch kurze Wangenstachel und deutlich segmentiertes Pygi- 
dium gekennzeichnet. 



^ Ch. Callaway. On a new Area of Upper Cauibrian Rocks in South Shropshire, wilh 
a Description of a new Fauna. The Quarterly Journal. S. 663. PI. XXIV. Fig. i 
London 1877. 

2 I.W.Salter. A Monograph of the British Trilobitcs. S. 165-166. PI. XXIV, Fig. 
6 — 12. London 1864 — 1883. 

3 W. C. Bkøgger. Ufiber die Ansbilditng des Hypostomes bei einigen skandinavischen Asa- 
phiden. Sveriges geologiska Undersöltning. Afhandlingar och Uppsatser. Ser. C. 
No. 82. K. Svenska Vet. Akad. Handhngar. Bd. 11. No. 3. Stockholm 1886. 

* Nyt Magazin for Naturvidenskaberne. Grundlagt af Den Physiographiskc Forening in 

Christiania. Bd. 36. 1897 — 1898. 
5 F. Frech. Ueber die Entwickelung der silurischen Sedimente in Böhmen und im Süd- 

ivesten Europas. N. Jahrb. f. Min., Geo), u. Pal. Jg. 1899. Bd. II. 
^ F. Frech. Die palaeozoischen Bildungen von Cabri'eres (Languedoc). Zeitschr. d. deutsch. 

geolog. Gesellschaft. S. 395. Berlin 1887. 



I92I.N0. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-MOBEFAUNA. I5 

Von den bisher bekannten AsapJiidcn aus dem böhmischen Untersilur 
entspricht überhaupt keiner dem Asapliclhis bohémiens von Plzenec. 

AsapJiellus iiisignis Zel. 
(Taf. Ill, P^ig. 2.) 

Eine 46 mm. lange Form, welche durch einen sehr breiten gleich- 
mäfaigen Randsaum des halbkreisförmigen Kopfschildes mit langen abste- 
henden Wangenhörnern gekennzeichnet ist. Die übrigen Merkmale des 
Kopfschildes sind verwischt, nur die Wangennähte sind schwach angedeutet. 

Die acht Segmente und Pleuren des deformierten Rumpfes sind schlecht 
erhalten. 

Das kreisförmige, abgerundete Pygidium ist gleichfalls teilweise be- 
schädigt. Die in die Hälfte des PN'gidiums reichende Achse nimmt kaum 
ein Viertel seiner Breite ein. Die Segmentierung ist nirgends sichtbar. 

Dafe es sich um eine neue Asaphellusart handelt, verrät schon auf 
den ersten Blick die eigentümliche Bildung des Kopfschildes und die Form 
des Pygidiums. 

AsaphelUis sp. 
(Taf III, Fig. I.) 

Einen isolierten Abdruck und ein Gegenabdruck der linken Wange mit 
einem sehr langen Stachel. 

Ob dieselbe einem großen Individuum von AsapJiellus bohémiens oder 
vielleicht einer anderen Form angehört, läfst sich auf Grund des fragmen- 
tarischen Körperteiles schwer nachweisen. 

Niobe (Ptychocheilus) sp. 
(Taf. III, Fig. 3.) 

Em Abdruck des rechten Teiles eines abgebrochenen Kopfschildes mit 
einigen deutlichen Rumpfsegmenten und Pleuren. 

Die Glabella ist abgeschält. 

Die Form des Randes mit vorhandenem schmalem und schräg ab- 
stehendem Wangenstachel, sowie der Verlauf der schwach angedeuteten 
Wangennähte erinnern an eme Niobe, wie sie z. B. Novak-Per.ner^ aus der 
Stufe dl;' von Nove Dvory abbildeten. 



Trtlobttt päsma D — d I y z okoli prazského. Palaeontographica Bohemiae. S.47. Taf.III, 
Fig. I. Nr. IX. Ceskd Akademie, Prag. 



r6 J. V. /KiJ/.Ko. M.-N. Kl. 

Cep II u lop o (l(t n . 

ürthoceras Iniitis Zki.. 

(Taf. IV, Fig. 6.) 

Eine schlanke, auf die Art der Hyolithf n ziemlich scharf zugespitzte, 
ca. 50 mm. lange Form, wie sie unsere Tafelabbiidung darstellt. 

Dieselbe war möglicherweise noch etwas länger; ihre Oberfläche ist 
glatt, die Wohnkammern reichen von der Spitze fast in die Hälfte der 
vorhandenen Schale. 

Aufser diesem Stück wurde noch ein Fragment mit fehlender Spitze 
und ohne Wohnkammern gefunden, von welchem man unmöglich sagen 
kann ob es auch der obigen Form angehört. 

Die Oberfläche beider Stücke ist teilweise vom Eisenoxyd rostig- 
braun geftirbt. 

Brachiopoden. 

Von den zahlreichen grofseren und kleineren Liugnliden und anderen 
Brachiopodcn habe ich nur folgende herausgenommen, welche eine nähere 
Bestimmung zulassen. 

Lingulella sulcata Barr. 
(Taf. IV, Fig. r. i a, 2, 2 a, 3, 3 a.) 

Diese bereits von Barrande (Syst. Sil. V PI. 106) aus der Stufe di/^ 
von Klabava angeführte Art, welche neuerdings nach Holub (1. c.) in dem 
Euloma-Niobe-Horizont bei Rokycan nachgewiesen wurde, kommt auch bei 
Plzenec ziemlich häufig vor, und zwar in dem halbverwitterten lichtgrauen 
Schiefer unterhalb des Kreuzes »U Blazeje«, sowie in den festen grau- 
grünen Graptolithenschiefern des am westlichen Ende der Lehne sich befind- 
lichen Steinbruches. 

Es sind schön geformte, mit feinen Radialrippen versehene Exemplare 
oder glatte Steinkerne vorhanden, deren Wirbel mehr oder weniger zu- 
gespitzt ist. Die konzentrischen Lamellen sind nur bei einigen Exemplaren 
erhalten (Fig. i a, 2 a). Die Schale ist schwarz und glänzend, der halb- 
kreisförmige Stirnrand hie und da mit einem engen Saum versehen. 

Die Form der von Barraxde von Klabava abgebildeten Exemplare 
(Sys. Sil. V, PI. 106, Fig. III, 2, 3) scheint mir mit den Originalen, die ich 
seinerzeit untersuchte, nicht genau übereinzustimmen. 



ICSI. No. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUN A. I7 

Liu gule lia rugosa Barr. 
(Taf. IV, Fig. 4. 4 a ) 

Es finden sich meistens Steinkerne mit nur teilweise erhaltener Schale, 
Dieselbe ist breit und verhältnismäfsig kurz, der Wirbel abgestumpft. 

Die Oberfläche weist eine Reihe scharf hervortretender, schütter situierter 

Radialrippen auf; konzentrische Lamellen sind nicht überall entwickelt. 

Kommt in beiden Lokalitäten »U Blazeje« vor. B.JiRRande führt diese 

Art gleichfalls von Klabava (Syst. Sil. V PI. 152) und Holub aus derselben 

Gegend an. 

Lingiilella amygdala Zel. 
(Taf. IV, Fig. 5, 5 a.) 

Die Schale der vorhandenen Art unterscheidet sich von beiden oben 
beschriebenen Linguliden durch eine mandelartige Form. 

Der Stirnrand ist breit, halbelliptisch abgerundet, die Schale allmählich 
verschmälert und mit einem stumpfen Wirbel endend. 

Es sind nur einige Stirnkerne mit stellenweise erhaltener Schalen- 
kruste vorhanden. Sonst ist die ursprüngliche, aus feinen Radialrippen und 
konzentrischen Lamellen bestehende Skulptur ziemlich gut sichtbar. 

Kommt in beiden Punkten »U Blazeje« vor. 

Lingulella pnsilla Zel. 
(Taf. III, Fig. 8, 8 a, 9, 9 a). 

Von den bei Plzenec häufig vorkommenden kleinen Linguliden sind 
hier zwei typische Exemplare abgebildet. 

Die ca. 3 — 5 mm. lange Schale ist schmal, glänzend und mit spärlichen 
Lamellen, welche nur unter der Lupe sichtbar sind, versehen. 

Der Umrifà ist elliptisch oder schwach verschmälert, eiförmig, die 
Schale flach gewölbt oder gegen den Wirbel zu ein wenig erhöht. 

Die besprochene Art wurde nur in den graugrünen Schiefern »U Blazeje« 
gefunden. 

Acrot/ie/e? iiov. sp. 
(Taf. Ill, Fig. 7, 7 a, 7 b.) 

Eine fast kreisrunde, ziemlich hoch gewölbte Schale, in einem Durch- 
messer von 8 mm., deren Oberfläche hornig, glatt, glänzend und unregel- 
mäßig konzentrisch gestreift ist. 

Eine kleine Einbuchtung des Schlofarandes erinnert, wie aus unserer 
vergröfaerten Abbildung (Taf. III, Fig. 7) ersichtlich ist, an eine ähnliche 
schlitzförmige Öffnung bei der Discina oder Trcmatis. 

Vid. Selsk. S'<rifter. L M.-N. Kl. 1921. No. ro. 2 



i8 J. V. ZKLizKo. M.-N.Kl. 

Daß das hier beschriebene Fossil jedenfalls zu der Familie Discinae 
gehört, ist auf den ersten Blick sichtbar. 

Dasselbe stammt gleichfalls aus dem graugrünen Schiefer. 

Ohohis'. sp. 

Zwei sehr kleine an Obolus minimus Barr. (Syst. Sil. V, PI. 5) erin- 
nernde Individuen aus demselben Fundort »U Blazeje«. 

Spirifcr sp. 

Ein ca. 6 mm. langer und 7 mm. breiter, hoch gewölbter Steinkern, 
welcher dem Habitus nach wahrscheinlich der oberwähnten Brachiopode 
anzugehören scheint. 

Dieselbe stammt aus dem lichtgrauen Schiefer unterhalb des Kreuzes 
»U Blazeje«. 

Pteropoden. 

CojiHlaria pygmaea Zel. 
(Taf. Ill, Fig. 5 ) 

Eine sehr kleine, unvollständig erhaltene Art, deren Länge ca. 17 mm. 
und die Breite bei der Mündung ca. 7 mm. beträgt. 

Die ursprüngliche Epidermis hat sich nur in einer Partie bei der 
Spitze erhalten. Sie besteht aus nur unter der Lupe sichtbaren, gleich 
grofsen und gleichmäßig aneinander situierten Wärzchen rundlicher Form. 
Diese Wärzchen waren senkrecht aneinander, also mit der vorhandenen 
parallel laufenden Medial furche gereiht. Der den Abstand zwischen den 
Wärzchenreihen bildende Raum längs der Schale ist schmal, jedoch ist 
derselbe durch den Umstand, dafe die Wärzchen gleichmäfaig dicht geordnet 
sind, gut sichtbar. 

Durch die geschilderte Schalenskulptur ist die Coiiularia pygmaea von 
allen aus dem älteren Paläozoikum Böhmens bekannten Formen sofort zu 
unterscheiden. 

Comilaria su/ca Zel. 
(Textabbildung. Taf. III, Fig. 4.) 

Auf demselben Gestein, wo sich die oben beschriebene Art befindet, 
kommt neben derselben ein Bruchstück einer 14 mm. langen Schalenvvand 
vor, deren Skulptur aus einer Reihe auftretenden Querrippen besteht, die 
in einem kaum i mm. messenden Abstand voneinander getrennt sind. 
Zwischen den Rippen findet man keine Granulation oder sonstige Ver- 
zierung. Diese Skulptur stimmt mit der mir im Jahre 1906 von Herrn 



192I.N0. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. I9 



Prof. HoRAK zur Bestimmung geschickten Conularia aus den bereits ange- 
führten d I /Î^-Schichten von Klahava überein. 

Da dieses Fossil bisher weder beschrieben noch abgebildet ist, möchte 
ich dasselbe hier kurz behandeln. 

Es ist ein Abdruck und ein Gegenabdruck einer teilweise erhaltenen, 
ziemlich großen Schale mit deutlicher Skulptur vorhanden. Die Verzierung 
besteht aus groben, ziemlich tiefen Ouerfurchen, welche in Abständen von 
I —1.5 mm. von einander entfernt sind. Die Granulation oder andere Ver- 
zierungen fehlen. Dem Schalenumrisse nach scheint, dafs das vollkommene 
Exemplar 8 cm. hoch war und seine Breite bei der Mündung höchstens 
3 cm. betrug. 

f 



\ 




\ 



Eine ähnliche, aus Querrippen und Furchen bestehende Skulptur wie 
sie das Schalenfragment von Pilsenetz und das Exemplar von Klabava auf- 
weist, besitzt auch die von mir seinerzeit beschriebene Conularia Purkynei^ 
von Sirâ-Cekov (di/). 

Die erwähnten Rippen und F'urchen, welche gleichfalls einfach und 
ohne Verzierung sind, treten sehr scharf hervor, und je mehr sie der Spitze 
zulaufen sind sie feiner und seichter. Bei der Mündung entfallen von 
diesen 15 — 17, bei der Spitze 28 — 30 auf 5 mm. Stellenweise fliefàen 
auch zwei Rippen in eine länglichrunde Leiste zusammen. 

Gleichfalls die von Osek (d i y) bekannte Conularia rohtista Barr- weist 
nur teilweise ähnliche Schalenskulptur wie unsere neue Form von Plzenec auf. 



^ Neue Pteropoden des älteren Paläozoikittns MittelböliMiens. Jahrb. d. k. k. geol. Reichs- 

anst. 1911. Bd. 61, p. 45. Taf. III, Fig. i a, ib. 
- Syst. Silur. Vol. III, p. 51. PI. 16, Fig. 10— 11. 



20 J. V. ZKI.iZKO M.-.\. Kl. 

Conularia sp. 

Ein zerdrückter positiver und negativer Abdruck einer unvollständigen, 
24 mm. langen, nach vorn gekrümmten Schale. Die Seitenleisten sind gut 
sichtber, die Medialfurche sowie die aus ursprünglich rundlichen Quer- 
streifen bestehende Skulptur ist nur schwach angedeutet Dieselbe Skulptur 
fand man auch bei einem Bruchstücke eines anderen Exemplars. 

Aufàerdem wurde noch ein Abdruck eines kaum i cm. langen Ge- 
häuses mit erst unter der Lupe erkenntlichen SeitenJeisten und Medial- 
furche gefunden. 

Ortliothecal n. sp. 
(Taf. III, Eig. 6.) 

Ein stark zerdrücktes, 17 mm. langes Gehäuse ohne Spitze. Der Mund- 
rand ist abgebrochen, die Längsleisten sichtbar und die Oberfläche glatt. 

Ein daneben befindliches isoliertes, 4.5 mm. langes und 2 5 mm. hohes 
Operculum scheint demselben Stücke anzugehören. Dasselbe bildet ein 
gleichschenkeliges Dreieck mit abgerundeten Ecken. Was die anderen 
Merkmale des Deokelchens, dessen Innenseite vorhanden ist, anbetrifft, 
ähnelt dasselbe dem von Novak abgebildeten ^ aus verschiedenen Fund- 
stellen des böhmischen Obersilur und Devon herrührenden Orthothecen 
am besten. 

Obwohl die untersilurische, aus den Quarzkonkretionen der d i /- 
Schichten von Särka stammende Orthotlicca Särkaensis'^^ von Novak als 
fraglich bezeichnet wurde, so ist es doch nicht ausgeschlossen, daß die 
Verbreitung dieser von Noväk gestellten neuen Pteropodengattung auch in 
das Untersilur hineinreicht. 

Da unser Exemplar allerdings vielmehr Verwandtschaft mit der Ortho- 
tJieca als mit den aus dem böhmischen Untersilur bisher bekannten Bactro- 
theca- oder Hyo/it/iiisgatiungen aufweist, so habe ich dasselbe auch provi- 
sorisch zur Orthotheca gestellt. 

Schliefelich sei noch bemerkt, dafe es nicht ausgeschlossen ist, dafe der 
von Brøgger-^ aus dem norwegischen Kambrium angeführte HyoUthus pH- 
catiis nach Novak ^ zur Gattung Orthotheca gehören kann, weshalb diese 
Gattung auch schon in primordialen Ablagerungen vertreten wäre. 



1 Zur Kenntnis der Fauna der Etage F—fi in der palaeozoischen Schichtengruppe Böh- 
mens. Sitzungsber. der königl. böhm. Gesellsch. d. Wiss. Prag 1886. 

•^ Revision der palaeosoisciten Hyolitliiden Böhmens. Abhandl. d. k. böhm. Ges. d. Wis«. 
VII. Folge, 4 Bd. Math. -nat. CI. No. 6, p. 44. Prag 1891, und 1. c. Taf. II. Fig. 29. 

3 Ovn par ado.xides-skij erne ved Krekling. Nyt Mag. for Naturvidenskaberne. XXIV. 
Christiania 1877. 

4 1. c. 



1921. No. I o. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. 21 

Graptolithen. 

Unter allen Fossilien der d i/î^-Schichten bei Plzenec sind die Grapto- 
lithen am häufigsten vertreten. 

Wie ich schon vorn angeführt habe, kommen dieselben besonders in 
dem nordwestlichen Teile der Lehne »U Blazeje« im festen, graugrünen 
Schiefer vor, dessen große spaltbare Platten mit langen, schön entwickelten, 
ruthenförmig verzweigten oder buschigen Exemplaren bedeckt sind. 

Da ich das ganze Material meinem Freunde, Herrn Dr. Ferner zur 
Bearbeitung übergab, möchte ich nur im Kurzen auf einige Graptolithen 
hinweisen, die ich seinerzeit Herrn Prof. S. L. Törnquist in Lund zur 
Vergleichung mit schwedischen, dem ältesten Ordovician entstammenden 
Arten eingesandt habe. Durch die Güte des Herrn Törnquist wurden 
folgende Arten bestimmt: 

Didymograptus. Scheint zwischen Didymograptus haïtiens Lapw. oder 
Didym. artus Elles and Wood zu stehen, ohne mit dem einen oder dem 
anderen identifiziert werden zu können. Nahestehende Formen kommen 
in schwedischer Zone mit Didymograptus haïtiens Tullberg vor ^. 

Jener von Ferner in seiner Monographie über die Graptolithen des 
böhmischen Untersilurs von Särka und Osek (di;') beschriebene Didy- 
mograptus iiamis weist einen anderen, von unser Art ganz abweichenden 
Charakter auf. 

Tetragraptus quadribraclnatus Hall und aufàer diesem entweder Tetragr. 
serra Brongn., nach der Auffassung Türnquists (= Tetragr. Ami Elles 
and Wood-), oder Tetragr. serra Brongn, Elles and Wood -^. Vielleicht 
ist es auch eine andere nahestehende Form. 

Eine verwandte böhmische Art Tetragraptus caduceus Salter beschreibt 
Ferner von Klabava (d i ß). 

Aufàer diesen Graptolithen wurden bei Plzenec auch Fragmente ent- 
weder von einem gabelförmigen Didymograptus oder Bryograptns gefunden. 

Nach der vorläufigen Mitteilung des Herrn dr. Ferner überwiegt in 
dem ihm von Plzenec eingesandtem Material Dieliograptits und Holograptus 



1 Sven Leünh. Törnquist. Researches into the Graptolites of the Loiver Zones of the 
Scanian and Vestrogothtan Phyllo-Tetragraptus Beds. \. Lunds Universitets Årsskrift. 
Bd 37. Afdeln. 2. Nr. 5. König!, fysiografiska sällskapets handiingar. Bd. 12. Nr. 5. 
Lund 1901. 

2 L Ferner. Studie o ceskych graptolitech. Cast II. Monografie graptolitu spodniho silurti. 
Palaeontographica Bohemiae. N. Ill b. Prag 1895. 

3 Sv. Leonh. Törnquist. Researches into the Graptolites etc. II. Lund 1909. — Derselbe. 
Några anntärkningar ont indelninger inotn Sveriges kanibro-silur. Geolog-. Foren. Fö--- 
handl. Bd. 35. H. 7, p. 416. Stockholm 1913- 



22 J. V. ZKLiZKO. M.-N. Kl. 

(Temnograptus); siehe Taf. V. Außerdem findet man darin Holograptus 
(Trnchograplus) sp. off. Deani Lapw. und einige schwer bestimmbare Reste, 
die an Loganograptiis und Azyograptiis erinnern. 



Die aus der Bande d — d 1 /^ 
angeführten Fossilien. 

Trilobiten. 

Asaphcllus boJiemicus Zel. 
Asaphellus insignis Zkl. 
Asaphellns sp. 
Niobe (Ftyclioclioilus) sp. 

Cephalopoden. 

Orthoccras tenuis Zel. 

Bracliiopoden. 

Lingulella sulcata Barr. 
Lingulella rugosa Barr. 
Lingulella amygdala Zel. 
Lingulella pusilla Zel. 
Acrothelel nov. sp. 
Obolus^ sp. 
Spirifcr sp. 

Pteropoden. 

Conularia pygmaea Zel. 
Conularia sulca Zel. 
Conularia sp. 
Ortliotl/cca? n. sp. 

Graptolithen. 

Didymograptus. 

Tetragraptus quadribraclüatus Hall. 

(? Tetragraptus serra Brongn. (= Tetr. Ami Elles and Wood) oder 
Tetr. serra Brongn. Elles and Wood). 
Dichograptus. 

Holograptns (Temnograptus) sp. 
Holograptus (TrocJiograptus) sp. off'. Dcani Lapw. 



igai. No. lo. äquivalente der untersilurischen euloma-niobefauna. 23 
Schlussfolgerungen. 

Die unmittelbar unter den schwarzen Schiefern der Stufe diy mit 
vorherrschenden Placoparia Zippei bei Sedlec auftretenden d i /îJ-Schichten 
bilden stratigraphisch sowie faunistisch einen selbständigen Horizont der 
Bande D — d i ^. 

Nach der in der Lokalität »U Blazeje« festgestellten Fauna scheint 
dieser Horizont jünger zu sein als der Eido)na-\\ox\zon\. bei Rokycan ^ 
Deshalb wird es bei weiterem Studium der Ablagerungen der Stufe à\l^ 
notwendig, eine ähnliche Zonengliederung derselben vorzunehmen, wie 
solche Kloucek- in den d i;'-Schichten durchführte und für deren Aufnahme 
für die d i /îJ-Schicheten auch die Graptolithenstudien Per.ners^ sprechen. 

Diesbezüglich können uns auch die von mir Herrn Dr. Ferner zur 
Bearbeitung übergebenen Graptolithen von Flzenec gewiß ein positiveres 
Resultat liefern. 

Was einen vorläufigen \"ergleich der seinerzeit von Herrn Frofessor 
TöRNQuiST von Flzenec untersuchten Graptolithen mit jenen von Schweden 
anbelangt, geht aus dem nachstehenden, mir von Herrn Frof. Törnquist 
zur Verfügung gestellten schematischen Gliederung des ältesten schwedi- 
schen Ordovician am besten hervor: 

d) Zone mit Isograpliis gibbendiis Xich. 

c) — » Pliyllograptus deiisus Törnq. 

b) — » Diciyniograptiis ballicus Tullberg. 

a) — » Tclragraptus pliyllograptoidcs Linrs. 



Unterer Didymo- 
graptus-Schiefer 



Niobe-Euloma- 
region, auch 
Dictyonema- 

region genannt 



Ceratopygenkalk als Kalkfazies und Dictyonemaschiefer 
als Schieferfazies. 



Der von Flzenec angeführte Tclragraptus serra kommt in der Zone a 
und Tetragraptus qnadribrachiatiis in der Zone b vor. 

Nach den vorhergehenden Erörterungen scheinen die fossilführenden 
Schiefer von Rokycan und Flzenec teilweise dem obersten Niveau des 
unteren Didymograptusschiefer (Zone a und b) zu entsprechen. 



1 1. c. 

- C. Kloucek. Predbesnä zpräva o dvoii novych horisoniech v piisiim D — diy. Vestnik 

kral. ces. spol. nauk. Prag 1908. — O vrstvdch Diy, jich trilobitcch a nalesistich. Roz- 

pravy Ceské Akademie. Ig. XXV. N.39. 1916. 
^ I. Ferner. Studie o ceskych graptolitech. II. Monografie graptolitu spodniho siluru. 

Ceské Akademie. Prag 1895. 



24 J- V. ZKÜZKO. M.-N. Kl. 

Daß die in Rede stehenden Schichten vom stratigraphischen und 
paläontologischen Standpunkt aus älter sind als die der Bande D — di;' 
und jünger als die der Bande D — iß, dafür sprechen unsere bisherigen 
Erfahrungen bei Plznec und noch besser die Fossiiienfunde Holubs ^ aus 
der Gegend von i^okycan. IIolub fand hier, wie bekannt, eine den Über- 
gang zwischen dem obersten Kambrium und untersten Silur bildende Fauna, 
die Brøgger als Euloma- Niobefauna bezeichnete-. Aufaer Brachiopoden 
(auch Lingula sulcata und L. riigosa wie bei PIzenec), neuer bisher noch 
nicht bestimmter Graptolithen, Pteropodcii und underen Fossilen, sind aus 
der Gegend von Rokycan bis heute folgende Trilobiten beschrieben: Agnostus 
splendens, Agnostus consors, Asaphcllus Pcrneri, Aspidaeglina miranda, Euloma 
Bohemicum, Euloma inexpcctatum, Lichas praccursor, Megalnspides cuspidatus, 
Nilcus pater, Aeglina Bröggeri und Barrendei primula. 

HoLUB, welcher die böhmische Fauna mit der norwegischen, französi- 
schen, englischen und bayrischen verglich, bemerkt, trotzdem sich die von 
ihm bei Rokycan gefundene Fauna mit der alten skandinavischen Euloma- 
Niobefauna nicht vollkommen identifizieren läfjt, ist es doch notwendig, 
dieselbe als ein Äquivalent der übrigen europäischen Eulomaschichten zu 
betrachten. 

Dasselbe gilt auch der von mir entdeckten Fauna bei PIzenec. 

Aus den Forschungen Brøggers und Törnquists geht gleichfalls hervor, 
dafa wir nur an wenigen Punkten der Erde eine mit der nordischen Euloma- 
Niobc-Fauna näher übereinstimmende Fauna vorfinden, nämlich: 

in Shropshire in den Shineton shales, 
» North Wales in den Tremadoc-Schichten, 
» Bayern die Leimnitz-Schichten bei Hof^, 
» Südfrankreich die Schichten von Caunes 
und St. Chinian in Languedoc. 

Aus dem Grunde, daß in Böhmen Äquivalente der Euloma- N lobe- Fauna 
jetzt nachgewiesen wurden, mufà natürlich die alte Definition Brøggers'*, 
T>daß ein unzweifelhaftes Äquivalent zu der Euloma- Niobe- Zone (Ceratopygen- 
kalkzone ja) des skandinavischen Nordens in Böhmen bis jetzt nicht nach- 
gewiesen ist und deshalb zu fehlen scheint«, entfallen. 

Brøgger selbst war nicht der einzige, der an die Existenz der älteren 
ordovicischen Ablagerungen in Böhmen zweifelte. 



^ I. c. I, p. 7. 

2 1. c. 4, p. 14. 

3 I. Barrande. Silurische Fauna aus der Umgebung von Hof in Bayern. [868. — 
I. F. PoMPECNj. Ein neuentdecktes Vorkommen von Tremadoc- Fossilien bei Hnf. 

* Über die Verbreitung der Euloma-Niobe- Fauna p. 224. 



1 92 1. No. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. 25 

Schon LiNNARSSON im J. 1873 hat auf die fehlende Übereinstimmung 
zwischen den böhmischen und nordischen Ablagerungen aufmerksam ge- 
macht; er sucht dies teils auf eine mögliche Landsperre zwischen den böh- 
mischen und dem nordosteuropäischen Silurmeer während der Zeit dieser 
Ablagerungen zu beziehen. 

Auch TöRNOUiST hat sich dafür ausgesprochen, dafe Böhmen mit Frank- 
reich, Portugal und Spanien, während dieser Zeit ein von dem nordischen 
ordovicischen Meer getrenntes Absetzungsgebiet ausgemacht haben mue, 
während aber die Graptolithcn beweisen, daß jedenfalls bisweilen i>das nord- 
europäische Graptolithemneer^ sich auch über dies Gebiet ausdehnte ^ 

Diese Vermutung hat sich jetzt durch die Funde bei Rokycan und 
Plzenec tatsächlich bestätigt. 

Bezüglich der Entwickelung der silurischen Sedimente in Böhmen und 
im Südwesten Europas sei hier auf die gleichnamige Arbeit Frechs- bzw. 
auf die Monographie Brøggers^ hingewiesen. 

Obwohl die Trilobitenfauna von Plzenec gegen die von Rokycan bis 
zum heutigen Tage nicht so mannigfaltig ist, scheint mir wenigstens das 
Vorkommen des Asaphellus bohémiens für den betreffenden Horizont nicht 
ohne Bedeutung, denn, wie bekannt, die ersten Vertreter der Asaphiden sind 
für die untersilurischen Ablagerungen ebenso charakteristisch wie z. B. 
Agnostus, Paradoxides und Olcnus für das Kambrium. Deshalb repräsen- 
tiert Asaphellus bohemicus auch bei Plzenec ein charakteristisches Leitfossil 
ähnlich wie der gleichaltrige Asaphellus Homfray in den Shineton Shales 
in Shropshire, und in dem Lower u. Upper Tremadoc in North Wales, 
sowie im Tremadoc von Cape Breton in Britisch-Amerika und Asaphellus 
IVirtlii in den Leimnitzschichten bei Hof. 

Jedenfalls aber ist es wünschenswert die d i /5-Schichten bei Plzenec 
einer weiteren Durchforschung zu unterziehen, was nach meinen bisherigen 
Erfahrungen eine längere Zeit in Anspruch nehmen wird, wenn die Fos- 
silienliste um weitere für die betreffenden Schichten ebenso charakteristische 
Trilobitenformen vermehrt werden soll. 

Wie schon oben erwähnt wurde, bilden die Ablagerungen der Stufe 
d I , j bei Plzenec einen faunistisch scharf abgesonderten Horizont gegen- 
über den schwarzen fossilienreichen d i ^'•Schiefern, die als Hangendes der 
grünlichgrauen d i /:?- Schiefer zu betrachten sind. 



1 Ibid. p 223. 

- Neues Jahrbuch für Min., Geol. und Pal. Jg. 1899. ß*^- ^^- — Auch Die geographische 
Verbreitung und Entwickelung des Cambriuni, von demselben Autor. Congrès géolo- 
gique international, 7me session Russie 1897. St. Petersbourg 1899. 

^ Die silurischen Etagen 2 und j im Kristianiagebiet u. s. zv. Kristiania 1882. 



26 J. V. ZKLÎZKO. M.-N. Kl. 

In welchem Verhältnisse die letztgenannten Schiefer zu den Krusnâ 
Hora Schichten (dia), welche neuerdings eine häufigere Fauna lieferten*, 
stehen, das festzustellen wird wiederum eine andere Aufgabe sein. 



Zum Schiufa dieser Arbeit fühle ich mich verpflichtet, Herrn Direktor 
Prof. A. JeiJnek in N. Bydzov, für den grööten Teil von ihm aufgenom- 
menen Originale meinen herzlichsten Dank auszusprechen. 

Wien, November 1917. 



' C. Kloucek. Novinky z Krusnohorskych vrstev. I — III. Rozpravy Ceské Akademie 
1915 u. 1917. 



I92I. No. lO. ÄQUIVALENTE DER UNTERSILURISCHEN EULOMA-NIOBEFAUNA. 27 



Erklärung zu den Tafeln. 



Tafel I. Asaphellus bohémiens Zel. Natürliche Giöfae. 

„ II. Fig. I. Asaphelltis bohémiens Zel. Negativer Abdruck. 

, 2. Asaphelltts bohémiens Zel. Positiver Abdruck mit erhaltenem Hypostom 
Natürh'che Größe. 
„ III. Fig. I. Aiaphellus sp. Isolierter Abdruck der Imken Wange. Natürliche Größe. 

„ 2. Asaphellus insigrtis Zel. Natürliche Größe. 

„ 3. Niobe (Ptyehoeheilus) sp. Natürliche Größe. 

„ 4. Comtlaria sulea Zel. Eine vergrößerte Schalenpartie. 

„ 5. Conularia pygmaea Zel. Eine vergrößerte Schalenpartie. 

„ 6. Oriholheeal n. sp. E'n wenig vergrößert. 

„ 7. Acrolhelel n. sp. Natürliche Größe. 

„ 7 a. Dieselbe ein wenig vergrößert. 

„ 7 b. Dieselbe vielfach vergrößert. 

, 8 — 9. Liiigulella pHsilla Zel. Dorsalschale. Natürliche Größe. 

„ 8 a — 9a. Dieselbe ein wenig: vergrößert. 
„ IV. Flg. 1 — 3. Liiigulella sulcata Barr. 1—2 Dorsalschale, 3 Dorsal- und Ventral- 
schale. Natürliche Größe. 

„ I a, 2 a, 3 a. Dieselbe vergrößert. 

„ 4. Lingiilella rtigosa Barr Dorsalschale. Natürliche Größe. 

„ 5. Liiigulella amygdala Zel. Dorsalschale. Natürliche Größe. 

„ 5 a. Dieselbe vergrößert. 

„ 6. Orthoeeras tenuis Zel. Natürliche Größe. 
„ V. Holograptus (Temnograptus) Holm. Auf dem Original sind die einzelnen Hydro- 
thecae gut sichtbar. 6/7. 



Gedruckt 29 Ok'dber 1921. 



Vid.-Selsk. Skr. I. M.-X. Kl. igoi. X 



o. I o. 



Tafel I. 




Vid.-Selsk. Skr. I. AI.-X. Kl. 1021. Xo. 10. 



Tafel II. 





1«^^ 



Vid.-Selsk. Skr. 1. iM.-X. Kl. 1921. No. 10. 



Tatll III. 




7b 



Vid.-Selsk. Skr. I. I\I.-X. Kl. 1921. No. 10. 



Tafel IV. 




Vid.-Selsk. Skr. I. M.-N. Kl. 1921. No. 10. 



Tafel V. 







THE STRAND FLAT 
AM) ISOSTASY 



BY 
FRIDTJOF NANSEN 

IWITH 170 ILLUSTRATIONS AND MAPS IX THE TEXT) 
(ViDEXSKAPSSELSKAPETS SkRIFTER. I. MaT.-NATURV. KlASSE. ICSI. No. II> 



ÜTGIT FOR FRIDTJOF NANSENS FOND 



KRISTIANIA 

I KOMMISSION HOS JACOB DYBWAD 
1922 



Fremlagt i den mat.-naturv. klasses mote den 3. juni 1921 



A. W. BRØGGERS BOKTRYKKERI A- 



To 

Dr, Hans Reusch 

who first described the Norwegian Strandflat 

as a distinct geographical 

feature 



m memoriam 



FOREWORD. 

The autbf)r much regrets the late appearance ri this treatise. It 
has been written and printed with many interruptions due to various 
circumstances over which he had no control. The greater part of the 
manuscript was ready for the press more than eighteen months ago. 

He wishes to express his sincere gratitude especially to the Norwegian 
geologists Professor Jacob Schetelig, Professor \ . M. Goldschmidt, and 
Mr. Adolf Hoel, Uni\-ersity Reader in geology, also to the Swedish geo- 
logist Professor A. G. Högbom. for much valuable information and advice. 

A special word of thanks is also due to the publisher Jacob Dyl)wafl 
by whose courtesy a number of illustrations from the author's book 
"En Ferd til Spitsbergen" are reproduced here. 

The letter o used in Norwegian names is e(iuivalent to oe and is 
pronounced somewhat similarly to the letter o in the English words 
"son" or "work". 

Lysaker, December 1922. 



CONTENTS. 

Page 

I. Introduction r 

II. Théories ot the .ü;enetic oris^in of the strandfiat 4 

III. 1 he effect of the suhat'i'ial denvidation in the coastal zone 16 

IV. (ilacial erosion 20 

The erosion of the glacier.s of the inland ice 20 

The erosion of local glaciers 25 

\' . Marine denudation. .Shore erosion by frost 28 

VI. The development of the Norwegian strandflat _)3 

When was the Norwegian strandflat developed 46 

The height of the Norwegian strandflat 49 

The partial absence of a strandflat along some ]jarts of tlu- Norwegian coast 49 

Summary 52 

\'II. rile strandflat of the Norwegian west coast from the region of Sogne Fjord to 

the region of Hardanger Fjord 55 

The strandflat of Sogne Fjord 57 

The genetic origin of the strandflat in Sogne Fjord 66 

The strandflat north and south of the mouth of Sogne Fjord 69 

The region of Lindås Peninsula 71 

The region of Radoi 73 

Region of Bergen 79 

Region of Sotra 8 r 

Islands between Kors Fjord and Hardanger Fjord 83 

Hardanger Fjord and Bommel Fjord 84 

VIII. The strandflat of the southwestern and southern coast of Norway 91 

The region of Karmoi gi 

The Stavanger region 93 

Relation between the geological structure of the coast and the occurrence of 

the strandflat 98 

Jæderen 100 

The high, steep coast between jæderen and Lister 102 

From Lister to the Naze (Lindesnes) 104 

The southern and south-eastern coast of Norway 105 

The inner end of Christiania Fjord E07 

IX. The strandflat along the coast from .Sogne Fjord to Vikten 113 

The region of Sondtjord, Nordtjord, and Stat 113 

The region of Ålesund 113 

The region of Smolen, Hitteren, and Froia 120 

From Trondhjem Fjord to Vikten 126 



Page 

X. Thi:- stiandtlat ol' northern Norway 129 

Helgeland " roq 

Træna 110 

Relation between the differences in the surface topography of the submerged 

plateaus and differences in the nature of their rocks 147 

Coast of northern Helgeland and the east coast of West Fjord 152 

Lofoten and Vesterålen 1^6 

Senjen to Ringvasøi 163 

Finmarken 163 

XI. The strandfiat of Bear Island 16^ 

XII. The strandfiat of Spitsbergen 178 

Literature i-j8 

Relation between the development of the strandtLit and the geological struc- 
ture of the land 185 

Prince Charles Foreland 1 85 

West coast of Spitsbergen south of Ice Fjord 192 

Cape Guissez in Cross Bay 193 

Cape Mitre Peninsula and the north-western part of Spitsbergen 193 

Reindeer Land 106 

The peninsula between Liefde Bay and Wijde Bav 197 

East coast of Wijde Bay, and Verleegen Hook .... 200 

North-west coast of North East Land 203 

The fjords 203 

The age and genesis of the strandflat of Spitsbergen 204 

Raised shore-lines 207 

Platforms at high levels 208 

XIII. The strandflat along the coast inside and outside the arctic regions 211 

The coasts of Siberia, Greenland, and Alaska 211 

Shetland Islands 214 

XI\'. The levels of the strandflat and their development 217 

Summary of the results of our investigations of the strandflat 217 

The emerged strandflat 217 

The submerged strandflat of Norway 218 

The three levels of the strandflat 219 

Causes of the changes in the level of the strandflat 220 

The nature of the rocks and the topography of the strandflat and the conti- 
nental shelf 225 

X\'. The continental shelf and its formation 227 

Fluctuations in sea-level caused by the formation of ice-caps 229 

The thickness of the ice-caps 230 

Shifts of sea-level and shore-line caused by changes in the volume of the 

Ocean 233 

The lowering of sea-level during the glacial periods and the surface of the 

continental shelf 237 

The deposition of sediment has gradually raised the general sea-level 239 

XVI. The strandflat and the late-glacial and postglacial submergence of Fenno-Scandia 245 

The late-glacial and postglacial upheaval of Sogne Fjord 246 

The late-glacial and postglacial upheaval in the region of Nord Fjord 248 

Gradient of the late-glacial and postglacial uphea\al of the west and north 

coast of Fenno-Scandia 25 1 

Relation between the inclinations of the two conspicuous raised shore-lines 

of Northern Norwav and the Kola Peninsula 2^6 



VIII 

l'a;.'«- 
Relation bctsvccn the altitudes ol' the two conspiciioiis raised shore-lines in 

the Tromso — Hämmertest districts 259 

l\( iatlon licivvi I II tlie altitudes of Tanner's shore-lines If and IIA in Finmark 262 
The ii|)|)c r limit ol late-glacial submergence and the Tapes level f>n the Kola 

Pcniiisiil:! 266 

The upper limit of submergence and the Tapes level along the Norwegian 

coast south 1)1 the Tromsø district 269 

The iinitunn rhaiaiti )• of the late-glacial and postglacial upheaval along the 

west and north loast of Fenno-Scandia 273 

The late-glacial and postglacial submergence and emergence in central and 

eastern Fenno-Scandia and in Jutland 275 

The present crustal moxcmeiils in the regions round the Baltic Sea and the 

( iulf of Bothnia 276 

The possibility of a strandtlat along the Baltic and Bothnian coast 276 

The reasons why the postglacial crustal movements have been retarded in 

the regions rinintl the Baltic Sea and tin- (iulf of I'othnia 280 

Relation between the heights of the Ta]j(s Icxcl ami of the ii])per marine 

limit in central and southern Fenno-Scandia 282 

The cause of the transgression of the sea in the Tapes-Littorina period 286 

What may have been the i-ause of a rise of the sea-level during the Tapes- 
Littorina period ? 287 

XVII. Isostasy 290 

The theory of isostasy 291 

The Crust's capacity of isostatic readjustment 294 

How are the isostatic vertical movements of iIk- lithosphère effected 297 

How long a time does the earth's crust require to reach its new isostatic 

le\el after a disturbance of its equilibrium V 303 

What is the extent ot the smallest area within which isortatic movements 

may occur? 304 

Isostasy and erosion 304 




Fig. I. Strandllat at Lille Molla (535 metres high, to the right) and Skråven (281 metres, 
to the left), south side of Lofoten. (July 2nd, 19 12.) 



. INTRODUCTION. 



The Norwegian "strandfiat" extends as a low flat foreland, and an 
often broad belt of thousands of low islands, skerries, and rocks (the 
■"Skjærgård") in front of the high, mountainous land, along the wes1 
and north-west coast of Norway, from Lindesness to Finmarken. It is 
backed by the escarpment of the mountains which often ascend abruptly 
from the inner margin of the strandfiat to altitudes of hundreds of metres. 

The width of this strandfiat may vary much, from a few k'lometres 
in some regions to 60 kilometres (37 miles) in the region of Hitteren and 
Frøia, and 46 to 50 kilometres (29 to 31 miles) in Helgeland (region of 
Hæroi, Donna, and Træna). 

The degree of development of the strandfiat may also vary a great 
deal. Along great parts of the coast it is extremely conspicuous and 
sharply defined, but in other regions it is less striking, and in some places 
it may even be difficult to trace out or to distinguish from the low hills 
of the land inside. 

Where it is well developed, its characteristic features are: (i) the 
remarkably horizontal plane formed by the flat summits of its thousands 
of low skerries, islets, and peninsulas, rising to a certain low height above 
the sea, looking at a distance as if they had been cut off and planed down 
to this level along a ruler. (2j The horizontal and sharply defined in- 
cision which this plane forms in the mountain side of the land, often 
oversteepened and rising abruptly from the inner margin of the plane. 




Fig. 2. Strandflat on Arsteinen, east of entrance to Raftsund, Lofoten. (July 2nd, 1912.) 
Vid.-Selsk. Skrifter. I. M.-X. Kl. 1921. No. 11. 1 



IklDllOK NANSKN. M.-N. Kl. 



(3) Hills, or mountains, or mountainous masses, surmount here and there 
the plane of tlie strandfiat as more or less isolated 'stacks' or *mf;nadnfjcks', 
often with very steep sides. 

No observant tra\ellar can avoid being struck by this peculiar for- 
mation so dominant in many regions, and Norwegian geographers long ago, 
c. g. H. Mohn [1877], called attention to it. But Hans Reusch was 
the first geographer in literature [1894] who described it as a uniform 
and important morphological feature in the topography of the Norwegian 
coast and propounded a general theory of its genetic origin. He gave it 
the name "strandfiat" ( /. c. shore plane). 

The English term "coastal plain" J rlo not consider appropriate for 
this formation, because, as a rule, it is not a plain, and in my opinion 
never has been. It is, and always was, composed of numerous low islands 
and peninsulas, separated by scjunds and fjords of varying, often con- 
siderable depth. Moreover, any plain in the coastal region, also formed 
more or less of loose material formerly deposited in off-shore water, is 
often called coastal plain. 

"Coast platform" might be a better term, but I consider it to be pre- 
ferable to keep the Norwegian "strandfiat" as a name for this formation, 
so characteristic of the Norwegian and other northern coasts, but which 
does not occur in its typical form outside regions which have formerly 
been glaciated or exposed to severe climates. 

After Reusch's important paper of 1894, much attention has been paid 
to the strandfiat in scientific literature, and a lively discussion has been 
going on about its nature, origin, and age. A brief account of the literature 
concerning this subject has been given recently by Hans W:son Ahlmann 
[19 19, pp. 93—98]- 

Prof . W. C. B r g g e r, wdio in 1893 accompanied us as far as 
Tromso on our way out with the Fram-Expedition, was struck by the 
peculiar, flat formation of the many low islands along the coast of Nord- 
land. He gave a lecture on the subject in the Geological Club of Christiania 
on December 14th, 1893, and expressed the view that this level formation 
is a plane of marine denudation. 

During the Fram-Expedition the writer also found that an extensive 
and strikingly flat, low foreland is a very dominant feature of the topo- 
graphy along the north coast of Sibiria, especially the west, north, and east 
coast of the Taimyr Peninsula. In a later report [Nansen 1904, pp. 20, 
39, 42, 71, 75, 90, 102 ff.] the strandfiats of northern lands were described 
and discussed at some length. 

Since that time I have occasionally studied the strandfiat in various 
regions. During cruises with my yacht "Veslemoy" in the summers of 
1904 and 1909 along the south and w-est coast of Norway as far as the 
region of Ålesund, observations of the strandfiat were made, although the 
main object of the cruises was océanographie research. 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. 




Fig. 3. Strandtlat near Kunna 1593 metres highi, Northern Helgeland. 
View north-eastwards from off Meloivser. (September 9th, 1912.) 

In the summer of 191 1 a cruise was made with the "X'eslemoy" along 
the south and west coast of Norway, as far as the region of Sogne Fjord, 
and also to Shetland, with the special purpose of studying the strandfiat 
and measuring its heights. 

A cruise with the "\*eslemoy" in 19 12 along the coast of Norway to 
Bear Island and Spitsbergen, gave an opportunity of studying the strand- 
flat in those northern regions. 

An expedition from Norway through the Kara Sea to the Yenisei river 
in 191 3, gave a new opportunity of studying the coast formations in those 
regions. 

In the following report I shall try to give tlie results of tlie obser- 
vations concerning the strandfiat, made during these cruises, especially 
that of 1911. I regret that other work has prevented that this was done 
before. ]Much that was in mind when the many observations were made, 
has certainly faded from memory now so many years afterwards. But 
nevertheless, I venture to hope that the observations as they are, may still 
be of some interest to geographical students. In a recent book on the 
cruise to Spitsbergen in 19 12 [1920, 1921" I have already described some 
results of my study of the strandfiat on Bear Island and Spitsbergen. 



KKIinjOF NANSEN. M.-N. Kl. 



II. THEORIES OF THE GENETIC ORIGIN OF 
THE STRANDFLAT. 

Tlie \'ie\vs as rcg'ards the genetic orijj^'in of the straiulflat liave varied 
iiuicli, and liave chiefly been the following: 

J\ e u s c h explains the strandfiat as being a plain of marine abrasion, 
chiefly due to wave erosion, but the general subaërial denudation, imme- 
diately near sea level, has also been of some importance. The plain was 
formed chiefly in Tertiary time along a but slightly dissected coast, but 
it was also to some extent formed during "that portion of the glacial 
period when our country was comparatively free from ice". 

E. Richter [1896] also maintained very decidedly that the strand- 
flat was formed by wave erosion, but in interglacial time. 

J. H. L. Vogt [1900, 1907] is strongly in favour of the view that 
the strandfiat has been formed by wave erosion along a practically un- 
dissected coast. He thinks that in Helgeland (and northern Norway) the 
strandfiat was cut in this manner into the massive block of land to a width 
of about 45 kilometres, long before this land was dissected by the numerous 
deep valleys, fjords, and channels now existing. The layer of solid rock 
thus cut away, had a thickness of at least 400 metres along the inner zone 
of the strandfiat, and in some regions even more. Vogt assumes that the 
development of the strandfiat may have gone on ever since the time of 
the Jurassic dislocation on Andoi till the beginning of the first Ice Age. 

The view that the strandfiat is a plain of marine abrasion (wave 
erosion) has also been held by Davis [1899], Re k stad [191 2, 1915], 
Sahl strøm [1917], D. W. Johnson [1919], and others. 

A n d r. M. Hansen [1894, 1898^ assumed that the Norwegian 
strandfiat had been formed during the later part of the first Great Ice 
Age, partly by wave erosion and partly by the scouring of the drifting ice 
along the shores of the sounds. The coast had then already been dissected 
by numerous fjords and channels, greatly increasing the line of attack. 
The mode of formation of the strandfiat was supposed to be practically 
the same as that of the shore-lines. The broad strandfiats are simply 
several shore-lines which have met and united, advancing from different 
sides of the islands and peninsulas. 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. 5 

Hansen considers that the glacial erosion has also been of much im- 
portance, particularly lor the denudation of the low level surface of the 
skerries. He also pointed out that during the time mentioned, when the 
margin of the first inland ice still remained in the fjords, the climate 
greatlv favoured subaërial erosion. 

A m u n d H e 1 1 a n d, and later F. X u s s b a u m [1909], maintained 
that the strandfiat and the low skerries have been formed by the erosion 
of the inland ice, advancing over the outer coast. 

A view different from most others is held by Gerard de Geer 
[191 2] who considers dislocations to have been of great importance for 
the formation of the strandfiat. This theory has been carried still further 
by J. J. S e d e r h o 1 m [1913] who sees dislocations everywhere along the 
inner margin of all shore-planes, and he tries to explain the 'stacks', sur- 
mounting the plane of the strandfiat, as horsts bounded by numerous lines 
of dislocations, in a most artificial and complicated manner. But strand- 
flats do with certainty occur in many regions where there are no traces 
of dislocations near their inner margin, and Hogbom [1913] has in 
several places in Nordland been able to trace conspicuous lavers, with 
gentle seaward dip. from the mountain slopes and on the sharply defined 
plane of the strandfiat without any l^reak, which proves finally that the 
formation of a strandfiat is not conditioned by dislocations. On the other 
hand, investigations of Norwegian geolists have proved that dislocations 
may often occur within the region of the Norwegian strandfiat, without 
showing any relation to its limitation, or to the bosder lines of the stacks 
On the contrary the plane of the strandfiat often continues across the lines 
of dislocation without a break. 

By bringing weaker rocks in the shore-region in level with more 
resistant rocks further inland, dislocations may naturally fascilitate the 
formation of sharply defined shore-planes, but thev cannot form them. 

In the report [1904] mentioned above, and in a later paper [1905], 
the present writer maintained that the Norwegian strandfiat had been 
formed by the joint action of subaërial denudation and marine denudation. 
after the time when the coast had been dissected by the numerous fjords 
and channels, which now split it up into its innumerable peninsulas and 
islands. The subaërial denudation was supposed to have been of the 
greatest importance for the denudation of the coast land, while the marine 
denudation had been important for the levelling of the strandfiat. During 
the glacial periods the subaërial denudation of the coast land was essen- 
tially increased by the disintegrating effect of the frost on land. The 
marine denudation was due to wave erosion and also to the disintegrating 
effect of frost on the shore that was wetted by tide and waves. The 
transport of matter by ice, formed on the beach, was also of some im- 
portance. 



KklDlJOK NANSEN. M.-N. Kl. 



Tt: was cinpliatically Diaiiitaincd that a hroarl straiulflat, like that of 
iiorllicrn Norway (Helg-claii'l i, cannot have been formed by wave erosion, 
(luring- a rcasfjiiahle time, before the coast had been dissected by deep 
fjords and channels, but it must have ])een formed after that time. This 
was especially for two reasons: 

On tlie one side, the erosive effect of the waves on the shore would 
be much reduced in the extremely shallow sea over a submerged, nearly 
horizontal strandfiat, along- an undissected coast, where the waste would 
have a difficult way to travel in order to reach deep water. While along 
a dissected coast the line of attack of the marine denudation is immensely 
increased, and the waste is easily washed into the deep channels and fjords. 
The subaërial denudation is also much increased because on all islands 
and peninsulas the waste has got a short way to travel to reach the sea. 

On the other side, the total quantity of rock that had to be worn 
away from the small islands and peninsulas of a dissected coast, like the 
present one, would be only a small fraction of the quantity that must 
have been cut away from a high and solid coast. 

For the above reasons, I reached the conclusion that the Norwegian 
strandfiat cannot be of preglaciale age, but must have been developed 
especially before, during and after each glacial period when the rough 
moist and cold climate highly favoured an active subaërial denudation, 
as well as an active marine denudation. 

Although I shall later go more fully into this question, let me here 
at once correct what I consider to be a mistake. The development of the 
strandfiat cannot, as a rule, be assumed to continue during the late part 
of a glacial period, when the glacial margin is retreating from the coast, 
and not for a considerable period after that time, because, near the end 
of a glacial period, the land along the greater part of our coast, has most 
certainlv been depressed by the weight of the extensive ice-sheet, far 
below any level of the strandfiat, and it took a long time before the land 
had again returned to its natural level. 

Tet me in this connection also correct another mistake. It was as- 
sumed that the slow oscillation of the shore-line during the glacial periods 
may have been of importance for the formation of the strandfiat. These 
oscillations w^ere "caused by the isostatic movements of the land under 
the pressure of the ice-caps" as well as "by the accumulation of water in 
the ice-caps by -which the volume of the ocean was altered; and also, 
though slightly, as a result of the raising influence exerted on the sea- 
level by the attraction of the ice-masses on land" [1904, p. 109]. 

A characteristic feature of the strandfiat is, however, the great width 
and the nearly perfect horizontality of its levels, which indicate that it 
must have been formed during several long periods, when the level of 
the shore-line remained fairly stable. It has not been formed during 
periods of isostatic movements of the land due to the load of the ice-caps, 



ig21. No. II. THE STRANDFLAT AND ISOSTASV. 7 

when, however, the raised beaches were formed. A characteristic diffe- 
rence between the strandfiat and the raised beaches is that, the former is 
horizontal, and indicates stable levels which the land has had during long 
periods, while the latter are tilting and indicate the levels of the shore-line 
during a temporary submergence of the land. 

As being of importance for the development of the strandfiat it was 
also pointed out that by each advance of the glaciers of the glacial periods 
the waste was carried away seawards from the land and the shores, leaving 
bare rock surfaces for attack when the glaciers retreated. 

I still hold similar views on most points regarding the nature and 
formation of the strandfiat, as I did at that time; but I now am of the 
opinion that the shore-erosion by frost is of even much greater importance 
for the marine denudation of the strandfiat than I thought then. 

Nearly the same views regarding the nature and genetic origin of the 
strandfiat, as put forth in my report, have also been held by T h o r o 1 f 
Vogt [1912, 1914], Otto Norden skjold [1912, 1914], A. G. 
Hogbom [1913], and others. In his admirable paper, Hogbom proves, 
by numerous convincing evidences, the untenability of the above mentioned 
tectonic theory of De Geer and Sederholm, explaining the strandfiat as 
formed by dislocations. He points out that the coast land must have been 
dissected by the fjords before the strandfiat was developed, and had been 
so much lowered by subaërial denudation that there was not very much 
left for the marine abrasion to cut away in order to form the strandfiat. 
He thinks that this last quantity of rock may thus have been at most 
ten per cent of the quantity calculated by J- H. L. \'ogt as having been 
cut away by marine abrasion. I would be inclined to reduce even this 
figure considerably. 

In his recent publication "Geomorphological studies in Norway" 
[1919] Hans W:s o n A h 1 m a n n has discussed the nature and genetic 
origin of the Norwegian strandfiat. As his views in this respect differ from 
those of previous writers I shall have to mention them at some length ^. 

Ahlmann arrives at the somewhat startling conclusion that a Nor- 
wegian strandfiat (or "coastal plain" as he calls it) does not really exist. 
In the first part of his paper he assures us that the formation, previously 
called so, has nothing to do with marine denudation. It is a base-levelled 
plain formed solely by subaërial denudation, in some places assisted by 
glacial erosion. He therefore thinks that the name of "coastal plain", or 
strandfiat, for this formation is inadequate, and proposes to call it "the 
distal base-levelled plain". 



^ Just as this manuscript is going to press I have received from Prof. A. G. H g - 
borna paper [1920] discussing Ahlmann's views. Hogbom's points are to a great 
extent the same as mine, but he has also mentioned some other sides of the subject 
which I have not paid attention to. 



IKIDTJOK SASHES. M.-X. Kl. 



Tlierc seems, liowcvcr, lo he no sufficienlly wcij^Hity reason for such 
a clian/^e of name, e\en if there lia'l been some prohahilit}' in favour of 
,'\.'s tlieor\' of the i^cnelic origin of the stranilflat. which is not the case 
according to the view of the present writer. 

A.'s theory is not so revolutionary as he seems to tliink. Some 
previous writers, especially A. (1. Høgbom [1913] and the present writer 
[1904, 1905], had suggested that the strandfiat had been formed chiefly 
by the conjoint action of subaërial denudation and marine denudation, 
and also to some extent glacial erosion. By far the greater part of the 
denuding work was suppcjsed to have been accomplished by subaërial 
denudation, while the marine denudation was supposed to have finally 
planed off the low coastal zone of islands and peninsulas, and had thus 
developed that nearly horizontal plane of the strandfiat which is so very- 
conspicuous in many regions of the coast. Now A. postulates that this 
finishing planing by the marine denudation is not necessary, as the sub- 
aërial denudation alone may have accomplished the whole work. His 
evidences are, however, hardly convincing, as we shall see. 

Like most geomorphologists A. considers marine denudation to be 
solely due to wave erosion, and as, according to his view, the w'aves can- 
not possibl}^ cut very broad and nearly horizontal plains along an open 
undissected coast, nor can have much erosive fores along the protected 
shores in narrow sounds, inlets, and fjords, a strandfiat or "coastal plain" 
cannot be formed by marine denudation. 

His argument is that the formation of a strandfiat in this manner 
would be such an extremely slow process, that, during the long time 
required, the coast land would be planed down to base level by subaërial 
denudation. He therefore considers it to be probable that, what is called 
the strandfiat, has been formed in the latter manner, during preglacial 
time before the coast had been dissected by the many deep fjords and 
channels, deepened during the glacial periods, and now traversing this 
plain. 

Already on this point A. seems to come into a serious conflict with 
himself, in as much as, in the last part of his treatise [1919, pp. 237 — 239], 
he describes very broad and nearly horizontal plains in northern Norway, 
in the regions of Vaeroi, Røst, and probably Træna, which he thinks are 
"unmistakable results" of marine denudation (i. e. wave erosion). On 
Væroi, these plains are cut as much as two or three kilometres landwards, 
into the mountain side, forming cirques one kilometre and a half broad. 

He says [19 19, p. 239] that on Rost "marine abrasion and other 
atmospheric agencies of destruction have broken down all land so that 
only isolated parts survive". He thinks that this "took place during the 
last Ice Age outside the inland ice. At that time too the subaërial destruc- 
tive agencies in the regions, situated just outside the inland ice, were 
undoubtedly extraordinarily powerful. The land area was rapidly broken 



I92I. No. II. THE STRANDFLAT AND ISOSTASV. 9 

down tlirough the combined force of atmospheric weathering and marine 
abrasion". 

A.'s views as regards the formation of these coast platforms are 
obvionslv irlentical with the views of the formation of the Norwegian 
strandfiat held by the present writer [1904] and by Hogbom [1914] and 
are expressed almost in the same words. It is, however, hard to see why 
A. assumes that the joint action of marine denudation and atmospheric 
weathering, so effective in this special region, and during so short a t'me 
as the last glacial period, should have had practically no effect during the 
same period, not to speak of the much longer preceding glacial periods, 
along the rest of the Norwegian coast, where he assumes that there was 
also a border lying "outside the inland ice". We may naturally ask, what 
have these agencies, so effective on \'æroi, Rost, and Træna, been doing 
in other regions during all that time with climatic conditions favourable 
for erosion? Is it concei^al)]e that they should have left no traces of their 
activity? 

As far as I can see, no answer to this question can be found in A.'s 
paper. He assures us that A'ærøi Røst, and Træna are very like each other, 
and are "markedly different from the rest of the coast region of Norway" 
without explaining what this marked difference chiefly is. If it is an 
exceptional evenness of the strand flat, this might seem to be sufficiently 
explained by the fact, also pointed out by A., that these islands were pro- 
bably not covered by ice, or have at least not been much attacked by ice 
erosion, during the last glacial period, while other regions of the strand- 
flat have l)een more or less eroded by glaciers. And what is then to be 
said about other parts of the Norwegian strandfiat, which are also very 
level — c. g. on Sandoi, south of the mouth of Sogne Fjord, or in the 
regions of Smolen and Froia, Bronoi, Heroi, Donna etc.? 

It is also difficult to see any marked difference between the strandfiat 
of the Lofoten Islands, Fig. i and 2, and the strandfiat along the coast 
of the mainland. Fig. 3. 

A.'s views as regards the importance of marine denudation, do not, 
however, seem to be quite consistent, for mentioning the region of Smolen. 
west of Trondhjem Fjord, he says that he does not "wish to deny that 
abrasion has at some time occurred here, but only as a final smoothing 
process. The inland ice has also been of great importance in the planing 
of the ground" [19 19, p. 197]. The question is how great the importance 
of this "final smoothing process" has been? It can hardly have been 
insignificant if it is chiefly of preglacial age, and has been able to survive 
the erosion of the glacial periods. But it is very difficult to understand 
why the marine denudation has only occurred "at some time", if it is due 
to wave action? One would expect that the waves have always been at 
work along the coast of Norway, where it was not covered by ice. It also 
seems liighly improbable that the inland ice has had a planing effect upon 



lO riilDTJOF NANSKN. M.-N. Kl. 

the grouiul. As a rule its effect has been the other way, as will be 
mentioned later. 

In his anxiety not to admit too much to the effect of marine denu- 
dation, A. assumes that only "the imiermost part" of the strandfiat 
("shelf") on Værøi (and Rost) has been thus formed, while "the extensive 
outer part, \\liich now lies beneath the surface i^{ the water" is a pc-rfectlv 
different formation, due to subaerial denudation, and belonging "to the 
initial topography as a foreland, equivalent to that round the islands of 
Vesterålen and Lofoten" [1919, p. 238]. 

How is this to be understood? Is the outer part of the strandfiat 
a preglacial formation while the inner part is late glacial, and is it to be 
supposed that the levels of these two platforms, formed in so entirely 
different manners, and during periods so remote from one another, should 
coincide to such a degree, that the one platform forms a direct continu- 
ation of the other? This does certainly not sound very probable. Or is it 
after all so that this initial foreland has also been "smoothed" b}- marine 
denudation? If so the effect of this smoothing process may have been 
quite considerable, unless we assume that the level of the shore-line during 
the long preglacial time happened to be nearly the same as, or slightly lower 
than, that of the late-glacial shore-line. 

In order to avoid misimderstanding it may at once be pointed out 
that I do not attribute so much planing effect to the wave erosion as A. 
does in the case of A^æroi and Røst, and probably Træna. Though im- 
portant the erosion of the waves may have been during the enormously 
long time they have had to work in, still I hold that during the glacial 
periods, and during the cold time preceding them, the shore erosion by 
frost has been much more effective for the planing of the strandfiat along 
the Norwegian coast, while the chief importance of the wave action has 
been its transport of débris from the shores. 

The topography of the low and flat Radøi, north of Bergen, which 
Ahlmann describes in much detail, is in his opinion a convincing evidence 
proving that the strandfiat is a base-levelled plain formed by subaërial 
denudation, without the aid of marine erosion. He describes, however, 
two distinct levels of this strandfiat which is in conformity with what 
has been observed in other regions of the Norwegian coast, and these 
levels are in some places very conspicuous as will be described later. 

Quite apart from the improbability that planes, as horizontallv level 
as these, can be formed by subaërial denudation alone, it is hard to see 
how two such distinctly different levels, in some regions appearing as 
nearly horizontal benches cut in solid rock, have been formed by base- 
levelling. If we imagine that the shore-line has remained fairly stationary 
during very long periods (in preglacial time?), first at the upper level, 
and later at the lower, and obviously younger level, it might be expected 
that the subaërial denudation, while base-levelling the land towards the 



1 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. II 

lower level, would wear away more or less the traces of the upper level, 
and would produce a gradual transition from it towards the lower one. 
But this is not the nature of the two levels. 

By marine denudation, working conjointly wnth subaërial denudation, 
we get a simple explanation of the two levels of the strandfiat. The plane 
of the lower one is cut backwards under the upper one, and both may exist 
simultaneously, because the latter one will not disappear till it is entirelv 
cut away by the lower level; and a more or less abrupt transition between 
them may be found, unless it has been much modified by later erosion. 

An important reason why A. thinks that the Norwegian strandfiat has 
not been formed by marine denudation, is that a plane thus formed could 
not slope so very gently from its inner margin, at the foot of the moun- 
tains, towards the outer skerries, as does the strandflat. Its inclination 
must have become steeper, for else the weaves would not have got suf- 
ficiently deep water to work in. Nevertheless he thinks that the innermost 
parts of the very flat platforms on A'æroi, Rost, and probably also on 
Træna, are formed by wave erosion. A. is obviously right in his view 
that w'ave erosion alone cannot, as a rule, form wide planes sloping as 
gently as these; but he has not considered the effect of shore erosion by 
frost. Where this process, as described later, plays the leading rôle in 
marine denudation, wide and nearly horizontal planes, like those of the 
Norwegian strandflat, may certainly be formed along a much dissected 
coast. 

On the other hand A. objects that, because in some places, c. g. in 
the region of Rorvik, north of the Trondhjem region [1919, p. 197], the 
low coastal border-land is not flat, there is no strandflat. For if this 
"had been formed, wholly or mainly, through marine abrasion, the plain 
would everywhere have the same main character," but "at Rørvik, and 
at many other places, there occur, in complete contrast to this, all stages 
from the most broken topography to the groups of small level islands." 

This is, however, just what might be expected to be a characteristic 
feature of a strandflat formed by the joint action of subaërial denudation 
and marine denudation, and also glacial erosion, along a much dissected 
coast. Its degree of evenness will naturally depend on the degree of 
maturity to which it has been developed. The less mature the more 
ridges, hills, and stacks will surmount the general marine level. Where 
the rocks are relatively resistant, or where initially relatively great masses 
of rock surmounted sea-level, many islands and hills mav remain more or 
less incompletely levelled, and their summits may vary much in height. 

On the whole, the surface of the strandflat may seem to be remarkably 
level, considering that it has also been exposed to considerable glacial 
erosion, perhaps during several periods, which has scoured the islands 
and hills, rounding off their edges and sunnnits, and also considering 



FRIDTJOF NANSEN. M.-N. KI. 



lliat the straiidflat has been dcvchjpe'l rlurinj^ l'J"^' pcriofls, with fliffcrciit 
levels (jf the shore-line. 

A most characteristic feature in tlic topography of the Norwegian 
strandfiat, is its incision into the mountain side of the lanrl behind, forming 
an often sliarply marked line of demarkation between the level strandfiat 
and tlie mountains ascenrling abruptly anrl steeply, often hundreds of 
metres. As far as T have seen, .Ahlmann has made no serious attempt 
to explain how such a horizontal incisifMi could have been formed by the 
vertically working sul)aërial denudation, although this would naturally 
have l)een of essential imi)ortance, if his theory should have been made 
more plausible. Nor does he try to explain how the vertically working 
subaërial denudation alone can possibly develope such extremely flat, 
and nearly horizontal i)lanes, as occur c. g. in Smølen, Frøia, Heroi, 
Donna, Risvær, and .Solvær. In the case of Smolen he himself has ob- 
viously had a notion that it was necessary to open some opportunity of 
a finishing touch by a horizontally working agency, the marine denudation, 
as was mentioned on a previous page (9), but, as if to weaken the effect 
of this admission, he also draws in the erosion of the inland ice, as having 
a planing effect, which I consider very unfortunate, as will be discussed 
in a later chapter. 

If the assumption that the strandfiat is a base-levelled plain, due 
to subaërial denudation, be correct, we might expect that this process 
would have had the greatest facilities in developing a broad strandfiat in 
southern and south-eastern Norway, where tlie initial land was low, and 
sloping gently towards the coast, — while it would require infinitely longer 
time on the initially high west coast, especially if it be assumed, with A., 
that his "distal base-levelled plain" was mainly formed before this coast 
was dissected by the many deep fjords, channels, and valleys of the glacial 
periods, i. c. at a time when a very much greater mass of rock had to be 
worn away before the land could be base-levelled. 

We find, however, just the contrary: an often broad and well developed 
strandfiat along the steep and high west coast of Norway, while there are 
only slight indications of such a formation along the southern and south- 
eastern coast. 

It is, however, not always easy to grasp what A. exactly means by 
his base-levelled plain, for in some regions, c. g. in southern Norway, 
it may have a relief of a hundred metres above the sea, and in that case 
it would still be much rock to plane away before we reach the strandfiat. 

If the Norwegian strandfiat is solely a "base-levelled plain," it would 
also be extremely difficult to understand why subaërial denudation has 
not developed similar strandfiats along anv old coast, where the coast-line 
has remained stationary during a time necessary for base-levelling, and 
why it is that, on the contrary, typical strandfiats occur chiefly in regions 
that have been covered by ice-caps, or have at least had very cold climates 



I92I. No. II. THE STRAXDFLAT AXD ISOSTASV. I3 

We might have expected to find this "distal base-levelled plain" most 
perfect in those milder regions, where its surface has not been attacked 
by glacial erosion. A. has made no attempt to help us out of this serious 
difficulty. 

It might be objected that along the east coast of India there is a very 
broad (up to j^ kilometres broad), extremely flat and low plain, backed 
bv steep mountain walls. But unfortunately, this is just a region with 
A-erv little rain-fall, and where therefore the subaërial denudation has been 
so insignificant that this magnificent plain of marine abrasion could be 
developed so perfectly, and remain relatively undisturbed, and backed by 
oversteepened hills, because it was only sligthly attacked by subaërial 
denudation. 

A.'s views as regards the formation of his "distal base-levelled plain" 
has been well expressed in his description of the region of Smolen, where 
he says [1919, p. iÇ/j: "On the mainland opposite Smolen there occurs 
a broad denudation surface with about the same height above sea-level as 
that on the island, and abruptly attached to a steep fell-side about 500 m. 
high. In certain places this surface seems to continue in islands and 
terraces at the side of fjords, which towards the east pass into a mature 
valley-generation." From this he draws the conclusion that the broad 
denudation surface of Smolen and the mainland has been formed by base- 
levelling in the same manner as the floor of his base-levelled valley 
generation, which he describes so well. His view is obviouslv that the 
broad valley floors have joined together in front of the h'gh land and have 
"formed a peripheric base-levelled plain" [1919, p. 221]. 

To me it would have seemed more logical to argue that, the low is- 
lands and ledges along the sides of the fjords have obviously been levelled 
by the same process as the flat and very even surface of the strandfiat 
of the coast outside, and that this process has been marine denudation 
of some kind, because the islands and ledges in the fjords have, in most 
cases, little resemblance to what might have been expected to be remnants 
of the floors of base-levelled valleys, that have been exposed to the erosion 
of several glacial periods. Besides in several places the floor-level of A.'s 
preglacial valleys differs distinctly from the level of the strandfiat in the 
same locality, e. g. in Sogne Fjord, as will be mentioned later. 

Another difficulty is also connected with A.'s views as regards his 
base-levelled valleys. For the same reasons which the present writer has 
pointed out [1904, pp. 44 f., 54 ff., 151 ff. etc.] A. also assumes that the 
preglacial fluvial valleys of the land has been continued across the floor 
of the now submerged continental shelf, e. g. outside the coast of Romsdal, 
Trondhjem Fjord, and Helgeland, at a time when the land stood about 
250 to 300 metres higher than now. A. is obviously of the opinion that 
tliese valleys too were base-levelled; but if so how is this fact reconcilable 
Avith his theory of the valleys and strandfiat as having been base-levelled 



14 FKID'IJOF NANSF.N. M.-X. Kl. 

at about present sea-level? If, after that great elevation (250 — 300 metres 
hif,Hier than now) when the valleys were l;ase-levelled on the continental 
shelf, the land was suhnierj^a'fl, the valleys were naturally als(j more or 
less submerged, and their floors would be jjelow sea-level near the coast. 
There would tlius be no possibility of base-levelling the valleys in this 
coast zone, as they were already deepened below base-level. 

It might then be assumed tliat the elevation of the land occurred 
after the valleys of the present land-surface had been base-levelled and 
'the distal Ijase-levellerl plain' had been formed. But if so the land must 
have stood at that higher level for a considerable time, sufficiently long 
for the base-levelling of the fluvial valleys of the shelf, and after that 
the land has again sunk to its previous level, or the shore-line has returned 
to the same level wdiich it had during the very long period wdien A.'s old 
valley generation was base-levelled. But what kind of movements is it 
that has changed the level of the shore-line in this peculiar way, and after 
such a long time brought it back to its original level? A. cannot be seen 
to have considered this difficult question, which his theory must inevi- 
tably raise. 

There is also another difficulty which he does not mention. He 
thinks that the inner part of the continental shelf "constitutes an imme- 
diate continuation of the coastal zone, thus l^elonging to the peripheric 
base-levelled plain" [1919, p. 211]. But how is this to be understood? 
Was the plain of the continental shelf base-levelled at the same time as 
his old valley generation further inland was base-levelled? But the level 
of the latter, is now^ near present sea-level, and is considerably higher 
than that of the former. How then could extensive formations, so sharply 
marked as the strandfiat, and the floors of his old valley generation, be 
so well developed at this base-level, when there was also another lower 
base-level (before or after?), represented by the inner part of the con- 
tinental shelf? A.'s ideas do not seem to be very clear on this point. 

His views of the genetic origin of the continental shelf shall not be 
taken up for discussion here. A factor of much importance for this and 
other questions dealt with in A.'s treatise, is the isostacy of the earth'.^ 
crust, which, however, he does not mention. 

The conclusions arrived at concerning Ahlmann's views of the nature 
of the Norwegian strandfiat, may be summarized as follows: 

A. is right in assuming that the subaërial denudation (i. c. atmo- 
spheric weathering, frost disintegration, and fluvial erosion) has been of 
chief importance for the denudation of the Norwegian coast land, as well 
as for the land slope within. This is in full accordance with the views 
of several previous writers. 

A. is wrong in assuming that the Norwegian strandfiat, forming often 
a sharply marked horizontal incision in the mountain slope, has been 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. I5 

developed solely by base-levelling of the subacrial denudation, without 
the aid of marine denudation. 

A. is also wrong in assuming that the present strandfiat has been 
formed mainly in preglacial time before the coast land had been dissected 
bv the numerous fjords and channels now traversing the strandfiat. 

Profiles of the strandfiat and the high land behind it. like those given 
in Chap. VII and \'III, and also the numerous profiles given in my report 
of 1904 [PI. s XII — XX], demonstrate how very level the strandfiat may be 
in some regions, and what sharply defined horizontal incision it may form 
in the steep mountain side of the land. It is hardly conceivable that level 
planes like these, can have been formed exclusively by a vertically working 
process, like the subaërial denudation, ^^"e must assume that they have been 
finally levelled by some process working horizontally. 

The writer has recently [1920, 1921, Chap. VIII] mentioned several 
processes that may have a planing effect upon a land surface, even above 
sea-level. But their levelling effect is small compared with that of marine 
denudation in cold regions, where an active shore erosion is produced by 
the disintegrating activity of the frost, in and just above the shore-line, 
as will be described later. 

Let us, however, first examine more closely what importance the 
various denuding processes may have for the denudation of the coast land 
and for the development of the strandfiat. 



i6 



IKIDIJOF NANSEN. 



M.-.\. Kl. 



III. THE EFFECT OF THE SUBAËRIAL DENUDATION 
IN THE COASTAL ZONE. 

If a land-surface sloping- f^Tarlually towards the coast, be exposed to 
subaërial deniulation duriniL,^ sufficiently long periods, and witliout being- 
disturbed by other processes, it is obvious that a base-levelled plain may 
finally be developed along the coast, and according- as the sloping land- 
surface is denuded, this plain will extend landwards. Fig. 4 demonstrates 
very roughly the development. S L is sea-level and gab the initial land 
slope, which by subaërial denudation is lowered to c d, and then to e f, 
and the base-levelled plain a c c is thus developed. It is here supposed 
that the rate of suliaërial denudation as well as the resistance of the rocks 
have been the same in all regions of the land slope, which of course is, 
not the case, as a rule. The transition between the land slope and the 
base-levelled plain would naturally be very gradual, and there would not 
be a sharp break as in the figure. 



d 




S 




a^ '^ c ^ 


^ e-^^^ 


L 


S' 


J- 


' ' ' ' '-«*^^ 


^^^^ 


L\ 


^" 


-o- 


^^^^^'^^^ 




L" 



Figs. 4 and 5. 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. I7 

It has, huwcver. to he considered tliat the suhaërial denudation is 
working, not only in the coastal region, like the marine denudation, hut 
over the whole land surface simultaneously. As will be mentioned in a 
later chapter, the resulting removal of rock from this great area, will 
necessarily cause a corresponding isostatic elevation of the land, and this 
will change the above process. In stead of the three stages a b, c d, and e f 
of Fig. 4, we will find that when the land surface was denuded, from a b 
to c d the land would be elevated very nearly the same he'ght, ;'. c. the 
shore-line in Fig. 5, would stand at S"L' in stead of at 5 L, in relation 
ID the land, and the coastal plain would be developed to form th? profile c /;. 
The rock-surface would then follow the line g'h c d. 

When the land surface is denuded to e f, the land will rise so much 
that the shore-line will be lowered to S" L" in relation to the land. The 
coastal plain will be developed to form a slope as indicated by the line i e. 
As the sea floor has not been eroded, it should not be elevated, onlv its 
inner part g"i will be raised with the rising coast, owing to the rigidity 
of the earth's crust. 

The final result will be that the slope of the land, the coastal plain, 
and the slope of the sea floor will be something like what is represented 
by the line g"icf with sea-level at S" L" . 

It might be objected that the coastal region g"i e will thus be elevated 
above its earlier level. It has, however, to be considered that this zone is 
lifted with the whole land, and the surplus of mass elevated in the coastal 
zone, above and below sea-level, is more than compensated for bv the 
defeceit of mass elevated in the region of the land inside. 

The coastal plain resulting from this process will rise inland, and 
will be continued in the land slope with a perfectly gradual transition 
between them, as demonstrated by Fig. 5. 

A last upheaval of central and western Norway, in relation to 
sea-level, may perhaps have ocurred in earl}- Tertiary time (Eocene), or 
may have been finished in Miocene. During the many millions of years 
since then, the subaërial denudation has been working continually on the 
surface of Norway. 

During this long time the coast land has probably been much denuded, 
and a plain with a fairly gentle slope may have been more or less developed 
by the process described above. But a low foreland, similar to the strand 
flat, could not be developed in this manner, except perhaps in regions 
where the coastal border is Iniilt up of rocks with relatively littl-? power 
of resistance. Rut even there the surface could hardlv be planed so verv 
flat by this process alone. 

If the shore-line oscillated vertically during this long preglacial time, 
and provided that there had been no marine denudation, the slope would 
become steeper in the outer coast land which by submergence had been 
less exposed to the subaërial denudation. 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 11. 2 



KRID'IJOl- NANSKN. M.-N. Kl. 



huriiii;- llic j^-laci.'il pcritjfls a ^n-ni change was, however, produce'] in 
these c(jn(Iiti(ins. T\\v \alleys were niiicli deepened by the j^-'lacial erosion, 
the coast land was nuudi dissected l)y fjords anrl sounds, and the outer 
coast was cut up into narr. 'w ]),iiiiisulas and tliousands of islets. Idie 
suhacrial denudation was also very niucli increased by the conditions of 
the glacial climates. 

The result must be ha\e been that, when llu- much dissected ceast 
l.'md was not covered by the inland ice, the subaërial denudation, before 
and during each glacial epoch, had a greatly increased power to denude 
the many islets and narrow peninsulas towards sea-level. The climatic 
conditions then greatly favoured the denudation, because the temperature 
was low, the frost was very active destroying the rocks, and w-as assisted 
in its disintegrating work by small local accumulations of snow in the 
hollows of the rocks on the coast laufl, remaining during a great part 
of the summer, and keeping the joints anrl fissures of tlie surrounding 
rocks alwa3's filled with melting-water which would split the rocks 
when again freezing during the frequent daily oscillations of temperature 
above and below freezing point in tliis region of melting snow and cold sea. 

As I have pointed out before (cf. p. 6), the waste would comparatively 
easily be carried away down the often steep slopes on these small islands 
and peninsulas of the much dissected coast, where the distance to the sea 
is very short. In addition to the transport of the waste by running water, 
there may also have been a considerable transport of matter by the down- 
ward movement of the soil, consisting of water-soaked mud or gravel or 
even stones, along the slopes. This movement is caused by the freezing 
and melting of the water in the soppy soil and gravel. This transport is 
very common in Arctic lands, as has especially been described by Joh. 
Gunnar Andersson and Bertil Hogbom [1914J, and may often be con- 
siderably more effective than the transport of waste by running water. 

In this manner the rock surface of the dissected coast land has been 
covered by comparatively little waste, and was therefore continually ex- 
posed to fresh attacks of the subaërial denudation, especially that caused 
by the frost. 

All these processes mentioned above may now be studied in full 
activity on Spitsbergen. 

It has obviously also been of great importance that each time the in- 
land ice or the local glaciers of the glacial periods advanced over the coast 
land all waste and loose material on the land, as well as in the fjords and 
sounds, would be carried away seawards, and when the glaciers again 
retreated all valleys and sounds had been reopened and fresh rock surfaces 
were exposed to new attacks of a vigorous subaërial denudation. During 
the various glacial epochs, while the margin of the inland ice was standing 
near the outer coast land, there were probably many such oscillations of 
the glaciers [cf. Nansen, 1904, pp. 109, 113]. 



1 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. I9 

The final denuding effect of the subaerial erosion during the glacial 
periods must therefore have been comparatively very great in the much 
dissected coastal regions, and the islands and peninsulas wer.^ compara- 
tively rapidly denuded towards sea-level. 

It may also be possible that this effect of the subaerial denudation, 
was somewhat greater in the coastal regions than farther inland. But it 
is hardly possible that, under otherwise equal conditions, an abrupt break 
in the land-slope could thus arise along the boundary between the coastal 
region and the mountains of the land inside. A strandfiat with a practi- 
cally horizontal level, like what we now find along the Norwegian coast, 
could therefore not be formed in this manner without the effective as- 
sistance of other processes. 



20 ll'.IDTIDT NANSFN. M.-X. Kl. 



IV. GLACIAL EROSION. 
The Erosion of the Glaciers of the Inland Ice. 

It wc sliall try to form some idea as to what the importance of the 
j^^lacial erosion may have l)een for the devehjpment of the strandflat, it 
would be necessary to examine the mode in which the glaciers erode under 
different circumstances, and their erosive effect upon the land surface in- 
different regions. There will be no opportunity of making such a com- 
prehensive investigation liere. We must lim't ourselves to some points of 
importance for our special subject. 

There is a striking difference between, on the one hand, the remarkab'v 
small denuding effect which the inland ices of the glacial periods have had 
on the relatively level surfaces of central and eastern Norway, and the 
greater part of Sweden, where, in many places, the depth of glacial erosion 
may only amount to several metres, and, on the other hand, the con- 
siderable dissecting work performed by the glaciers of the ice ages in the 
coastal regions of Norway, where the valleys and fjords have been deepened 
hundreds of metres. 

This remarkable difference of effect luust be chiefly due to differences 
in the velocities with which the eroding ice has moved over the rocks, and 
not appreciably to differences in its thickness and the pressure therebv 
exercised on the underlying ground. 

A. G. Hogbom [1910, p. 435 f.] and other writers are obviously right 
in assuming that the erosive power of a moving glacier is much less due 
to its scouring effect upon the underlying rock tlian to its power of 
breaking loose (plucking) stones from the rocky ground, and this power 
obviously increases rapidly with the velocity of the ice motion. The ero- 
sive effect of a glacier will also to some extent depend on its power to 
move and transport stones and rocky material along the ground. ]Moreover. 
a sufficiently thick glacier will have a disintegrating effect on the under- 
lying rock by the alternate melting and freezing of water on its under sde, 
caused by the frequent changes in the stress and pressure of the ice movirg 
over the rough rock surface. 

I have tried to investigate quantitatively the various manners in which 
moving glaciers may erode the underlying ground. The results of these 



19^1- No. II. THE STRANDFLAT AND ISOSTASY. 21 

investigations will be given in a later paper. It is difficult to estimate 
the total erosive power of glaciers, as it will greatly depend on the tectonic 
structure of the underlying rock. Where the rock surface is relatively 
easily smoothed, and few edges and projecting stones are left for the 
flowing ice to attack, the erosive work will to a great extent be reduced 
to grinding, the effect of which will depend on the quantity of stones and 
grit transported by the glacier along its under side. 

The erosive effect will in that case be very much less than in places 
where the rock surface remains rough, and where there are always pro- 
jecting rocks to be attacked and broken loose by the flowing ice. 

I have come to the conclusion that, on the wdiole, the erosive power 
of a moving glacier may increase opproximately with the third power of 
the velocity of the motion at its under side. 

Hence we may expect that if this velocity be increased ten times, 
the erosive power of the glacier may be increased about a thousand times. 
A glacier which, with a velocity of ten centimetres in 24 hours, was able 
to denude the underlying ground to an average depth of ten centimetres 
during a certain long period, would then be able to excavate a hundred 
metres of rock during the same period, if its velocity be increased to one 
metre in 24 hours, and provided that the glacier were working onlv 
vertically. 

The observed velocities of the moving glaciers of Norway and the 
Alps are generally between o.i and 0.4 metre in 24 hours, while on the 
big glaciers of the Himalaya velocities of between 2 and 3.7 metres in 24 
hours have been observed during the summer. At the end of the big 
glaciers of northern Greenland, however, velocities of as much as 20 to 
30 metres in 24 hours have occasionally been measured by Amund Hei- 
land, Stenstrup, Hammer, and Ryder, during the summer. The normal 
movement during the whole year is less, but still very considerable. There 
must consequently be an enormous difference between the erosive powers 
of these various glaciers. 

If we take a European glacier, moving with a velocity as great as 
0.5 metre in 24 hours, and a Greenland glacier, moving with an average 
velocity of 10 metres in 24 hours, we find that, while the former is able 
to erode 10 centimetres of rock, the latter will erode 800 metres. 

This explains why the inland ice has had so remarkably little erosive 
effect in its more central areas, in Norway and Sweden [cf. Høgbom, 19 10], 
while its outflowing glaciers have had such enormous effect in the narrow 
valleys and fjords of the coast land, just outside these regions. 

We also understand why there are so sudden and striking differences 
between, on the one hand, the great deepening effect of the glaciers in the 
narrow fjords where they may have eroded a thousand metres of rock or 
even more, and, on the other hand, the quite insignificant erosive effeci 
of the inland ice which covered the high land on the sides of these deep 



I KlinjOK N'ANSKN. M.-X. Kl. 



fjords, where the inilial i)rej^lacial ("palæic"; mountain surface rnav still 
be recognizable. 

In the above manner we alsf) obtain a natural ex|)lanation why the 
fjords are so very narrow, deej), and often winding-, in regions of hard 
and resistant Archæan and igneous rocks, in western Norway and in 
Greenland, while in regions with weak'cr rocks, tliey are wider, less deep, 
and often radiating (Ait from the central region of the land, as they do 
on Iceland, in Finmarken, and partly on Spitsbergen. 

In regions composed of resistant rocks, the big, inoving glaciers of 
the glacial periods eroded chiefly vertically. The horizontal erosion, by 
frost on their sides, has been comparatively small. They were guided by 
the initial preglacial valleys, formed by the subaërial denudation greatly 
along lines and zones of fracture and weakness. The glacial erosion 
followed the same lines. As the outflow from the inland ice converged 
into these narrow channels, the velocity of their ice streams became very 
great, and the vertical erosive power of the latter enormously intensif ed. 

In regions of less resistant rocks, the initial preglacial valleys were 
broader, the glaciers would be more able to take their own course, and 
their erosion would not be so exclusively vertical, but also to some greater 
extent horizontal, especially because the mountain slopes on their sides 
were much more rapidly destroyed by frost. The glaciers would thus 
widen out, be less deep, and would obtain much smaller velocities. 

A common effect of this erosion in such regions will therefore be 
that the fjords become comparatively shallow and wide, and they often, 
as on Iceland and in Finmarken, become wider towards their mouth, while 
in regions of resistant rocks they are generally narrower towards their 
mouth than further inland, wdiere they are excavated into much deeper 
troughs, or true rock basins. 

It has already been pointed out on a previous page that the glacial 
erosion has been of much importance for the development of the strandfiat 
by dissecting the coast and splitting it up into innumerable peninsulas, 
islands, and skerries. Thus the coast line, or the line of attack, has been 
enormously increased, and the subaërial denudation, as well as the marine 
denudation, have been greatly facilitated. 

Otherwise the erosion of the inland ice, and its outflowing glaciers, 
can have been of no direct importance for the planing of the horizontal 
level of the strandfiat, as far as I can see. 

When the big glaciers carved or deepened the fjords and sounds tra- 
versing the present strandfiat, the islands and peninsulas between them 
may also have been denuded to some small extent where they were covered 
by the moving ice. But the general denudation thus accomplished, cannot 
have been considerable, as the movement of these marginal ice-sheets be- 
tween the glaciers of the deep channels must have been relatively slow. 
As the velocities of this motion must have varied much locally, the depth 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 23 

of this erosion must also have varied very much; and no plane could be 
developed in that manner. 

It has been maintained that the many low skerries of the strandfiat 
may have been planed off to their present level by the inland ice. There 
is, however, no reason why this erosion should stop at sea-level. 

It has been a quite common mistake amongst geologist to assume 
that the erosive power of a glacier should be reduced by a partial sub- 
mergence of the glacier into the sea, because the pressure of the ice on 
the sea floor, as compared with that on dry land, should bs reduced bv 
the boyancy of the submerged volume of the glacier. But, as was pointed 
out by G. K. Gilbert [1904, pp. 10 f.] and myself [1904, p. 163, footnote], 
the pressure of the ice on the underlying ground will naturally be the same 
whether the ice be submerged or not, and provided that the velocity of its 
movement be the same, there can be no difference in the erosive power of 
the glacier, as long as it actually rests on th? ground. The difference will 
come in the moment that the glacier begins to float in the sea, and a layer 
of liquid water actually intervenes between the under side of the glacier 
and the sea floor. The chief effect of submergence will otherwise be 
that the outward pressure in the mass of the glacier will be counteracted 
by the pressure of a layer of sea-water equal in depth to that to which 
the glacier is submerged, and its velocity may thus be reduced. 

Provided, however, that the velocity of its motion remains unaltered, 
a glacier will have the same erosive effect upon the underlying ground, 
whether above or below sea-level, down to a depth below the latter equal 
to nearly nine tenths of the tlrckness of the ice, at which depth the glacier 
begins to float and is lifted from the sea floor. 

This erosion cannot therefore be limited by any definite level, except 
that determined by the depth at which the glaciers float. But as this depth 
must have varied much, owing to the varying thickness of the glaciers, 
no horizontal level of erosion, like that so distinctly exhibited by the sker- 
ries and islets of the strandfiat, could possibly be produced by this glacial 
erosion. 

On the whole, it is hardly conceivable that an inland ic2 can have a 
planing effect, even where it moves over a fairly flat land surface. It may 
break away projecting rocks, and more or less smooth the rock surface; 
but it will soon move with greater velocities along certain lines than along 
others, especially along the initial fluvial valleys, even though they may be 
ever so broad and shallow. As the erosive power of the moving ice in- 
creases so much with its velocity, this will lead to a deepening of these 
depressions or valleys, with an increasing rate as time goes on. The final 
result will be that the initially flat land surface is dissected by deep glacial 
valleys and channels. 

In the region of the Norwegian strandfiat a planing effect of the 
moving inland ice seems to me to be especially impossible. The inland ice 



24 



IKIDTJOF NANSKN. 



M.-.\. Kl. 




---Tf°S&fc=s:--^^^ 



Fig. 6. Cirquc-gUiciers on the west coast of Cross Bay, Spitsbergen [from Nansen, 1920]. 




Fig. 7. Future picture of the coast in Fig. 6 [from Nansen, 1920]. 

would here move with consirably greater velocity along the fjords and 
sounds then over the islands between them, and it would go on deepening 
them at a very much greater rate than it could erode the surface of the 
islands. The velocity of its movement along these channels may have gone 
on increasing with increasing depth, and its erosive effect would increase 
very much more. How then would it be possible that the inland ice could 
have planed the surface of the islands and peninsulas to practically the 
same level, while it has excavated the sounds between them to very 
different depths? 

The inevitable effect of the erosion of the inland ice in this region 
must be that, not considering its smoothing and rounding of the local rock 
surfaces into roches moutonés, it will make the wliole lanrl surface much 
more uneven than it was before. 

For this reason, according to my view, and as has also so clearly been 
pointed out by Hogbom [19 13, p. 57], it is impossible that the deep sounds 
and channels, traversing the strandfiat and dissecting it into its many 
islands, can have been formed after this plane had been levelled. If the 
sounds were deepened, to some considerable degree, by the glaciers of the 
inland ice, this deepening process could not in that case have left the plane 
between the sounds as undisturbed as it actually has. A glacial erosion, 
to that degree selective, is not conceivable on a flat plain. 

If the sounds and channels have been formed chiefly by atmospheric 
and fluvial erosion, then as Hogbom has pointed out, it becomes still 
more absurd to think that they could have been eroded without the inter- 
vening plain being dissected by deep valleys, sloping towards the floors 
of these sounds. But this is not the case. 



1921. No. 1 1. 



THE STRANDFLAT AND ISOSTASY. 



^D 




Fig. 8. Inner Nor\va\- Island, north coast ot" Spitsbergen [from Nansen, 1920]. 

Any profile of the Norwegian strandfiat gives the impression that the 
land, rising between the sounds and submerged channels, has been truncated 
and i)laned after the latter had been formed, and that there have been 
relatively slight modifications of the relief after this happened. 



The Erosion of Local Glaciers. 

Another form of erosion is due to the activity of small "botten"- 
glaciers or cirque-glaciers, and of snow and ice accumulated in depressions 
and hollows of the rock surface, and remaining during the summer, or 
at least a greater part of it. Along the edges of these glaciers and snow- 
accumulations, the frost will have a strong disintegrating effect on the 
rocks during the season when the snow is melting, and when there are 
perpetual oscillations of temperature between frost and thaw. All joints 
and fissures in the rocks are here kept full of melting wtaer, ready to burst 
the rocks at each frost, that may even occur several times during the day 
and night. The rocks are therefore rapidly eaten away and thumbling 
flown on to the glaciers or snow-heaps w^hich carry them off. 

This erosion is thus a combination of glacial erosion and atmospheric 
weathering (mainly by frost) which is especially powerful in glacial cli- 
mates, and may be studied now at full work in Spitsbergen [cf. Nansen, 
1920, 1921, Chap. VIII]. 

Fig. 6 represents cirque-glaciers now at work on the west coast of 
Cross Bay, Spitsbergen. Fig. 7 illustrates the probable future effect of 
this erosion when the cirque-glaciers have got time to cut through the 
mountain walls beliind them. 

As was already pointed out by Amund Hellanrl and Lorange, there 
must also be a very active vertical erosion under the cirque-glaciers 



26 



IKIDTJOK NANSKN. 



M.-N. Kl. 









YWs. 9 & lo. The Buchanan and Murray Glaciers along tlic est 

Tilis may, to some great extent, be due to the alternating melting and 
freezing of the ice and water, at the under side (^f the glaciers, caused by 
sudden changes in pressure, as was mentioned on a preceding page. 
It may also be caused by melting water coming from above in the summer, 
and again freezing under the ice. As Werner Werenskiold has first 
pointed out to me, the pressure under water filling holes in the ice, will b2 
greater than under the ice at the same level, owing to the considerably 
higher density of the liquid water. If the temperature is at melting point, 
the ice may thus melt under the water, and this may continue down to the 
bottom of the glacier, provided that the holes remain full of water. But 
as soon as the pressure is reduced, the water will again freeze under the 
glacier. Such holes, more or less filled with melting water, may often be 
formed along the edges of the glaciers, between the ice and the rock, and 
in this manner the disintegration of the rock may be increased. 

At the bottom of crevasses often formed near the inner edge of the 
cirque-glaciers there may also be frequent changes of temperature above 
and below freezing point, causing rapid disintegration of the rocks 
[cf. W. D. Johnson, 1904]. 

It is, however, obvious that the cirque-glacxrs also erode the ground 
by their movement. Although in most cases the movement of these small 
glaciers mav not be considerable, we find that they have scoured and 
polished the rocks on the floor of their cirques, they have smoothed the 
edges and have plucked stones. Moreover their motion is of importance 
by carrving away débris and detritus formed by the erosion. 

The whole process described above, may have a general denuding 
effect upon the land surface where it works, which is considerably greater 
than the average erosion of the inland ice. There is also this difference, 
that it will be mainly limited to the land surface which is above the sea. 
Given sufficient time, it might therefore be able to denude the coast land 
towards sea-level and m'ght transform it into a fairly low, though uneven 
plain, provided that the climatic conditions allow the cirque-glaciers to 
descend to the sea. In Lofoten the cirques have actually been eroded to 
some depth below present sea-level [cf. Heiland, Thorolf \'ogt, 1912. Ah!- 



192 1. No. II. 



THE STRAXDFLAT AND ISOSTASY. 




;oast of Prince Charles Foreland. Spitsbergen [from Nansen, 1920] 



mann, 1919]. It is conceivable that, where the cirque-glaciers are suf- 
ficiently thick, they may be submerged to some extent, but. as a rule, 
they will not descend much below sea-level. 

It is obvious that, in the manner described above, the local erosion 
of small cirque-glaciers and accumulations of snow on the mountain sides 
along the shores, may help to form low coastal borders, backed by over- 
steepened mountain walls, as demonstrated by Fig. 8. 

I have observed in Spitsbergen that several cirque-glaciers, after 
having eroded themselves down nearly to sea-level, may be widened bv 
horizontal erosion, until they meet and their glaciers unite and form one 
continuous, nearly horizontal ice-sheet, covering a fairly flat shore land 
in front, with a mountain wall ascending steeply behind. Figs. 9 and 10 
illustrate some glaciers of this kind, covering the low shore land along the 
east coast of Prince Charles Foreland. Spitsbergen. 

In the same manner as a cirque-glacier, the flat shore glaciers may have 
an ability to eat themselves backwards into the mountain side, by the frost 
erosion along their inner edge. The remaining ridges between the original 
cirques may thus gradually disappear, and few traces of them may be left. 
As the flat glacier, by the pressure of the snow, each year accumulating on 
its surface, will be moved slowly towards the sea, it may gradually carry 
away the débris. 

In this manner we might imagine that a kind of strandfiat, with an 
abruptly ascending, oversteepened mountain acclivity behind, can be 
formed. But the plain thus arising cannot be expected ever to become 
very broad, nor very level and regular, and there are in fact no indications 
that the strandfiat of Norway, has to any appreciable extent been formed 
in this manner. In southern Norway there are very few traces of cirques 
having ever been formed near present sea-level; though this does not prove 
that it may not have been the case during the Great Ice Age. 

Any how, it cannot be doubtful that, in regions with the necessary 
climatic conditions, as it now is in Spitsbergen, the above mentioned form 
of glacial erosion may have been of importance for the development of a 
low shore land, although it has not been able to form a typical strandfiat. 



28 FRIDTJOF NANSEN. M.-X. Kl. 



V. MARINE DENUDATION. SHORE EROSION 

BY FROST. 

There remains then no other process for the planing of the strandfiat 
but the marine denudation. Most writers assume that this denudation is 
chiefly due to \\ave erosion. But if the wave erosion has been able to 
cut the plane of the broad and nearly horizontal strandfiat along the coast 
of Norway, it must certainly have been able to cut similar planes, at least 
to some extent, along more southern coasts, exposed to the full fury of 
the ocean during sufficiently long periods, even though those coasts were 
less dissected. But, as a rule, no real strandfiat occurs along the coasts 
of milder regions. 

Considering the extremely slow progress of wave erosion alone on a 
coast built up of solid rock, it is conceivable that during the very long time 
wdixh the wave erosion will need for cutting a fairly broad strandfiat, the 
land will be so much denuded and dissected by the atmospheric weathering 
and fluvial erosion, and will be so much raised by the isostatic elevation 
thus caused, that the traces of the strandfiat may be more or less ob- 
literated. 

The striking fact is that the typical strandfiat is a characteristic 
feature of high northern and southern latitudes. It occurs preeminently 
in regions that have formerly been subjected to glacial conditions or have 
at least had very cold climates, like the coast of Siberia. It seems, there- 
fore, to be a natural conclusion that the formation of typical strandfiats 
have, as a rule, had some connection with low temperatures. 

S. W. Gushing [1913] has described a remarkably smooth emerged 
plane of marine denudation along the east coast of India, wh'ch is like a 
strandfiat. It is backed by "an ancient sea wall," rising steeply to an 
average height of over 650 metres above sea-level, in some regions even 
to 2300 metres in an almost vertical wall. Numerous remnants of quartzite 
rise in the shape of steep-sided ridges or stacks, often with flat summits, 
above this plane. Their bases are not unfrequently marked by sea caves, 
and there seems to be room for little doubt, but that this plane, in places 
about 75 kilometres broad, is actually formed by wave-erosion [cf. D. W. 
Johnson, 1919, p. 231]. The plane has been developed over meta 



1921. No. II. THE STRAXDFLAT AND ISOSTASY. 29 

morph.c structures mainly, represented l)y numerous schists, gneisses, and 
quartzites. "Because of its low lying attitude and meager rainfall it 
has been little dissected." 

Pushing does not give the height above sea-level of the inner margin 
of the plane, but it might be inferred that it cannot be much over 70 or 80 
metres; and the plane dips gently seawards to the low, flat shore. 

It is conceivable that, in a region with very meager rainfall, and con- 
sequent slow subaërial denudation, the wave erosion may, in the length of 
time, be able to develop sharply defined planes backed by oversteepened 
sea walls. But in regions with a more abundant rainfall, the subaërial 
denudation will work faster than the wave erosion. During the long time 
which the latter would require for the development of a broad plain of 
marine abrasion, the subaërial denudation would certainly denude the land 
surface much more, and it would wear down the sea wall (the cliff), and 
give it a gentle slope. If the plain of marine abrasion be elevated (by 
isostacy) above sea-level, its surface would gradually be more or less dis- 
sected by the subaërial denudation, and after some time there may be no 
very sharp boundary between the marine plain and the more undulating 
surface inland, especially as the whole land would gradually be elevated, 
by isostatic movement, at about the same rate as it was denudet, and n:-arly 
the same amount. 

Hence, as a rule, we will find no typical and sharply defined strandfiat 
in regions with an active subaërial denudation, unless there have been 
especially favourable conditions for its development, like those prevailing 
in regions exposed to severe climates. 

After having had an opportunity of studying the process of marine 
denudation and its effects more closely, especially in Spitsbergen, I have 
modified somewhat my earlier views [1904, 1905] as to the manner in 
which the strandfiat has been developed (mentioned on pp. 5 f.). I have been 
lead to the conclusion that the wave erosion has been of but little direct 
importance for the planing of the strandfiat of northern regions, compared 
with the erosion effected by frost in and just above the shore-line, wdiich 
process I have found to be even much more effective in an arctic climate, 
than I formerly believed (cf. above p. 5). 

As the general expression 'marine denudation' is bv most writers 
combined with wave erosion, I prefer to use the expression shore erosion 
by frost for this special form of marine denudation. 

When sea-water freezes (at — 1° to — i.9°C.) it does not expand so 
suddenly in the moment it is transformed into ice, as fresh-water does 
at o" C. But at sinking temperature the ice of sea-water goes on expanding 
gradually [cf. O. Pettersson, 1883]. According to my observations, this 
continues as long as the sea-ice holds in its pores liquid brine which is 
gradually transformed into ice. New sea-water ice is also soft, porous, 



30 



IKlDljOF NANSKN. 



M.-N. Kl- 




Fig. II. Shore at Cape Elisabeth, Bear Island (after a photograph by J. G. Andersson). 



and flexible, and very different from the hard, solid, and brittle fresh- 
water ice. 

For this reason the sea-water, when it penetrates into the fissures of 
the rocks and freezes there, has not as great a disintegrating power as the 
fresh-water, and there is a material difference in this respect even if the 
water contains comparatively little salt [cf. O. E. Schiotz, 1894]. 

One might, therefore, expect that the sea-water cannot by freezing 
produce a very powerful erosive effect on the shores, especially if it is not 
much diluted with fresh-water. It is an obvious difference in this respect 
between the shores of a lake and the shores of the sea. 

It has, however, to be considered that in the fjords, and sounds, and 
enclosed parts of the sea, especially in the Arctic or glacial regions, the 
sea-surface is generally covered by layers that are very much diluted by 
river-water, and by the melting water of glacier-ice carried into the fjords, 
often producing nearly fresh surface-layers. 

Moreover, nearly fresh surface-layers are quite commonly formed in 
Arctic seas by the melting during the summer of the ice, formed in the 
sea during the winter. If the sea is sheltered against waves by floating 
ice, or by islands, this fresh surface-layer may remain more or less un- 
stirred during the summer and autumn, till it again freezes next winter. 

In this manner, the disintegrating power of the sea, caused by 
freezing, will become more vigorous along sheltered coasts, either in fjords 
and sounds, or where the coast is sheltered by floating ice. 

In a somewhat different manner the sea has, however, a much greater 
erosive effect along the shores in cold climates. As was mentioned before, 
glaciers or patches of snow may, by alternate thaw and frost, have a 
vigorous disintegrating effect upon the rocks along their edges. 

Similar conditions prevail along coasts in an Arctic or glacial climate. 
Ice is formed on the beach or along the shore-line, just above high-tide 



1 92 1. No. II. 



THE STRANDFLAT AND ISOSTASY. 



31 




Fig. 12. Northwest shore of Reindeer Islands (Kjellman Islands), coast of Siberia. (August 20, 18931. 



level, and is covered by snow-drifts during the winter. This ice and snow 
remain in greater or smaller patches on the beach, above high-tide level, 
during the summer, or at least during the greater part of it (cf. Figs. 1 1 
and 12). Along their edges, especially on their inner side, a very effective 
erosion goes on during the melting season in the manner mentioned on p. 25. 
As the dark shore cliff absorbs the light heat-rays, the melting will begin 
very early at its base. All cracks, fissures, and pores in the rocks are kept 
full of melting water, ready to freeze at the slightest frost, and will split 
the rocks [cf. Nansen, 1920, 1921, Chap. VIIIj. 

As the masses of melting snow and ice, as well as the cold sea surface, 
keep the air temperature near the freezing point of water, a slight fall of 
temperature may be sufficient to cause frost. Alternations of thaw and 
frost may therefore occur almost every day and night during a great part 
of the year. 

Owing to its dark surface the rock will al)sorb the light heat-rays to 
a much greater extent than the snow and ice surface. Every time the rock 
is exposed to direct sun-shine it may, therefore, be heated to temperatures 
much above the freezing point of water, although the air-temperature in 
the shade is very low. The snow and ice, especially along the vertical 
always dark rock-walls, will then be melted, and the fissures and cracks 
of the rock kept full of liquid water. As soon as the direct sun-shine 
disappears from the rock surface, this water will freeze at once. 

This process may be repeated several times during the twenty four 
hours, and may begin very early in winter or spring, as soon as the sun 
rises sufficiently high to have an appreciable effect. In not too northern 
latitudes it may even occur, more or less, during the whole winter. In 
polar regions I have observed drops of liquid water being formed on the 
surface of white ice cliffs in this manner when the air-temperature was 
below — IS*' C 

In regions where the sea is covered by ice during the winter, it is 
also of importance in this respect that, during late winter and spring, the 
atmosphere is generally very clear, and there may be continual sun-shine 



32 



FRI DI JOK NANSK.N. 



M.-N. Kl. 




Fig. 13. Ice-foot' (shore-ice) formed above the upper fucus-hmit, and above 
the average sea-level, at Holstensborg, Greenland. [K.J. V. Stcenstrup, 1907.] 



(luring- a great deal of tlie time, till the sea is opened, and fogs l.ecome 
frequent. 

The result is that the rocks are rapidly disintegrated and crumble 
to pieces. 

The tide may also be of importance by alternately submerging some 
part of the shore twice every day and night, even during the cold winter. 
This may cause alternate melting and freezing of the water in the fissures 
and depressions of the rocks, wherever the shore is not covered bv more 
permanent layers of ice. Thus an active disintegration of the shore rocks 
may be produced even by the sea-water. 

During the time of the year when the sea is more or less open, the 
wave action will wash away from the shore the debris formed by the frost 
disintegration. This is of the greatest importance for the continued shore 
erosion by frost. The waves and the tide may also break loose the ice 
formed on the beach, and may carry it away with the débris, accumulated 
on its surface from the cliffs above. 

The ice formed at low water, on small water pools left by the tide 
on the beach, may also be of some importance in this respect. Mineral 
particles and small stones enclosed in this ice, mav be carried along with 
it when the tide rises [cf. O. E. Schiøtz, 1894]. The waves may help to 
break the ice loose, and thus the shore erosion may be increased. 

The socalled 'ice foot', or thick layer of solid ice, formed along Arctic 
shores during the winter, may also carry along with it enclosed stones, 
or stones and débris fallen from the cliff on to its surface, where it is 
broken loose in the spring or summer [cf. Knutsen, 1889, p. 249]. But 
according to my observations, this transport by the Mce foot' is less than 
might l)e expected, because on the one hand, a great deal of the 'ice foot' 
lies above the average high-tide level, and melts gradually in situ, without 
being carried awav l)v the sea, and on the other hand, the ice on the lower 



1921. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY. 



33 




Fig. 14. Picture of the future, showing how the ice-foot' in Fig. 13 
ma}- erode a shore-ledge in the length of time. 

beach may be so firmly frozen in between the stones, and boulders, and 
rocks, that it cannot be broken loose by the waves, but is gradually worn 
away or melted in situ. I have seen ice remaining during the summer in 
this manner between the stones on the sea bottom along submerged beaches. 

In a different manner a transport of débris may be effected. The 
stones loosened by the disintegration of frost from the cliff and the steep 
mountain side, will fall down on the sloping snow surface of the 'ice foot' 
and thumble, or gradually glide, down this slope to the shore where they 
may reach the surface of the sea ice and be finally carried away by this ice. 
Or they are. deposited in or near the water, and when they have been suf- 
ficiently disintegrated by frost and waves, are finally washed away. In 
this manner a terrace of loose stones and débris may often be formed along 
the shore outside the bench cut in solid rock. 

The accumulations of ice and sno\v along the shore will year by year 
eat themselves landwards, making the shore-bench broader, and forming 
a higher and higher cliff, or mountain wall, of crumbling rocks inside. 
Thus the typical shore of Arctic lands is developed. 

This may even occur along steep coasts where the mountain side falls 
abruptly into the sea, as is demonstrated by Figs. 13 and 14. Figs. 13 
illustrates a small *ice foot' formed above the upper sea-tang (fucus) limit 
and above the average high-tide level at Holstensborg in Greenland 
[K. J. V. Steenstrup, 1907]. 

Fig. 14 illustrates how a shore-bench might be formed in the length of 
time by the frost erosion of the accumulations of ice and snow at this level. 
The rate of the erosion will greatly depend on the resistance of the rock 
to the disintegrating effect of the frost and ice. In places where the reck 
is relatively less resistant, small circiues may be formed. 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 11. 3 



34 



I lUirrjOF NANSEN. 



M.-X. Kl. 




Fig. 15. An old shore cliff at the inner margin of a strandflat of low level, on Verleegen 
Hook, Spitsbergen [Nansen, 1920]. 



During storms with high-water, especially in late summer and autumn, 
the débris accumulated on the bench may be more or less washed away. 
The débris will also be moved outwards by the creeping motion in the 
accumulations of ice and snow themselves, and by the transport caused 
by the alternate melting and freezing of water along their inner margins. 

It might be objected that the ice formed on the beach is sea-ice and 
will not therefore have much erosive power. But on the one hand, the 
sea in these regions is often covered by nearly fresh surface-layers, as 
mentioned above, and on the other hand, when the sea-ice grows old, 
especially during summer, the brine gradually sinks out of it, and its 
melting water becomes practically fresh, so that, for instance, it makes 
excellent drinking water. Moreover, the ice on the beach is generallv 
covered by deep snow-drifts which remain during a great part of the 
summer. 

This form of shore erosion may be studied at work in its different 
phases in Spitsbergen. We may there see how it is now forming ledges 
along the present shores, and we may see how it has l)een able to form 
broader benches or small strandfiats as demonstrated by Fig. 15. where 
there is an old vertical shore cliff of disintegrated rock, with an erod ng 
accumulation of snow at its base, on the inner side of a quite low strandflat. 

After this chapter on the formation of the strandflat, had been written, 
my attention has been called to a paper by T h o r o 1 f \^o g t [1917] which 
he has kindly sent me. It is of great interest for our subject as it de- 
scribes recently formed shore-ledges, cut just above high-tide level in very 
steep rock walls along the coast of Kvænangen Fjord in northern Norway. 



1 92 1. No. 1 1. 



THF STRAXnFLAT AND ISOSTASY, 



35 




Fig. 1 6. Shore-ledge cut in gabbro with bands of syenite, at Simalanga, 
Kvænangen Fjord [after a photograph by T. Vogt]. 

A more convincing demonstration of the manner in which the shore-erosion 
by frost actually works, could hardly be found. 

The ledges are 8 and 12 metres broad, the one is cut in gabbro with 
bands of syenite (Fig. 16), the other in dolomite. Their outer edges rise 
generally between half a metre and one metre, or a little more, above the 
level of their floor inside, which lies perhaps half a metre above the upper 
fucus limit, and slightly above average high-tide level. The floor of the 
ledges may probably be flooded at spring-tide. 

As A^ogt maintains, these ledges are obviously formed by frost dis- 
integration in recent time and at present sea-level. He thinks that the 
eroding work has to a great extent been performed by accumulations of 
snow on the ledges, which have frequently been soaked with spray from 
the waves, and have afterwards been frozen. In this manner he finds an 
explanation of the height of the ledges above the average high-tide level. 
I think, however, that it is the alternate melting and freezing of the ice 
and snow accumulated on the ledges, which have had the chief erosive 
effect on the rock in the manner described on a preceding page. The ice has 
been formed on the ledges at spring-flood and at exceptional high water, 
as well as by water washed up by the waves in cold weather, and on top of 
this ice snow-drifts have been accumulated (cf. Fig. 14). Because of the 
salt contained in the sea-water the soaking of the snow by the spray from 
the waves may reduce somewhat the disintegrating effect of the melting 
and freezing snow. 



36 



KRIUTJOK NANSr.N. 



M.-N. Kl. 




Fig. 17. Shore-ledge 8 metres broad cut in argilaceous shist on the Fornebo Peninsula. 



The actual occurrence of such shore-ledges, recently eroded by frost 
in solid rock and at present sea-level along- shores where a considerable 
wave action has been working simultaneously, as is the case in Kvænangen 
Fjord exposed to the stormy Loppen Sea, is a convincing evidence that, 
Avhere the necessary conditions are present, the shore-erosion by frost is 
much more effective than the wave erosion, and it is able to cut shore- 
benches in solid rock in a much shorter time. 

It might be objected that the mountain sides of Kvænangen Fjord, 
where Vogt's ledges have been cut, are so very steep with so deep water 
just outside (a depth of 22 metres only 4 metres outside the shore) that 
the wave erosion would, in any case, have very little effect on such a 
nearly vertical rock wall. But the fact is, as Aogt points out, that the 
loose stones as well as the projecting solid rocks on the rough surface of 
the ledges are angular, and show no appreciable traces of having been 
rounded by the waves. While the disintegrating effect of the frost has 
been able to cut these ledges 8 and 12 metres broad into the steep mountain 
wall, the wave erosion has not even been able to produce any appreciable 
effect upon the outer edge of these ledges. The wave action has, however, 
obviously been of importance by helping to carry away the débris. 

Similar shore-ledges, formed by shore-erosion by frost, also occur in 
more southern latitudes. Along the shore outside my house on the Fornebo 
Peninsula (at Lysaker), in the inner end of Christiania Fjord, relatively 



1921. No. II. 



THE STRANDFL.AT AND ISOSTASY. 



37 




Fig. 18. Shore-ledge 18 metres broad, cut in argilaceous shist near Fornebo Harbour. 



broad ledges have been formed in this manner, in recent time and at present 
sea-level. Figs. 17 and 18 represent some parts of these shore-ledges. 
They are cut in solid rock and are from 8 or 9 metres (Fig. 17) to 18 and 
20 metres broad (Fig. 18). The rock consists chiefly of ordovician clay- 
slates, alternating with bands of lime-stone, occurring in series of lenti- 
cular nodules with intercalated shales [cf. Werenskiold, 191 1]. This 
formation is intersected by numerous dykes of diabase. 

The general floor of these ledges is perhaps half a metre above 
average sea-level. The tide is insignificant in the inner end of Christiania 
Fjord, and the sea does not often rise sufficiently to flood the ledges. In 
som places the outer part of the ledge, near its edge, may be somewhat 
higher than the floor inside. In Fig. 17 the outer edge of the ledge is about 
0.6 to 0.8 metre above average sea-level, while the floor inside is 0.2 to 0.3 
metre lower. In some places rocky nabs occur near the outer edge rising 
as much as i metre above the floor of the ledge (see Fig. 18). These pro- 
jecting nabs and ridges near the edge of the ledges, are in some cases 
formed by dykes of diabase, which has been more resistant to the erosion 
than the argilaceous shists and lime-stone, but they may also, as in Figs. 17 
and 18, consist of the same kind of rock as the inner part of the ledges. 
In the nab in Fig. 18, there is relatively much lime-stone. 

These ledges have obviously been formed by shore erosion by frost, 
in a similar manner as that described on the preceding pages. When the 
fjord was covered by ice in the early winter, as it always used to be except 
in late years when the fjord is kept open by ice-breakers, the ledges were 
covered by thick layers of ice formed at high water. When this ice melted, 
all fissures and depressions of the rock surface would be kept filled with 
water, which would freeze and disintegrate the rock at each frost. 



38 FKIDIJOF NAN.SKN. M.-X. Kl. 

TIic arj^ilaccous sliists aii<l sliaics arc easily dcniulcl iii lliis manner. 
'J'lie nodules of lime-stone may he more resistant, and tliey llierefore often 
project somcwliat above the surface oi the surrounding shales. 

As J/'i^. 17 shows, the rock of the higher parts of the ledj^es has the 
very rough surface typical for rocks disintegrated by frr^st. In some 
places where the rock surface is extremely rcjugh the relief of the many 
projecting ridges and blocks may amount to nearly one metre. The edges 
of the projecting rocks, on the outer part of the ledges as well as further 
in, are but slightly rounded by the wave action, which has obviously been 
of very little direct importance for the original forination of these ledges. 
The surf at high water lias, however, been of great indirect importance 
by washing away the débris formed by the frost erosion. 

Where the ledges are low, and slope gently seawards below average 
sea-level, as in some part of the shore represented in Fig. 18, the solid rock- 
surface is more or less covered by a layer of stones of different sizes. It 
is possible that the wave erosion may here have denuded somewhat the 
outer part of the ledges. The stones are rolled up and down the beach 
by storms from the south and south-east, which always cause high-water 
and may have great effect by a vigorous surf. The rock-surface of the 
lower levels show indications of wave erosion, though it has also to some 
extent been disintegrated by frost. 

The stones have obviously been broken loose from the solid rock by 
frost disintegration. There are all stages from a great number of angular 
stones recently disintegrated, and consisting to a great extent of diabase, 
to rounded, wave-worn pebbles, which greatly consist of the more resistant 
parts of the argilaceous shist, especially the calcareous nodules. 

These ledges have been developed in recent time after the land had 
been elevated very nearly to its present shore-level ; and there has obviously 
been no appreciable elevation of the land after their formation. The sur- 
face of the ledges shows no traces of glacial erosion, but the surface of 
dykes of diabase in several places projecting in the shore-line at the outer 
edge of the ledge is rounded and striated by glacial erosion, and so is the 
upper surface of the dyke of diabase to the right in Fig. 17, forming the 
cliff backing the ledge. The low skerry seen in the background, behind the 
bath-house to the right, is also rounded and striated by glacial erosion. 

In many other places along this shore one also finds rock-surfaces, 
chiefly of dykes of diabase, which are rounded and striated by glacial 
erosion. They are sometimes situated near sea-level outside the shore- 
ledge, and sometimes above it. 

This shows that the rocks of the shore had been scoured and rounded 
by the glaciers of the last glacial period, before they were attacked by the 
shore-erosion, after the shore had been elevated to its present level. The 
argilaceous shists were then easily disintegrated by frost, and were cut 
back bv the shore-erosion, while the dvkes of dial)ase were much more 



1921. No. II. THE STRAXDFLAT AND ISOSTASV. 39 

resistant. The ledge illustrated in Fig. 17, was cut back till the erosion 
stopped against the nearly vertical wall of the dyke of diabase to the right 
in the picture. This wall rises about i .7 metre above the floor of the ledge. 
The level upper surface of this dyke is about 2.2 metres above the sea, 
and seems to be part of a shore-ledge formed before the last glacial period. 
A little further inland there is a higher ledge at about 3.1 metres above 
the sea, which is 40 to 50 metres broad, and partly cut in solid rock, 
partly formed of loose material. 

In the region of Fig. 18 there was no dyke of diabase to stop the 
shore erosion, and tlie shore-ledge has here been cut back about 18 metres 
to the foot of the cliff of argilaceous shist, which in some places is nearly 
vertical, 4 to 6 metres high and even 10 to 12 metres high. 

In one place (at Godthåb) there is a broad dyke of syenite-porphyry 
along the shore, in which the shore-ledge has been cut, and it is to some 
extent backed by a cliff of argilaceous shist. This porphyry is traversed 
by numerous fissures and was easily disintegrated by frost. 

The effect of the shore erosion differs much with the height above 
sea-level. At the lowest levels of the shore, where the rocks emerge only 
at low water, the recent shore erosion has been insignificant. One may 
even find rocks of argilaceous shist scoured by the glaciers, which have 
kept their rounded polished surface, where the glacial striation is still 
visible. The reason is obviously that the rocks at these low levels have 
been protected by the water against the disintegration by frost, w^hile the 
wave-erosion has had no appreciable effect. They are a convincing demon- 
stration how the former process has been essentially more effective than 
the latter one. 

The disintegrating effect of alternate thaw and frost may be most 
vigorous just above average sea-level, where ice is formed at high-water, 
but where the freezing of the water during frost is not too frequently 
disturbed by submergence. The shore-ledges have therefore been developed 
at this level.* 

In some places the ledges are, however, as much as a metre, or even 
1.3 metre, above average sea-level, and are rarely submerged. It is there- 
fore a question whether the shore may not have been slightly elevated 
since their formation. If so it can, however, only be a fraction of a metre. 

These ledges on the Fornebo Peninsula demonstrate on a small scale 
very clearly how the shore erosion by frost, assisted by the wave action, 
works, and how it is able, in relatively short time, to cut fairly broad 
shore-ledges. The planing effect of this erosion may differ much ac- 
cording to the resistance of the rocks and other circumstances. The result 
is an often very uneven rock surface, varying somewhat in height in dif- 
ferent places. We cannot therefore expect that a plane of abrasion thus 
formed, will ever become quite level. 



40 



IKIDI \(>y NANSKX. 



M.-N. Kl. 



If W. C. Brøgger [1905] is right in assuming tliai the coasts of 
Christiania Fjord were still rising fluring the first part of the Bronce- 
age, but attained their present level towards the end of that age, we may 
assume that the ledges of the present shore of the Fornebo Peninsula 
have had about 2500 years for their formation. 

The ridges along the outer edge of the shore-ledges, rising above their 
inner floor, which are so conspicuous on the ledges in Kvænangen Fjord 
(Fig. 16) and of which there are also traces on the ledges of the Fornebo 
Peninsula (Figs. 17 and 18), appear to be a common feature on shore- 
ledges formed by frost erosion. They occur frequently on the lateglacial 
and postglacial raised beaches of Norway, where especially O. E. Schiøtz 
[1894] has called attention to them. They are also mentioned by other 




Fig. 19. 

writers. J. H. L. Vogt and J. Rekstad [\'ogt, 1900, p. 73] state that near 
the outer edge of the shore-ledges in Helgeland, 10 to 20 metres broad or 
more, rocky nabs frequently rise one or two metres above the general floor 
of the ledges (c. g. on Leka in Helgeland). They are obviously formed by 
the greater effect of the disintegration by frost on the inner part of the 
ledge, where the accumulation of ice and snow remains longer, while it is 
more easily washed away by the waves along the edge. On the inner part 
of the ledge the accumulation of ice and snow (Fig. 19, i) may thus erode 
the rock down to a lower level without being disturbed by the waves, as 
is demonstrated in Fig. 19, A and B. 

The raised beaches of Norwa}^ have obviously, as a rule, been formed 
by shore erosion by frost, in a similar manner as the strandfiat, but during 
relatively short periods. 

Already F e i 1 d e n and De Ranee [1878] pointed out very clearly 
the great importance which the disintegration by frost and the ice-foot 
must have had for the formation of the many raised beaches or shore- 
ledges in Arctic regions. 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 4I 

B ] y 1 1 [1881] considered the elevated shore-lines to have been formed 
by the freezing of the sea-water in regions with a great range of the tide. 

H. K n u t s e n [1889, p. 249] thought that the shore-lines were chiefly 
formed by the transport of the ice-foot (see above), and J. H. L. Vogt 
joins this view [1900, pp. 75 f.; 1907, p. 25], he also thinks that the dis- 
integration by frost in the shore is of great importance. 

O. E. S c h i o t z [1894] maintained that the raised shore-lines have 
been formed by the disintegrating effect of the frost, and also by the 
transport of mineral material by the ice formed on the shore-ledge. He 
pointed out that sea-ice must have less erosive effect than fresh-water ice, 
because the sea-water does not expand so suddenly when freezing as fresh- 
water. 

P. Sehe i [1904, p. \'IIIj strongly maintained that the ice-foot, in 
Ellesmere Land up to 100 metres broad, was of the greatest importance 
for the formation of shore-lines. 

T h o u 1 e t has mentioned the erosive effect of the freezing sea-water 
on the shore-line in Newfoundland. 

J. Re k s tad [igoy, 1916] maintains that the elevated shore-lines 
have been formed by frost and the ice-foot, and by the transport of stones 
with this ice-foot when loosened. 

He also points out that there are often small cirque-like depressions 
in the mountain slope, above the raised shore-lines, and the thinks they 
have been formed by small cirque glaciers [19 16, p. 16]. 

Some writers have held the opinion that the raised beaches should 
have been formed chiefly by the erosion of drifting ice. Judging from 
my observations of the effect of the drifting ice on Arctic shores, I hold 
it for impossible that shore-ledges can be formed in this manner. The 
drifting ice has in fact an insignificant erosive effect. During ice pres- 
sures the ice-floes may be piled up against the shore, and may push up 
stones and gravel on to the beach, but will have very little effect upon the 
solid rock. The transporting effect of this drifting ice is also insignificant, 
except that it may transport detritus and gravel carried down on it bv 
w^ater or small snow-slips from the shores, or it may occasionallv carry 
stones enclosed by freezing. 

It has also been maintained that the raised beaches have been formed 
by wave erosion; but we have already seen that the waves work extremely 
slowly on the shore, compared wåth the frost. The wave action is, however, 
of much importance for the formation of the shore-ledges, bv washing 
away the débris formed by the frost erosion. But in some places where 
the coast is open and tlie surface water of the sea is salt, the wave action 
may have been a hinderance to the development of the shore-ledges by 
washing away the ice and snow on them. This may especially have been 
the case in postglacial time when the climate w^as less severe. 



42 IKIDIJOF N'ANSKN. M.-X. Kl. 

Oli the (jtlicr liaiid, tlic waves will lia\e a f^realer erosive force on 
exposed coasts. P)Ut as the \\a\e erosion wiii attack a vertically more 
extended surface of the sliore cliff than does the shore erosion by frost, 
it will not cut a sharply markerl incision in the mountain side, as has been 
pointed out by T. Vogt [19 17]. 

Thus we may have an exf^lanation why raiscl beaches are less con- 
spicuous and seldom observed along the open coasts of Norway, while they 
are often very conspicuous in the fjords and sheltered sounds, especially 
in northern Norway, where the climatic conditions have been favourable 
for tlieir formation. The lower salinity of the surface layers of the sea 
in the fjords and sounds may also have been of some importance in this 
respect. 

A striking difference between the strandfiat and the lerlges of the 
raised beaches, besides the difference of extension and width, is that the 
latter are tilted, having always a dip seawards, while the plane of the 
strandfiat is very nearly horizontal. It is obvious that the strandfiat is 
a formation that has been developed during long periods when the shore- 
line was standing at a very nearly stable level, while the ledges of the 
raised beaches have been formed during relatively short periods when the 
shore-line had only temporary levels. 



1 92 1. No. II. THE STRANDFLAT AND ISOSTASY. 43 



VI. THE DEVELOPMENT OF THE NORWEGIAN 
STRANDFLAT. 

It seems to be probable that the shore-ledges have been formed by 
the same process as the strandfiat, and that they are in a way embryonic 
strandfiats, that might have developed into more mature ones, if they had 
been allowed the necessary time. 

But how may these narrow ledges develop into the broad planes of 
the strandfiat? 

The wave action is obviously of much importance for this develop- 
ment. While the frost is disintegrating the rock by the alternate melting 
and freezing of ice and accumulated snow above the average high-tide level 
— widening the ledges into broader benches and at the same time lowering 
their floor — the waves attack the outer edge of the benches, assisted to 
some limited extent by the frost, and in the length of time they will manage 
to wear down this edge, even below low-tide level, and will thus get access 
to the inner floor of the benches. There may here be plenty of débris and 
loose stones for the surf to work with, rolling them forwards and back- 
wards, and thus gradually wearing down the outer part of the benches, 
and make them slope gently into the water, in the manner which may be 
studied on a small scale on the shore-ledges of the Fornebo Peninsula 
(cf. Fig. i8). 

In this manner the sea, headed by the sharp teeth of the frost of an 
Arctic climate, and following with the roaring surf, may eat itself land- 
wards and create a gradually broadening low and flat shore-plane, sloping 
very gently seawards, and with its inner part emerged above high-tide level. 
Landwards this plane will be bounded by a more or less vertical cliff, con- 
tinually retreating before the attack of the frost. 

When the shore-plane attains a certain width, the waves may lose 
their force in the shallow water over the outer submerged part of this 
shore-plane, before they reach the emerged shore, and their kinetic energy 
will then to a great extent l)e expended in transporting the gravel on the 
flat sea-floor [cf. Nansen, 1904, p. 182]. But by so doing they will erode 
the sea-floor, and will continue to do so as long as the shallowness of the 
water prevents them from attacking the emerged shore with their ful! 



44 P^KIDJJOP^ NANSEN. M.-N. Kl. 

energy. 1'lnis both processes, the liorizontal shore erosion by frf>st and 
waves, and tlic vertical wave erosion on tlic- sul;nierge'l j^art of the shore - 
plane, will go on simultaneous!}', and will continue llieir gradual march 
landwards. 

This is the probable genesis of the typical Arctic shore, which we find 
in many regions where the sea-level has remained fairly stable for a suf- 
ficiently long time. I may instance the north coasts of Russia and Siberia. 

Wilhelm Ramsay describes [1900, pp. 59 f.] an illustrative example of 
such a flat shore formed at present sea-level on the northern coast of the 
Ribachi Peninsula (Fisher Peninsula, Murman Coast). This shore-plane 
(Fig. 20) is emerged at low-tide to a breadth of 50 metres, and is cut in 
black clay-slates, wdiich is easily disintegrated by frost. This coast has a 
severe climate, but is at the same time exposed to the violent surf of the 
open and stormy northern ocean. 

Although the frost is the chief causal agent for the formation of the 
shore-ledges, the aid of an effective wave erosion is thus necessary in order 
to develop these ledges into broader shore-planes, because on the one hand 
the waves must wash away the débris of the frost erosion, and on the other 
hand they must denude the outer part of the planes sufficiently to get 
access to their inner parts. 

Though the shore-ledges may be easily formed in fjords and sheltered 
sounds, it is thus obvious that the best conditions for the development of 
extensive shore-planes must be along the open, outer coast, exposed to the 
full force of the wave action of the open sea, and where also the land 
generally is lowest and most dissected into islands, so that there is much 
less rock to be cut away. 

The rate of the wave erosion increases very much with the storminess 
of the sea, it being proportional to something between the third and the 
si.vtii power of the velocity of the wind [cf. Nansen, 1904, pp. 181 f.] pro- 
vided that the latter has a steady landward direction. 

Thorolf Vogt, in his valuable paper on recent shore-ledges [1917, 
p. 1 21 J, has already given expression for a view of the formation of the 
strandfiat which is similar to what has been hefe described. He says: 
"The forces developing shore-lines, work qualitatively very sharply limited 
without removing very much rock, while the surf combined with frost 
disintegration removes much greater quantities of rock without working 
so sharply limited as to level. Where there is a vigorous surf in connection 
with an intense disintegration by frost the necessary conditions prevail for 
an abrasion of great dimensions, and this may be the point of view from 
which one has to consider the development of the strandfiat." 

In the above discussion of the development of the strandfiat by the 
joint action of frost disintegration and the breakers, the sea-level was 
supposed to remain stable. If, however, the coast be slowly submerged 
during this process, it is obvious that the frost disintegration as well as 



192 1. No. II. 



THE STRANDFLAT AND ISOSTASY. 



45 




Fig. 20. Shore-plane formed at present sea-level, on Ribachi Peninsula. The picture is 
taken at low water [after W. Ramsay, 1900]. 

the wave erosion on the shore may be much increased horizontaUy. The 
sea win then get freer access to the shore-ledge. It will gradually raise 
the level of the disintegration by frost, and the breakers will reach the 
actual shore with greater force, less energy being expended in deepening 
the submerged part of the shore-plane. Hence the combined erosion of 
the frost and the breakers may advance landwards at a greater rate. The 
result will be a wider but more sloping shore-plane. The effect of the 
erosion will be extended more horizontally, but the vertical erosion will 
be less. 

If the coast be slowly emerging during the process, the result will be 
the opposite. The effect of the combined erosion will be extended less 
horizontally, and more vertically. It will lower the seaward part of the 
shore-plane, and the result will also in this case be a more sloping plane, 
but narrower, standing on the whole at a lower level. 

During those periods when, according to rny view, the Norwegian 
strandfiat was mainly developed, the sea-level was for a long time fairly 
stable, or the coast was being quite slowly submerged, owing to the weight 
of the accumulating ice-caps (cf. infra) . During the periods of emergence 
of the coast, after the ice-caps had been removed, the land was much sub- 
merged. Then the raised beaches were formed, while the strandfiat was 
lying below sea-level. The climatic conditions may then, to a great extent, 
have been less severe and less favourable for the shore erosion by frost. 

The ability of the shore erosion, described above, to develop a strand- 
flat along a coast, will depend on the relative length of the shore-line and 
on the heigth of the land (i. e. the quantity of rock to be removed), not 
considering the power of resistance of the different kinds of rock. 

Along a relatively undissectcd and high coast the effect of the shore 
erosion in this respect, will be very little, and unless the rocks be very 



46 FRIDTJOF NANSEN. M.-X. Kl. 

weak, it woulil lianlly l)C aljlc t(j form a straii'lflai of much uirllli, even 
thoLig'h it could wdik (lurinj^- llic loiif^-est ^l-( )\< )^\c'd\ time we mij^^lit possibly 
imagine. 

Alonj^- a nuicli dissected coast, like the west and nortli-wcst coast of 
Norway, tlie conditions are, liowever, entirely flifferent. The length of 
tlie shore-line has been enormously increased, the islands and peninsulas 
will be attacked by the shore erosion from all sides, and its effect will be 
increased accordingly. 

Moreover, along this rlissectcd coast the greater part of the denudation 
towards sea-level has already been accomplished beforehand by the sub- 
aerial denudation and the glacial erosion, which have worn away enormous 
(|uaiitities of rock, and there is comparative!}- little work left for the shore 
erosion, in order to plane off the small and low islands and peninsulas 
more or less to sea-level, and thus form the strandfiat; especially as during 
the cold periods, when this plane v\'as in my opinion mainly given its pre- 
sent flat surface, the atmospheric weathering and the local glacial erosion 
by frost denuded the surface of the dissected coast land vigorously. 

On the greater and higher islands, or wdiere the rocks were more 
resistant, it would take longer time for the shore erosion, assisted by the 
atmospheric weathering and the local glacial erosion by frost, to denude 
the land to sea-level. In such places mountains or smaller knolls and ridges 
may still be left, surmounting the level of the strandfiat, often as isolated 
stacks or 'monadnocks'. 

The result of the whole process, gradually developing the planes of 
the strandfiat by the erosion of the shores of the islands, will be that a 
quite common shape of many islands and even of small skerries, is more 
or less similar to that of a hat swimming on the sea with the brim near 
sea-level and a rounded crown forming the central part. 



When was the Norwegian Strandflat developed. 

The conclusions arrived at on the preceding pages regarding the mode 
of formation of the strandflat, make it probable that the Norwegian strand- 
flat has been developed to its present form chiefly before and during the 
quaternary glacial periods. 

We do not know what the climatic conditions may have been along 
the coast of Norway in late tertiary time, and whether they may have been 
favourable for an active shore erosion by frost, so essential for the 
formation of a strandflat. This is, however, hardlv probable, as the 
climate may have been too mild. 

Towards the beginning of the first quaternary ice age, the temperature 
sank, and a more effective shore erosion began when frosts became more 
frequent. As time went on the climatic conditions grew more favourable 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. 47 

for the shore erosion by frost, and at the same time the subaerial de- 
nudation became more effective, as the disintegration of the rocks by frost 
increased. Provided that the level of the shore-line was then approximately 
the same as during" later interglacial periods, a more active formation of 
the strandfiat may have begun. 

But as the coast had not yet, at that time, been dissected by the glacial 
erosion, it is hardly probable that the development of the strandfiat could 
make much progress. 

After the valleys of the coast had been deepened by the erosion of 
the Great Ice Age, and the coast land had been dissected by the numerous 
deep fjords and channels, intersecting the coast and splitting it up into the 
many islands and peninsulas the conditions were essentially changed, as 
has been pointed out on a preceding page. It therefore seems probable 
that the Norwegian strandfiat has chiefly been developed to its present 
shape after that time. 

It is possible that in interglacial times there may have been periods 
with climates sufficiently severe for shore erosion by frost so essential for 
the formation of the strandfiat; but tlie climatic conditions preceding, 
accompanying, and succeeding each glacial period were especially favour- 
able for this erosion. At the same time the subaerial denudation of the 
outer coast land was also very effective. 

For the development of the strandfiat it was, however, also essential 
that the shore-line was stable during long time and standing at or near 
the level of the strandfiat. This was probably the case during a great part 
of the intergiacial times and at the beginning of the glacial periods, while 
during the glacial periods the land was gradually depressed by the weight 
of the growing ice-caps, as was pointed out above. The strandfiat cannot 
therefore have been developed during the late part of the glacial periods 
or during the time of submergence after these periods, even though the ice 
had retreated from the outer coast. The development of the strandfiat 
could not begin until the land had again been elevated nearly to its norinal 
level. But at that time the climatic conditions would not, as a rule, be 
favourable for the shore erosion by frost. 

We are thus lead to the conclusion that the strandfiat has been de- 
veloped chiefly during interglacial periods with cold climates, and especially 
during the very cold time preceding each glacial period, and during its first 
part, before the outer coast was covered by the inland ice, and as long as 
the level of the shore-line still remained nearly stable. 

This may have lasted some time, because the level of the ocean was 
lowered by the accumulation of water in the ice-caps on land. The gradual 
sinking of the land caused by the growing load of the ice-cap, may thus 
for a long time have been more or less counterbalanced by the sinking of 
the shore-line. It should also be considered that the depression of the land 
began in its interior parts where the ice masses first accutnulated, and did 



48 I'KIDTJOF NANSEN. M.-X. Kl. 

not cxtciifl lo 1I1C oulcr coast before after some time. iJurinj,'- tlie last 
glacial period scjine part of the outer coast was probably never depressed, 
as will be mentioned later. 

We do not know how many (|uatcrnary ice ages there have been in 
Norway. But if there have been only two, as is generally assumed, we are 
lead to the conclusion that the present plane of the strandfiat has been 
developed chiefly during the cold time preceding the last glacial period, 
wliile its formation may already have begun during the time preceding the 
first, great Ice Age. There are, however, indications which, according to 
my view, make it probable that there have l)cen several glacial periods in 
Norway, and that the strandfiat has been developed during several dif- 
ferent periods, as will be mentioned later. We must then assume that 
during each interglacial time the ice-caps of Norway have almost entirely 
disappeared, so that the coast has been able to rise to its normal level 
at which the strandfiat has been developed. Oxaal [1914, pp. 42 f.] has 
maintained that there were three glacial periods in Norway, and that the 
upper level of the Norwegian strandfiat, in about 40 to 50 metres above 
sea-level, has been planed by marine abrasion during the first interglacial 
period, while the planes of the lower levels of the strandfiat, in about 
10 metres or even less above the sea, have been developed during the second 
interglacial time. My investigations have lead to similar conclusions, 
which will be mentioned later, after the observations and measurements 
of the strandfiat in various regions have been described. 

Some writers have maintained that the strandfiat must have been 
formed before the fjords, because there are no traces of a strandfiat along 
the coasts of the latter. The answer is: Firstly, that there are actually 
strandfiats in the fjords, not only in their outer parts where there are 
often beautifully developed planes, c. g. in the mouth of the Hardanger 
Fjord; but there are small strandfiats also in their inner parts, as will be 
described later. 

Secondly, the shore erosion will need a relatively very long time for 
the development of a strandfiat in most fjords, especially in their inner 
parts, because the fjord sides are as a rule very steep and high, and consist 
of resistant rocks. 

Thirdly, the fjords have bien much excavated and deepened by the 
glaciers during each glacial period, and the strandfiat along their sides 
would then be more or less obliterated by this erosion, especially in the 
inner parts of the fjords. 

Fourthly, if the strandfiat was to some extent developed during the 
first part of each glacial period, before the land \vas too much submerged 
— the inner parts of the fjords were then soon filled with glaciers, while 
the planing of the strandfiat may still have continued along the outer coast. 

The fact that nevertheless small strandfiats do occur along the inner 
coasts of the fjords, is a convincing evidence that the fjords had been 
formed before the strandfiat was developed. 



192 1. No. II. THE STRAXDFLAT AND ISOSTASV. 49 

The Height of the Norwegian Strandflat. 

Several writers, like J. H. L. \ ogt, Thorolt \ ogt, and A. G. Høgbom, 
make the height of the inner margin of the Norwegian strandflat, at the 
foot of the steep mountains, especially in Helgeland and Lofoten, to be 
near 40 metres above sea-level, Oxaal says between 40 and 50 mstres, 
perhaps nearest the former height, while H. Reusch and Andr. M. Hansen 
put it at 100 or even 120 metres. The reason why the latter writers have 
got such great heights is obviously that they have taken the benches of 
raised beaches as belonging to the strandflat. 

As was pointed out on p. 42, there is, however, this striking difference 
between the strandflat and the raised beaches, that the planes of the former 
are very nearly liorizontal and have obviously been developed during times 
with a more or less stable sea-level, while the raised beaches are tilted and 
have been formed during periods of submergence of the land, when the 
shore-line was staying at temporary levels for relatively short periods. 
It ought therefore to be sharply distinguished between the two kinds of 
formation, although they are formed more or less by the same process of 
erosion; and may often lie nearly at the same levels. 

But even if we stick to the planes of the real strandflat it may be 
difficult to determine the exact height of their upper limit because, on the 
one hand, this limit between the strandflat and the steep mountain side is 
not often very sharply marked, as the upper planes of the strandflat are 
old formations, which have been much modified and dissected by later 
glacial erosion, and on the other hand, the level of the shore-line has not 
remained quite stable during the development of the strandflat, and its 
planes are obviously formed at somewhat different levels. Amongst other 
things it may then to some extent depend on the resistance of the rocks 
which plane has become most conspicuous. 

Nevertheless there is a fair agreement between the estimations of 
most writers of the height of the upper limit of the strandflat, but at the 
same time distinct lower planes of the emerged strandflat have also been 
described. We shall return to this subject after our investigations of the 
strandflat have been described. We will then also have to consider another 
question: Why it is that these nearly horizontal levels of the strandflat are 
now raised above present sea-level? 

The partial Absence of a Strandflat along some Parts 
of the Norwegian Coast. 

A difficult question of much interest is: Why there are so great dif- 
ferences in the development of the strandflat along the various parts of 
the Norwegian coast? 

While the high and steep west coast of Norway has a well developed 
and broad strandflat, in some places nearly 40 kilometres broad, there are, 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 11. 4 



50 FRIDTJOF NANSEN. M.-N. Kl. 

for instance, only slif^lit indications of siicli a formation alonj^ its southern 
and soiitli-castern coast, altlion^ii the land is iIhtc very much lower, 
sloping"" gently towards the coast, and there w(juld c(jnse(|uently have been 
much less rock to denude in order to form a strandfiat. 

As far as I can see, (jne main reason ma}' have been that during the 
cold interglacial periods, when the strandfiat was to a great extent de- 
veloped, the climatic conditions were much less favourable for the shore 
erosion I)y frost along the south and south-east coast of Norway than they 
were further north, and therefore the strandfiat couid not be developed to 
any considerable extent in those southern regions. It is the saine reason 
why in postglacial time, shore-ledges, cut in solid rock by shore erosion 
bv frost, have been so well developed in northern Norway wjiilc they have 
not been formed along its southern coasts. 

In the inner part of Christiania I'jord, where there was more of an 
inland climate w ith colder winters, a strandfiat has been developed, as will 
be described later. 

During the glacial periods the climatic conditions may have been 
favourable for the shore erosion also in southern Norw^ay, but then the 
south and south-east coast was relatively soon submerged, while the outer 
west coast was only slightly submerged, if at all. As Andr. M. Hansen 
[1895] has already pointed out, the south and south-east coast was also 
soon covered by ice, and during most part of the glacial* periods there was 
no border of bare land similar to that existing outside the fjords of the 
west coast. 

There is still the possibility that the big glacier that filled the sub- 
merged Norwegian Channel round along the coast, and excavated it to 
depths of 500 and even 700 metres below present sea-level, may also have 
eroded the outer part of the coast, and may more or less have cut away 
parts of any strandfiat that had been developed, provided that there have 
been several ice ages, when the Norwegian Channel was filled by a big 
glacier. 

It is, however, much more difficult to account for the absence of a 
strandfiat, near present sea-level, along the coast of Finmarken. In this 
region the strandfiat suddenly ends, just with the extension of the igneous 
and possibly Archæan rocks outside Ringvadsøi, Rebbenesøi, Grotøi, and 
northern Kvaløi. These islands consist of igneous and so-called Archæan 
rocks, extending north-eastwards along the coast from \^esterålen and 
Senjen. Outside these islands as far as north of Kvaloi, there is a 
submerged strandfiat w^ith a great many skerries, sunken rocks, and 
shoals. But north-east of Kvaloi this strandfiat suddenly ends. Vannoi, east 
of Kvaloi consists partly of sedimentary rocks (schists) and partly of 
'Archæan' rocks. Fugloi, east of \'annoi consists entirely of schists. North 
of these two islands there are some scattered submarine platforms with 
sunken rocks and shoals (in 70° 36' N. Lat.), which may be considered as 



I 



I 191. No. II. THE STRAXDFLAT AXID SOSTASY. 5I 

belonging to tlie strandfiat. and wliich arc i)rul)al)ly built up of harder 
'Archæan' rocks [cf. Nansen, 1904. pp. 42, 117 ff.j. 

East of this region, whicli coincides with the western boundary of 
Finmarken, the typical strandfiat, near present sea-level, entirely dis- 
appears. The outer coast is here chiefly built up of sedimentary rocks 
and crystalline schists, less resistant to shore erosion than the igneous and 
so-called Archæan rocks to the west and south-west. The coast suddenly 
changes character at this boundary line between the geological formations. 
There is no typical 'skjærgård' along the Finmarken coast; only in some 
few places, round Ingoi and outside the north-west corner of Mageroi, 
where more resistant 'Archæan' rocks occur, are there indications of a 
'skjærgård' and a strandfiat at sea-level. 

The only explanation I can find of the absence of a strandfiat, near 
present sea-level, along the coast of Finmarken, and of the remarkably 
close relation of this absence to the change in the geological structure of 
the coast, is that the sedimentary rocks and schists of Finmarken have 
offered relatively little ristance to the vigorous shore erosion of this 
northern region and to glacial erosion. During periods with a low shore- 
line, the rocks of the outer coast have therefore been denuded to deeper 
levels. They have also been cut away by glacial erosion. 

The outer coast of Finmarken is to a great extent very steep and 
precipitous, forming high shore cliffs, with much oversteepened, some- 
times almost vertical mountain walls, often some hundred metres high. 
This is especially the case in East-Finmarken wdiere the coast consists 
chiefly of series of dolomite-bearing sandstones and shales of great thick- 
ness [cf. O. Holtedahl, 19 18]. This coast has obviously got its typical con- 
figuration by a vigorous and very effectixe shore erosion. If in this region 
the formation corresponding to the strandfiat along the Norwegian coast 
to the south-west occurs, it may therefore be looked for at lower levels, 
below the sea-surface, where submerged platforms actually occur at depths 
of between 40 and 90 metres [cf. Nansen, 1904, pp. iijff.l. Tt may, 
however, be difficult to distinguish between these platforms and the inner 
part of the continental shelf. 

It is obviously a general feature that coasts built up of relatively weak 
rocks, offering little resistance to shore erosion, never have a typical 
'skjærgård', while this formation always occurs along coasts consisting of 
harder Archæan or igneous rocks, and formerly exposed to glacial erosion. 

As examples of coasts without a 'skjærgård' besides Finmarken, I may 
mention the coasts of Bear Island, Spitsbergen, and Iceland. These coasts 
are mostly built up of rocks which offer relatively less resistance to shore 
erosion, such as sandstones, shales, dolomite, limestones, basalts, etc. But 
wherever harder igneous rocks occur, there are generally indications of a 
'skjærgård', c. g. outside the north-west corner of Mageroi, at the north- 



FRIDTJOF NANSEN. M.-N. Kl. 



\\cst corner of Spilsbc-r^'-cn, in the rc^Mon of X'jrlli TajÆ on Xf^rth-east 
Land and llic Scxcii Islands, al tlic soutli-w est corner of i'arents Island, &c. 

It might I)c objccled that Franz Joseph LanrJ, the Faroes as well as 
Shetland are not built up of very resistant rocks, but are none the less 
dissected inio a great number of islands. Ihe fact is, however, that they 
arc not surrounded b}- a typical 'skjærgård', with numerous low islands, 
skerries, sunken rocks, and shoals, like those along the coast of Norway 
and Greenland. 

The above mentioned coasts, consisting of relatively weak rocks, 
c. g. on Bear Island, Iceland, and the Faroes, have as a rule high pre- 
cipitous shore cliffs, showing that the coasts have been exposed to a 
vigorous shore erosion. 

By our above discussion we are led to the conclusion that the absence 
of a strandfiat near present sea-level along the outer coast of Finmarken 
in northern Norway is chiefly due to a too vigorous erosion on a shore 
of relatively weak rocks, while the insignificant development of the 
strandfiat along the south and south-east coast of Norway, has partly an 
opposite reason, a too ineffective shore erosion in regions where the 
climatic conditions as a rule were too mild. 



Summary. 

Summarizing the results of the above discussion we may describe 
the cycle of the formation of the Norwegian strandfiat as follows: 

During i)reglacial, probably postmiocene (or possibly still longer) 
time the coasts of Norway, as well as its land-surface on the whole, were 
much denuded by subaërial weathering and fluvial erosion. The land- 
surface and the coasts were to some extent dissected by fluvial valleys 
which may even have descended below present sea-level, on to the now sub- 
merged continental shelf the plain of which they have probably traversed 
[cf. Nansen, 1904, pp. 54 ff., 58, 151 f., Ahlmann, 1919, pp. 209 f.]. When 
these submerged valleys were formerl is difficult to decide. 

During a great part of this very long preglacial time the marine de- 
nudation was not by far so effective as during the quaternary glacial and 
interglacial periods, because probably milder climates did not favour the 
shore eros'on by frost, at least not along the greater part of the Norwegian 
coast, and the marine denudation was therefore chiefly limited to wave 
erosion. 

When the time of the first (|uaternary ice age approached, the climate 
became colder, and more favourable conditions for shore erosion by frost 
as well as for a more active subaërial denudation (greatly increased bv 
frost) extended southwards along the coast of Norway. A more active 
development of a narrow low foreland along this coast may have begun. 



1 92 1. No. II. THE STRANDFLAT AND ISOSTASY. 53 

An essential condition for the formation of the present strandfiat of 
Norway did, however, not exist before the coast had been eroded and dis- 
sected by the glaciers of the Great Ice Age. The preglacial fluvial valleys 
were then much deepened, the coast was dissected by numerous deep fjords 
and channels, and was split up into thousands of peninsulas and islands, 
the length of the shore-line, /. c. the line of attack of the shore erosion, 
was enormously increased. The waste on the land was swept away, and 
bare rock-surfaces were exposed to the attack of the erosion, especially 
the disintegration by frost. 

During interglacial times there may have been cold periods favouring 
erosion by frost, at least along some parts of the Norwegian coast. But 
especially the climatic conditions preceding and accompanying each glacial 
period, greatly increased the subaërial denudation as well as the shore 
erosion, in all regions not covered by the inland ice, i. c. especially the 
outer coastal border along western and north-western Norway. During 
these periods, as long as the level of the shore-line remained fairly stable, 
before the land was too much submerged by the weight of the inland ice, 
the strandfiat was chiefly developed, along the much dissected coast, by 
the joint action of subaërial denudation, particularlv by frost, and local 
glacial erosion on the peninsulas and islands outside the inland ice. and 
finally by the shore erosion by frost and wave action. 

The two first mentioned processes were of main importance for the 
denudation of the land towards sea-level; because they attacked the whole 
land-surface, while the shore erosion gave the thus denuded uneven islands 
and peninsulas their nearly horizontal plane. By the repeated advances of 
the margin of the inland ice and its glaciers all waste was swept away 
seawards from the land, and bare rock-surfaces were exposed to fresh 
attacks of the erosion when the glaciers retreated. 

But as the shore-line was not quite stable, at least not during some 
part of the time, when the final planes of the strandfiat were developed — 
because, for instance, the sea-level was gradually lowered by the accu- 
mulation of water in the ice-caps on land — the strandfiat was not finally 
planed at a perfectly fixed level of the shore-line. Besides this, it is also 
probable that when the land had again attained its level of isostatic equi- 
librium, after each glacial period, the shore-line may have stood at a level 
somewhat lower than that of the preceding interglacial time, owing to the 
removal of a great deal of rock material anrl waste by the eros'on during 
each glacial period, as w^ill be mentioned later. 

It is not, therefore, to be expected that the strandfiat as we now see it, 
should form one definite anrl sharply marked plane. The level of its 
surface may vary somewhat according to circumstances. 

Then it has also to be considered that during each glacial period the 
strandfiat has been largely exposed to a vigorous glacial erosion which 
may have modified its surface more or less, and made it more uneven. 



54 KRIinjOK NANSKN. M.-N. Kl. 

'Flu- (lcj4T(,c of mauiriu to wliicli llic ^iramlflai mav have bjcii fle- 
velopcrl in tlic \arious rcj^'ions, will naiurally 'lc|)cii'l uj)on the tiine fluring 
which llicsc various processes of (kinulation, especially the sliore erosion, 
have had to work, aiul also upon the initial heij^^ht of the land and the power 
of resistance of the rocks. Some islands may have been entirely levelled, 
others more or less, leaving*- hij^dicr hills an'l mountains in their interior 
parts, etc. We may therefore expect to finrl all different stages, from 
almost perfectly flat and level planes of low islanrls and peninsulas, to 
regions with more \arying and undulating- lieiglits, aiul onh' some islands 
and peninsulas here and there lia\e Keen nearly ])laned, or have planed 
borders round tlieir shores. 

As the strandfiat lias been eroded by glaciers, at least during one, 
atid probably during two glacial periods, we cannot expect to find very 
sharj) lines of demarkation between the planes of the strandfiat and the 
sliore cliffs of the surmounting ridges and stacks, as these have been 
rounded off ])y the glaciers. On \'æroi and Tvost in Lofoten, and proba])ly 
also on Træna, the strandfiat was not eroded l)y glaciers during the last 
glacial period. Hence the emerged strandfiat on some of these islands 
is very level, but hardly more so than we inay also find in other regions. 



1921. No. II. THE STRANDFLAT AND ISOSTASY. 



VII. THE STRANDFLAT OF THE NORWEGIAN WEST 

COAST FROM THE REGION OF SOGNE FJORD 

TO THE REGION OF HARDANGER FJORD. 

The heights of the strandtlat g-iven in the following- descriptions 
were determined by levelling-. In the summer of 191 i, in the Sogne Fjord, 
along the Norwegian coast, and on the Shetland Islands, a small levelling 
telescope and a levelling rod (4 metres long) were used. In the following 
summer, on Spitsbergen, the same rod was used, but the levelling instru- 
ment simply consisted of a U-shaped glass tube containing coloured water. 
Where there was an opportunitv, the level was also determined by the 
horizon of the sea. This is not a very accurate method, but as the 
distances within which the measurements were made were never con- 
siderable, the final error of a levelling thus performed hardly ever ex- 
ceeded a metre. If the distances are greater the use of a telescope w'ould 
greatly increase the accuracy of the measurement. 

The accuracy with which the heights of the strandfiat may be deter- 
mined, does not, however, as a rule depend on the levelling method 
employed. The real difficulty is in most cases to decide where the actual 
level of the strandfiat is. Seen at a distance this level may look at if it 
were quite sharplv defined; but coming near one often finds the surface 
of the strandfiat to be uneven, with depressions and elevated ridges, 
causing doubt as to what actually indicates the general level, and whether 
it should be put perhaps one metre higher or lower, and sometimes 
even more. 

As the strandfiat has been denuded by glacial erosion after its final 
formation, it may be preferable in most cases to be guided more or less 
by the elevated ridges, if their summits are lying approximately at equal 
heights. But on the other hand, the surface of the strandfiat may originally 
have been fairly rough, and ihe ridges may have been higher than the 
general level. It will, therefore, often be a question of personal judgement 
what the correct height should be estimated to be. 

Another difficulty is the determination of the present shore-line on 
which the levelling is based. Most observers have used the uppsr fucus 
limit in the shore as a base, whicli nia\' have great advantages. But this 



^ y I ■ ■ 

Vi/L^t ■front iKrUUuitxa. 




i^MiJa«^ 



. ,Kii,om^cs 



SO 

Fig. 2 1. 



100 



I 



I92I. No. II. THE STRANDFLAT AND ISOSTASV. 57 

Fig. 21. Map of Norwegian west coast between Sogne Fjord and Bommel Fjord. The 

Strandtlat is marked black where the author has investigated it. A is the inner continuation 

of Sogne Fjord. Scale i : i 000 000. i mm. equal to r kilometre. 

I Sogndal P'jord. 2 Norum Fjord. 3 Tjugum. 4 Ese Fjord. 5 Vangsnes. 6 Balestrand. 
7 Kvamsoi. 8 Tangen. 6 Hoiang Fjord. 10 Fuglset Fjord. 11 Ardal. 12 Holmen. 
13 Akre. 14 Sjrevik. 15 Matsnes. 16 Eike Fjord. 17 Oppedal. 18 Risne Fjord. 
19 Brekke. 20 Vadheim Fjord. 21 Afsnes. 22 Mjelleli. 23 Torven. 24 Rå. 25 Ringer- 
eide. 26 Lavik. 27 Brendingsdal. 28 Bo. 29 Næsje. 30 Rutletangene. 31 Dingenes. 
32 Inner Sulen island. 33 and 34 Two southern peninsulas of the latter. 35 Næs Island. 
36 Northern peninsulas of Inner Sulen island. 37 Buskoi. 38 Færoi. 39 Sakrisoi. 
40 Luten Island. 41 Skivenes. 42 Hisoi. 43 Store Vatsøi. 44 Sandoi. 45 Bjortnes 
Island. 46 Kversoi. 47 Grimen Island. 48 Vardetangen, northwestern extremitv of Lindås 
Peninsula, 49 Store Tangen. 50 Lindås Peninsula. 51 Fens Fjord. 52 Rautingen. 
53 Kjelgaulen. 54 Bakoi. 55 Njotoi. 56 Fosen Island. 57 Lygre Island in Lygre Fjord. 
58 Radoi. 59 Feie Island. 60 Henoi. 61 Lyngoi. 62 Forhjelmen. 63 Sæloi or Hjelmen. 
64 Alvoi. 65 Hjelte Fjord. 66 Fedjeosen or Fedje Fjord. 67 Bredviksoi. 68 Blomsoi. 
69 Holsenoi. 70 Askoi. 71 Herløi. 72 Toftoi. 73 Gjernes on Holsenoi. 74 Salhus. 
75 Bergen. 76 Nordåsvand. 77 Lille Sotra. 78 Store Sotra. 79 Litårnet. 80 Hakensund. 
81 Glesvær. 82 Toftoi. 83 Kors Fjord. 84 Hundevågoi. 85 Horge Island. 86 Mogster 
and Akre Islands. 87 Hufteroi. 88 Fugloi. 89 Selbjorn Island. 90 Reksteren Island. 
91 Selbjorn Fjord. 92 Slotteroi Lighthouse. 93 Fondoi. 94 Stord Island. 95 Tysnes 
Island. 96 Langenuen Sound. 97 Lervirk. 98 Bommel Fjord. 99 Hardanger Fjord. 
100 Hamaren. lor Haukanes. 102 Silden Island. 103 Ænes. 104 Akrehavn and Akre- 
holm. 105 Ulvanes. 106 Steinane. 107 Ölvc. 108 Snilstveitoi. 109 Husnes. iioHuglo 
Island. Ill Halsnoi. 112 Fjelbergoi. 113 Borgundoi. 114 Tittelsnes. 115 Valestrand. 
116 Moster Island. 117 Hjartholm. 118 Ekelandsnes. 119 Nesheim and Nordbo. 
120 Southern peninsula of Bomlo Island. 121 Bomlo Island. 122 Ry varden Lighthouse. 

123 Lyngsøi. 

line is not always easy to determine, especially not in the fjords. The 
upper limit of high tide has also been used, but that is still more uncertain, 
as the marks of high water may be much influenced by the waves and bv 
storms. I have, therefore, started from what T estimated to be the mean 
level of the shore-line, as marked by the colour of the rock-surface and by 
the organic life in the shore, balanes, snails, &c., and also the sea-weed. 
I admit, however, that this way of estimating may be more or less arbi- 
trary, and I consider the upper fucus limit to be preferable as a l)ase 
wherever it is fairlv distinct. 



The Strandflat of Sogne Fjord. 

Along both sides of Sogndal Fjord and Norum Fjord, in the inner 
region of Sogne Fjord (see Figs. 21 and 22), there are very conspicuous 
low, flat points (promontories) and benches, being obviously parts of a 
strandflat, cut in solid rock at the base of the steep mountain sides. They 
have an ccjual altitude of about 10 metres above sea-level. The rocks in 
tliis region are j^artly crystalline schists overlying phyllite, partly gabbro. 



58 



KKID'IJOF NANSEN. 



M.-N. Kl. 




Sita/ietfUU at Mflmam- 
veti , LeikariffeA, 



Fig. 22. Sogntlal Fjord and Noruin Fjord, in the inner part of Sogne Fjord. 

At Kjørncs, the flat, rocky point (crystalline scliist) between Sogndal 
Fjord and Eids Fjord, there is a \ery level plane cut in solid rock, and 
forming a distinctly marked incision in the mountain slope (see Fig. 23). 
The level of its flat surface is 9.84 metres above the sea. 

At Pallenes, in Norum Fjord (Fig. 24), a distinct horizontal ledge 
lies at about 10 metres above sea-level, anrl is the same formation as the 
l)lane ol>served at Kjørnes. The rock is here gabbro. 

At Vines (gabbro), farther out in Norum Fjord, a distinct plane is 
cut in s(jlid rock forming an incision in the luountain slope at about tlie 
same level as at Pallenes anrl at Kjornes. The small islets, the \'ines 
Holms, outside the shore, form a continuation of this plane (see Fig. 25). 

The small peninsula, on the other side of the fjord, just opposite the 
Vines Holms, forms a vcrv distinct strandfiat at an average level of some- 




Fig. 23. Kjørnes Point, between Sogndal Fjord and F.ids Fjord. Strandtlat 9.8 metres above 

sea-level. 'July 29, 1911). 



1921. No. 1 1. 



THE STRANDFLAT AND ISnsTASV. 



59 




^^P^^^_ — 



Fig. 24. Fig. 25. 

Fig. 24. Pallenes in Norum Fjord. Strandllat about 10 metres above sea-level. 

After photograph. ijiil\- 29, 191 1). 

Fig. 25. Vines and the Vines Holms in Norum Fjord. In the background to the right 

the low peninsula at Nordnes. After photograph. <July 30, 191 1). 

what less tlian 10 metres above the sea, but the hiqlicst ridg-e of tliis pen- 
insula approaches 10 metres. 

The low peninsula (crystalline schist) at Xorditcs (see Fig^. 25) further 
out in the Norum Fjord, shows no distinct horizontal level, but the outer- 
most portion which has obviously formed a round islet, has on its outer 
side indications of the same level nearly at 10 metres above the sea. 

At Öz're Slindc (crystalline schists), near the mouth of Norum Fjord, 
the level of the strandfiat is 12.4 metres above the sea. Along^ its inner 
marg-in it forms a distinct incision in the mountain slope, indicating that 
its plane has been developed by sliore erosion (Fig. 26). 

At Hcrmansverk, in Leikanger, there is a very distinct strandfiat at 
a level somewhat lower than 10 metres above the sea (see Fig. 22, lower 
right corner). 

The rocky point at Hanirc (phyllite), west of Hermansverk, shows 
the same distinct strandllat at nearly 10 metres above sea-level. 

This strandfiat at about the same level is observed on most points 
and headlands in tliis region more or less irrespective of their geological 
structure. Both sides of the outer part of Sogne Fjord, outside Leik- 
anger, consist entirelv of Archæan rocks. 





^v^ v\ «V H^ 






^ 




Fig. 26. Fig. 27. 

Fig. 26. Strandllat at Ovre Slinde in Sogne Fjord, near mouth of Norum Fjord. 

After photograph. 'July 30, igri). 

Fig. 27. Vegarnes in Tjugum, seen from Ekedal. After photograph. 'Aug. 10, 191 1 1. 



6o IKIUTJOK NANSKN. M.-N. Kl. 




Fig. 23. P.kctlal witli the tower of the chiircli in Tjugum and the Ese Fjord behind, 
seen from Vegarnes. (Aug. lo, 191 1 1. 

In TjiigiDii, Balestrantl, on tlic north side of Sogne Fjord, tlie strand- 
flat is represented by the peninsula Vcgarncs (Fig. 27) and by the flat 
narrow foreland (Fig. 28) on which the church of Tjugum is situated, 
under the steep mountain slope. In the outer part of Vegarnes there is 
a wide plain or level rock-surface about 11 or 12 metres above sea-level. 
The ridge of the peninsula reaches 25.05 metres above the sea. 

The platform at Tjugum church (Fig. 28) has a level surface at 
about 12 metres above sea-level, but the rock rises gently higher and 
has its highest level near the foot of the mountain slope, at 20 metres 
above the sea. A projecting rock-knoll rises 3.8 metres higher, giving 
a total height of 23.8 metres. 

The islet at Kvauiso, between Målsnes and Kvamme, shows a distinct 
strandfiat at between 12 and 20 metres above sea-level. The same plane 
is continued along the coast inwards along Balestrand, and the houses and 
farms are lying on this ])lain. 

Vaiigsiics (Framnes), on the other side of Sogne Fjord, rises to a 
flat rocky plain at 25.82 metres above sea-level. But the top of the pen- 
insula is 32.15 metres higher, at a level of 57.97 metres above the sea. 

At Tangen, on the south side of Sogne Fjord, opposite Hoiang Fjord, 
there is a well-marked strandfiat (Fig. 29) on which the houses are situ- 
ated. The level of this plane at the houses of Tangen, was measured 
to be 17.2 metres above the sea. Farther to the west the plane is a couple 
of metres lower, whilst to the eastward there is a well developed plane 
I or 2 metres higher. 

Along the coast westwards from Tangen, platforms of a strandfiat 
occur at many places, forming often sharplv marked incisions in the steep 



I92T. No. II. THK STRAXDFLAT AND ISOSTASV 



6r 




Fig;. 29. Tangen on south side of Sogne Fjord. A. -Seen from the sea. B. The platform 
of the strandflat at Tangen, 17.2 metres above sea-level. (Aug. ir, rgiil. 




Fig. 30. The ncss at Ardal, west side of entrance to Fuglsct Fjord. (Aug. ir, 191 i). 



62 



KUIDIIOK NANSKN. 



M.-N. Kl. 




Fig. 3r. Strandllat at Holmen west of Fuglset Fjord. A. Eastern side. B. Western side. 

(Aug. II, 191 1 1. 

mountain slope, and the strandfiat is perhaps still better developed on the 
opposite side of Sogne Fjord, westwards from Hoiang Fjord. The farms 
are lying on this plane on both sides of the fjord. 

Illustrations are here given of the ness or rockv point at Ardal, on 
the west side of the entrance to Fuglset Fjord (Fig. 30), of the point with 
sharply marked strandfiat at Höhnen, west of Fuglset Fjord (Fig. 31), 
and of the strandfiat at Akrc (Fig. 32). The height of the level of the 
strandfiat is at all these places very nearly the same as at Tangen. 

Further west there were similar platforms on each ness and point, 
and also at some places along the shore between the points, c. g. at Sørcvik. 

At Mafsncs, east of Fike Fjord on the south side of Sogne Fjord, 
there is a well developed strand flat, forming a sharp incision in the 



' \ 







Fig. 32. Strandllat at Akre. (After photograph. Aug. ii, 191 1 1. 



1 92 1. Xo. 1 1. 



THE STRANDFLAT AND ISOSTASY, 



63 




Fig. 33. Strandfiat at Matsnes, east of Eike Fjord on southern side of Sogne Fjord. A. The 
point seen from the west. B. Surface of strandtlat, looking southeastward. (Aug. 11, 191 1 1. 



mountain slope (Fig. 33). There are three low points with ridges parallel 
to one another. The two ridges have exactly the same level, measured 
to be 1 7. 1 metres above the sea. At this height there is a distinct plane 
cut in the solid rock-surface. This plane lies a little lower on the western- 
most point. But here a ridge rises to about 25.5 metres above sea-level. 
A similar ridge rises to the same level on the middle point. 

The strandfiat at Alatsnes is relatively broad and extends a con- 
siderable distance westwards along the coast of the fjord. Its surface 
is scoured by glacial erosion (see Fig. 33 B). The projecting edges of 
the rocks are rounded on the thrust-side or eastern side, and are more 
angular on the lee side or western side (Fig. 33 B). But the surface has 
on the whole the appearance of having been originallv formed by shore 
erosion by frost, and not by glacial erosion. 

Further west at Brekke, west side of Risne Fjord, there is a verv 
conspicuous strandfiat forming a distinct incision in the mountain side 
(Fig- 34)- 



64 



FRIDTJOF NAN.SKN. 



M.-N. Kl. 




I""ig- 34- 




Fig. 35- 




Fig. 36. 




Fig. 37- 



Fig. 38. 



fig- 34- kisni- Fjord with well 
marked strandfiat at Brekke on 
west side (right hand side) of 
entrance to the fjord. (Sketch 
July 27, 191 1). 



Fig- 35- Strandflatalong the north 
side of Sogne Fjord from Vikholm 
and Mjelleli (to the right) and 
westwards past Tor\-en towards 
Rå. (Sketch Aug. 10, 191 1). 



Fii;. 36. Strandfiat at Brendings- 
dal and Bekken, where Sogne 
F'jord bends northeastwards to- 
wards Lavik. (After photograph. 
July 27, 191 [). 



Fig. 37. Næsje Peninsula and 
Næs Holm, in Bo, on the north 
side of Sogne Fjord. (After photo- 
graph. July 27, 19 I 1 1. 



Fig. 38. Rutletangene seen from 

the east. (After photograh. July 

27, 1911). 



At Oppedal, between Eike Fjord and Risne Fjord, fluvial terraces 
have been deposited on the strandfiat. 

On the north side of Sogne Fjord we find equally well developed 
coast platforms all along the shore, e. g. at Afsnes on the west side of 
\^adheim Fjord, at Torven (Fig. 35) where there are conspicuous platforms 
in many places along the shore from A^ikholmen and Mjelleli westwards 
to Vareleite, Rå and farther west at Lavik and Brendingsdal (Fig. 36). 
The average height of these platforms seems to be the same (about 17 
metres above sea-level) as at Matsnes and Akre on the opposite side of 
Sogne Fjord. 

At Torven there is a conspicuous plateau, the height of which I esti- 
mated to be something like 200 metres above sea-level (see Fig. 35). At 



I 



ICSI. Xo. II. THE STRANDFLAT AND ISOSTASY. 65 

Ringereide further west, near Lavik, there is a hanging valley at about 
the same height, and a mountain ridge slightly higher. 

This plateau and hanging valley may possibly indicate' the approxi- 
mate height of an initial Palæic valley of this region, which had been 
developed to great maturity before the excavation of the present Sogne 
Fjord began. 

Xear the mouth of Sogne Fjord the platforms of the strandfiat on 
both sides of the fjord become conspicuously wider and more developed 
than further in the fjord. 

The whole peninsula at Xasjc in Bo. with the Xæs Holm outside on 
the north side of Sogne Fjord, is low and flat, and belongs to the strand- 
flat (Fig. 37). 

At Riitlctafigcuc opposite this place, on the south side of the fjord, 
there is an exceptionally well developed and level strandfiat, on the pen- 
insula as well as on the islets outside (see Figs. 38 and 39), and this 
plane is very distinctly marked as a formation different from the mountain 
slope inside. Fig. 39 illustrates well the evenness of the plane. 

The surface of this strandfiat is seen in the panorama Fig. 39 taken 
from a hill (28 metres above the sea) surmounting the plane. The surface 
is somewhat undulating with small hills or knolls. The tops of most of 
them are at the same level, measured to be 16 metres above the sea. Two 
hills farthest out reach an altitude of 20 metres. Another level at which 
the tops of some ridges are lying, is a little lower than the hill from which 
the panorama was taken; they are about 25 or 26 metres above sea-level. 
Only one ridge on the westernmost point reached an altitude of 32.5 
metres. By far the greater part of this wide plane is, however, lying 
between 15 and 16 metres above sea-level. The whole surface of the 
strandfiat at Rutletangene is well rounded by the erosion of the ice. 

At D i ligenes south of the entrance to Sogne Fjord there is a well 
developed and unusually level strandfiat (Fig. 41). Its average altitude 
was measured (on Aug. 12th. 1911) to be 11.28 metres above sea-level. 
A round knoll rises 3.38 metres above this plain to an altiti.de of 14.66 
metres. The surface of the strandfiat is scoured and rounded by glacial 
erosion as seen in Fig. 42. 

On the northern side of the entrance to Sogne Fjord indications of 
a strandfiat, forming incisions in the steep mountain sides of sandstone, 
were observed in the sound between Inner Sulen island and Losneoi 
(Fig. 42 A), and well marked planes of the strandfiat were observed on 
the two southern peninsulas of Inner Sulen and on the greater part of 
Xæs Island (Fig. 21, no. 35) between them. A hill is rising above the level 
of the strandfiat in the northern part of the latter island. The rock is 
sandstone on these islands, with a border of phyllite along the south- 
eastern side of Inner Sulen and the southern side of Losneoi. 

Vid.-Selsk. Skrifter. I. M.-X. Kl. 1921. No. 11. 5 



66 



FRIDTJOF NAN.SKN. 



M.-N. Kl. 




Fie;. 30. Rutletangcne. A. Seen from the west (Sketch Aug. 12, !■ 



Fig. 40. The islands on the southern side of" the Sogne Sea or entrance to Sogne Fjord. View t 

The Genetic Origin of the Strandflat in Sogne Fjord. 

Along the Norwegian west coast, north and south of the mouth of 
Sogne Fjord, numerous formations of tlie strandflat were observed, per- 
fectly similar to those which have been described above along the shores 
of the fjord. All these formations must obviously have had the same 
genetic origin, and before proceeding further in our description, let us 
examine what this origin might have been. 

Dr. Hans Reusch has described similar low rocky benches in front 
of the mountain slopes along the shores of Hardanger Fjord, especially 
m its inner part [1901, p. 189]; but strange to say he has not observed 
similar features in Sogne Fjord. He states that the height of these 
benches is often about 20 to 30 metres. I have not investigated these 
formations in the inner part of Hardanger Fjord, but in the outer part 
of the fjord I found their height to be about 19 metres above sea-level 
and even more (see later). They seem accordingly to be perhaps slightly 
higher there than in Sogne Fjord, although I have also observed a lower 
level there only some few metres above the sea. 

Reusch thinks that these rocky benches are remnants of the floor of 
the initial "Hardanger valley system" in which the deeper channel of the 



\ 



1 92 1. No. 1 1. 



THF. STRANDFLAT AND ISOSTASV, 



67 




totographic vie^v of Rutletangene and islands outside. lAug. 11. igri). 




a the sea west of the coast between Rautingen and Kversoi, see Figs. 21,52, & 46. ( Julv 27, 191 ii. 

fjord has been excavated later. These benches should consequently be of 
preglacial origin. Reusch also points out that the strandfiat along the 
coast outside the fjord may perhaps have been formed at the same time 
as this valley system. 

Reusch's explanation of the origin of these benches seems to me im- 
probable for several reasons. Firstly it is hardly credible that these 
benches and flat rocky points on the sides of the fjord could possibly 
have survived the violent glacial erosion by which the fjord was excavated 
to its present depths, 600 or 800 metres deeper. 

Secondly if the strandfiat along the coast outside the fjord was 
formed simultanuously with the initial Hardanger valley system, the floor 
in the inner part of that valley — in the inner part of the present fjord 
region, in Sorf jord — must certainly have been considerably higher than 
the sea-level indicated by the strandfiat along the coast outside. But the 
level of the benches and rocky points along the sides of the inner Hard- 
anger Fjord is no higher than the strandfiat of the outer coast, and the 
floor of the initial valley must obviously have been lower along its middle 
than now indicated by these benches along its sides. In Sogne Fjord 
we found that the flat rocky points had even a somewhat lower height 



68 



^ FRII)Tjr)r NANSKN. 



M.-X. Kl. 



Kig. 41. I>inf((;n<:s. (After photo- 
grajdi All?;. 12, 191 i). 

than 1I1C slraiiflflal aloiüL; ilic outer cr)nst. If ihese points and benchis 
were iHMiiiiaiits of an initial \allcy floor, tonncl at the same time as the 
.slrandflal of xW outer coast, wc would therefore have to assume tliat the 
land in the inner part of tlie fjord is now stanrhnt,'' coni[)aratively a good 
deal lower than the lanrl alonj.^' tlie outer coast. 

Tliirdiy the flat j)lalforms of these rocky points and benches often 
form \ery distinct incisions in the steep mountain slope's on the sides of 




42. View landwards from top of Diogenes. Surface of strandllat, rounded by glacial 
erosion. (Aug. 12, 191 1). 



the fjords, indicating that these platforms and benches have been formed 
after the mountain slopes and not simultaneously with them. The com- 
paratively level surface of the platforms also prove that, after their final 
formation, they cannot have been exposed to any verv effective glacial 
excavation like that which gave the fjords and their sides their present 
main features. 

There is no possibility that these low and flat points along the sides 
of Sogne I'jord can have been formed by glacial erosion, transforming 



Fig. 42 A. Southern entrance to 

the sound between Inner Sulen 

(to the leftl and Losneoi. (Sketch 

Aug. 12, 191 1 1. 




I92I. No. TI. THF. STRAXDFLAT AND ISOSTASY. 69 

the projecting spurs on the sides of a valley in a manner similar to that 
suggested by Ahlmann [1919, p. ~2), Fig. 35]. The very nearly uniform 
height of the points must have been determined by the sea-level at the 
time of their formation, while the erosion of the glaciers that filled th^ 
Sogne Fjord could not be fletermined by the level of the sea to any 
similar extent. 

Without considering whether spurs may be eroded by glaciers in the 
manner suggested by Ahlmann, the distinct incisions frequently formed 
in the mountain slopes by the flat platforms of the rocky points along the 
sides of Sogne Fjord also prove that they cannot have been originated 
m such a manner. 

There seems to me to be no other feasible explanation of the genetic 
origin of these flat rocky points and benches along the sides of Sogne 
Fjord, as well as Hardanger Fjord, but that they have been formed by 
shore erosion, chiefly by frost, and that they are formations identical with 
the Strandfiat of the outer coast. The fact that this strandfiat in the fjords 
chiefly occurs on the rocky points and headlands, and much less along the 
shores between the points, is what might be expected, considering that 
the shore erosion by frost has a much greater surface for attack along the 
sides of a headland, than along the straight coast. 

Summary. We may summarize our researches in Sogne Fjord as 
follows: numerous indications of a strandfiat occur along both sides of 
Sogne Fjord and in its side branches, chiefly on most of the points and 
promontories. The altitude above the sea of the horizontal level of this 
strandfiat varies somewdiat in the different regions of the fjord, it being 
about 10 metres in the inner part in Sogndal Fjord, Norum Fjord, and 
Leikanger, about 12 metres in Tjugum and Balestrand, and about 16 or 17 
metres in the outer part of Sogne Fjord from Tangen and Hoiang Fjord 
to its mouth at Rutletangene. 

A higher level of the strandfiat, about 25 metres above the sea, was 
observed in Tjugum (on \>garnes and at Tjugum church), in Balestrand, 
on Vangsnes, at Matsnes, and perhaps on Rutletangene. This is the ap- 
proximate height of the inner boundary line of the strandfiat at Matsnes, 
where it forms a distinct incision in the mountain slope. 



The Strandflat north and south of the Mouth of Sogne Fjord. 

North of the mouth of Sogne Fjord a strandflat was observed on the 
northern peninsula of huicr Siilcn, and especially well developed on the 
islands Buskoi, Færoi, and Drevoi to the west. The strandflat is very 
conspicuous on Sakrisoi and Luten Island to the northeast of Inner Sulen, 
and also on the mainland at Ski venes, east of Sakrisoi (see Fig. 21). The 
rock is chiefly phyllitic schists in the whole of this region. 



70 



■Kl 1)1 I OK NAN.SKN. 



M.-N. Kl. 




Fig. 43. Panoramic view of the islands seen between E, S, W to N from a iiill, 42 metres above sea-le\ 

upper ones 

On all the islands to the south of the entrance to Sogne Fjord the 
strandfiat is well developed and its horizontal plane is ver}^ conspicuous. 

A well developed strandfiat extends alonij the western side of 
Ilisfli (His Island, Fig. 21, no. 42) continuing southwards on Store Vatsoi 
(Fig. 21, no. 43) and other small islands, and along the western side of 
Sandoi (Fig. 21, no. 44). The rocks are here Archæan. 

A photographic panorama of the strandfiat extending over the is- 
lands in this region, was taken from a small hill surmounting the plain 
on Figenes, on the southw-estern corner of His Island. The altitude of 
the hill was measured to be 42.05 metres above sea-level (Fig. 43). 

The altitude of the ridges of a good many islets seawards is some- 
what less than 28.0 metres above sea-level, while a great deal of the other 
ridges approach a level of about 36 metres. Many of the highest tops of 
the islands rise to nearly 42 metres above the sea, the altitude from w'hich 
the panorama was taken. 

The strandfiat cut in Archæan rocks along the western side of Sandoi 
(Fig. 44) forms a very conspicuous flat platform on which the farms are 
situated. I estimated its height to be about the same as that measured 
at Pigenes, i. e. between 28 and 36 metres above sea-level, perhaps chiefly 
about 30 metres. The steep slope of the mountain inside ascends abruptly 
from the nearly horizontal platform. 

A great number of erratic blocks, partly very big, are scattered about 
on the surface of the strandfiat on this island and on the other islands 
in this region. 

All the islands west of Hisoi and Sandoi form a very conspicuous 
strandfiat, which is a continuation of the plane extending along the 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASY 



lUceï/^û^^ 



5 /^jo-yne^J-^, 




r: 



^^^^^f^iØ^t^r -6i,^^4^^ 



-^JSK*^ 



mes, Hisoi. E, S, W,X are East, South, West, and North. Lower pictures are a direct continuation of the 

%. 12, 191 II. 

western sides of those two islands, and over Store Witsoi and the other 
islets between them. Fig. 40 gives an impression of the flatnes of the 
strandfiat in this region west of Hisoi and Sandoi. All the islands are 
very flat and low. and here are few higher hills. They consist of Archæan 
rocks. On M jouicii Island (the island between Sandoi and Bjørtnes Is!., 
Fig. 21, no. 44 and 45) there is a mountain in its northern part, while 
tlie southern part is (|uite flat. On the east side of Bjortncs Island 
(Fig. 21, no. 45) a mountainous hill rises above the strandfiat. which is 
otherwise extended over the whole island. 

The whole of Kz'crsai and Grimen Island (Fig. 21, no. 46 and 47) 
form parts of the level strandfiat, and so do the two Vats Islands, Store 
Hille Lsland. (S:c. north and northeast of (irimen Island and Kversoi. 



The Region of Lindås Peninsula. 

The strandfiat continues over the many islands to the south, and is 
especially well devclopefl and level on the southern side of Fens Fjord, 
where the rocks are chiefly crystalline schists (Ulriken gneiss) and a band 
of gabbro or labradorite rocks along the northeastern side of the Lindås 
Peninsula. On the northern side of the Lindås Peninsula the plain is 
extremely flat as seen from the sea (Fig. 46) with all the farms lying on 
it. very nearly at the same level between 16 and 18 metres above the sea. 

A panorama was taken from an isolated hill on \^ardetangen, the 
northwestern end of Lindas Peninsula, 2j.2^ metres above sea-level. The 
first part of the panorama (Fig. 45 A) begins in the north-east, looking 
towards the flat Håvarden Island (of labradorite rocks) in Fens Fjord 



72 



KKIDIJOF NAN.Sr.N. 



M.-N. Kl. 




Fig. 44. Strandfiat along the western side of Sandoi. A. In northern part ol" island, seen 

from Undelandssund south of Skjergehavn. B. Middle part of island. C. View northward 

from Fens Fjord through sound along west coast of Sandøi. lAug. 12, 191 ij. 



and the southern end of Sandoi in the background on the northern side 
of the fjord. It goes southwards with the sun to east, south-east; &c.. 
over land consisting of LHriken gneiss, with labradorite rocks in the far 
distance. The second part of the panorama (Fig. 45 B) extends over is- 
lands (of Ulriken gneiss) from south-southeast and ends in north-north- 
west looking towards Store Stangen in Fens Fjord. Årsøi and Børtnes 
Island are seen in the far distance. 

This panoramic view shows that the strandfiat has a remarkably uni- 
form height in this region, which was measured to be on the average about 
17 or 18 metres. The farms are lying at this altitude. 

The pictures prove that the hill, 27.25 metres high, from which the 
photographs were taken, is higher than the plane of the strandfiat on 
Lindås Peninsula as well as on the islands to the west. 

Further south, on the Lindås Peninsula and on the islands along its 
western side (Bakoi, Njotoi, &c.), the strandfiat has very nearly the same 
height of about 18 metres. The rocks are Ulriken gneiss. At Kjclgaiilcii 
(Fig. 21, no. 53), on the southern end of the island between Bakoi (Fig. 21, 
no. 54) and the northern part of Lindås Peninsula, the height of the plain 
was measured to be between 17 and 22 metres above sea-level. Many of 
the ridges were lying at a level of 28 metres, but at this height one was 
above the general level of the strandfiat on the islands to the south. But 
many ridges on the islands towards the west or seawards, on Bakoi. 
Njotøi, and Fosenoi (Fig. 21, no. 54 — 56), rise to altitudes about 42 m. 
The ridges on the islands towards the south are on the average lower, 
and I looked down upon them from the height of 42 metres (see Fig. 47). 



ig2I. No. TI. THE STRAXDFI.AT AXn ISOSTASY. 73 

A level and well developed strandfiat extends along the eastern side 
of Lygre Fjord over the many islands, Risoi, Lauoi, Dragoi, Bragøi. 
Spjotoi, &c. and along the whole of the Lindås Peninsula (Fig. 48). The 
rock is Ulriken gneiss in this region. 

There is also a well developed strandfiat on the land on the western 
side of Lygre Fjord, on the Lygre Islands (Fig. 21, no. 57), and on the 
peninsula to the south of them (Fig. 49). The level plane is sharply de- 
fined against the steeply ascending hills on the southern portion of 
Radoi behind. 

The general height of the extremely level strandfiat in the whole of 
this region, on both sides of Lygre Fjord and northwards towards Kjel- 
gaulen, seems to be between 20 and 2^ metres above the sea, but possibly 
rises slightlv westwards towards the sea. 



The Region of Radøi. 

The conspicuously level strandfiat of the northern Lindås peninsula 
is continued westwards over the islands Bakoi and Fosenoi (Fig. 21, 
'"■o- SZ — 56), and southwestwards over Radoi (Fig. 21, no. 58) which is on 
the whole very flat, especially in its northwestern part. Fig. 50 is a view 
of the northwestern end of Radt)i with the many small islets outside and 
Fosenoi to the north, taken from the sea southeast of Feie Island (Fig. 21, 
no. 59). The height of this outer, very flat land I estimate to be about 
17 metres above the sea, or very much the same as that of northwestern 
Lindås Peninsula. Small hills rise here and there above the plane of the 
strandfiat, but this is much more frequent in the southeastern part of Radoi. 

Seawards outside Hjelte Fjord and Fed je Fjord (Fig. 21, no. 65, 66) 
there is a series of low islands: Feie, Sulen, Staksøi, Hennoi, Lyngøi, 
Forhjelmen, Sæloi, Alvoi, &c., forming the so-called Öigaren. Seen at a 
distance the surface of this series of islands exhibits an almost uniform 
horizontal level or strandfiat at about the same height as that of the 
land inside. 

The lighthouse of Helliso, at the southwestern end of Feie, is situated 
on a rock-surface 17 metres above the sea, which is very nearlv level with 
the general surface of that island, to judge from what I saw from the sea. 
Here and there more or less isolated hills rise above the plane of this 
strandfiat, c. g. Hesten and Feieb Jörnen on Feie, Sælstakken (59 metres), 
and Asdolcn (49 metres) on S.'eloi. On the islands Bredviksoi and Blomøi 
to the south there arc also a few hills rising to 51 and 52 metres above 
the sea, and one hill on Blomoi even to 70 metres. 

It is noteworthy that these outer series of islanrls have such a very 
sharply defined strand flat in very nearly the same level as Radoi inside 
of Fedje Fjord and Hjelte Fjord although they are built up of Archæan 



74 



KRlD'ljOF NANSKN. 



M.-N. Kl. 




Fig. 45. Panoramic views taken from X'ardetuva, on Vardctangcn, the north-western end of Lindas Peninsula 

and north-northwest 

rocks, wliile Radoi clv'efly consists of young-er crystalline schists (I'lriken 
gneiss) and in some i^arts of labradorite rocks. 

Ahlmann has visited Sæloi or Hjelmen, and says [1919. p. 102^ that 
there is a general level between 20 and 25 metres above the sea. "In 
certain places along the western side we can also distinguish a lower level 
of denudation at 10 — 14 m." 

Tn the Manger district, in the middle portion of Radoi, Ahlmann 
distinguishes "two levels: one at 12 — 16 m. above sea-level, and the other 
at 33 — 40 m. al)Ove sea-level. The first-namerl is considerably smaller m 
extent than the latter, but seems to be somewhat more level. It extends 
c|uite near the shore as an open plain, luit also stretches in as bays into 
the higher level. Perhaps we may include in this level als(T the narrow 
ledges which in certain places occur along the valleys and hollows at the 
height of about 12 m. above sea-level" [19 19, p. 99]. I know the Radoi 
and the Alanger district very well, l)ut have not been ashore there for 
many years, and not with the purpose of studying the strandfiat. Ahl- 
mann's description agrees, however, exactlv with my recollection of it, 



d 



II 



1921. No. II. 



THE STRANDFI.AT AND ISOSTASV 




A. View eastwards bet »veen north-east, Havarden Island, and south-southeast. 
Aug. 12, 191 II. 



B. Between south-southwest 



and I tliiiik that these are just the characteristic features one might ex- 
pect to find along a coast where the shores have been cut l)ack l)v frost 
erosion, forming broad benches or narrower ledges backed by steep cliffs, 
or shore-walls, along the outer coast as well as in the bays. The débris 
formed by this erosion has been carried away to some extent by the wave 
action, but much more by the glaciers which have afterwards moved across 
the lanrl. The steep sides of these bays and small valley channels on Radoi, 
which Ahlmann considers to have been formed by glacial erosion, and of 
which he has given a very good picture [1919, p. loi, Fig. 47I, have just 
the characteristic features of shore-cliffs formed by frost erosion, and 
have not, as far as I can see, the typical surface of rock-walls originally 
formed by glaciers, although they are to some small extent scoured by 
the glaciers of the last glacial period. 

Ahlmann says that his higher level (jf between 33 anrl 40 metres, 
comprises the greater part of thé Manger district, and he thinks that in 
the southern part of the island, at Sæbo, this satue level may possiljly be 
45 — 50 metres above the sea, and lie draws the conclusion that this "upper 



76 



FRIDTJOF NANSKN. 



M.-N. Kl. 



H 








1 




fjH^ 








■ 






1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASV 



t 

■ i ' 


: 


j 


4 




'S^ 




-<î> - 



^- 



H J. 



78 



KIDIjf)!'' NANSKN. 



M.-N. Kl. 




P'ig. 52. Strandflat at Salluis north of Ik-rgcn. «Aug. 14, 191 1). 

(Icmiflatioii altitiulc slopes j^cntly towards NW, or from 45 — 50 m. at 
Sælxifjcll to 25 — 30 m." ill the nortli-w estern part of tlie island. 

He furthermore considers it indisi)iitahle that the denudation surface 
at 20 — 25 111. above sea-level on Hjelmen (or Sæloi, Fig. 21, no. 63) in 
Oigaren outside Radoi "corresponds to the upper level on Manger The 
slope which this level proves to have at Manger thus continues as far 
out as Oigaren". 

Although this denudation surface may have a slight slope, especially 
on extensive islands, I do not ])elieve that this slope can be as great as 
assumed by Ahlmann. The reason why he finds the heights greater in the 
region of Sæbo is obviously because the labradorite rocks of this region 
are more resistant to the erosion than the gneiss in the other parts of the 
island. The strandflat has not therefore been developed to such a degree 
of maturity, and more hills and ridges rise to higher altitudes. 

On Holsenøi, at Gjernes (Fig. 21, no. 73). just opposite the Sæbo 
region, he says himself that inside a hill quite near the shore "there ex- 
tends a well-marked denudation plain at 25 — 30 m. above sea-level". 
Hence, according to his own measurements, the height of this surface is 
no higher here than in the Manger region, on the contrary it seems to be 
8 or 10 metres lower, or very much the same as the average altitude 
he found in his most north-western profile across Radoi, in the Bo region 
where it was between 24 and 30 metres above sea-level. 

On the other hand it may be noticed that, occording to my ob- 
servations, the general level of Feie, Fosenoi, and the north-western end 
of Radoi is probably somewhat less than 20 metres, or about 17 metres. 




•åjR'»^- „.. 



Létal 




f'R- 53- Strandtlat on the southwestern end of Fane Peninsula and the island 



ig2T. No. II. THE STRAN'DFLAT AND ISOSTASY. 79 

and is obviously very nearly the same as that which I found in the north- 
western part of the Lindås Peninsula. But this surface may correspond 
to Ahlmann's lower level which he found both at Manger and on Sæloi 
(Hjelmen), where it was, however, lo — 14 metres above the sea, and 
which he thinks, "does not show any inclination in any definite direction". 
Fig. 51 is a profile across Radoi and the Lindås Peninsula from 
Lyngøi and Forhjelmen (Fig. 21, no. Or and 62) in öigaren (in 60° 40' 
N. Lat.) to Skårefjell on the mainland (in 60° 42' N. Lat.) east of Öst- 
fjord. This profile demonstrates the remarkably level plane of the very 
brt)ad strandfiat, above which the mountain sides rise abruptly. It also 
shows how the plane of the strandfiat extends uninterupted and without 
nny marked change in height across regions of different geological struc- 
ture. Some of the hills surmounting the plane of the strandfiat are built 
up of comparatively resistant rocks, e. g. Kaisås consisting of mangerite. 



Region of Bergen. 

In the region round Bergen the land is more mountaincjus than along 
the coast farther north, and the strandfiat is much less developed, and has 
not such great extent as in the region of Radoi and Lindås. But it is 
quite conspicuous along the sides of the sounds and fjords in some places. 

At Salhiis, in the Salhus Fjord, north of Bergen, the strandfiat forms, 
for instance, a well marked incision in tlie mountain slope (Fig. 52). 

Bergen is to some extent situated on a narrow strandfiat, which 
extends more or less southwards through the valley to Fjosanger and 
Nordåsvand. 

On the Faiic Peninsula, between Nordåsvand and Fane Fjord, south 
of Bergen, the strandfiat has a wide extension west of the steeply 
ascending mountains. Seen from the sea to the west (Fig. 53), it appears 
very flat and conspicuous. A great many of the islands to the west of 
Fane are also low and form parts of the strandfiat. 

In the region of Os at the southern extremity of tlie whole peninsula 
between the Bergen By Fjord and Samnanger Fjord, the strandfiat is 
well developed and sharply defined in the region of Rotingen (Fig. 54). 




^Etmøm 



i 



island rii^lit in front and to the lel't is l.croi. I lulv 20, 1911). 



FRIDTJOF NANSEN. 



M.-N. Kl. 









'■< 



Nii 



;'i 




r. ~ 

-r o' 



1 92 1. No. 1 1, 



THE STRAXDFLAT AND ISOSTASY 



- :,Mair7! 


B, 


A^.v/ 


IUÎC Sa.rU^- 


m^ 


1 





Fig. 56. Northward view iVom the summit (71 metres above the sea) ot" the peninsula at Ekerhovde on 

Store Sotra. l.\ug. 21, 191 1. 1 




Fig. 57. Strandtlat at Hakensund, southern end of Store Sotra. (Aug. 21, 191 1. 




Fig. 58. The strandtlat seen northwards from the hill (42 metres above the sea) north of Glesvær. 

(Aug. 21, 191 1.) 

Region of Sotra. 

The strand flat has a wide extension on the l)ig island Store Sotra 
and also on Lille Sotra ( h i^^. 21, no. 78 and ~~) to the west of Bertrcn. 
But it is comjiarativcly uneven on these islands with manv ridges and 
liills risin«,»- al)o\e the <j;encral le\el of the strandfiat, to heights more than 
(.0 metres. The strandfiat comjirises the greater part of Lille Sotra. and 
the whole western and northw'estern part of Store .Sotra with tlie ex- 
ception of the manv above-mentioned ridges and hills rising above its 
general level, in the eastern and south-eastern i)art of Store Sotra the 
land is more mountainous, rising to heights of 341 metres at Titarnct 
(see Fig. 21, no. 79). Fig. 55 gives a panoramic view of the land towards 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. ir. 6 



82 



FRIDTJOF NANSF.N. 



M.-N. Kl. 




Fig. 59. Panoramic view between east and west-sout: 



tlic west, north, and cast, taken from tlie to]) of tliis mountain. We hero 
look down upon the extensive stranrlflat, wliicli is especially flat and 
cons])icuous towards the southwest and west. A part of the above- 
mentioned stranrlflat in the region south of Nordàsvand i also seen to- 
wards the east, to the right. 

At Rkcrlun'dc, on the east coast of Store Sotra, just west of the 
soutliern end of Lille vSotra, and under Litårnet, the height of the general 
level of the strandfiat was found by levelling to be about 30 or 32 metres 
above the sea. At a height of 52 metres one was decidedly above the 
general level of the strand flat. 

The summit of the small peninsula at Ekerhovde, west of the southern 
extremity of Lille Sotra, was 71 metres above sea-level. Many ridges 
and hills on Lille Sotra anrl the islands northwards, rise approximately 
to this altitude (cf. Fig. 56), and the highest hill on Lille Sotra even 
higher. But the average height of these ridges is somewhat lower, per- 
haps about 50 and 60 metres above the sea. 

At Håkcnsiind, at the soutliern end of Store Sotra, the strandfiat 
has an unusually level surface at about 30 metres above the sea, and 
distinguishes itself sharply from the ascending mountain slopes to the 
north ( Fig. 57). 

Li the region of Glesvær, at the southwestern extremity of Store 
Sotra, the strandfiat is well developed (Figs. 58 and 59). Its altitude was 
found by levelling to be 29.9 metres above the sea. \n some places on 
the peninsula to the north of Glesvær (Kausland) it mav be a little higher, 
but it does not reach 40 metres. The altitude of the highest hill just north 
of Glesvær is 42.5 metres above sea-level, buti this hill rises decidedly 
above the general level of the strandfiat, as is shown by the view north- 
ward, taken from this hill, see Fig. 58). This picture gives an idea of the 
levelness of the strandfiat in this region. In the Ijackground to the right 
is seen the mountain Litårnet from which the panorama Fig. 55 was taken. 
The altitude of the general level of the strandfiat mav l)e assumed to be 
between 30 and 2)i 'iietres above the sea in this region. 



I 



192 



I. No. I ] . 



THE STRANDFI.AT AND ISOSTASV 



«3 




Hi the hill above Glesvaer harbour. lAug. 21, 1901.1 

It is interesting" to note that tlie strandtlat in this region just north 
of Glesvær and across th.e land westwards is cut in flioritic rocks, while 
the land both north and south, on Store Sotra and the other islands, is 
built up of Archæan gneiss and partly of granite [Reusch, 1901, p. 106]. 
This n^.akes apparantly no difference in the height of the level strandfiat. 



Islands between Kors Fjord and Hardanger Fjord. 

South of Kors Fjord (Fig. 21, no. 83) I observed a distinct strand- 
flat extending over the southern parts of Hundevàgoi (granites), and over 
Horge Island (granites) along the eastern side of Hufteroi (granites), and 
on the small islands to the west of the latter: Mogster Island (schists and 
ether rocks), the Aker Islands (where it is very conspicuous). Fugloi 
(gabbro), and the many surrovmding islets. 

On the southern side of Hufteroi the strandfiat is sharply incised in 
the mountain slope consisting of gabbro. 

Along the southern side of Selbjorn Fjord a well developed strand- 
flat, al)out 30 metres above sea-level, extends from Slotteroi Lighthouse 
over all the islands eastwards as far as the western side of Fondoi. These 
islands consist of granites, while the eastern part of Fondoi consists of 
gabbro. 

On l)oth sides of Langenuen Sound a strandfiat is well marked, on 
the northern end of Stord Island (gabbro) as well as along the south- 
western side of Reksteren Island (granites) and along the west side of 
Tysnes Island (gabbro), see Fig. 60. 

Along the western side of Stord Island the strandtlat is distinctly 
defined, but at an altitude perhaps somewhat higher than the usual level 
of the strandfiat. The rocks are here granites, and in the southern part 
chlorite schists or similar schists, and also some conglomerate. 

Along the eastern side of Stord Island there is a -fairly distinct, but 
relati\'ely high strandfiat on many points, especially in the southern part 
near I-eirvik (Fig. 61). The rocks are phyllite anrl in the northern part 



84 



KUIinjOK NANSK.N. 



M.-N. Kl. 




60. 



Slran.l:lal in tr.,nl ol llu- hi,t;h mountains on Rckslcren l.slan.l and TyMics l.land do ihu right). 

(Aug. 27, 1911)- 



oaI)I)r(). TIktc arc indications of a kind of shore lcdi,-e at altitudes of 
between 45 and 55 metres above sea-level. A well developed shore-line 
was observed at about (lO metres (measured with the aneroid barometer), 
which is more or less continuous with the hii,diest level oi a fairly ex- 
tensive plane. Th.ere is, however, also a verv low level of less than 20 m. 
above the sea, which is rjistinctly inarked on the ].oint north of Leirvik 
harbour (Fit,-. 62) and on the islands outside. The reck is here phyllite. 
A strandfiat with ab.,ut the same hci-ht is als(. founrl further eastwards 
in Hardant^er Fjord. 



Hardanger Fjord and Bømmel Fjord. 

In Hardanger Fjord the strandfiat is developed in the same way 
as in Sogne Fjord, though at a perhaps slightly higher level. It is con- 
spicuous at the points of the promontories and peninsulas along l)Oth sides 




Fi-. 61. East coast of Stord Island, north of I^t-irvik. lAug. 28, 191 1. 1 



I 



ig2i. No. 1 1. 



THK STRANDFLAT AXD ISOSTASV, 




Fig. 62. View along the east coast of Stord Island towards Leirvik. Two levels are seen : the general 
higher level at 50 to 60 metres above the sea, and a lower one at less than 20 metres. (Aug. 28, 191 1 1. 




Fig. 63. Hamaren in Hardanger Fjord. 'Aug. 30, 191 1 1. 




Fig. 04. East side of Varaldsoi, near Haukancs, with the Sild Island (to the right), Hardanger Fjon 

(Aug. 30, igr 1 1. 




Fig. 65. Point and small islet at Akrehavn on Varaldsoi. Sild Island to the left. (Aug. 30, 191 1 1. 



86 



I RU) I |OK NAN.SKN. 



M.-N. Kl. 




F'\g. 63. Southern end of Varaldsoi. 'Aug. 30, rçi i ). 




Fig. 67. Ænes in Hardanger Fjord, at the entrance to Mauranger Fjord. 'Aug. 29, 191 1). 



of the fjord, c. g. at Hamareii (Fig. 63), 011 X'araldsoi (Figs. 64, 65, 66), 
Ænes (Fig. 67), and Ulvanes (Fig. 68). But it also occurs along the shores 
between the promontories. Tt is everywhere cut in solid rock, and forms 
often very cofispicuous and sharplv marked incisions in the mountain 
slopes (cf. Figs. 63, 64, 65, and 66). 

The altiturle of this strandfiat was measurerl by levelling at Akrc- 
liaz'n on \'araldsoi. The rock is here phyllite. The ridge of the point 
(Fig. 65) is 19 metres above sea-level. The level of the strand flat was 
also observed on Sild Lsland and on other islets along the shore, though 
the latter are as a rule somewhat lower. 

The summit of the point at Ænes, opposite \'araldsoi, was found by 
levelling to be t^t, metres above tlie sea, and belongs obviously to a higher 
level than that on X'araldsoi and Hamaren. The rock is here Archæan. 

On Siiilstz'citol, Kvinherred, there is also a well-developed strandfiat 
in Archæan rocks. 

1 ha\-e not had an opportunitv of investigating the occurrence of the 
strandfiat in Hardanger Fjord inside Hamaren (Fig. 21, no. 100), but as 
mentioned on p. 66, Dr. Reusch has pointed out the occurrence of what 
he calls a "mountain-foot" or a low rocky i)latform along the side:^ 
of the inner parts of Hardanger Fjorrl, and in the map Fig. 21 I have 



1 92 1. No. 1 1. 



THE 5TRANDFLAT AND ISOSTASV. 




Fig. 68. Ulvanes, Hardanger Fjord. lAug. 30, 1911). 




Fig. 69. Southward view along the eastern side of Hardanger Fjord, towards Husoes, where the strandflat 
begins to extend and the high mountains withdraw. lAug. 30, 191 1). 



marked the places with strandflat according to his map of the inner 
Hardanger Fjord [1901, p. 189]. Reusch states the heights of his plat- 
forms to be about 20 to 30 metres. I expect thev are chieflv about 
20 metres, which agrees with mv irieasurements in tlie outer part of the 
fjord at \'araldsoi. 

As in Sogne Fjord so also in Hardanger Fjord the strandflat is con- 
spicuously wider and better developed along lx)th sides of the fjord in 
its outer part, and in Bommel Fjord, than in the inner fjord. It is 
especially outside the region of Husnes and Huglo Island (Fig. 21, no. IC9 
and 1 10) that the strandflat begins to extend more widely over the islands 
as well as on the mainland, and the mountains rctiie from the shores 
of the fjord, leaving the strandflat as a wide level foreland (cf. Fig. 69). 

A distinctly developed strandflat extends along the northwestern side 
of Bommiel Fjord cut in rocks varying much in their power of resistance. 
It extends over the southern end of Stord Island (Fig. 70) consisting of 
phyllite and chloritic schists, over Moster Island (Figs. 70 and J2) con- 
sisting of chloritic schists and gabbro, along the eastern side of the 
southern peninsula of Bomlo Island, and over Espevær (Fig. 75) con- 
sisting of schists and gabbro. 

Along the southeastern side of the outer Harrlanger Fjord and 
Bommel Fjord, the strandflat is well developed along the whole coast from 
Husnes (Archæan rocks. Fig. 69), on Hâlsnoi, and Fjelbergoi (Archæan 
rocks and phyllite on both islanris). along \'alestranrl from Tittelsnes 
(phyllite. Fig. 71), and southwestwards along the coast, consisting of 
Archæan recks, towards Ryvarden Lighthouse and Haugesund (Figs. 73, 



88 



I-RI DI' I OK NANSKN. 



M.-N. Kl. 




l'if.'- 70. Slran'ltlat on .southern cn'l 'A 



■frr^ff.'-Lliy--Mc 



Tf chj.',^nt^ 



.mmmmmifmilfm 




Fig. 71. South-eastern side of Bommel Fjord, from Tittel 



i 



'"^CS-tJ^ ^^LcLTld 



B,rr U M 



*■'. iû.1 'J^io/r^ai, 



màÊÊmimmÊÊÊmam 



^tr\/A '}4jn.y\M 



Fig. 72. Panoramic view extending from Moster Island to the northwest aero.'-- 



74 anil JJ). and further southwards along Karmsuiul, where the rocks are 
chlorite schists or similar schists. 

I did not g-et an opportunity of measuring the height of the strandfiat 
in this region of the fjord outside Stord Island. But on the whole it gave 
tlie impression of being higher than in the region further in the fjord, 
being as a rule perhaps a1)out 30 metres above sea-!evel, and sometimes 
even higher. 

As mentioned above I found a fairly well developed plane at between 
45 and 55 metres along the eastern and southeastern side of Stord Island 
and especially at Leirvik. The distinct plane extending over the southern- 
most part of Stord Island (Fig. 62) has an average height approaching" 
this level, or perhaps more correctly between 40 and 50 metres. 

On iVIoster Island the average level of the strandfiat is between 30 
and 35 metres, but some few hills rise to 45 and 51 metres, and a hill in 
its northern part even to 55 metres. Whether the rocks are gabbro or 
weaker chloritic schists or similar schists makes no appreciable difference 
in the heights of the hills or the strandfiat. The latter is lowest in the 
southeastern part of the island, near Mosterhavn, where it is less than 
30 metres high. It is somewhat higher further west. 



I 



i 



I92I. 


yo. 1 1 . 


THE STRAXDFLAT 


AND ISOSTASV. 


89 












^^^^9 


-- 




^ÊÊÊÊÊÊÊÊÊÊÊÊ 


HHH 


IHHIHiiiliHHiifei 


■■■mÉHHÉfi^^H 






■Sto'tä. IslAnd. 











u aiiu un .Muster Islaiai. lAutiUst, 1904». 




mk 



MMUiM 



nh-west\vards along; the coast of Valestrand. (Aug. 30, 191 1 1. 



. iJjL'UXiuf.A 




alanger Fjord to Valestrand and the coast towards the east. 'Aug. 30, 19111. 



On the opposite, southeastern, side of Bommel Fjord the strandflat 
seemed to have a similar heig-ht of between 30 and 40 metres, or in some 
places perhaps slightly higher. The rocks are here Archæan, only at 
Tittelsnes they are phyllite. The islands out in the sea to the west, at 
Lyngsoi (Fig. 21, no. 123), also seem to have a fairly considerable height 
approaching this higher level. 

On Bomlo Island the strandflat has a wide extension along its 
western and over its whole northern part and on the islands outside. 
The height of the strandflat is on the average between 30 and 40 metres, 
perhaps nearer 30 metres, but many hills and ridges rise above its general 
level even to altitudes of more than 100 metres. 

The Bomlo Island shows great variations in its geological structure. 
The whole northern part consists of granites, the middle portion of gabbro, 
while the southern peninsula is built up of schists, with spots of gabbro 
and also quartz porphyry. 

On the whole the strandflat in the region of Hardanger Fjord seems 
perhaps to have a similar tendcncv towards rising slightly seawards from 
the inner regions as we found in Sogne Fjord. But it has to be noted 
that there are obviouslv at least two different levels: a lower level less 



90 



I-kliri |01- NANSF.X. 



M.-N. Kl. 



Fig. 73. Strandlhit along soiilh-eastern side of li.jmmc! I'jord, south-westwards from Ekelandsnes, in continuation 

from rigilt end of Fig. 7;. lAug. 30, 191 i). 




Fig. 74. Strandllat along sonth-eastern side of Hammel Fjord, near Nesheim and Norbo. Aug. 30, 191 



Sspt-l/rfi- .Ti/a . 






"Byyn/o fiuA 


'ßc'hn/r- Js/n.^ r^ 






AriRpH 












■ •--•,,. ^^^^. 



Fig. 75. Northward view towards the fairly high southern peninsula of Bomlo Island, and the low islands 
of Espevaer to the left, taken from the sea north of Ryvarden Lighthouse. <Aug. 30, 1911). 



tlian 20 metres liigli observed at Leirvik, X'araldsoi, Hamaren, &c., and a 
higher level of about 30 metres, or lietween 30 and 40 metres. 

The strandfiat of the Hardanger and Bomlo region seems on the 
whole to have very nearly the same average height as in the Bergen region 
and the Sogne Fjord region, perhaps slightly higher in some places. On 
Stord Islaiifl the strandfiat seems to be especially high. 



ig2I. Xo. II. THE STRANDFLAT AND ISOSTASY. 9I 



VIII. THE STRANDFLAT OF THE SOUTHWESTERN 
AND SOUTHERN COAST OF NORWAY. 

The Region of Karmøi. 

On Karmoi there is a well developed strandflat extending over nearly 
the whole island. But here and there hills rise above the level of its plane 
(cf. Figs. 77 B and 78), and this is especially the case in the middle and 
southern part of the island, mostly on its eastern side. 

The northern part of the island consists of chlorite schists, the middle 
and southeastern part of gabbro, and the southwestern part of granites 
or gneisses. The strandflat extends equally over these different formations 
without any appreciable difference in height. But the highest hills occur 
chiefly in the gabbro region. The coast of the mainland along the eastern 
side of Karmsund consists of chlorite schists. 

According to levelling made north of the bay at Angvaldsnes, the 
general level of this strandflat is about 29 metres above the sea. But a 
great part of the plain, on which many farms are situated, is between 16 
and 20 metres, and this is also the general level of the strandflat on the 
mainland along the eastern side of Karmsund. The town of Haugesund 
is situated on this plane at the same height about 17 metres above sea-level. 
The heights were measured by levelling with a theodolite from a hill at 
30.5 metres above sea-level, north of Angvaldsnes Bay. 

Fife- 78 gives a panoramic view taken from this hill. By careful 
levelling with the theodolite it was ascertained that the ridges of the 
undulating plain of Karmoi were very nearly at the same level as this hill, 
or a metre or two lower. E. g. the height of the platform on which the 
Angvaldsnes church is lying, was found to be 29.5 metres above sea-level. 
The bases of several houses in Haugesund were measured to be at about 
16 metres above sea-level. 

To the north Bjorjene Hills on Karmoi are seen rising above the 
plane of the strandflat. On the mainland hills are also seen rising above 
this plane, but e. g. east of Haugesund they distinguish themselves sharply 
from the level of the strandflat. 

A sharply defined and widely developed strandflat extends south- 
wards aloner Karmsund on the mainland (Fig. 80) as well as on Karmoi 




Hilemettes 



/ûû 



Fig. 76. Map ol" the Norwegian coast between Haugesund and The Naze (Lindesnesl. 
I Haugesund. 2 Karmoi. 3 Angvaldsnes Bay. 4 Karmsund. 5 Høie Varde. 6 Eastern 
Bokken. 7 Western Bokken. 8 Rennesøi. 9 Fjoloi. 10 Mosteroi. 11. Kvitingsoi. 12. Stav- 
anger Peninsula. 13 Stavanger. 14 Tananger. 15 Hàstein. 16 Roth Islands. 17 Hellesto 
(Halandl. 18 Feisten Lighthouse. 19-19 Region of Kiep. 20 Region of Nærbo. 21 Ogne. 
22 Egersund. 23 Egeroi. 24 Presteskjær Lighthouse. 25 Josing Fjord. 26 Sogndal. 
27 Hitteroi. 28 Lister Fjord. 29 Lister. 30 Farsund. 31 Southeastern part of Lister. 

32 The Naze (Lindesnesl. 



I92I. Xo. II. THE STRAXDFLAT AND ISOSTASV. 93 






Fig. 77. Strandflat north of Haugesund. lAug. 30, 191 1). 

(Fig. 79) but, as mentioned above, a good many hills rise more or less 
abruptly above its plane. The geological structure differs much on the 
two sides of the sound, the rocks being gabbro on Karmoi and chlorite 
schists on the mainland; but there is no appreciable difference in the 
appearance or height of the strandflat. 

A well marked strandflat. with a height approaching 30 metres, and 
forming a sharply defined incision in the mountain slope extends along 
the western and southern side of Eastern Baku island consisting partly 
of phyllite partly of Archæan rocks (Figs. 81 and 82 K 

On the southern side of the island another level was noticed at about 
double that height or a little more (Fig. 82). 

On the southern part of Western Bokn island (chiefly Archæan rocks 
and some phyllite), there is also a distinct strandflat. 



The Stavanger Region. 

The northwestern end of Rennesoi. north of Stavanger, forms a very 
flat and low strandflat (cut in granite), with a height of less than 20 
metres (Fig. St,). 

Ahlmann in his attempt to prove that the strandflat has not been 
formed by marine denudation, points out that the occurrence of the strand- 
flat or rocky bench with a steep cliff behind it, on the southern side of 
Rennesoi, coincides with the tectonic difference in the geological structure 
of the island where igneous rocks, forming the cliff, rest on a base of 
weaker sedimentary rock (phyllite), forming the strandflat. He thinks 
that in such a case the formation of the bench "can be well explained as 
a result of subaërial and glacial erosion, but not by marine abrasion". 
But what is the explanation in the manv thousands of other cases of 
similar shore-benches or strandflat formations where there is no such 
tectonic difference? And why have the horizontal strandflat and the 
shore-benches so very similar heights, or to a great extent practically 
identical heights, in the different regions so widely separated, although 
the rocks may differ much in their power of resistance to erosion? It 
seems that Ahlmann, in his anxiety not to admit the effect of the marine 
abrasion, has been compelled to seek for different explanations of the 
same formations in the different cases. 



94 



I KID I I OF NANSKN. 



M.-X. Kl. 




Fig. 78. Panoramic view of Karmoi and the mainland, taken from a hill north of the bay at Angval; 

of the u 



aBÊÊÊÊÊl 



Fig. 79. Karmoi south of Hoge Varde lighthouse, with hills rising above the plane of the standfla 

(Aug. 31, 191 1 1 



On the islands south of Renncsoi (Ulfstcin Island and Fjolloi), there 
is a well-developed strandfiat. The sketch Fig. 84 demonstrates how the 
almost horizontal plane of the strandfiat extends to the foot of the moun- 
tains Byrefjell and Kneberg-, which rise abruptly above this plane. The 
rock is granite on both islands. 

The strandfiat has a wide extent over most islands in this region 
north of Stavanger and out in the sea to the west where all islands are 
low and very flat. 

Kz'itiii[^soi with its hundreds of surrounding islets and skerries has 
a very conspicuous low flat level cut in chlorite schists (Fig. 85). In the 
eastern part of these islands where the view Fig. 86 was taken, the average 
height of the plane was measured by levelling to be about 9.5 metres above 
the sea. But the level at which the farms are lying farther westwards (see 



1921. 


No. 


1 1. 


THE STRAXDFLAT 


AND ISOSTASY. 


95 










sw 


3 c 


'' 




i-v^^r^ 






;^U ot" camera 32 metres above sea-sevel. (Aug. 31, 291 1. 
sat a. 



Th 



nvcr pictures arc a direct continuation 




hi'j:. 80. I he mainland along the eastern side of Karm-sund, seen from the sea south of 

Hoçe Varde lighthouse. 



Fig. 86 right side) is higher. I estimated it to be about 20 metres or per- 
haps a little more.^ 

The general level of the ver\' flat Rott Islands further south, west of 
Tananger on the Støvanger Peninsula (see Fig. 89), was found by levelling 
to be about 17.7 metres above the sea, or perhaps between 16 and 18 
metres, l)ut the western and southern part of the islands is somewhat 
lower. They consist of phyllite. 

The many islets between Kvitingsoi and Rott are mostly low and 
flat, marking a low plane (largely less than 10 metres high) above which 



' Alter this was written I learn from kind information I have received from the .Super- 
intendant of Lighthouses, that the base of the lighthouse of Kvitingsoi is 22 metres 
above sea-level, which agrees well with my estimate, as this base is level with the 
plane mentioned above. 



96 



IKlDTjOF NANSKN. 



M.-N. Kl. 




l'ig. 81. l'oints at Ovrcbu im noi-tli-wc-stern side ot Kastcrn I)okn island. (Sij)!. i, 19111. 

Håsteincn (i)liyllite) rises al)ru]>tly lo n lici<(ht of 45 metres ('according 
to "Tojjoj^ra fi.sk Kart 15Iarl (> I), Stavanj^er"'). Tlie islaiul Storkjor, soutli- 
west of Jvolt, lias a lieij^iit of 18 metres. 

The laiul on llie Stcwangcr l'cniusula forms on the whole an un- 
usually level anrl well-developerl stramlflat, ciu in solifl rock, mostly phyl- 
lite, but in some places, at Tananger and round I lafs Fjord, also granite. 
The difference in the rocks causes, however, no a{)j)recial)le difference in 
the plane or the height of the strandfiat, though a difference in the rough- 
ness of the surface is easily perceptible. Hills, like Randeberg (Fig. 87) 
rise more or less abruptly to heights of 79 metres above this almost per- 
fectlv horizontal plane, extending to the foot of the hills. 

North of Stavanger, at F^inncstad near Dusevik, I found bv levelling 
the height of the plane to be 29.5 metres above the sea. This plain is to 
a great extent covered by moraine material, but tlu solitl rock (phyllite) 
protrudes at the surface of the plain in many places. 

On the northwestern side of the peninsula, south of Tungenes I.'ght- 
liouse, the strandfiat is conspicuously level and low, less than 10 metres 
above the sea (Fig. 87), and is cut in solid reck (phyllite) forming th? 
shore. The base of Tungenes Lighthouse is about 8 metres above sea-level. 

At Tananger the general level of the very even plane of the strandfiat. 
extending landwards anrl cut in granite, was founrl by levelling to be al)Out 
14 metres above the sea. Near the shore it was somewhat lower. The hill 
witli a cairn, north of Tananger harbour, rises to about 22 metres above 
sea-level. The panoramic view I'ig. 89 was taken from this hill. It shows 
that the level of the plane is much below that height. The height of the 
small Tananger Lighthouse seen in the middle of the picture is 15 metres 
above the sea, and that of its base al)out r i metres. The height of Fladho'm 




Fig. 82. Eastern Bokn island, seen from the south, with strandflat extending along the whole coast from the 
point to the north-west, which is the same as in Fig. 8i. (Sept. i, 191 il. 



1 92 1. Xo. 1 1. 



THE STRANDFLAT AND ISOSTASV. 



97 






^j 

g 
^ 






CS 



1 



' ^ 



O 



Ç 



Vid.-Selsk. Skrifter. I. M.-N Kl. 1921. Xo. ti. 



ts. — 




Fig. 89. Panoramic view of Tananger Bay and the islands outside, taken 1 

Lij^litliousc is 15 metres above the sea, and its base about 8 metres. 
The picture demonstrates tlie remarkable flatness of all the islands out- 
side the coast. These islands consist of phyllite. The picture also shows 
the roughness of the granite surface of this strandfiat north of Tananger. 

The ridge on the peninsula south of Tananger Bay (to the left in the 
picture) has a fairly rough contour, although it is built up of phyllite. 

Further south, at Hellesto (Håland) at the northern end of Jæderen, 
^\e find the same strandfiat cut in solid rock at the foot of steep mountain 
slopes, rising to heights of 90 metres al)ove sea-level (Fig. 90). According 
to Reusch's geological map [191 3] the rocks in this region are argilaceous 
schists (phyllite). 

The height of the rock of Feisten, in the sea southwest of this place, 
is about 12 metres above sea-level. 



Relation between the Geological Structure of the Coast and the 
Occurrence of the Strandflat. 

In the preceding description of the strandflat along the Norwegian 
west coast, occasional remarks have been made on the geological structure 
of the coast and its relation to the occurrence and extension of the strand- 
flat. The geological dates are taken from the geological maps of the coast 
anrl from Reusch's description with map of the geology of Sondhordland 
and Ry fylke [191 31- 

As has been pointed out on several occasions, it is striking how the 
strandflat often extends eciually over regions with very heterogeneous 
geological structure, without showing any appreciable differences in its 
altitude or in its whole appearance. The plane of the strandflat is. for in- 




ig2i. 


Xo. 1 1 . 


THE 


STRANDFLAT AND ISOSTASY. 


99 






Wi^^^m^^ä 


«Muni 


g^g;35«a^^ç-:j;??7'^.^. .. -:xa».'g5» 


,1 1 gTft,^,, ^ 











1 ig. 88. 1 h^' jtrandiîau cul in graniic nurih ol lananger. i-Scpl. i, igi i i. 



7 OLyrtjO/na^'t' JÙa^*"- 



I ^ 



.7ptr nu-rvci 




Il 122 mttrcs above sea-leveb north of Tananger harbour. i.Sept. i, igii 



Stance, very often continued horizontally without a break from regions 
built up of very resistant rocks, like granites and gabbro, into regions 
with weak schists (chlorite schists, phylüte, &c.). This is the case in the 
regions of Bomlo Island, Karnitii, anrl Stavanger Peninsula. A difference 
may be that in the regions of the more resistant rocks there are often more 
and higher hills and mountains surmounting the plane of the strandfiat, 
and the latter may not have as wide an extension there as in the regions 
of weaker rocks. But even this is not always the case, c. g. on Bømlo Is- 
land, and the height of the real plane of the strandfiat does not as a rule 
differ. 

The onlv simple explanation of this striking feature seems to me 
to be that although the principal processes for the lowering and sculpturing 
of this coast have been the subaërial denudation and the glacial erosion, 
the final levelling of the plane of the strandfiat has been accomplished by 
the marine rlenudation. ;'. e. chiefly shore erosion by frost. After this plane 
had been thus formed, there cannot have been any great amount of sub- 
aerial denudation nor glacial erosion within the outer regions of the 
strandfiat. for otherwise greater differences in the heights of th? plane 
of the strandfiat would necessarily have been created, especially wh.ere the 
power of resistance of the rocks differ much. On the otlicr hand the 
strandfiat has been exposed to some glacial erosion, wliich has to some 
extent broken the level surface of the strandfiat and made it less even 
than it was originally. The effect of this glacial erosion has naturally 
differed somewhat with the structure of the rocks and their power of 
resistance. 

It has, however, to be considered that the power of resistance to shore 
erosion and also to glacial erosion depends less on the hardness of the recks, 



lOO FKID'IJOK NANSF.N. M.-X. Kl. 

lliaii Uli tlicir tciiflciicy tu he split hy tlic frost or lo form n rouj,''!) surface 
lo l)c att.ickcil hy tlic iiio\ inj^- icc. it is for instance striking that the 
granites haw in sonic rc-j^ions been (juitc as nnicli croflc-fl as the much 
softer schists, |)roI)ahly because they have as a rule a r^Ai^her surface. 



Jæderen. 

ja'fleren or j.'eren is the low l;iiif| exteii'lini,'' aloiii,'' tlie coast between 
the Stavanj4"er Peninsula to tht- north aiul the J'-f^a-rsuiul rej^^ion to the 
south. It is II to 13 kilometres l>road and bounderl along its eastern side 
by a mouiUainous land (of Archæan rocks) rising- more or less abruptly 
above the plain, w ith comparatively steep mountain slopes, to altitudes of 
100 to 200 or even 250 metres. 

The plain of Jæderen is formed to a ^"reat extent by quaternary, 
chiefly glacial, accumulations which have filled u]) the depressions and 
hollows of the rocky ground, often to great thickness. In numerous spots 
here and tliere tlie bare rock appears, however, in tlie surface of the plain, 
and in the northern and southern part of the plain this is even the case 
near the outer coast line [cf. the map by Grimnes, 19 10]. 

The plane indicated Ijy these exposures of bare rock has a height 
above the sea of about 15 to 25 metres, near the coast, in the region of 
Sele and Byberg and eastwards to Hegre in the northern part of Jæderen, 
south and southeast of Hellesto (Hâland). According to Bjorlykke [1913. 
p. 16] the protruding rocks southeast of Sele are hornblende- and mica- 
schists. According to Grimnes's map the protruding rocks rise to higher 
levels, al)Ove 50 metres, in the region further inland towards east-south- 
east, at Svensvoll, Lea, and Skjæveland, where the rocks consist of a grey 
gneiss [Bjorlykke, 1913I. 

Further east towards the foot of the mountains forming the eastern 
boundary of the plain, the level of the rocks is again lower, about 25 metres 
or even less, and at Åse, north of Holland railway-station, Bjorlykke ob- 
served hornblende-schists and layers of brown granular limestone (marble). 

Further south, between Bore and Kiep, there are rocks of mica-schists 
and gneiss rising to 50 and even 75 metres above the sea. at a distance 
of only 3 to 6 kilometres from the coast. 

Rocks of hornblende-schist, gneiss, and amphibolite rise to similar 
heights of between 50 and 75 metres in the region south of Kiep towards 
Tu, where the hill Tua or Tinghaug (hornblende-schist and amphibolite) 
is even higher. 

An almost continuous region with numerous exposures of Archæan 
locks, chiefly gneiss, and hornlilende-schists [Bjorlykke, 19 13, p. 15], ex- 
tends from Nærbo towards east-northeast. The rocks (gneiss) in the 
western part of this region, at Bjårland north of Nærbo. 4 kilometres from 



ig21. No. II. THE STRANDFLAT AND ISOSTASV. 




Fig. 90. Kock.\- coast with strandtlat at HcUestn at the northern end eil Jæderen. i^îept. 2, 19111. 

the coast, have heights of less tlian 20 metres, or about 17 metres above 
sea-level. But further east the heights rise towards 50 metres and more. 

Further east and northeast, in the region of Tuneim, Mossige, Lende, 
Folland. and Hoiland near the foot of the high eastern mountains, the 
protruding rocks, observed by Bjorlykke, consist of phvllite. and their 
heights rise to 75 and lOO metres above sea-level. 

In the southern part of Jæderen, north of Ogne, the level of the 
distinct plane of the low protruding rocks, of granite and labradorite, in 
front of the higher hills and muuntains (of labradorite rocks) is between 
15 and 25 metres. 

It is of course impossible to study in detail the topography of the 
rocky ground where it is covered to such an extent bv quaternarv accu- 
mulations as is the case on Jæderen. It seems, however, probable that if 
these accumulations were removed, we would have a low somewhat uneven 
rocky surface to some extent perh.aps broken up into peninsulas and is- 
lands separated by shallow fjords and sounds. But this rocky surface 
would probably to a considerable extent have low heights of about 10 or 
15 to 25 metres above sea-level, as indicated by the rocks especially in the 
northern and southern parts of the plain. 

I see no reason why this low rocky foreland should not be a strand- 
flat of the same kind as the strandfiat on the Stavanger Peninsula 
(forming its direct, northern continuation) and along the coast further 
north. 

The striking topographical difference between the plain of Jæderen 
and the higher mountainous land to the east mav to some extent be due 
to the difference in geological structure, the rocks of Jæderen being 
largely crystalline schists and partly phyllite which may have been more 
easily eroded to low levels than the Archæan and igneous rocks of the 
higher land to the east and south, although we have seen that fairly 
resistant Archæan rocks (gneiss, amphibolite, &:c.) also occur on the low 
land of Jæderen. 

But howsoeA'er this may be, it seems to me obvious that the low and 
nearly horizontal level, or perhaps levels, marked by the rocks protruding 
in the surface of the plain of Jæderen, cannot have been formed solely by 
subaërial denudation, and still less by glacial erosion. The simplest ex- 
j^lanation is here as elsewhere, that these low horizontal levels of the 
coast-land have been finallv created bv shore-erosion levellin<^ the hills 



IKIDTIOK NAN.SKN. 



M.-N. Kl 




I''ig. 91. The land nortli-wi'stvvard along the coast, seen I'rom a hill (135 intlrtsi (jn the peninsula north of 

l'resteskjær Lighthouse. (Se[)t. 3, 1911). 

of llie low coast, which had already beforehand been much Icnvcred to- 
wards sca-Ievel by tlie other agencies. It is also a striking feature here 
on Jæderen, that the plane of the stranrlflat extends with ahnost uniform 
heights along the coast over regions with rocks differing very much in 
their power of resistance, a feature wdiich is characteristic of the result 
of the shore-erosion. 

Ahlmann suggests that the erosion of the inland-ice in a previouslv 
flat region, like Jæderen and the Stavanger Peninsula, would result "in 
soft and still more levelled forms than l)efore" [1919, P-481. Although 
the glacial erosion may grind the forms rounder and softer, I do not con- 
sider it possible that it can make any extensive rocky ground more level 
than it was before. C)n the contrary T think it will always make it less 
level, and it is, in mv opinion, probable that it is just the glacial erosion 
to wdiich Jæderen has been exposed after its strandfiat was finally levelled, 
that has made its rocky ground so uneven in height as it now seems to be. 

There is no reason why the inland-ice shouhl have had any special 
tendency to form such nearly horizontal planes just above sea-level. As 
pointed out before, the erosion of the inland-ice and the big glaciers is 
not limited in any way by the sea-level, but goes on equally well at all 
levels down to the depths below the sea reached bv the moving ice. \\'hy 
then are these planes always limited to the coasts? Whv has not the 
inland-ice formed similar levels and horizontal rocky plains under the 
water, and also far inland, awav from the coasts? 



The high, steep Coast between Jæderen and Lister. 

Everybody travelling southwards along the coast of Jæderen must be 
struck by the sudden change in the whole character of the coastland in 
the region of Ogne, about 17 kilometres north of Egersund. where the 
mountainous land suddenly approaches the coast, leaving only a com- 



1921. No. 1 1. 



THK STRAXDFLAT AND ISOSTASY. 



103 




Fig. 913. \'ic\v south-westwards towards Sogndal and the moiilh of Josing Fjord, troni a hill I135 nn'tres) on 
the peninsula north of Presteskjær Lighthouse. (Sept. 30, igri). 



paratively narrow, hut low and well marked .strandfiat, cut in solid rock, 
between the .shore and the fairlv steeply rising hills. This strandfiat may 
he a couple of kilometres broad, with a level perhaps about 20 metres or 
less above the sea. It extends along the coast southeastwards as far as 
Egeroi, on the western peninsula of which it is fairly well developed. 
But further southeastwards from this island the coast becomes high and 
often precipitous as far as Lister, and there are only few indications of 
a very narrow strandfiat in front of the steep mountain slope along this 
coast. 

The sivdden change in the character of the coast is obviously due to 
the difference in geological structure, the coast in this region being built 
up of resistant igneous rocks, the so-called Egersund labradorite rocks 
[Kolderup, 1914]. This high coastland is dissected bv narrow vallevs, but' 
its general surface is fairly level or undulating, rising gently inland from 
a fairly equal height of about 130 to 150 metres, above the sea, near the 
coast. Tt is j)robably the same kind of formation as the fairly level coast- 
land of southern Norwav, about 100 metres high or somewhat more, 
which Ahlmann calls the base-levelled plain of southern Norway. I think 
it represents more or less the Palæic mountain surface of this region 
which has not been substantially lowered bv glacial erosion because the 
inland-ice has had very slow movements over these mountain plateaus, 
although the rocks have everywhere 1 een rounded b\- the moving ice (see 
Fig. 91). 

In the .region northwest of Presteski;cr Lighthouse I found, by baro- 
metric reading, the fairly level mountain surface (Fig. 91) to be between 
140 and 180 metres above the sea, and to the southeast, in the region of 
vSogndal and losing Fjord (Fig. 92), it was ])erhaps about 170 to 200 
metres liigh, with a steep declivity down to the shore, where in several 
places, under the headlands and elsewliere, there are ledges cut in the 



I04 



rUIDTJf)!" NANSKN. 



M.-N. Kl. 



hQO 



^zoo 




c 

'•''S- 93 



'»ZOO 



l'rulilr ol tlic coast and the islets outside, at Dybland and IratoddL-n, the promontory 
south of Jøsing Fjord (see Fig. i6, 25). 



solid rock (see l'ij:^s. 91 and 92). Alonj^ the coast nortliwest of Preste- 
.skjcer, as well as outside .So^-ndal, Josinij;" Fjord, and some other places, 
liiere are a number of low rocky islets and skerries (Fig'. 92) consisting of 
the same kiml of rock (lahrarlorite) as tlie mainland. In connection with 
the ledges along the shores inside, they form a low and distinctly marked 
strandfiat in front of this high and steep coast (see the profile Fig. 93). 

The many islets and skerries along the steep coast west and southwest 
of Hitteroi, west of IJstcr Fjord, form a similar strandfiat. 

It can hardly be dou1)ted that these shore-ledges and the low islands 
outside, have been cut by marine denudation (shore-erosion). They are 
in appearance perfectly identical w ith those we have founrl in Sogne Fjord 
and Hardanger Fjord. Here along this outer coast 1 suppose all ideas 
have to be given up that these formations could be due to base-levelling 
by subaërial denudation, or that they might l)e remnants of the floors 
of some preglacial valley-generation. 



From Lister to the Naze (Lindesnes). 

Like Jæderen the low j)lain of Lister is to a great extent formed 
by quaternary (glacial j accumulations, but especially in its southeastern 
part much rock rises to the surface of the plain, indicating a low and 
fairly horizontal level, or probalily tw^o levels, the lower one being also 
marked by the many low rocky islands and skerries outside the coast in 
this region south of Farsund (Fig. 94). 

The rocks of the low land and the islands outside are chieflv Archæan 
gneiss and granite, partly also diorite and gabbro [Reusch, 1901. p. 92]. 
The higher mountains behind the low foreland consist chieflv of Archæan 
gabbro or labradorite rocks further inland. 

Though there is a marked rlifference l)etwcen the low foreland, and 
the higher land behind, the mountain-slopes do not on the whole rise so 
abruptly from the plain as for instance in some regions of Jæderen. but 



I92I. No. II. THE STR.WDFLAT AND ISOSTASY, 







Fig. 94. South-eastern part of tlie Lister Land, from Einarsnos north-westwards towards the valley of Fram- 
waren, inside Helvig Fjord. View taken from the sea near Asegrund west of Svaneflii. Ijuly 25, 191 1 1. 

nevcrtlieles.s I see no reason wliv the low rocky levels should not be parts 
of a strandfiat of the same nature as the one we have found extendinj^" all 
along- tlie coast to the northwest and north. 

The mountainous land insifle the low land of Lister has a fairly level 
surface at about 200 metres or more above the sea, rising- in Ravncheien 
to 350 and 366 metres (see Fig. 94) . 

Between Lister and The Naze (Lindesnes) and along the coast further 
eastwarfls where the rocks arc ])artly gabbro or labradorite (between Far- 
sunrl and The Naze) and partly Archæan, there are a great many low rocky 
islands anrl skerries indicating distinctly the lower level of a strandfiat, 
and to some extent perhaps two levels. 

The promontory of the Naze itself (consisting of Archæan rocks) 
belongs to the strandfiat and exhibits two marked levels, a general liigher 
level at about 16 to 18 metres above the sea. and another very low level 
at the outer point, only a few metres above the sea (Fig. 95). The level 
of about 16 to 20 metres was also observed further east, c. g. at Kje- 
holmen and Kalvehaue on Imsa Island at Spangereid (Fig. 96). 



The Southern and South-eastern Coast of Norway. 

Along the southern coast of Norway, between Mandai and Christiania 
Fjord, there is a Ijorder of numerous islets, skerries, and rocks, but there 
are comparatively few perfectly certain indications of a strandfiat. Ahl- 
mann goes, however, decidedly too far when he asserts [1919, p. 36] that 
"there is no sign whatever of marine abrasion or the presence of any 
marine coastal plain". 

Where the general land surface is low and slopes gradually towards 
the coast, as it does in this region, it is very often difficult to decide 
where there is a coastal plain or strandfiat cut by shore-erosion, and 
where it is only the general low level of the land surface that slopes gently 
into the sea. It seems to me, however, that a plane so nearly perfectly 
horizontal as is exhibited by many series of islands along this coast (see 
for instance Fig. 97 and especially l*"ig. 98) cannot have been formed solely 
by subaërial denudation and glacial erosion, but it must have been finally 
levelled by shore-erosion, in the manner I have previously described. 
Sometimes one mav observe a distinct break between the horizontal level 
of this narrow stranrlflat ami the slope of the higher land inside. The 



io6 



FRH)r.|OF NAXSr.N. 



M.-X. Kl. 




ï"'êi- 95- l'i*-' Nazi- (Lindi-snc's). Tlic liylil ol tlu- lif^lithousc is 50.1 metres above sea-level. 

(July 25, 191 I). 



§ 



Fig. 97. Land nortliward Irom I'K'oisund, east ol C hrisliansand. (Sipt. 4, i g 1 1 1. 



islets anrl skerries in some ])laces show inflicalions of two levels of shore- 
erosion. 

Some observations in the region of Langesund may he of interest 
to prove the existance of the strandfiat along this coast. 

The low islands in the mouth of Langesund Fjord, when seen from 
the sea (Fig. 99), exhibit a low level distinctly different from the much 
higher surface of the mainland on both sides of the fjord. The Langesund 
Peninsula shows a higher level. 

Fig. 100 represents the southern point of the peninsula on which 
Langesund is situated. Tliere can be no doubt that the flat points and 
ledges, a few metres above present sea-level, with the steep cliffs behind, 
have been cut by shore-erosion. They have obviously the same level as 
the islets outside, seen in the picture. Tt is hardly probable that these 
ledges and islets have been formed in postglacial time. We may b? con- 
vinced of this by looking at the coast on the western side of the peninsula, 
a view of wdiich is given in Fig. 10 1. The level of these low flat islands 
with rounded, ice-worn surfaces is approximately the same as that of the 
flat points and islets of Fig. 100, and it is also the same as the low. flat 
points forming distinct horizontal incisions in the slopes of the two pro- 
montories seen behind the low islands in Fig. igt. These formations can- 
not l)e due to postglacial erosion. 

Hence, we have here a low level or strandfiat along the coast, due to 
shore-erosion, and distinctly different from the higher general level of the 
land behind, which is seen as a fairlv even surface on the peninsulas in 
Fig. loi and also in Fig. 99 on the left hand side. 

The peninsula of Langesund (Fig. 100) is built of sedimentary rocks 
(phyllite) and so are the low islands and the nearest promontory in 
Fig. 101, but the distant promontor}- in this picture, which has an 
especially distinct low point or strandfiat under the steep shore cliif. is 
built of Archæan rocks. 

Whatever the genetic origin of the higher, fairly level surface of this 
land may be, there can. in mv opinii^n, be no doubt that the lower level 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASV 




Fig. 96. Indications of a strandllat outside Spangereid, east of Lindcsnes. Kjeholm in the foreground. 
To the right the point Kalvchaue on Imsa Island, with a well-marked strandtlat. 'July 25, 191 il. 




Fig. 98. Islands north of the entrance to Kragero. iScpt. 4, igiil. 

of the small islands and the points in this region is a regular strandfiat 
of the same nature and origin as the strandfiat of the Norwegian west coast 
where it has, however, a much wider extent. 

It mav be that this low level of the strandfiat along- the southeastern 
coast of Norway corresponds to the lowest level (12 — 17 metres high) of 
the strandfiat of the west coast. The reason why it is here so much lower 
might be, in that case, that this southeastern coast has been less elevated 
than the west coast after the last development of the strandfiat, because 
the denudation of this low land, and the amount of waste carried away 
from it after that time, have been considerably less than what was removed 
from the much higher and steeper west coast. 



The Inner End of Christiania Fjord. 

As Prof. W. C. Brogger's well known investigations have proved, 
the Christiania Fjord is a fault trough, in the inner part of which the 
Silurian or Ordovician clay-slates have sunk down to levels below those 
of the Archæan region to the east and of the igneous rocks forming the 
hills to the north and west of the fjord. ]t is to a great extent b\' the 
subaërial and glacial erosion of these clay-slates, much less resistant than 
the Archæan and igneous rocks on the sides, that the depression of the 
fjord has been formed. 

Tf one looks down upon the Christiania valley and the fjorrl from 
one of the heights round this vallcv, it must strike one that the surface 
of the many islands in the inner part of the fjord, inside I^ågoi, Lang- 
åren, and Steilene, and also the peninsulas, Bygdoi, Fornebo, Snaroi, and 
Konglungen, forms a remarkablv level j^lane. where the ridges generally 
rise to between 20 and 30 metres above sea-lc\-cl, and very seldom abtn-e 
40 metres (cf. Fig. 102). 

The lanfl of Archa\'in rocks to the east, i. c. the Nesodd Land and the 
land east of Runde h'jord. ri.scs abruptlv, Vvith a sharpl\- marked fault 



io8 



lUIDTJOF NANSKN. 



M.-N. Kl. 






«■»1^. 



<-ft>.<»/—.^ "«w^ol«^« tr'i^tmer 9 i-Jfftiu^t ",1,', 




Fif;. 99. l'iiiti-anci- lu I .anf,'cMinil Ijuril. iSijit. 4, 191 ii. 

cscarpninit. aixivr tins plane, lo 150 aiul 200 nielres above sca-level, while 
llic land in Akcr, lla-runi, and Asker, rises more gently in ridges to the 
foot (jf the liiglier liills lo tlic north and west of the fjord. But even here 
there is often a distinct rHfference l)elween the undulating slopes of these 
\'alley sides and the lower ])lane of the islanrls and peninsulas of the fjord, 
as is demonstrated l)\- the profile l*"ig. 103. 




t2S«Ê^. 



Fi.t;. 100. Lhlls with low puintb on the buutlicrn end of the Langesund 
Peninsula. (July, 1904). 

Fig. 104 gives a photographic view of Foruebo Peninsula and the 
islands and hills towards the south-southwest and west-southwest taken 
from the top of the writer's house on the hill marked X i'l Fig. 102 in the 
northeastern part of Fornebo Peninsula. The peninsula as well as the is- 
lands and the low land to the right in the picture, form a nearly horizontal 
plane, distinguishing itself sharply from the steeply ascending hills be- 
hind: Skogumsås, N'ardeås, &c. As the land is covererl with pine wood, 
the plane appears in the picture higher than it actually is. The Xesodd 
T.and to the left in the picture, rises very steeply from the fjord, with 




Fig. 101. Indications of a strandflat on the eastern side of Langesund Peninsula. 'July, 19041. 



1921. An. 1 1. 



THE STRAXDFLAT AND ISOSTASV. 



109 







I TO 



FRIDTJOF NANSEN. 



M.-X. Kl. 




K o L s å- s 



fltiilcum 



Rhomè- and EsSijiLtxi-Poi^phl/i-l^ \Y''..'y.-^ Dci^ofLLuH SaJlcUton 



^^---^-^^y^^--- Os 

I 
SUu-tUui HcJtLsCs iifCU, I 



Fig. [03. Profile across the inner end of Christiania 



indications of a strandfiat formed by the islands of Tldjernet (seen in the 
picture) and Steilene at the fof)t of the steep mountain side (cf. Fig. 102). 

This plane cannot, in my opinion, have been finallv formed bv sub- 
aerial denudation, and still less by glacial erosion. These processes would 
have made the general surface of this land more sloping or more bowl- 
shaped. The plane must have been finallv levelled by marine denudation, 
/. c. shore-erosion by frost, in a similar manner as we have seen (p. 37) 
that shore-ledges have been formed at present sea-level in postglacial time 
in this same region. 

The shore-erosion has found very favourable conditions for work in 
this region of easily crumbling argilaceous schist, dissected bv glacial 
erosion into numerous islands, peninsulas, and points with a verv long 
shore-line for attack. It is therefore not surprising that, although the 
inner part of Christiania Fjord was probablv filled bv glaciers during 
the greater part of the various glacial epochs, there has still been time 
enough for the shore-erosion to cut this plane, extended over so wide an 
area, in a low dissected land where there was only a comparatively small 
(|uantity of easily crumbling reck to remove in order to level the surface 
down to sea-level. 

The plane thus formed has afterwards been eroded bv the glaciers 
of the last glacial period which have denuded more or less the weaker 
parts of the rocky surface, while the more resistant parts have b?en less 
affected and form ridges largely going in the direction of the strike of the 
folded schists. The fairly flat summits of these ridges have to a great 
extent nearly uniform heights of between 20 and 30 metres above sea- 
level. In some places, especially on Snaroi, Ostoi, Bronnoi, and Xesøi 
(see the white spots in Fig. 102) they rise above 30 metres, and only in 
a few cases they rise to 40 and 50 metres. 



^ 



1 92 1. No. 1 1. 



THE STRA.NDFLAT AND ISOSTASY 



"fOO 



200 



Kil-omttt-L^ 




Ch iLstiama 

\0 tclovxcian. Schists conia^ning mucJi Luize^ 



^ Aicha cort Gneiss 



-1 Kolsas to Nesodd Land, from A to B in Fie;. 102. 



Tt is difficult to decide what the exact height of this plane may 
originallv have been, because after its final planing, it has been denuded 
by glacial erosion. Most of the land on the Fornebo Peninsula, for in- 
stance, is lower than 20 metres, and a great deal of its area is even lower 
than 10 metres, but this chiefly consists of fields formed of loose material. 
The many long rockv ridges running in the direction of the strike (about 
WSW to ENE) [cf. Werenskiold, 1911] between the flat fields are mostly 
between 10 and 20 metres high, with their fairly flat summits slightly above 
20 metres. They show, however, distinct traces of glacial erosion. Their 
summits are rounded and the dykes of diabase rise above the argilaceous 
schist with rounded, polished, and striated surfaces (cf. Fig. 17). These 
rockv ridges have, therefore, obviouslv been somewhat lowered bv the 
glacial erosion, but it is difficult to say exactly how much. I think, 
however, that the general plane may be assumed to have been about 25 
metres above present sea-level, or perhaps a few metres more. 

The sides of the rocky ridges are frequently so steep and abrupt that 
they look like shore formations or shore cliffs somewhat modified by later 
'dacial erosion. 




Fig. 104. Tlie plane of Fornebo Peninsula and the islands and land to the south-west, seen against the surroundung 
higher hills. View from south to west-southwest from the top of the writer's house near Fornebo. (.April, 1922). 



FRIDTJOF NANSEN. M.-N. Kl. 



Acor'IiiiL;' \'> wlial lias hccii sai<l al)(»\c, llic low lc\cl rc^\(i]] al the 
iiiiKT cml of ( liristiaiiia l-jonl, may he considcrcfl to be a rej,^ular slranrl- 
flat formed in a manner similar lo llial of the west coast. 

W'liellier there are two différent le\'els, or even more, in this rej.^ion, 
is ili ff icult to decifle, l)ecause the (jrig^inal plane or planes have been mfjre 
disturbed bv the late j.^lacial erosion than those of the strandfiat of the 
Norwej^-ian west coast. Another flifficulty hampcrinj:^ the investigations 
is also that the strandfiat in the Christiania reg-ion is almost everywhere 
wooded. 

Although it mav look as if there are indications of at least two levels 
in several places, \ dare not express a definite opinion without having 
made more careful imestigations. 



ig2I. No. 11. THE STRANDFI.AT AND ISOSTASV. II3 



IX. THE STRANDFLAT ALONG THE COAST 
FROM SOGNE FJORD TO X'IKTEN. 

The Region of Søndfjord, Nordfjord, and Stad. 

The Norwegian west coast in the reg^ion of Askevoll and Floro north 
of Sogne Fjord has a distinctly developed strandfiat in the shape of 
numerous low islets, skerries and sunken rocks, and in many places there 
is a low foreland in front of the mountainous coasts of the higher islands 
and the mainland. The rocks are here Silurian schists and later (upper 
Silurian and Devonian) sandstones and conglomerates; and in some few 
places (near Askevoll and on the southern side of Bremanger Land) 
granite. I have not had an opportunity of investigating the strandfiat 
in this region. By a study of the charts of the coast one will notice that 
in many places the submerged platform with shoals and rocks is more 
widely developed in this region than outside the coast farther south. 

Dr. Reusch gives [1901, p. 189J an illustrative drawing of the narrow 
strandfiat (10 metres above sea-level) at the foot of the steep mountain 
side on the outer coast of Askroven Island in Sondfjord. Small cirques 
with nearly flat floors have there been carved I)y the shore-erosion into 
the mountain sides. 

North of Bremanger Land the coast is built up of Archæan recks, 
and the strandfiat is less conspicuous. 

Neither on nor outside the high Stad Promontory are there indi- 
cations of a strandfiat, excepting perhaps a few sunken rocks. In this 
respect the Stad Land forms an exception from almost the whole rest of 
the west and northwest coast of Norwav. 

Along the coast northeast of Stad Land there is again a strandfiat 
consisting of a submerged platform with numerous skerries and rocks, 
and often a low foreland on the islands. The rocks are presumably 
Archæan. 

The Region of Ålesund. 

The high islands Godoi and \'alderoi, outside Ålesund, have the 
typical shape of a hat swimming on the water, the low strandfiat forming 
the flat brim above sea-level (Fig. 105). This shape, which was already 

Vid.-Selsk. Skrifter. I. M.-X. Kl. 1921. No. 11. 8 



114 (KIDI.Idi- N.WSKN. M.-X. Kl. 




Fig. 105. Islanc 



(lcscril)C(l l)y i )r. Kcuscli as cliaraclcristic <»f llie stranilflat, is often seen 
even on small islets ami holms. 

T measure»] the heii^'hts of tlie strandflat on the southern and s'juth- 
westcrn side of Haroi to be between 17 and 25 metres ai)ove the sea, 
on the west coast of Alien Island 28 metres, and on the small Dronen Is- 
land, witli the mountain Droneliatten in the nuddle, 18 metres. These 
islands are northeast of Alesunrl. and the heij^hts were measured with 
sextant from off the coast of J^^jortoft Island, where I was lying at anchor 
on Aug'ust 22nd, 1909. Owing to the distance the measurements cannot 
be expected to be very accurate. 

The Borgund Fjord, just south of Ålesund, is narrow and fairly well 
sheltered against the open sea outside. Tliere is none the less a well 
developed strandflat along the northern coast of Suloi, and on the islands 
in the fjord (Fig. 106). 

The mainland as well as the islanrls in the region of Ålesund are 
formed of fairly resistant rocks, generally supposed to be Archæan, but 
lately by J. Schetelig [1913] maintained to be younger eruptive gneisses 
of the great Caledonian "folding-ditch" of Northern Norway. The low 
parts of the row of islands north of Ålesund — including Giskoi, Aalderoi, 
\"igra, Lepsoi, Haramsoi, Flemsoi, Fjortoft, and Haroi — are to some 
extent built up of quaternary (glacial) material. But the strandflat de- 
scribed above, is largely cut in solid rock. 

From the seaward coast of these islands submerged platforms, with 
hundreds of rocks and shoals in or slightly below sea-level, extend 7 to 
10 kilometres out into the sea. 

It is a noteworthy fact that the sul)merged platforms with sunken 
rocks and shoals are much broarler in these northern regions than any- 
where along the Norwegian coast further south. Advancing along the 
coast northeastwards and northwards we will find the submerged strand- 
flat increasing still more in wirlth. 

This striking difference in the development of the submerged strand- 
flat cannot be explained l)y a difference in the geological structure of 
the coast, because in this region the coast is built up of very much the 
same kind of what is supposed to be Archæan rocks, as in many places 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND I50STASV. 



I olden 



outside Ålesund. 



further scutli; tlie rocks of tliis reqion are at anv rate not less resistant 
to erosion than those further soutli. 

The natural explanation of the increasing development northwards 
of the submerg-ed platforms of the strandfiat is in my opinion that in these 
northern réglions the marine erosion, and especially the shore erosion by 
frcist. has been much more favoured l)y severe climates than further south. 
Hence, the (|uant!ty of rock cut away bv shore erosion during" past ag'es 
much increases northwards with increasing- latitude. 

There is also the likelihood that in these northern regions the climates 
liave. pro])ablv always, been more stormy than farther south. As I have 
pointed out before, the transporting and erosive capacity of the waves 
increases with something bewecn the third and sixth power of the velocity 
of the wind. Along a coast like this, exposed to the full power of the sea, 
this may also be of much importance for the marine denudation. 

The severer climates of the northern regions may also increase the 
subaërial denudation, but not sufficiently to explain the striking difference 
in the development of the strandfiat and its submerged platforms. It is 
a striking fact that in these northern regions the submerged parts of the 
strandfiat have considerable areas as compared with those of its emerged 
parts. 

In the region of Hiistad there is a conspicuous level strandfiat ex- 
tending over the low skerries outside the coast, as well as over the outer 
border of the mainland, where the steep, often precipitous mountain sides 




Fis. loo. .Strandtlat in Bortcund Fjord, iiiskIc Alc;>und. 



,Y l/Vcst i±om ClvUstiutiia 



l:'iOO 000 



Stiandflat qf Smcletti[),Hmci€tt{IIl 
rtëi/tdll), and Ftxran lslaruts(lV>. 

mH La^nd Umm^ than. 30 mettes abc2>c SUl-IovcI 

j^@ Land helAffm 30cmd GOvidtM ahwt sai-kvel. 

^^^Piain on Hateten kighct- Uuin. COmcUu. 

\ \ LancL highet thcun, SOmcl^es about sca-kvel. 

^^^Subei^ed Fluffeiins hssthan TSniePics hdcw ota-ltvd-. 











Fig. 107. Map showing the strandllat in the Region ol 




^»olen (I), Hittenen (II), Froia (III), and Froan Islands (IV 



KKIDIJ'il' NANSKN 



M.-N. Kl. 




Fi"-. 1 08. Profile across 



Smulen and Tusteren [islands, iron. pouU A to south o. po„U B .n Fig. xo8^ J 

gives the natural relati 



rise abruptly above the plane to he.gbts of 600 and 700 nutres al,o. te 
sea This coast gives a convmcing hnpression of how tlic plane ot the 
strandfiat ,nust have been cut hor.zontally into the oversteepenc^d mountam 
s.des [cf. Nansen. .904, PL V, F,g. 5I. The h,gh mountain Stemshesten 
is"consp,cuous. nsh,g as a -stack" above the level plane wh.ch however 
is to a great extent formed of quaternary loose n,ater,al on both s.des 
of the mountain. 

The maniland as well as the islands of this region is built up of rocks, 
supposed to be Archæan, but Schetelig [1913] considers them to be younger 
eruptive gneisses and igneous rocks. 

The strat.lflat extends as submerged platforms with innumerable 
shoals and rocks 9 to 15 kilometres out to sea, a region much feared by 

all seafarers. 

The surfaces of these sulnnergcd platforms are fa.rly uneven and 
are dissected l,^■ nunu-rous depressions making the depths varv. As^" ' 
be nientioned later, this ,s generallv the case where the sul>n,erged p at^ 
forms are bnilt up of resistant rocks l.ke granite or gneiss, &c. There 
are manv scattered shoals and rocks near sea-level, or about 5 to .0 "letres 
below it, some rocks and skerries also nse above the sea. It is d,tf>c 
to decide exactiv at what level the plane ot these subn,erged plattur.n 
stands. There .s no sharply def.ned edge of the plane, as the surfaces o 
the platfor,ns slope towards the deeper sea on the sides, as a rule w.thou 
anv sharp break. If, however, the tops of the many shoals and sunken 
rocks mav be taken to indicate approximately a plane, tins stands ^er^ 
near present seadevel, or perhaps five to ten metres below ,t. 

OutsKle the coast in the region ot Chr.sfansund, there ,s a series of 
submerged platforms w.th shoals and rocks and small islands, situated, 
as it were, on a submarine r>dge running parallel ,0 the average d.recon 
of the coast, and forndng a continuat.on of a ridge on which are situated: 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASV 



119 



Jotçjuisvàij ScUia 907th. 
TmUinciksLa 675 r/i. 



rIOOO 




ff'^^\ CongLometcUc-Samistom ^^^ Auk<kxuL otyoiinge^ Igmous Rocks 



500 



deal scale is 5 times exaggerated in relation to the horizontal scale. The proSle in black on the land 
height to length. 

Orskjærene north of Hustad, Ona (with (Jna Lig"hth()use ). and sur- 
rounding submerged platforms, Sandoi. and the before mentioned series 
of islands, Haroi. Fjortoft, FlemscM, Haramsoi, Lepsoi. and \'igra. The 
high islands of Smolen. Hitteren, &:c. are situated on the same ridge to 
the northeast (see later). 

This ridge was probablv formed during the great Caledonian mountain 
folding of this part of Norway. It is separated from the coast of the 
mainland, and the mountainous islands near it, bv a channel (the fairway 




Fig. 109. The surface of the strandflat on Smolen. Northward view from the Nel\-ikberg 
167 metres above sea-level I towards Roksvag. (Photograph by P. Schei. July, 1Q041. 



I20 I RIDTjOK V.WSCN. M.-N. Kl. 



lit;, iio. Siranilllat aloni; tin- west coast of Hittcren, with tlic Skar P'jcll behind, seer 

cjr 'k'flcn"), oflcii more than 2uü niclrcs flccp. Tlic distance from the coast 
to the outer c(h^c of these suhmerj.^cd jjlatforms is in the region of 
Christiansund about 12 or 13 kilometres. The submerged plateau of Grip 
with nmiierous islets and skerries, outside C'hristiansund, may be especially 
nient iontMl. 

These submerged i)lat forms ha\e uneven surfaces similar to those 
of the platforms, just described, to the southwest. The CJrip Plateau 
seems, however, to have in parts somewliat more level surfaces with more 
distinctly defined edges. It may pro])al)ly be Ijuilt up of somewhat less 
resistant rocks, like those of the Smolen Plateau to the ncjrtheast. 

The Region of Smølen, Hitteren, and Frøia. 

The region of Smolen, Hitteren, and PVoia is interesting because of 
the complexity of the geological structure, and the unusually well developed 
strandfiat which cuts horizontally through the various geological forma- 
tions, frequently without any appreciable break in its level plane (cf. Figs. 
108 and III). 

Especially on Smolen and its neighbouring islands (see the map 
Fig. 107) we find the most perfectly developed plane of the emerged 
strandfiat which occurs along the coast of Norway. 

Smolen island is 16 kilometres from south to north and 20.7 kilo- 
metres from west to east. The whole of the island is one unbroken plain, 
which is especially even in its northern part wdiere only two small hillocks 
(Dyrnestuva and Måbergtuva) rise slightly above the plane. The height 
of this almost perfect plane is 15 to 20 metres above sea-level. In the 
southern part of the island the plain is somewhat more undulating and 
a few isolated hillocks rise above it to 26 metres (Storsetberg), 33 metres 
(Hoaåsen), 35 metres (Torshaug)* 39 metres (Rambergtuva), and one 
even to 45 and 67 metres (Nelvikberg) in the southeastern corner. 

The northern part of Smolen consists of diorite with a belt of pressed 
igneous rocks along the northwestern coast [cf. Schetelig, 1913]. In the 
southern part there are various rocks, probably of Silurian agz [Reusch, 
1914, Schetelig, 1913] with enclosures of a limestone probably of Ordo- 
vician age [Holtedahl, 1914]. On the islands to the south of Smolen, on 
Kuli Island, Edoi, &:c. there are conglomerate and sandstone, probably 
Silurian. 

The profile Fig. 108 from the mainland (Saksen), across Tusteren 
Island, Kuli Island, Smolen, and the small islets to the north, gives the 



I92I. No. ir. THE STRANDFLAT AND ISOSTASV. 






from the eastern side of Smolen. (After photographs by P. Schei, July 11, 19041. 

various rocks according to Schetelig's map [1913;. The strandfiat is here 
about 36 to 39 kilometres broad, from the foot of the Tusteren mountains 
to the edge of the outer platform with skerries (see the map Fig. 107). 
This profile demonstrates the remarkable evenness of this strandfiat, its 
plane cutting horizontally through the geological formations, southwards 
to the foot of the steep mountains. The depressions between the islands 
north of Smolen, at Hvalpoi, Hammeroi, Inveien, &c., are drawn much 
too deep in the profile (Fig. 108). These depressions are not deeper than 
10 to 20 metres below sea-level, generally less. The same is also the case 
south of Smolen, at Rosvolløi and Kuli Island. 

On the islands south of Smolen, built up of conglomerate-sandstone, 
there are a few small hills rising above the plane: two hills to 44 and 
78 metres (on Kuli), two hills to 39 and 47 metres on Edoi. and one hill 
to 40 metres on each of the small islands Orten and Glasoi. 

Fig. 109 is a photograph taken by Mr. P. Schei in Juli 1904 from 
Xelvikberg (67 metres high) on southeastern Smolen. It gives an illustra- 
tive picture of the even surface of the strandfiat on Smolen. The' rocks 
are here diorite and are rounded by glacial erosion. The flat plain of 
northern Smolen is to a very great extent covered with peat-bogs. 

As our map Fig. 107 shows, Smolen is surrounded bv a submerged 
strandfiat which is aI)out 12 kilometres broad at its broadest on the 
southwestern and the northwestern sides of the island. The area of this 
submerged strandfiat is considerably greater than that of the whole island. 

Our map gives the submerged strandfiat at levels less than 25 metres 
below sea-level. Unfortunately lack of time has not allowed the writer 
to flraw more detailed maps showing its surface topographv. It is 
dissected by channels and depressions but not as much as the sub- 
merged platforms to the southwest which were described above. Its 
surface is more level than the surfaces of those, and in great parts it forms 
extremely level horizontal planes, near present sea-level. Hundreds of 
islands and rocks, rising above the .sea, are scattered almost over its whole 
area, also near its outer edges. 

If in a detailed chart the isobaths be drawn for 10, 20, and 50 metres 
of depth, it will be seen that the isobath for 20 metres has very compli- 
cated shapes, it comes almost everywhere close to the isobath for 50 metres 
near the outer edges of the platform, and even the isobath for 10 metres 
comes very near them to a great extent. This indicates that the edge of 
the submerged strandfiat is fairly sharply defined, and is in most places 



122 



I Kil. I loi- NAN.SKN. 



M.-N. Kl. 




Üimitr ftrid C(itnil-i.hi<riih 



Fig. in. Profile across Frøia and Hitteren from point C to point \) in Fig. ro8. The vertical scale is 5 times 

between height 

less than 10 tiiftrcs below sca-lcvcl, ;. c. the cd^œ of the horizontal plane 
iiirlicatol l)v Ihi- summits of the manv shoals and sunken rocks, where 
it is houndeil by the stecjKT sideslopes of th.e j)latform. 

'The t()p<),t;raphy of the larg-e island llittcrcii shows a consiflerably 
o-reater variation than that of Smolen, but there is a well developed strand- 
flat, especiallv in its western (Fi.y. no) and northern part (cf. Fi^. 107). 
The strandfiat is, however, not so perfectly level as on Smolen, and is on 
the whole somewhat higher, to a great extent above 30 metres. It cuts 
through a variety of geological structures. 

The northern and northeastern part of the island is according to 
Schetelig [19 13] built up chiefly of pressed igneous rocks, gneisses and 
gneiss-granites, with a belt of diorite passing across this region from west 
to east. The middle and southeastern part of the island consists of diorite, 
with a border of conglomerate and sandstone (prol^ably of Silurian or 
early Devonian age) along the eastern half of the south coast. The western 
part of Hitteren is built up of comparatively young granite [cf. Schetelig. 
1913, Reuscli, 1914]- 

In the profile (Fig. in), across Hitteren and Froia, the occurrence 
of the various rocks is indicated according to Schetelig's map. The strand- 
flat is seen cutting through the various geological formations to some 
extent, but in the southern half of Hitteren the land is higher (cf. Fig. 
107). This region consists to a great extent of diorite. There are a good 
many liills rising to 160 metres above sea-level or more, some even to 220 
and 227 metres, and Els F jell to 319 metres. The strandfiat extends al- 
most horizontally to the foot of the higher hills (cf. Jamt F jell in Fig. in). 

In the western granite region of Hitteren there is also a small sharply 
defined mountainous area, rising abruptly above the plane of the strandfiat 
to heights of 150 and 200 metres, Skår F jell even to 306 metres (MorvoUs- 
tuva) and 369 metres (Morkdalstuva). Otherwise the strandfiat is well 
developed in the coastal region of this western part of the island and 
on the manv smaller islands outside the coast to the west and southwest, 
likewise consisting of granite, as well as on the small islands to the north- 
west (Sæbuoi, Stromsoi, Gjedsoi, Skåroi, <K:c.) consisting of diorite. 

Although as was said above, the stranrlflat of Hitteren is not on the 
whole as remarkablv level as that of Smolen. still it is extremely even in 



I 



I92I. X(>. II. THE STRANDFl.AT AND ISÛSTASV. 1 23 




exaggerated in relation to thj horizontal scale. The pro.ile in black on the mainland i;!ves the natural relation 
and length. 

many i)laces. for instance in tlie region of Kvenvær. as Scheteligf has 
j)ointcd out to me. 

y ig. 112 is a photOL^rapli of the mountain Tonningen (231 metres 
above sea-level) in the western granite region of Hitteren, taken bv Prof. 
J. Schetelig. It demonstrates the abruptness with which the mountains 
rise with tlieir steep sides above the denudation plane of the strandfiat. 
This plane as well as the mountain is formerl of the same granite and tliere 
is no flifference in the geological structure to account for the configuration. 
Forms like these cannot therefore be formed solely by subaërial denudation, 
which would necessarily give to the mountains bounding the plane less 
abruptl}' ascending sides. They might be formed l)y glacial erosion, and 
the ridges of the surface of this strandfiat have obviouslv been rounded 
by glacial erosion, as Fig. 112 shows. But if this glacial erosion had lasted 
long enough to produce mountain forms like Tonningen, the strandfiat 
would necessarily have been deeplv dissected into a much more uneven 
surface than we now find. The only natural explanation is that before 
the last glacial erosion (of the last glacial epoch) the marine denudation 
finally planed the surface of the strandfiat which rises gently from the 
shore to the foot of the mountain slope probably at about 30 metres 
above sea-level. 

Between the western granitic mountain area and the more extensive 
dioritic mountainous region to the east, a broad and flat plain extends 
across the island from the south ccjast to the northwest coast, forming a 
continuation of the strandfiat, rising gently to somewhat more than 60 
metres above the sea, but its greater part, especially in its northern and 
northwestern area is less than 50 metres high. According to kind in- 
formation from Schetelig, this low region is built up of crystalline schists, 
mica-schists, and hornblende-schists which have been more easilv denuded 
than the granite to the west and the more resistant diorite to the east. 

The strandfiat continues along the northern coast of Hitteren, formed 
of pressed igneous rocks, gneisses, and some granite and amphibolite, &c. 
The plane is well developed at heights of l)etween 20 anrl 35 metres above 
the sea with some few hills rising to about 50 metres. The long Dolm 
Island to the north is also a continuation of the low strandfiat with two 
isolated hills rising to 46 metres (Storvarden) and 70 metres (Hjertås) 
above the sea. 



124 FRIDTJOF NANSEN. M.-N. Kl. 

/■ruta is cliic-fly huilt u|» <>\ ,L;raiiilc wliicli in soiiiu places is pressed. 
Aloiij^- its soullicni coast thcrr is ilioritic f^Miciss. J'Voia is very low and 
iiKist of its area beloiif^rs to the strandflat, with some hills anrl riflges rising 
above its plane, especially in the southern or s^nitheastern part of the is- 
lanrl (see ¥\<^. 107), to lieights of al)(Ait 50 t(j 70 metres. 

1'he height of the strandflat of i'Voia is 20 to 30 metres above the 
sea in its western part an'l along the northern and eastern coast. In its 
imier and southern i)arts llic land is mostly more than 30 metres high. 
The surface is considerabJN- less even than the surfacj of Smolen, and 
has many ridges and de[jressions. This is obvitnisly due to its geological 
structure, the granite giving (jften a rough and broken surface. As 
Schetelig has pointed out to me, the unevenness in this case may especially 
have been caused 1)\- alternations of regions with more resistant porphyria 
granite and regions with pressed granite or gneisses, less resistant to 
erosion. 

Along most parts of the coast of Hitteren anrl also along the south 
and west (or northwest) coast of Froia the submerged strandflat is cotn- 
parativelv narrow and is much dissected and uneven. This is also the case 
along the coast of the mainland (see Fig. 107). But to the north of Froia 
a submerged strandflat with hundreds of islands and skerries extends 
20 kilometres into the sea, and continues 50 kilometres towards the north- 
east, comprising the extensive region of the many low and flat Froan Is- 
lands (see Fig. 107). The surface of this submerged platform is much 
dissected as the map may give the impression. But in some parts, c. g. in 
the region of Sillen, Bu Skjær, and Gjeit wSkjær northwest of Froia (see 
map Fig. 107), the surface is extremely level over considerable areas, 
forming horizontal plains near sea-level or only a few metres below it, 
and having well defined edges at depths of less than 10 metres. 

The width of the strandflat from the edge of this partly submerged 
platform to the foot of the higher land on Hitteren is 43 kilometres. But 
if we reckon it to extend across the Hitteren to the foot of the mountains 
on the mainland the width will be 58 kilometres. The width of the strand- 
flat from the edge of the outermost submerged platform outside Smolen 
to the foot of the mountains on Tustern is 40 kilometres. The distance 
from the outer edge of the submerged platform of the Froan Islands to 
the foot of the high mountains on the mainland is about 46 kilometres. 

The strandflat is here conspicuously much wider than in any region 
along the Norwegian coast to the south, and is partly developed to fuller 
maturity. The reasons may be several. 

On the one hand the geological structure has favoured the formation 
of a strandflat. By the Caledonian mountain folding a fairly low land 
was made to emerge along the top of one or two folding ridges far out 
in a stormv sea, where it was exposed to the full furv of the marine 
denudation. This low land was to a great extent luiilt up of rocks, dioritic 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASY. 




Fig. 112. The mountain Tonningen, seen trom the north-western shore ot Laugen Lake, which 
is seen in the foreground. (Photograph bj' J. Schetelig, 1909). 



schists and pressed schistose igneous rocks, wliich were fairly liable to 
erosion by subaërial denudation as well as l)v marine denudation. 

On the other hand the erosion, the subaërial as well as the marine 
erosion, has been much more favoured by a severer climate in this northerly 
latitude than it was along the coast further south. The severe and frequent 
frosts of a beginning glacial period would commence much earlier here and 
last much longer than along the southern coast of Norway. The subaërial 
denudation would be much increased by the disintegrating effect of the 
frosts, and the shore erosion (by frost), assisted by the stormy sea, would 
become very vigorous, planing down the low land, which was already 
dissected into numerous small islands during previous glacial periods. 

Finally the probability is that these islands out in the sea, were 
covered much later by the inland ice than the coast of the mainland, and 
they were therefore exposed to the destructive forces of a glacial climate, 
combined with the attack of the sea, during a much longer period than 
other parts of the coast. 

It seems to me that these are reasons giving a satisfactory explanation 
of the occvirrence of a very wide and fully developed strandfiat in this 
region, as well as along the coast of Nordland to the north, as we shall 
see later. 

It has been maintained, l)y Ahlmann and others, that the glacial 
erosion might have helped to plane the strandfiat. Such a view seems 
contrailictory to what ma\' now be considered as an esta])lished fact, namely 
that it is greatlv the glacial erosion which has dissected the coast into its 
thousands of islanrls and skerries. If, for instance, we look at the map 
Fig. 107, it woulrl indeed l)e rlifficult to unrlcrstand why the glacial erosion 
should lia\X' helped to plane the surface of Smalen and Froia while it has 
dissected the surroundinsj- lan'l into those swarms of islets, often with 



J 26 FKIOTJOK NANSEN. M.-X. Kl. 

(Iccp s<-un'!s l)ci w ecu iliciii. Il \\(,ulil seem abMir*! {' > think that the saino 
asi'ency cdiiM lia\r ha<l sucli culircly oijposite effects in the same rej^'^ion, 
where in(,re(,\-er there wouM he no apparent differences in the geological 
structure tu aecoiuit for it. The fact i^ (ih\ iously that the geological erosion 
has general!}' and e\ery\\ liere a ijronoin.ccij tt-ndency to dissect the land 
surface and not to plane it. 

It nii,L;hl he asked why the strandfiat of Smolcn is somewhat lower 
and much more e\en than that of Hilteren anrl als(j to some extent than 
that of I'roia. ddie ri'ason ma\- chiefh- lie differences in the geological 
structin-e, the r(;cks of Smolen haxini;- been more easilx' and more evenly 
eroded. The situation of the islands ma\- also have been of som2 itn* 
portance. Altliout^ii in m\ oi)inion, the slK.Te eio>ion hv frost has been 
the most vigorous factor for the marine denudation, tlie work of the 
waves (the surf) has also been very important esi)ecially for the transport 
of the waste, and it is obvious that Smolen and to some extent Fpoia has 
alwa\s been exposed to a more effective wave action tlian the more pvo- 
tected Hitteren, as in this region the j)revailing winds were probably 
alwaws southwesterly, as long as the Norwegian Sea was not ice-covered. 

1 liave pointefl out before that the wave a.cüon on a coast exposed 
to the full hn\ of the open sea, may also have a tendency to wear down 
the strandfiat cut by the shore erosion by frost to a level somewhat lower 
than the initial one. I do not believe, however, that this would be a 
feasible explanation of the comparatively low flat level of Smolen. be- 
cause in that case its plane would naturally be expected to slope somewhat 
seawards, and coulrl not be so almost perfectly horizontal as it actually is, 
differing only some few metres in height from north to soutli. 

It is, however, a question whether there are not two levels of the 
emerged strandfiat in this region as in other regions of the Norwegian 
coast. In that case we mav expect the plane of Smolen to belong to the 
lower le\el, it ha\ing almost exactly- the same height as the lower level 
of the stranrlflat at the mouth of Sogne Fjord. 

From Trondhjem Fjord to Vikten. 

Along the coast between Trondhjem Fjord and Folia Fjord there is 
a distinctlv developed strandfiat in front of the steeply ascending moun- 
tains, and there are many submerged platforms with swartns of islets 
anfl skerries outsirle tlie coast. The width of the strandfiat from the foot 
of the mountains to the edge of the submerged platforms, at less than 
20 metres below sea-level, is not very considerable in this region, and 
decreases northwards from about 17 kilometres in the region of Melstein 
and Lovoi (63° 56' N. Lat.) to a few kilometres in Folia Fjord. 

The coast of this region is built up of pressed igneous rocks generally 
considered to be of the Archæan age, but which Schetelig [1913I holds 
to be vounirer. 



I92I. No. II THE STRANDFLAT AND ISOSTAS^ . I27 

To the nurth of Folia Fjord a g^roup of numerous islands, with a 
submerged platform outside, extend far seawards from the coast of the 
mainland. It is X'iktcn with its innumerable islets and skerries built up 
of granite. These islands and platform are probably situated on a con- 
tinuation of the large folding ridge or anticline on which the Froan Tslands 
are situated to the southwest [cf. Xansen. 1904, PI. XI]. 

A strandfiat with heights less than 30 metres above sea-level, extends 
over a great part of Inner \'ikten island. But many small hills rise above 
the level of the stranrlflat to altitudes of 40 and 50 metres and more, 
and several greater hills or ridges even to between 100 and 150 metres. 
On the whole the strandfiat has not been developed to any high degree 
of maturity on this island and that is still less the case on Outer \Tkten, 
but the islands to the north: Kalvoi, Borgan. Rodoi. &:c.. are very flat 
with a conspicuous strandfiat, and only some few isolated hills rise above 
30 metres. 

It is ol)vious that the general level of the stranrlflat in this region, 
wherever it has been developed, has a height of less than 30 metres or 
perhaps nearer 20 metres above sea-level. The area with heights between 
30 and 60 metres is only a very small part of the low land, and the foot 
of the hills, surmounting the plane of the strandfiat. mav as a rule be 
considered to be below 30 metres above the sea. This is still more marked 
further north along the coast of Helgeland. 

In tliis respect there is a difference between the stranrlflat of these 
northern regions, and the strandfiat of the Xorwegian west coast further 
south. For. instance, in the regions of Radoi, Store Sotra, Bomlo Island 
and Stord Island, &c. the greater part of the low land forming the strand- 
flat is somewhat higher than 30 metres above sea-level, on Stord Island 
even higher than 40 metres. In the regions of Karmoi and Stavanger 
a considerable part of the strandfiat is higher than 30 metres, although 
a great part of it is about 20 metres high or even less. On Smolen tlie 
whole strandfiat is less than 30 metres high, to a great extent about 
20 metres or less. On Froia a great part of the strandfiat is less than 
30 metres high, but about an ecjual area is between 30 and 60 metres or 
largely between 30 anrl 40 metres. On Hitteren Island the stranrlflat is 
largely somewhat higher than 30 metres, though along the northwestern 
coast of the island a considerable part of it is lower. The many islands 
north of Froia are lower than 30 metres and largelv about or lower than 
20 metres, and so are the Froan Islands, where onlv three or four solitary 
hills on \'æroi, Risoi, aiul Kimna rise to 31, 35, and 49 metres above 
the sea. 

On the whole the emerged stranrlflat seems consei|uentIv to be some- 
what lower along the northern part of the Xorwegian coast, north of the 
Romsdal region, than we have found it to be south of the Sogne Fjord 
region. 



128 KRIirrjOK NANSFN. 



()ulsiilc tin- \ ikten i^lainls, to llic soulli, west, aiifl nortliwest, there 
is a suI)nK'rj4X'fl platfcn'in, kj to 17 kiloniclrcs broarl, with a pcrfectlv 
Ik'w iMcriiijj;" nunihcr of low isUts, skerries, aiul slioals. The j^'^reat majority 
of the thousauils of shoals ami sunken rocks on this platforni is very 
nearly at sca-le\'el or onl\ a few metres below it. 

]W far the i^reater i)art of this suhtnerjj^ed {>latform has depths of 
less than 15 metres below sea-level, and its general level is less than 
10 metres below the water. It forms a fairly v\v\\ aiir] horizontal plane 
extending over many kilometres. It is traversed by many narrow chan- 
nels, sometimes 50 to too metres deep, and at its outer borrjer the sea- 
bottom falls abru])tly towards depths of 150 tf) 200 metres of the sea 
outside. 

The edges of the {)latform along the traversing channels as well as 
along its outer borrler are as a rule very sharply defined at depths of 
less than 10 metres. 

It is the same kind of platform as those extending northwest of 
Smolen, and north of Froia, with a general horizontal level at only some 
few metres below sea-level and sharply defined edges along the outer 
borders as well as along the many traversing channels. 

It seems obvious that after these channels and depressions had first 
been formed, the initial projecting peninsulas and islands have been 
truncated at a level some few metres below present sea-level, to form these 
very even horizontal planes, which may have been still more even at first 
before they had been exposed to the subsequent erosion of glaciers. 

Along the coast of Hclgeland to the north, similar platforms have 
a still wider extent, and when describing that coast we shall return to 
the question of their formation. 



1921. No. ir. THE STRANDFLAT AND ISOSTASV. I29 



X. THE STRANDFLAT OF NORTHERN NORWAY. 

Helgeland. 

The straiidflat is very conspicuous and extremely well-developed 
along the coast of Helgeland between \'ikten and \'est Fjord. It consists 
partly of an as a rule sharply marked plane or planes extending over 
the hundreds of low islands and peninsulas, and to a still greater extent 
of very flat submerged platforms with thousands of skerries, rocks, and 
shoals. It has an average breadth of 40 to 45 kilometres as stated by 
J. H. L. A'ogt [1900]. In some places it may even be 50 kilometres wide, 
or more, from the steep mountain sides on the mainland to the cuter edg^ 
of the submerged platform out to sea at less than 20 metres b,dow sea- 
level. 

The strandfiat of Helgeland has been described by J. H. L. \^;)gt 
[in 1900 and in 1907], and after him by several authors, especially Hog- 
bom [19 13], Oxaal [1914], Sahlstrøm [1914], and Rekstad [19 15]. In the 
two latter papers there are given some most illustrative pictures of the 
strandfiat and the isolated mountains rising abruptly like 'stacks" above 
its plane. The present writer has also previously [1904] described the 
strandfiat along the coast of Helgeland. 

This coast has a very complex geological structure, and is built up 
of a great variety of rocks. They are chiefly mica-schists, granites, voung 
gneisses, lin:estone, gabbro, some svenite, and some serpentine. Th S2 
rocks vary much in their power of resistance to erosion, the granites and 
gabbros and also the serpentine (on Leka) and the gneisses are as a rule 

Fig. 113. Map of the coast of Helg-eland from Bindal Fjord to Donna Island. Scale i : 350,000. 
Black areas on land are lower than 30 metres above sea-level. The dark hatching on land 
indicates areas between 30 and 50 metres above sea-level. The light hatching in the sea 
indicate submerged plateaus with depths less than 25 metres below sea-level. Isobaths are 
drawn for every 50 metres of depth, i. e. for 50, 100, 150, 200, 250, &c. The black spots 
Indicate rocks and skerries above sea-level. The small crosses indicate sunken rocks and 
shoals, crosses zvith spots indicate rocks near sea-level, and (lotted areas shoals near sea- 
level. The figures on the land and in the sea give the heights and depths in metres. 
I Vega Island. II Sola Island. Ill Ulvingen Island. IV Hamnoi. V Tjotta Island. VI Alsten 
Island. VII The mountains „Syv Søstre" (Seven Sistersl. VIII Heroi. IX Donna Island. 
Vid.-Selsk. Skrifter. I. M.-N. KI. 1921. No. 11. 9 



p 




132 



FKIDI |OK NANSEN. 



M.-N. K.. 




1 92 I. No. 1 I. THE STRANDFLAT AND ISOSTASY I 33 

Fig. 114. Map of coast of Northern Helgeland, northward Continuation of the map Fig. 114. 
Besides the Isobaths for every 52 metres of depth, the Isobath for 25 metres is drawn 

with dotted line. 
I Vandved Island. II Donna Island. Ill Lokta Island. IV Lifjell on the mainland. V Hugla 
Island. VI Handnesoi. VII Tomma Island. VIII Hestmannoi. IX Nesoi. X Sornesøi 

and L^Migvaer. 

more resistant than tlie limestone and the mica-schists. None the less 
the strandfiat often cuts horizontally through the various rocks, without 
anv ap[)reciable difference in height, straight to the foot of the mountains 
and hills rising abruptly and steeply above its plane. 

The region of Solvær and Lovunden (see Fig. 114) may be mentioned 
as an illustrative example. The many islands of Solvær consist of mica 
schist and limestone on the southeastern islands, gabbro on the south- 
western islands, granite on the northern, and mica-schist, limestome, 
gabbro, and granite on the northeastern islands. Lovunden mountain, 
as well as the low islets to the east, south, and west of it, are built up 
of gneiss, while the low islands of Lovundvær to the north and northeast 
consist of granite, gneiss, mica-schist, and limestone. All these many 
islands, built up of rocks varying so widely in their power of resistance, 
are cut down to an almost perfectly horizontal plane at about 10 metres 
above sea-level, and the plane extends at the same level to the foot of 
Lovunden, which rises abruptly as an isolated 'stack' to 619 metres above 
the sea [cf. Xansen, 1904, PI. \', Fig. i]. 

This is a convincing piece of evidence that the plane of the strandfiat 
has been cut by an agency working horizontally, /. c. by the shore erosion, 
and cannot have been formed by the vertically working subaërial denu- 
dation, nor by glacial erosion, which would necessarily have produced a 
marked difference in height between the regions of more and of less 
resistant rocks. The plane cuts in many cases so evenly across areas of 
entirely different rocks that it can only be explained as an effect of the 
shore erosion by frost. The wave erosion cannot have had very great 
direct effect, because it would produce greater difference in the level of 
the rocks with different power of resistance. In other cases there is such 
a difference, which may be accounted for by the wave erosion. 

The mountains rising as 'stacks' above the strandfiat consist very 
often of granite, e. g. on Vega and Søla islands, but in some cases, e. g. on 
Lovunden and Hestmannen (Hestmannoi), they are built up 'of gneiss, 
although the gneiss in other regions has been much denuded, and forms 
parts of the very low and level plane of the strandfiat. 

The reason why the strandfiat is so well-developed along the coast 
of Helgeland may be especially two fold: On the one hand the severe 
climate of this northern region has been especially favourable for the 
subaërial denudation as well as for the shore erosion. It is for this latter 
reason that in postglacial time the now raised shore-ledges have been 



134 



FRIATJOF NANSEN. M.-N. Kl. 



SO \vcll-flcvcl(jpc'l in tliis iiortlicni region, aii'l stil! Ijctler further iiorlli 
in Tromsø Fylke and I'"inniark, wliile tliey are liardly seen alonj,^ the 
coasts of southern Nc^rway. 

On the other hand the rocks forniin;^ ihc oiUcrniost CÆSt of Hclf^elanrl 
are on the whole less resistant to erosion than the rocks of the Romsdal 
and Sondmore region and also on the whole less resistant than the igneous 
rocks of T.ofoten and Vesterålen. It is furthermore probable that the 
initial outer ctvist land 'of the Helgeland region was comj^aratively low. 
and that no great quantity of rock had to be planed down for the formation 
of the strandfiat, after this land, consisting of the low outer folding ridges 
of the great Caledonian mountain foUling of Northern Norway, had been 
dissected into thousands of islands bv tlie glacial erosion. 

The stage of development of the strandfiat varies somewhat in the 
different parts of the coast. In some regions, c. g. in Heroi, Donna, and 
Solvær, the emerged surface is (juite unusually level, whilst in other region?, 
c. g. on the \'ikten islands, in the region of Tcjrghatten, &c., the surface is 
undulating with more hills and ridges rising to different heights above the 
plane of thé strand flat. This depends naturally on the degree of maturity 
to which the strandfiat lias been developed. In regions with less resistant 
rocks, or where the land was more dissected into smaller islands, and 
where there was much less rock to be planed down, the strandfiat was 
sooner developed to full maturity than in regions where the conditions 
were less favourable. Along a coast where the conditions differ much 
in this respect we may therefore expect to find the strandfiat in all stages 
of maturity. 

It has to be admitted that if one consid:rs the stra-.dfiat to have been 
formed solely by wave erosion, it might be difficult to understand how 
any strandfiat at all could have been formed in regions where numerous 
hills and ridges are still rising above the plane of th? strandfiat, with 
no great distance between them. One would expect that the waves would 
have had to wear down the hills on the seaward side first before they 
could obtain the necessary force to erode a strandfiat further inland. 

But in a severe climate the effecti\'e shore erosion l)y frost works 
simultaneously along all shore-lines, on the outer as well as the inner sides 
of the islands and peninsulas, and in all small creeks and bays. The islands 
will thus be attacked from all sides, and by the joint action of subaërial 
denudation, the shore erosion by frost, and the waves carrying away the 
débris, the strandfiat may gradually be developed to different stages of 
maturity in the whole indented and dissected coastal region at the same 
time. 

Fig. 1 13. gives a map of the strandfiat along the coast of Helgeland 
from Bindal Fjord to Donna Island. As material for the drawing of 
this map, I ha\e used the excellent detailed charts in the scale of i : 50.000, 
and the maps ("Gradavdelingskarter") in the scale of i : 100,000, published 



I92I. No. Il- THE STRANDFLAT AND ISOSTASY I35 

bv the Norwegian Geographical Survey i "'Xorges Geografiske Opmåling";. 
The black areas on the land have heights lower than 30 metres (100 feet) 
above sea-level, the dark hatching on the land indicates areas with heights 
between 30 and 50 metres above sea-level, and the light hatching in the 
sea indicate submerged plateaus with depths less than 25 metres below 
sea-level. The black spots in the sea indicate islets and skerries rising 
above sea-level, and the small crosses mark sunken rocks and shoals. 
Isobaths are drawn for every 50 metres below sea-level. The figures on 
the land and in the sea give the heights and depths in metres. 

}klaps of the same coast have been published by J. H. L. ^ ogt [1900, 
pp. 36 — 2)7 g^i"^!"? the emerged and submerged strandfiat in less detail, 
and by J. Rekstad [1915] gi'^'irife the isobaths of the sea. 

Our map Fig. 113 shows that the greatest part of the emerged strand- 
flat has heights less than 30 metres, and the areas with heights between 
30 and 50 metres are comparatively so very small that it may seem doubt- 
ful whether they can be considered as actually belonging to the strandfiat, 
the base of the steeply and abruptly ascending hill-sides being as a rule 
lower than 30 metres above sea-level. 

It has already been (p. 127) pointed out that the general level of 
the emerged strandfiat is on the whole lower in this northern region than 
along the coast of Norway further south. I have not had an opportunity 
of actually measuring the height of the strandfiat along the coast of 
Helgeland, but as far as I can make out from the maps, from my photo- 
graps, and from the most accurate measurements made by previous in- 
vestigators, especially Sahlstrom, the general level of the emerged strand- 
flat of Helgeland seems to be between 10 and 20 metres above sea-level, 
and in some places even somewhat lower [cf. Sahlstrom, 1914 . 

I shall return to this subject later, but wish first to draw attention 
to another striking feature in the formation of the strandfiat of this region 
brought out very clearly by our map. It is the wide extent of the very 
flat submerged plateaus of the strandfiat. In this respect the strandfiat 
of Helgeland differs entirely from that of Southern Norway. Along the 
coast south of Sogne Fjord, the submerged strandfiat is insignificant as 
compared with the extent of the emerged strandfiat. Along the coast 
northwards the submerged strandfiat increases in extent. In the region 
of Smolen and Froia the area of the submerged part of the strand flat 
at least equals that of the emerged part (cf. Fig. 107) and in Helgeland 
the area of the emerged strandfiat is verv small as compared with that 
of the submerged strandfiat. 

The depth below sea-level of these submerged plateaus, indicated by 
the light hatching, is less than 2^ metres, anrl by far the greater part of 
them have depths less than 10 metres, as is demonstrated by the detailed 
map Fig. 115 of the Hysvær Plateau to the northwest of \'ega and Sola 
islands. This map is based upon a small part of No. 54 of the Norwegian 




/50 100 SO 20 10 BO METRES 



MicxL-schlsl ^%A GtcLïdtt 

Fig. 115. Explanation 




•^ TzauÙ: ??u^s 



next pagre. 



r^S FRiin.ioi- NANSKv. M.-N. Kl. 

Fig. 115. Cliart ol tlu- Hysvær Plateau, northwest of Vega and Søla islands, based upon 
the Chart No. 54 publislicd by „Norges geografiske Opmåling". The figures in the sea 
indicate the tl(|)»lis in nurtres below low-water level, which is about 1.5 metres below 

mean water-level. 



rletailefl charts of the coast of Norway. It is given here as a characteristic 
illustration of the detailed surface relief of these submerged plateaus. The 
figures give the depths in metres below lowest water-le\ol, which is alxjut 
1.5 metre below mean water-level. Isobaths are drawn with flotted lines 
for 5 metres, with broken lines for lO metres, with thin lines for 20 metres, 
and with thick lines for 50, 100, anrl 130 metres of depth below lowest 
water-level. 

The isobath for 10 metres follows as a rule very closely and at a 
very short distance the isobath for 20 metres, so that the area bounded 
bv the latter contour line is not substantially larger than that bounded 
by the 10 metres line. The isobath for 5 metres demonstrates that the 
greater part of the surface of the plateau is even less than 5 metres below 
lowest water-level, and over great areas the de])ths are between 0.5 and 

3 metres. The general horizontal level of this plateau may be said to be 
between 2 and 7 metres below mean sea-level, and in its nortlieastern part 
near the islands of Hysvær, it is almost in the sea-surface. 

Fig. 116 gives a profile across this plateau along the broken line in 
Fig. 115, passing across Sola Island northwestwards to Skjærvær and 
thence seawarrls. This profile demonstrates the remarkably horizontal 
evenness of the plane of this plateau, in which the channels and hollows 
form sharply defined depressions with well marked edges, mostly at 3 to 

4 metres below mean sea-level. 

The even surface of this sul)merged plateau has olniously much 
resemblance to the even planes of the emerged strandfiat, c. g. on Dønna 
and Heroi, described by Sahlstrom and mentioned later. 

The other submerged plateaus shown in Fig. 113, have a similar 
surface relief, but their outer parts in the northern and southern regions 
of this map are somewhat more broken and irregular. The Hysvær Plateau 
has a width of about 15 kilometres, and the other submerged plateaus 
have a similar extent. 

Several writers, the present one included, have stated that the edge 
of the siihiiicrgcd strandfiat of Helgeland is at 20, 30 or 40 metres below 
sea-level. This is hardly correct and is apt to give an entirely wrong 
impression that the plane of the strandfiat slopes to such a low level. 
As we have just seen the almost perfectly horizontal plane of the sub- 
merged straiulflat lies very near present sea-level, and. as Fig. 115 demon- 
strates, the edge of this plane is sharply defined at depths of less than 
10 metres below the water surface. 



I92I. No. II. THE STRAXDFLAT AND ISOSTASV. I39 

Il niav be asked whether these extensive horizontal plains have 
actually been cut in solid rock. The many hundreds of islets and skerries, 
scattered over their surface and rising above sea-level, prove, however, 
that they must be rocky plateaus, although this does not make it impos- 
sible that their surface may to some extent be levelled by loose material. 
The sharply defined channels traversing them, with well markel edges 
and the sharply marked edges on their outer seaward borders where they 
are exposed to the full fury of the ocean, creating a violent surf, se?m, 
however, to prove that the horizontal plane of these plateaus is actually 
to a very considerable extent cut in solid rock. 

It is obvious that horizontal, rocky plateaus, with wide dimensions 
such as these, cannot have been formed in postglacial time, nor can they be 
cut by wave erosion. They must have been finally planed by shore erosion 
by frost at some period or periods before the last glacial epoch. As they 
have so much wider an extent than the plains of the emerged strandfiat 
on the islands and the mainland they may possibly have needed a longer 
time for their formation than the latter, although we have to consider 
that the previous strandflat over these now submerged plateaus was 
probably low, and that it was much dissected. 

In the region of Donna and Heroi Sahlstrom [1915' has studied the 
emerged strandflat. and has by very accurate levelling constructed some in- 
structive profiles across the northern flat part of Donna, across Southern 
Heroi. and across Blomsoi (Fig. 113). The strandflat forms her^ quite 
remarkably flat and extensive planes. 

On the northern part of Donna (see Fig. 114. II) the plane is 9 to 10 
metres above sea-level, and in some places (Rølvag) 1 1 metres. There are 
numerous small depressions and small valleys, but they are as a rule only 
some few metres below the uniform level of the solid bare rocks projecting 
in the surface of the ground, and are often not so large that in my opinion 
a great many of them may not more or less have been formed originally 
when this strandflat was planed by the frost in the shore and by the waves, 
and have only been slightly modified later by subaërial denudation and 
by glacial erosion. As mentioned before, there is a great resemblarce 
between these planes and the level surface of the submerged plateaus in 
the sea outside. 

Some few isolated hills rise above this plane to heights of 21 to 35 
metres above sea-level, and near the east coast of the island to 42 and 50, 
or a few even higher. The plane extends horizontally to the foot of these 
hills and there is as a rule a quite sharp demarkation line, almost like 
a shore-line, which, however, is often covered by shore gravel, and talus 
débris. 

The region of the strandflat of the northern part of Donna is built 
up of mica-schist, gneiss, and to some smaller extent of crvstalline 
limestone. 



140 FRIDTJOF NANSEN. M.-N. Kl. 



.---•T^'V. 




,..,,^^..y-.f,,.^A 



Tuf F ^^^ M icoL- scktst ^^^P Gt^anitf 



Fig. 116. Profile along the broken line in the map P'ig. 115 north-westwards from Søla Islands aero- 

the horizontal scale. The profile in black on 

The northern part of Southern Herøi, south of Donna (see Fig. 113), 
consists of gneiss-granite, mica-schist, and limestone. This region also 
forms a strandfiat with a very level surface 8 metres above the sea. The 
same level plane with the same height of 8 metres also occurs on Northern 
Herøi (Fig. 113, A^lll) to the east and, as Sahlstrom points out, it is note- 
worthy that the small rock in the sound beween the two islands has a flat 
surface at just the same level. This is convincing evidence that the plane 
has been formed by marine denudation (/'. c. shore erosion by frost) after 
the sound had been deepened approximately to its present shape by glacial 
erosion. The relief of this level surface cannot have been much modified 
by glacial erosion after this small rock had been truncated, for otherwise 
its flat top surface would certainly have been rounded and worn down to 
a lower level. Across the middle part of Southern Heroi the strandflat 
forms a very level plane between 6 and 7 metres above the sea. The 
ground is composed of gneiss and limestones which are planed to exactly 
the same level. 

In the southern part of Southern Heroi the plane of the strandflat 
is between 4 and 5 metres above the sea, but not quite so level as further 
north, ridges of gneiss-granite often rising slightly above the crystalline 
limestone which forms the greater part of the surface. 

As Sahlstrom points out, it is noteworthy that on the remarkably 
even strandflat described above tw^o kinds of rock so very different as 
to their power of resistance as granite and limestone, are in some places 
planed to exactly the same level, and in other places the difference of 
denudation is only a couple of metres. 

According to Sahlstrom's measurements, as mentioned above, the 
plane of the strandflat slopes slightly southwards, from about 10 metres 
above the sea in the northern part of Donna to 4 or 5 metres in the 
southern part of Southern Heroi. Sahlstrom considers it possible that 



I92I. No. II. THE STRANDFLAT AND ISOSTASV. I4I 




^:^—-d- 



[•svær Plateau to Skjær\-ær and thence seawards. The vertical scale is 5 times exaggerated in relation to 
Und gives the natural relation of height to length. 

this might be due to the fact that the limestone of Southern Heroi is less 
resistant to erosion than the harder rocks in the north. 

It is of much interest that real shore formations, striated by later 
glaciers, were actually observed by Sahlstrom on the surface of this level 
strandfiat. The lines of demarkation along the foot of the hills rising 
above the plane of the strandfiat are in some places so sharply defined, 
that they might be called shore-lines. 

The surface of the many hundreds of very low and flat islands and 
skerries in the region of Sol vær and Lovunden. north of Donna (see 
Fig. 114), is obviously at the same low level which v.as measured by 
Sahlstrom on Donna and Heroi. Pictures illustrating the even flatness 
of the islands of Solvær and the islands surrounding the solitary mountain 
of Lovunden are given by Rekstad [1912. Pi. I, Fig. 1], Sahlstrom [1915, 
Fig. I4J, and the present writer [1904, PI. V, Fig. i\ Rekstad [1915, 
Pi. II, Fig. i] gives a most illustrative view of the many islands in the 
region of Heroi, Husvær. and Skålvær (cf. Fig. 113), west of Alsten Is- 
land with De Syv Sostre (Seven Sisters). It demonstrates the extreme 
flatness and low altitudes of the many islands in this region. 

There is a striking difference between the low altitudes of these 
level planes of the emerged strandfiat of Helgeland and the heights, 
generally stated to be between 30 and 40 metres, of the inner boundary 
or upper limit of this strandfiat. Sahlstrom could find no traces of this 
higher level in the region studied by him. He found the plane with a 
height of about 8 to 10 metres extending to the foot of the mountains. 

J. H. L. ^'ogt [1907, pp. 20 f.j says that, according to his investi- 
gations along the coasts of Helgeland and the Lofoten Islands, the upper 
limit of the strandfiat, or the demarkation-line ("Knickpunkt") between 
its nearly horizontal plane and the steeply ascending mountain-sides stands 
everywhere at almost exactly the same level. In numerous profiles taken 



T42 



KKIDTIOI' NANSFN. N.-X. Kl. 



hv fduiul il sli^lilly liij^Iicr lliaii ,-50 metres above the sea, anrl he thinks that 
it may be j)ut aijproximately at 40 metres above sea-level. Froin occasionnai 
ol)servati(jns durin^^ short visits to the coast further north (Anrloi, 
Bjarkoi, (Sjc. in (i</'X.Lat.) as Avell as further s<*uth (in the Ivr>nis'lal 
region, Karnioi, I laugesinif!, l->omlo Island, &.c.) \ ogt draws the con- 
clusion that the ui)i)er limit of the straiulflat is fairly exactly at the same 
level alon«,'- the whole of tiie Nor\ve<;ian coast. He adds, however, that 
it is impossible to determine the height of tlie upper limit of the strandflat 
within an accuracy of some metres, anfl that the estimate will to a certain 
degree dej^iend on the observer. But he thinks "that greater errors than 
15 to 20 metres are excluded". As he docs not say that his estimate of 
the heights of the stranclflat is based on accurate measurements by level- 
ling, we may conclude that his figures are not meant to be very accurate. 
and with possible errors of 15 to 20 metres they give us a broad margin. 

Though the emerged strandflat of Helgeland may possibly to some 
extent, like the strandflat of southern Norway, have two levels, a widely 
extended lower one, like that rlescribed l)y Sahlstrom, and a much less 
extended higher one, still I think we ought to receive these statements of 
the higher levels of between 30 and 40 metres with some caution as long 
as they are not based on actual measurements by levelling on the spot. 

There has obviously been a tendency towards establishing general 
rules for the heights of the strandflat along the entire coast of Norway, 
and the limits of these heights have been put at 30 to 40 metres above 
the sea, and 30 to 40 metres below sea-level. We have seen that along 
the coast of Helgeland as well as in the region of Smolen. Hitteren, and 
Frøia, where there are well-developed submerged plateaus, this estimate 
of the lower limit is not correct, the outer edge of the submerged strand- 
flat being there rather less than to metres below sea-level. We have not 
yet obtained sufficiently accurate measurement to establish the height of 
the inner edge, or the upper limit, of the emerged strandflat of Helgeland. 
It seems at any rate to be somewhat lower there than along the west coast 
of soutlicrn Norway. 

In Fig. 118 from northern Helgeland the demarkation-line between 
the low flat strandflat and the very steep mountain-sides is seen. The 
inner part of the strandflat is to a great extent covered by the scree or 
talus heaps, formed in postglacial time by stones tumbling down from 
the mountain-sides, but it is obvious that the plane of the strandflat. 
hardly 15 metres above the sea, continues under the scree to the foot of 
the rocky walls at about the same height. J. H. L. \'ogt's description 
[1907, p. 14] of the strandflat, as a plain sloping gently seawards from 
its inner higher parts 30 to 40 metres above sea-level to its outer sub- 
merged edge 30 to 40 metres below sea-level, is also in my opinion some- 
what misleading. The strandflat is not really one sloping plane, but con- 
sists rather of several more or less horizontal planes. As we have seen, 



1921. No. II. THE STRAXDFLAT AND ISOSTASV. 1 43 

there is in Helgeland one almost perfectly horizontal plane of the sub- 
merged platform at a level of some few metres below the present sea 
surface — and there is at least one plane above the sea, measured h} 
Sahl Strom on Heroi and Donna at a level of about 8 to lo metres 
Whether there are more planes in this region is not quite certain. 



Træna. 

Far out to sea, northwest of Lovund and Lovundvær, separated from 
their plateau by a deep sea, the Træn Fjord, up to 438 metres deep and 
12 kilometres broad, is a submerged plateau with the island-group of 
Træna (see Fig. 114). We may call this plateau the Træna Plateau. It is 
31 kilometres long from SSW to NXF^, and about 9 to 12 kilometres 
broad. Its islands are much scattered, and subdivided into several smaller 
groups: the Træna Islands proper, in the southwestern part of the plateau, 
with the biggest islands Sanda (with the high peak Trænstaven, 338 m.) 
and Husoi, built up of pressed granite, gneiss-granite, Sandavær just 
north of this group and built up of the same kind of rock, Sandoi (Sandoy, 
of gneiss-granite) and to the north of it Torvær (young gneiss) in the 
middle of the plateau, Dorvær to the northeast, built up of not pressed 
granite, Arvær and Båsan or Rosoian (young gneiss) to the northwest 
of Torvær and Dorvær, and finally Selvær in the northeastern part of 
the plateau, built up, in its southern part of mica-schist with layers of 
crystalline limestone, and in its northern part of gneiss [cf. Rekstad, 1912, 
geological map]. 

The geology of the islands of the Træna Plateau has been described 
by J. Rekstad [191 2]. John Oxaal [19 15] has given an interesting general 
description of the island-group and its strandfiat. 

There are on these islands many evidences of the vigorous marine 
erosion, or shore erosion, to which the land has been exposed. As such 
evidence may be mentioned the precipitous sides of the mountains rising 
abruptly above the strand flat, the many big caves, the frequently cirque- 
like valleys ('botten' valleys) with almost vertical sides. 

The erosion by frost, especially along the shores, has obviously been 
of the very greatest importance for the development of these formations. 
When the shore-erosion by frost is greatly intensified during periods with 
a severe climate, and is assisted by the violent wave action of a stormy 
sea on an exposed coast, as in this region, this erosion will become 
extremely effective, and will have a great ability to cut away the land. 
Thus the shapes of the mountains of Træna are simplv explained: 
the shore-erosion has cut away the land, but on the higher and more 
resistant islands it has not been able to plane down the whole islands, 
and the high mountainous parts with their precipitous sides, or shore 
walls, especially on Sanda Island (pressed granite) and Buøi in Dørvær 



144 



I"KIt)TJOr NANSF.N. M.-N. Kl. 



(j^-raiiitc), riMiiaiii. iii<licat in- llic initial lici.c^Iil oi' tlic- laiul, and haviiij^ oii 
their lop often llie old surface, or l'alæic lanrl surface, extendinj^^ almost 
unaltered lo the ed^^cs of the j)recij)itous mountain sifles, which is a con- 
vincing,^ proof that, durini;- ihe time when ihe slrandflal was foriïied, the 
effect (jf the suhaerial denudation was of but little importance as com- 
pared with that of the sliore-erosion, where the latter was as effective as 
in this rej^J'ion. 

Other convincin,«;- evidence of the efficacy of the shore-erosion are 
the many big caves on the islands of Træna. As the floors of these caves 
are at levels between 29 and 56, and some even at about 70 metres, above 
the sea [cf. Oxaal, 1915, P- 72], and conse(|uently above the level of the 
strandfiat, they must have been formed during- comparatively short periods 
when the land was temporarily submerged to these levels. 

The fact that caves such as these formed in granite or gneiss, only 
occur near the coast and especially on the most exposed parts of it. indi- 
cates that they have been formed by shore erosion. 

In mv opinion, the chief agency which lias helped to form them i- 
the frost. This is already indicated by the fact that such caves are 
especially numerous in the northern parts of the Norwegian coast, and 
the great majoritv of them do not occur near present sea-level, but at 
levels to which the land was submerged during the glacial periods, and 
when there were cold climates. In more arctic latitudes, e. g. on Bear 
Island and Spitsbergen the caves occur mostly at the present sea-level. 
As has been pointed out by Rekstad and Oxaal and previous writers, 
the caves are formed along fracture lines, or line,s of weakness, in the 
rock. Here the frost had easy work by breaking loose blocks along the 
many fissures of the rock. Where the water in the rock was permanently 
frozen, its temperature would be near melting point where the rock sur- 
face was near the sea water, and frequent daily changes in temperature 
above and below freezing point would be caused by the temporary con- 
tacts with the warmer sea water, especially due to the tide, but also 
occasionally to storms. 

The stones thus loosened by the active frost will tumble down from 
the roof and the walls of the caves. They will also easily be broken away 
by the waves, especially during storms, and where the caves have the 
required shape for it, the alternate compression and expansion of the air 
m the caves by the waves, may have a considerable effect in breaking 
loose the stones, as was pointed out by Rekstad [1912, pp. 59 f"-. cf. also 
Oxaal, 1915. pp. 74 f-1- The stones thus broken loose, will accumulate on 
the floor of the caves, where they will be exposed to a vigorous dis- 
integration by frost when the floor is alternately submerged and left dry 
by the tide. Especially during winter, ice is also formed on the floor, and 
when shifted and broken loose in summer, it may help to carry away 
material. That the wave action generally is of little direct importance for 



I 92 I. No. II. THE STRANDFLAT AND ISOSTASY. I 45 

the disintegration of the stones is proxed hy the tact that as a rule little 
rolled material, pebhles or boulders, are found in the caves. 

The often cirque-like valleys with precipitous sides on the Træna 
Islands have also, in mv opinion, been formed by the shore-erosion to a 
verv great extent. Oxaal [1915- I'- 79] thinks they have been formed by 
the wave erosion, but I consider it j^robablc that in their formation also 
the frost has been the chief causal agent, although it has been effectively 
assisted by the wave action. It may be doubtful whether it is justifiable 
to make such an absolute difference between these kind of formations 
and the ordinary cirf|ue valleys. I think the first commencement of these 
valleys may often have been small regular cir(|ues eroded by small local 
glaciers or accumulations of snow and ice, in the manner I have discussed 
before (cf. p. 2' , and Fig. 8). This erosion has then been continued by 
the frost erosion in the shore, of the kind described in C"hap.\' (pp. 28 ff.), 
which is to some extent of the same nature as the erosion of the small 
cirque g;laciers. The shore-erosion l\v frost has at times been vigorously 
aided bv the violent wave actioit, breaking and carrying away all loosened 
material. In this manner the combined erosion of frost and waves ad- 
vances comparativelv rapidlv along- the lines of fracture and weakness of 
the rock, and forms small cir(|ue vallevs which may develop into narrow 
passes breaking through the mountain ridges. 

Considering the vigorous marine denudation, to which the Træna 
Plateau has been exposed, it was to be expected that it would have a well- 
developed strandfiat, although it is to a great extent built up of compara- 
ti\'ely resistant rocks. 

It would have been of much interest to know exactlv the height of 
the emerged strandfiat and especially its upper limit in this region, as 
we might expect to find it fairlv sharply defined at the foot of the steep 
mountain sides. Oxaal mentions this strandfiat, but unfortunately he does 
not seem to have actually measured its heights, and neither his description 
nor his illustrations give any clear indication of the actual altitude of its 
level i^lane or of its upper limit at the foot of the mountains. 

According to the maps ("Cjradavdelingskarter", in scale 1 : 1 00000) 
marked "Trænen" and "Luroy", the many islands are in general less than 
30 metres above sea-level, and only on some few of them does the low 
lanrl rise to heights of between 40 and 45 metres. The probabilitv seems 
to me to be that on the many low islands there is, to a great extent, 
a low le\el similar to that measured by Sahlstrom on 1 )onna and Heroi. 
Whether there is also a somewhat higher le\el approaching 30 or 40 
metres is doubtful, but may be possible. 

The method employed by Oxaal, consisting in measuring with a 
planimeter the areas between the contours for each thirty metres of height 
above the sea, and for each ten metres of depth l)elo\v sea-level, and then 
drawing a profile of the heights and depths of the plateau accordingly 

Vid.-Selsk. Skrifter. 1. -M.-N. Kl. 1921. No. 11. 10 



146 FRIDTJOF NANSFN. M.-.\. Kl. 

will ,L;i\i' an idea of \\liii-li lici^^lils ami dcplli^ prcrlomiiiatc on llic slrand- 
flal, liut cainiot as a rule \n- of uiucli \;iluc for finding'- the indicatifjns of 
its orij^inal, ncarh' lioii/ontal lc\( Is or planes. In cxceiAional cases where 
the contours run more or less concentrically and parallel lo each other, 
it nia\' ,L;ive tairh satisfactory results, i'.ut where, as in nifjst cases, the 
planes of the stranrlflat are rjissectcfl hy numerous channels and depres- 
sions of varvini^' depths, and where only a part, and perhaps even a 
conii)arati\ely small part, of the area of the strandflal actually indicates 
its plane or planes, altliouj.(h these may he extended over ridges a{jproaching- 
tlie outer Q(]^c (cf. the Træna Plateau, Fig". 114), there Oxaal's methfjd 
will i^i\e results entirely misleadin«,^ for the deternnnat ion of these planes 
and their lieit^hts. The prohahility is tliat it will generally give a more 
or less graduallN' slojiing profile of the strandfiat with no distinct breaks, 
and wdiere the ])lancs disappear more or less. 

From his computations of the kind mentioned above, Oxaal draws 
the conclusion that the outer edge of the submerged strandfiat of the 
Træna Plateau is at 30 to 32 metres below sea-level. This is hardly 
correct, and the actual e(]ge of the original plane of the submerged strand- 
flat is probably nearer the water surface (cf. Fig. 114). But its depth is 
difficult to determine with accuracy along the outer side of this plateau, 
as there is no sharply marked break between the plane of the submerged 
strandfiat and the slope of the sea bottom outside, which slopes gently 
towards the surface of the continental shelf. 

In this respect there is a striking difference between the outer edge 
of the strandfiat in this region, and the sharply defined edge of the almost 
perfectly horizontal plane of the submerged strandfiat to the south, in the 
region of Vega (see Fig. 113). We have seen that in the latter region the 
extremely level surface of the extensive submerged plateau stands only 
some few metres below- the sea surface (see Fig. 115) and extends very 
nearly to the isobath of 50 metres, which is often quite close to the isobath 
of 100 metres, there being a sharply marked edge at less than 10 metres 
(or near 5 metres) below mean sea-level, and a distinct l^reak between the 
horizontal plane and the slope of the sea bottom outside.. 

But the surface of the submerged Træna Plateau is much more 
irregular, with greater and more varying depths, and especially on its 
outer side there is often a considerable distance between the very irregular 
isobath of 25 metres and that of 50 metres, and the latter is largely more 
distant from the isobath of 100 metres than from the islands (see 
Fig. 114). 

Along the eastern or northeastern side of the Træna Plateau, at 
Selvær and northeast of Dorvær, it is different. The isobaths of 2^, 50, 
100, 150, 200, and 250 metres are there closer together (see Fig. 114), 
and the surface of the submerged strandfiat is more level and nearer the 
sea surface, with a sharply defined edge at depths of about 10 to 20 metres. 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. I47 

Oxaal [19 1 5. p- 87' thinks the prol)able explanation of this fact to be 
tliat the eastern, lower part of the submerged strandflat has been cut away 
bv the deep channel, excavated along the inner side of the Træna Plateau, 
and this channel should then to some extent be younger than the strand- 
flat. The probal)ility of this explanation might seem to be supported by 
the fact that the channel is deepest, with deptlis of 420 metres, just east 
of Selvær. where tlie submerged strandflat is especially level and high, 
with a sharply marked edge at about 10 metres below sea-level, while 
further south, southeast of Dorvær. where the channel is less deep 
(28S metres), the edge of the strandflat is at depths of about 30 metres, 
and still further south or southwest, south of Dorvær and east of the 
Træna Islands (Husoy). where there is no deep channel, the submerged 
strandflat has no sharply defined edge. Along tlie south-eastern side of 
the Træna Plateau tliere is also a very deep channel with depths more 
than 300 metres (and even 438 metres), but the submerged strandflat has 
no sharply defined edge in this region. 

Relation between the Differences in the Surface Topography 

of the Submerged Plateaus and Differences in the 

Nature of their Rocks. 

A\'hen we come to look at it. however, it is striking that the dif- 
ferences in the surface relief and the depths of the edge of the submerged 
stranflflat of the Træna Plateau coincide to some extent with differences 
in its geological structure. Selvær and the northern part of the Træna 
Plateau, where the eastern edge is so sharply defined at about 10 metres 
below sea-level, are built up of mica-scliist and young gneiss, while Cor- 
\ær is built up of granite and the islands of Sandøi and Træna to the soutli- 
west of pressed granite, and here the surface of the submerged strandflat 
is deeper and it has a less distinctly defined edge. There is a similar 
difference in the slope of the sea bottom outside the western side of the 
plateau. It is much steeper off its northern part built up of young gneiss 
than west of the southern part built up of pressed granite. 

It may l)e difficult to find a satisfactory explanation of these features. 
It might seem prol)able that the greater depths and the more sloping 
surface of the submerged strandflat of the Træna Plateau, as compared 
with that of Lovundva^r and Solvær region inside and the regions of 
Gåsvær, Lyngvær, Flovær, Skjærvær. Fuglvær. &c.. to the south (see 
Fig. 113). may, to some extent, be due to the effect of a violent wave 
erosion, which has been especially effective on the much exposed Træna 
Plateau, and which has eroded the strandflat after its first planing by 
the joint effect of the shore-erosion by frost and the wave action. 
It might be objected that the plateaus, for instance at Lyngvær. Flovær, 
Skjærvær. (Sic, have been almost equally much exposed to the furv of 



148 !■ kil) I JOK nansi:n. M.-X. Kl« 

ihc waves, ami still tlurc is a strikinj,^ dif fcrciicc in the rJepths anrl even- 
ness of llic surfaces of the suhiiier^«') plateaus aii'l their outer eHjifes, 
which are so \vv\ sliarpU iiiarkc <1 111 tin- latter rej^ions. 'J'herc is thi.s 
difference in llie situation that hetween I .ynj^j-vær in 66" 4' X. I>at. anrl 
the h'tij^-elslx) Skerries ( kjij^elsboskjæraii ) in CiS" 3 1 ' X. J.at. the outer 
C(»ast or edf^e of the now suhinerq-ed plateau has been almost continuous 
(sec ^i^- 113). while the isolated 'l"ra-na I'laleau has been exposed to the 
full fur\' of the sea almost on all sides, hut least on the nortlieastern side 
where tlie cdye of the plateau is sliarpest. It nni^lil therefore he expected 
that the effect of the wave erosion has been j^-reater on this plateau and 
that especialh' all loose material has been more completely swept away, 
while it may to some extent ha\e accumulated in the hollows anrl de- 
pressions (jf the extensive ])lateau to the south. It is notewortliy that 
the more isolated submerj^erl plateaus in the northeastern jiart of the ma]> 
Fifi;'. I 13, at Slaj,,^grunnen and still more at I'loholman, have less horizontal 
and more irrej^ular surfaces than the plateaus to the south anrl their depths 
are _s4rcater. Tlicy resem])le the surface on the fuiter sirle of the Træna 
riateau. The surface of the platform north of Skibatsvær in 66'Mo'N.Lat. 
(see map Fi^-. 113) and at jonsg^runnan in 66" 12' N. Lat. (see map Fij^^. 
114) is still more irregular. Outside Slaggrunnen the sea is less deep 
and the bottom is also sloping less steeply seawards than along the edge 
of the submerged ])latcau to tlie sf)uth (see Fig. 1131. This is also the 
case outside Jonsgrunnan (see Fig. 1 14), but not outside Floholman 
\vhere there is a fairly steep slope (see Fig. 113). 

It is, however, a striking fact that just the last mentioned regions at 
Lyngvær, Slaggrunnen, and Floholman, where the submerged surface has 
so great a resemblance to that of the southern part of Træna Plateau, are 
also built up of granite, and so are Skibatsvær and Jonsgruiman [Rekstad, 
19^5' PP- ^9 ^O- Tt is noteworthy that the submerged plateau west and 
southwest of Lovund and Lovundvær has an irregular surface like that 
of the w^estern side of the Træna Plateau. It is built up of comparativelv 
resistant gneiss and partly of granite. The much rlissected plateaus of 
Nesoi, Sornesøi, Lyngvær, and Mavær, east of Træma (Fig. 1 14, IX and X) 
are built up of granite. 

On tlie otlicr liand, the most le\el submerged plateaus with the most 
sharply defined outer edges, like those of Flovær and Lånan — and Sor- 
vær to the east-soutlieast of tliem as well as the plateaus at Ytre Flesan 
to tlie north — are built up of limestone, and so are the northern parts 
of Vega and the plateau to tlie north of it (at Kilvær). The extremelv 
level plateaus at Husvær and Sandvær, and at Hvsvær are built up of 
mica-schist, and at Skjærvær of a kind of metamorphic tuff (see Fig. 115). 
The plateaus to the soutli of the latter region, at Fuglvær and westwards 
to Sjola and southwards to Steinan and Fngelsbo Skerries (see Fig. 113), 
cire largely built up of granite, and so are Sola, the southern part of Vega, 



1 92 I. Xo. II. THE STRANDFLAT AND IsOSTASY. I 49 

and the group of islands called Mudvær to the soutli of the latter. At 
Alflesa north of Engelsbo Skerries there is granite and limestone. 

Although the inner parts of the submerged plateaus of this granite 
legion, c. g. south of Fugl vær, may be almost as level as the plateaus of 
mica-schist and limestone to the north, their outer parts, towards west 
and southwest, are less regular, as the map Fig. 1 13 shows, and their outer 
edge, c. g. at Sjola, Steinan, and Engelsbo Skerries, is less sharply defined 
than in the regions of mica-schist and tuff, at Skjærvær, and of limestone, 
at Flovaer and Ytre Flesan to tlie north. 

The outer parts of the submerged plateaus of the granite regions 
have thus a resemblance to the Træna Plateau built up of granite. 

On the whole as regards their surface relief there seems to be a 
typical difference between the submerged plateaus built up of granite, 
or similarly resistant rocks, and those built up of limestone, mica-schist 
or other less resistant rocks. 

But irrespective of the differences in geological structure it is natural 
that the surface of more or less isolated submerged plateaus far out to 
sea and exposed to the full effect of the breakers, should be somewhat 
lower and more outward sloping, with a less sharply marked edge We 
may also find this in the region of the mica-schist and limestone, c. g. on 
Skjærværgrunnan outside Skjærvær (see Fig. 113) where the smallest 
depth is 17 metres and there is no sharply defined edge. At the outer- 
most edge the plateau southwest of Skjærvær (and west of Hysvær at 
Flesa, Langtaran, Einarfall, &c., see Fig. 115) also exhibits similar fea- 
tures. We do not know, however, whether the rock may not be granite 
in this region, although the island Flesa consists of highly metamorph- 
osed tuff. We find granite quite near to the south at Sjola which 
is obviously situated on the same ridge as Skjærværgrunnan (see 
Fig. 113). 

The plateaus at Onsteinen, Ertenbraken, and Storbraken (northeast 
of Ryggefallan, see Fig. 113) are built up chiefly of mica-schist and to 
some extent of limestone (e. g. Storbraken). They have the very level 
and horizontal surfaces near sea-level, with sharply defined edges and 
very steep side slopes, which are typical of plateaus cut in these rocks. 

At H or svar to the south the rock is to some extent gneiss, and the 
islands are more scattered, and the surface is less regular (see Fig. 113). 
West of Horsvær and south of Ryggefallan the submerged surface is very 
irregular and resembling that of a granite region, and some rocks rising 
above the sea actually consist of granite. 

The banks at Hogbraken (southwestern corner of map Fig. 1131 may 
possibly also consist of granite, although they have perhaps more the 
features of plateaus of mica-schist, with fairly level surfaces, well-marked 
edges bounded by steep side slopes. 



150 FRIDTJOF NANSFN. M.-N. KI. 

South of I l();;l)r,'ila'n, ami far oui lo s<a, in ahout 65" 12' X. l-at. and 
I 1" I"". Lon,!4-., is an isolalc-(| plaicau with the small island-j^roup Sklinna 
which is huill U]; of -granite. 'The islands arc- soinc-what scattered as is 
<^cncrall\- the case on j^M"anitc plateaus, hut on the outer s'juthwestern, 
western, .and northwestern side of the islands tlie surface of the plateau 
is very le\el. foiminj;- a horizontal plane sf^me few metres below the sea- 
surface with a shariilx defined outer ed.s^e and very steep outer side 
slopes descendin^j;- abruptly to depths of uKjre than 300 metres. This is more 
like the typical features of plateaus of mica-schist or lirnestone. But on 
the inner side of the islands, towards northeast, cast, southeast, and south, 
the plateau has a more typical slopin«^- i^Tanite surface, with ^^-reater depths 
and with no definite horizontal plane, and no sharply marked edge. 

d"he i^lateau of Hnrta, east of Sklinna, in about 65° 12'X.Lat. and 
1 1" 25' JL. Lonj^-., is Iniilt up of g-abbro containing a great deal of carbo- 
nate of lime. Tts submergefl surface is extremely level forming a hori- 
zontal plane a few metres below the sea-surface with a very sharply 
defined c(\<f;Q at tlie same depth, and with a great number of low small 
islands, skerries, and rocks scattered over its whole area. 

7\s was mentioned on p. 128, the island-group of X'ikten built up of 
granite, and extending far into the open sea, is surrounded on its outer 
sides by broad submerged platform, exhibiting an almost horizontal 
plane only a few metres below the sea-surface, with no appreciable 
seaward slope, and with a sharply defined outer edge at about the same 
depth, or at least at depths less than 10 metres. The surface of this 
plateau is, however, less level and more dissected by clianncls and depres- 
sions than the very level surface of c. g. the Hysvær Plateau l)uilt up of 
mica-schist. 

As a result of the above cursory investigation (jf the relation between 
the surface topography of the submerged plateaus and their geological 
structure we may establish the following general rules: 

The typical plateaus built uj) of granite (and also of gneiss) have an 
uneven surface, the islands on them are Iving scattered, tlieir surfaces are 
much dissected by channels and depressions, and their depths vary much 
and are often comparatively great. Their surfaces slope outwards, with 
no well-marked edge, and often the sea outside is not very deep and has 
an outward sloping bottom, wdth no very deep cliannels or hollows, and 
there is no very sharp difference between the sloping bottom of the 
shallow parts of this sea and the submerged plateaus of the strandfiat. 

The typical plateaus built up of mica-scliist or limestone (or similar 
less resistant rocks) have a very different surface topography. The is- 
lands on them are flat and lying close together, onlv separated bv narrow 
and shallow sounds. The submerged surface is very level with small 
depths, forming a horizontal plane near present sea-level or only some 
few metres below it, and the outer edges are sharply defined at about 



I92I 



. No. II- THE STRAXDFLAT AND ISOSTASV 



the same depth. The side slopes are g-eneraU}' vei-\ ^tccp and there are 
often channels and hollows with considerable depths just outside or on 
the sides of these plateaus. 

It mav seem difficult to find a satisfactory explanation of this 
difference between the two kinds of plateaus. One mig^ht have expected 
that the plateaus of less resistant rocks should have been more attacked 
bv recent erosion, and more denuded, and consequently lower, than the 
granite plateaus. 

The only explanation I can find, is that the plateaus of rocks with 
little power of resistance have formerly had a greater extent, and have 
formed fairly compact land masses rising above the sea. After the land 
had been more or less worn down, the outer part of these plateaus have 
been much cut back, also during the last glacial period, by the glaciers 
excavating the deep channels and hollows outside them where the weak 
rocks offered especially favourable conditions for the glacial erosion. 
This is the reason why there are often such deep channels and hollows 
with very steep side slopes outside these plateaus. 

It is probable that the erosion of glaciers in rocks with comparatively 
little power of resistance to frost erosion, produces a very different sculp- 
turing above and below the water surface. Above the sea the mountain 
slopes on the sides of the glaciers are much attacked In- the frost erosion. 
and the result is comparatively broad valleys with sloping sides. Below 
the sea this is entirely different, the rocks are protected by the sea against 
the frost erosion, and there will onlv be erosion on the under side of the 
moving glaciers. They will therefore cut narrower channels with steeper 
side slopes, and sharply defined edges bounding the flat plateaus which 
will not in interglacial time be attacked and rounded off by frost erosion 
or subaëria! denudation against which they are protected by the sea. 
The still remaining middle parts of the plateaus were truncated by 
the shore erosion comparatively recently. They are, therefore, very level, 
standing near present sea-level, as there has been but little time for the 
wave erosion or for glacial erosion to wear them down to greater depths. 
It might be asked where are the older plateaus cut in this kind of 
rock? and why do we not find them at lower levels? The answer may be 
that where the older plateaus were cut in weaker rocks, thev have b3en 
more or less cut back by the glaciers during the subsequent glacial periods, 
and they no longer exist as parts of the submerged strandfiats. 

The map Fig. 113 also shows that the outer edges of the plateaus of 
mica-schist and limestone, e. g. at Flovær and Skjærvær, are cut back 
more than the edges of the granite plateaus at Lyngvær, Slaggrunnen. and 
Floholman to the north, and at Sjola and Steinan to the south. The 
plateaus at Ertenbraken and Storbraken are also cut back with deep 
channels and hollows round tiiem. 



152 IKIDIJOI' NA.NSKN. M.-N. Kl. 




Fig. 117. Soullivvard view Iroin Mcloivær towards Amoi. Stranclllat cut in granite. 

(Sfpt. 9, r 9 I :il. 

Tlic plaU'aus hiiilt uj) of j^ranilc, ov lianl i^iieiss or similar resistant 
rocks, lia\c not hccii cut hack 1)\- tlic t^laciers likr the plateaus of weaker 
rocks. Tlieir surfaces are therefore to a great extent ohler, anrl the sea 
outside tlieni less deep and with more gradual slopes, as the glaciers had 
more resistant rock to work in. Tlic surfaces of the plateaus are more 
uneven because, during the length of time after they were first formed, 
they have been exposed to much glacial erosion, dissecting them and 
making them more irregular and sloping. The channels between the islands 
have also l)een wirlened, producing the ajjpearance of more scattered is- 
lands. The wave erosion must also have had some appreciable effect 
upon the surfaces, especially of the outer parts of these plateaus, during 
the long time they have existed; and may have lowered them. Smaller 
and more isolated plateaus outside the greater ones, may have been 
lowered to various depths by glacial erosion, as well as by wave erosion, 
and owing to the resistance of their rock they have not been cut away 
bv the glaciers like those of weaker rocks. 



Coast of Northern Helgeland and the East Coast of West Fjord. 

The outer coast of northern Helgeland, north of Sol vær and Lovund. 
and the coast fartlier north along the eastern side of A'est Fjord is chiefly 
built up of granite and to some small extent of gneiss, /. c. rocks with 
much power of resistance to erosion. It is therefore in accorrlancc with 
what might be expected, that neither the emerged nor the submerged 
strandfiat are developed to any great width in this region. But the 
emerged strandfiat is seen almost everywhere along the shores, forming 
a flat, low foreland in front of the steep mountains (cf. Fig. 118) on the 
peninsulas of the mainland and on the high islands, and its low plane 
extending seawards over the various groups of small low islands (see 
Figs. 3, 117, and 119). Being largely cut in granite the surface of these 
islands is often .somewhat uneven, with small rounded knolls (see Fig. 117). 
Rekstad gives [1913, PI. I, Fig. il a most interesting illustration of the 
uneven surface of the strandfiat cut in granite on Briksvær Island, in 
67'' 16' N. Lat. and 14" E. Long. It consists oi a great manv rounded 



I92I. Xo. II. THE STRANDFLAT AND ISOSTASY. I53 




Fig. 118. Strandflat cut in granite at the foot of steep granite mountains near Kunna 
south of Salt Fjord. iSept. 9, 19121. 

knolls with steep sides and fairly deep depressions between them. But 
the summits of the knolls are to a great extent at the same level of about 
30 metres or somewhat more above the sea. 

I have had no opportunity of measuring the heights of the strandflat 
along this coast, but judging from the impression made when seen from 
the sea (c-f. Figs. 117, ii8, 120. and 121 ), I believe that there is. at least 
to some extent, a low level similar to that measured by Sahlstrom on 
Donna and Heroi. There is possibly also a higher level as. for instance, 
indicated by the summits of the many rounded knolls on Briksvær Island, 
mentioned above. 

Ahlmann says [1919, p. 205" that on the seaward side of the fairway 
between Sandnessjoen and Bodo "there occur continuous level surfaces 
at 5 — 10 metres and at 20 — 30 metres altitude above sea-level". 

The picture Fig. 119 of the island Landegode, north of Bodo, shows 
two levels, the low level of the south-westernmost point of Landegode 
and of the islets to the west, and a higher level on the southwestern part 
of the island in front of the steeply ascending mountain side (partly 
covered by a fog in the picture). As pointed out by Rekstad [1913, p. 15 
this plateau is about 100 metres above sea-level, and cannot therefore 
belong to the strandflat we are discussing in this paper. But it has ob- 
viously been cut by shore erosion. It is noteworthy that it has very nearly 
the same height as a similar plateau (or "shore-line") at Torghatten 
which is about 109 metres above the sea [cf. Rekstad, 191 5, p. 45 and 
Pl.\ II. Fig. 2 . I consider it probable that these plateaus may be remnants 
of an earlier strandflat, perhaps formed at the beginning of the first 
Great Ice Age. They are in both these cases cut in granite and have thus 
been able to survive later erosion. Remnants of an old strandflat at a 
similar level may perhaps also be found at other places, c. g. on the \'iker 
Mountain (granite) northwest of Torghatten where there is a fairly ex- 
tended plateau, on the coast of the mainland inside on the southern side 
of Sonnesviken, on the southwestern corner of Vega (granite), &c. 



154 


IKII/IJ( 


1 NANSKN. 




M.-N. Kl. 










.>> — " 


- 










■rtfM 



Fig. 119. Lüw land un .soulli-wosUrn cml ol Landigodc Island, witli stranddat on small 

islands outside. (Sept. 9, 1912). 

The .suhnuT.^iMl stramltlal roinnl the- inlands has no threat extent 
aloii^- northciii Mcl,i;elanfl ami the coast to the north, anrj it rajii'lly 
dimishes in width northwards towards \'cst h'jord, forming only a nar- 
row strip alon.t;- the coast of tlie mainland, and small ])latforms round the 
island-s^roups out in the sea. 

Its surface is on the whole une\-en anrl irregular anrl has much the 
same type as that of the southern part of the Træna Plateau. There may, 
however, be some difference: All the submers;ed plateaus with small 
island-groups built up of <j;ranite have very une\en surfaces, dissected by 
numerous channels and depressions, and especially far out to sea their 
surfaces are slopini;' outward towards the dee]X'r sea on the sides without 
an\- sharply defined edi^e, anrl the depths of the plateaus vary much, so 
that it is difficult to find any si)ecial depth which mi^-ht be said to 
indicate the levels of their initial planes. Outside these plateaus of 
granite there are also as a rule a great many isolated small shoals standing 
at various flepths anrl indicating no definite ])lanc. As examples of sub- 
merged granite plateaus of the above kind may be mentioned the plateaus 
of Mykcn, Valvar, and Skjarrar, northwest of Træna, with numerous 
isolated shoals, and the plateaus of Rorstapvar and Gronna far out to sea 
in about 66^ 54' to 67^' 3' N. T.at. and 13- 3' to 13" 18' E. Long, with very 
typical surfaces. Some parts of the submerged surface of these plateaus 
stand very near present sea-level with numerous shoals and rocks almost 
in the water surface or only some metres below it. One might get the 
impression that a plane has been cut approximately at this level, but then 
the ground slopes outward on all sides without any marked edge, and 
this makes it extremely difficult to decide what the depth of the level 
actually is. In manv cases the surfaces of the plateaus slope more gradually 
outwards on their outer, seaward side than on their inner, landward side, 
where there may even sometimes be more of a well marked edge. This 
might seem to indicate that the surfaces of these granite plateaus have 
been eroded to some extent bv the wave action, wliicli has lowered their 
initial levels and made them more sloping. 

It is very characteristic that the submerged plateau of Fiigloirar 
(consisting of granite) in 67"3'N.Lat. and 13^' 36' E. Long, has the 
typical outward sloping surface of a granite plateau, while Fleiiiz'a'r only 
6 kilometres to the northeast which is built up chiefly of crystalline lime- 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. I55 




Fig. 120. Engeloi with the strandtlat on Lundoi in front, cut in granite. (Sept. 8, 1912). 

stone [Rekstad, 1913, p- 4'. has a comparatively horizontal surface, with 
a great number of low islands, and a fairly well marked edge with steep 
slopes outside not far from the outer border of the islands. 

The plateau of IIcIlig:'(C)\ in about 67° 24' XLat. and 13^' 54' E.Long., 
east of Landegode, has a quite similar surface with a sharply defined outer 
edge near present sea-level and steep side slopes. It is built up of mica 
schists [Rekstad, 19 13. p. 13'. But the plateau of Lyngvar just inside 
and separated from Helligvær, only by a narrow channel consists of 
granite and has the typical outward sloping surface of granite plateaus. 

The small plateaus of Kjarvar (chiefly granite'» and Stcinsvar 
(granite), between Fleinvær and Helligvær, have the tvpical surfaces of 
granite plateaus. It is also a quite common feature with the granite 
plateaus that their islands are more scattered, while on the plateaus of 
less resistant rocks, like limestone and mica-schist, the islands lie as a 
rule closer together with narrower sounds between them. cf. for instance 
the striking difference between Lyngvær and Helligvær hing close 
together. 

The small plateau of Terra is built up of a kind of schistose horn- 
blende rock. Its surface is most like the granite surface. 

The plateau of Giver far out to sea north of Fleinvær is built up 
chiefly of a coarse mica-schist. It has a fairly horizontal submerged plane 
with well-marked outer edges some metres below sea-level. 

The plateau of Karlsokar, in about 67'^'33'X.Lat. and 14'' 38' E.Long., 
on the southern side of the entrance to Folia Fjord, is built up of mica- 
schist [Rekstad, 1917a, p. 16 and map', and has the typical surface of a 
plateau of rocks of little resistance, with the numerous islands close 
together and a sharply defined outer edge some metres below sea-level, 
and comparatively steep side slopes, while Slovccr consisting of gneiss, 
and situated just to the northeast and separated from Karlsoivær only 
by a narrow channel, has more the surface of a granite i)lateau with more 
scattered islands and less sharply defined outer edges. 

Still more characteristic in this respect is the plateau of Husvar 
(with Husoi), in about 67"43'X.Lat. and 14" 23 — 44' E. Long., north 



T56 FRIDTJOF NANSKN. M.-N. Kl. 



Fig. 121. .Strandtlal rut in granite and syenite at Tranoi Lighthouse on Hamaroi, near 
tile inner end of Vest Fjord. (Sept. 8, 1912). 

of Folia Fjord, ft is chiefly built up of j^niciss [Rckstarl, 19 19. \>. 27 and 
mapl. It has a threat many scattered islands, skerries, and sunken rock.s. 
and its submerged surface slopes outwards with no .sharply marked edge. 



Lofoten and Vesterålen. 

The islands of Lofoten ar.d \>sterålen are to a great extent built up 
of very resistant igneous rocks: gabbro-monzonites, granites, &c. And 
as the initial land was high the development (jf the strandfiat must have 
been a very slow process in this region in spite of the severe northern 
climate and the exposed situation of tlie coasts out to sea. 

The strandfiat is therefore naturally narrow along the coasts of 
Lofoten and Vesterålen, but in many places it is very conspicuous and 
sharply marked at the foot of the high steep mountains (cf. Figs, i and 2), 
fre(|uently forming deep horizontal incisions in the mountain sides, with 
precipitous rock walls or cliffs behind them. 

Descriptions of the strandfiat of this region have been given by 
J. H. L. Vogt [1907] and Th. \'ogt [1912] and also by Ahlmann [1919] 
who, however, holds the view that whilst in some places like Værøi its 
plane is, at least ]~)artly. formed by wa\e erosion, it is along the rest of 
the coast, what he calls the "distal base-levelled plain". 

Th. Vogt gives [1910, PI. II and III] some very illustrative photo- 
graphs of the strandfiat at Gaukværoi and ITasseloi in A'esterålen. which 
seem to me to demonstrate clearlv how utterly impossible it is that these 
planes extending horizontally to the foot of the steep mountains can have 
been formed solely by subaërial denudation (see also Fig. i of this 
treatise). 

The previous writers have given no accurate heights of the emerged 
strandfiat of Lofoten and Vesterålen, based upon actual measurements 
by levelling, nor has the present writer had an opportunity of measuring 
its height. J. H. L. Vogt [1907, p. 20], obviously basing his estimate 
on the official maps ("Gradavdelingskarter" in scale i : 100,000), says 
that the inner edge of the strandfiat in Lofoten lies in numerous profiles 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. I57 

(probably made from the mapsi just a little higher than 30 metres above 
sea-level. 

ludg-iiig from the impression which the emerged strandfiat gave, 
when seen from the sea in various regions of Lofoten and \ esteràlen. 
and also judging from the many photographs taken, I estimate the height 
of a great part of it to be about 20 metres or less above sea-level. It is 
possible that the height of its inner margin, at the foot of the mountains, 
may be somewhat higher and frequently approach 30 metres above the sea, 
but its actual height will generally l)e difficult to d.etermine, as it is to a 
great extent covered l)y the scree. 

In his excellent description of the topography of the Sfjuth-western 
part of Lofoten Th. \'ogt expresses himself in a very similar manner. 
He says [1912, p. 15] that at the foot of some hearllanrls of \'æroi. there 
are low points the inner boundary of which at the foot of the almost 
vertical cliffs is about 35 to 40 metres above sea-level, but he did not 
directly measure them. "As a rule the demarkation line between the pre- 
cipitous cliffs and the strandfiat is much lower both on \'æroi and on the 
Røst islands, in some places nearly at sea-level, but it is frequcntlv hidden 
by the heaps of sto?ies and gravel fallen from the cliffs." 

The subiiicrgcd strajidflat of the Lofoten and \'esterâlen Islands has 
a remarkably small extent, considering the exposed situation of the is- 
lands. Its surface topography is on the whole most similar to that which 
we liave found to l)e typical for submerged ])lateaus built up of granite 
or other resistant rocks. Il has in general fairlv great depths, is sloping- 
more or less outwards from the islands, and its outer edge is less sharply 
defined than that of the submerged strandfiat of southern Helgeland. 

The submerged stranriflat of Lofoten is most perfectly developed on 
the two small submerged plateaus farthest out to sea towards the south- 
west. On the one are situated the two islands Varai and Mosken and 
some scattered skerries, and on the other the islands of Rost. These is- 
lands are built up of gneiss and other crystalline schists. Only the rockv 
island Mosken near the north-eastern end of the \'æroi Plateau is built up 
of gabbro and granite. 

Th. \'ogt [1912^ has given a most interesting description, with a 
sketch-map and illustrations, of the emerged as well as the submerged 
Strandfiat of these plateaus. Their submerged 'topography mav be studied 
in much rletail in the charts Nos. 70 and 71 (in the scale i: 50.000) of 
"Norges geografiske Opmåling". 

Both plateaus are o]:)long with their longitudinal axes in the direction 
S\V to NE. They are much alike as to shape and size. If w^e take the 
isobath for 25 metres as boundary the Rost Plateau is about 25 kilometres 
long, from SW to NE, and about 11 kilometres broad at its broadest. 
The \'æroi Plateau is about 19 kilometres long (from SW to NE) and 
about 1 I kilometres broad. 



158 FRIDTJOF NANSEN. M.-N. Kl. 

'I'licir suhiiK rt^c'l surfaces arc comparât i\cly even, and cc^nsi'Urahh 
more su llian iliat of ilic 'I'ra-iia I'laleau, lo wliicli they otherwise have 
resemblances in se\eral respects. Alonf,^ tlieir middle parts the surface 
is near sea-level or only some few metres helow it forminj^ horizontal 
planes wliicli on hoili plateaus extenrl towards the north-western sifle, 
where i1k'\' come near the outer slopes with fairly sharply defined edges 
at depths of ahout 5 or (> metres below mean sea-level. Towards the 
opposite, /. ('. the south-eastern, sides of the plateaus their surfaces are 
more slopint^-, with less sharply flefincd edges, perhaps at about 20 metres 
or more below sea-level. Towards the south-west the surfaces of both 
])lateaus slope more or less gently, and in most j)laces with no well marked 
edges at any special depth. Tn their north-eastern parts they have, how- 
ever, more sharply defined edges at some few metres below sea-level. 

Rostoi, the biggest island of the Ivost Plateau, is situated near its 
north-eastern end and is ([uite flat, its highest part l)eing only i i metres 
above sea-level. Most of the many other islands on this plateau are also 
quite low and flat, but some of them form isolated 'stacks' with steep 
sides rising abruptly above the plane of the strandfiat to heights of about 
100 to 167 metres above the sea [cf. f. H. I.. \'ogt, 1907. p. 12, Th. \'ogt, 
1 91 2]. They are obviously the last remnants of the old land, which the 
shore erosion has not managed to plane down to sea-level. On the fairly 
flat tops of most of these 'stacks' (called 'nyker') the old, Palæic, land- 
surface still remains, almost intact, extending with its rounded undulating 
forms to the abrupt edges of the precipitous side walls [cf. Th. A'ogt, 
1912, i)p. 7 ff.]. 

This is still more striking on the island \'æroi, where this old land- 
surface is of a greater extent with sharplv marked edges above the 
precipitous side cliffs [cf. Th. \ ogt's illustrative drawing, 1912, Fig. 8 . 

Here again we have thus convincing proof that, during the last period 
when the strandfiat of this region was planed to its present shape, the 
effect of the subaërial denudation has been insignificant as compared 
with that of the shore erosion. 

As was already pointed out by Th. A ogt, it is striking that on the 
Rost Plateau as well as the A æroi Plateau, the high islands, being the 
remnants of the initial land, are all of them situated along the eastern or 
south-eastern sides of the plateaus, with the greater part of the submerged 
strandfiat on the outer side towards the west and north-west. The natural 
explanation might seem to be that this is due to the marine denudation 
wdiich has attacked the plateaus most vigorously on their outer sides. 
But the above mentioned fact that the surfaces of the plateaus are most 
level, and nearest the sea-surface in their north-western parts and have 
the most sharply defined edges along their north-western sides might seem 
contradictory to this explanation. 



I 92 I. X". ri. THE STRANDFLAT AND ISOSTASV. I 59 

Another feature is also of interest in this connection. Both plateaus 
are dissected by shallow submerged bays and channels along their south- 
eastern sides, while there are hardly any such formations along their outer 
north-western sides. These channels generally indenting the plateaus ir 
northerly or north-westerly directions may be 15 to 17 metres deep and 
in some places even deeper than 20 metres. The emerged land is to some 
extent indented in the same manner. The island Rostoi has several similar 
narrow bays on its southern and south-eastern side. But it is especially 
conspicuous on A'æroi. The north-western coast of this island is high 
and steep and is not indented while its whole south-eastern coast is in- 
dented by two or three cirque-like bays, which Th. \ ogt assumes to be 
formed chiefly by marine erosion. I think he is right, but it is the shore 
erosion bv frost and not the wave erosion which has been of chief im- 
portance, the same as on Træna (see p. 145). I also consider it to be 
probable that these bays may originally have been more or less cirques 
formed by local glacial erosion. 

But if the marine ersosion (i. c. the shore erosion) has formed these 
bays and channels along the south-eastern sides of the plateaus, why has 
it not produced similar formations along their north-western sides and 
why is not, for instance, the north-western coast of \'aeroi indented? 
It might be answered that on the latter side the marine erosion has been 
so vigorous that it has cut back the coast sufficiently tc obliterate these 
formations, but this answer would hardly be satisfactory, first, because 
an increased marine erosion might rather be expected to increase the bays 
if they are partly formed by it, and secondly, because it might at any rate 
be expected that traces of these bays and channels should occur on the 
submerged strandfiat outside the coast. This strandfiat is, however, very 
level especially on the north-western sides of the plateaus, and stands 
nearer sea-level there than on their south-eastern sides. 

It might be assumed that these channels and bays have to some extent 
been sculptured by local glaciers on these plateaus, at an earlier time 
before the land was so much cut back by marine denudation, and the 
level submerged strandfiat, especially on the north-western sides of the 
plateaus, may tlien be expected to have been formed by shore erosion 
during later periods, and to have become so perfectly even because the 
land was not previously as much dissected in those inner areas. In that 
case we might, however, expect to find traces of similar submerged 
channels along the outer north-western slope of the plateaus, but this is 
not the case, and the edges are well defined near the side slopes of the 
plateaus in these regions, and are near sea-level. 

E^ven if we could assume that because of the meteorological con- 
ditions, the glaciers especially occurred on the south-eastern sides of the 
plateaus, this could hardly give a satisfactory explanation. 



l6o FHIFiTjOI- XA.N.SKN. M.-N. Kl. 

'J'o ]]\v il siHiiis iiiosl |)r<.l)al)lr lli.il llicsc fcalurcs in tin- l'ijjo^^raijhy 
of llic suhmcrt^c'l si rand flal of these two jilaU-aus are rlue to tliL- sj.ecial 
slructure of llie .gneiss foiinni;; tlieiii. 'i'lie facl thai at least some of the 
sul)inef,mil cliannels ha\e more or less ])araliel 'lirections may also iiKÜcate 
their ilepenileiice on the structure ot the rock. 

It ma}- be pointed out that on Moskeuesoi lo the north-east, there is 
a siiuilar (Hfference between the coasts, the inner, eastern coast of the 
island heiiiq- indeiiteij 1)\ fjords aiKJ dissected by \alleys. while ajonj^ the 
outer, western coast there are no fjorrls, and the coast is very steep with 
an almost continuous mountain-rid!L;"e alon.L;' this side of the island [cf. 
d h. \ i)o-t, i()i2, [""i^. I I j. ddie fjords and \allevs of this island are cirques 
or cir(|ue vallex's sculptured b\ local cir(|ue f^laciers. The ex])lanation that 
the absence of fjords along the outer coast of Moskeuesoi is due to the 
fact that this coast has been cut back by tlie marine flenurlation is hardly 
satisfactory, because we mi^ht then at least expect to find traces of the 
deej) fjords on the submerged platform outsirle the coast. 

If the difference in the topography of the two Cfjasts cannot be ex- 
plained b\- differences in the structure of the rocks, I think it probal)le 
that the meteorological conditions may have been more favourable for 
the formation of cirque glaciers and for tlie cir(|ue ercjsion on the south- 
eastern and eastern side of the initial land than on its outer, western 
and north-western side. 

The strandfiat of Moskeuesoi is entireh' fiiffercnt from that of the 
\''æroi and Rost Plateaus. This may to some considerable extent be due 
to the difference in the rocks, which are very resistant, consisting of 
monzonite (augite-syenite), ga])l)ro, and labradorite rocks. As pointed 
out by Th. \ ogt, a narrow but well marked emerged straiulflat occurs in 
several places along the inner, eastern coast of the island, e. g. at A, Sør- 
våg, and Reine, and on its northern side, at Mevold and \'alle. But alorg 
the outer, western coast there are liardly any indications of an emerged 
straiulflat, the mountains falling steeply into the sea. 

In the case of the submerged strand flat it is quite different; there 
are hardly any indications of it along the inner coast of the island, the sea 
bottom sloping without any appreciable break from the coast towards the 
deep hollow of the Vest Fjord. Outside the outer coast there is. however. 
a submerged platform, which at least in some places lias a fairlv well- 
marked edge. But the depths of this p.latform and of its outer edge are 
as a rule much greater than those of the submerged strandfiat of Helge- 
land. In some places there are somewhat higher banks on the platform, 
with^depths of about 20 to 28 metres below sea-level, but to a great extent 
its depths arc al)out 35 to 40 metres or even more. In some places the 
surface of tlie platform slopes gradually from the coast to the depths of 
the continental shelf without anv noticeable edce. 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 161 

If the isobath for 40 metres of depth be assunied to form the boundary 
of the platform its width varies between 7 and less than 2 kilometres. 

As pointed out by Th. \ ogt the submerged strandfiat is most 
distinctly developed outside the coast at Refsvik (near the southern end 
of the island) where the land is built up of labradorite rock. He thinks 
this might l)e due to the fact that this rock is somewhat less resistant 
to erosion than the other gabbros of Lofoten. There is here a platform 
nearly 2 kilometres broad, with numerous shoals and rocks near sea-level 
and some emerging above it, and the outer edge of the platform is sharply 
defined. 

It seems doubtful whether the greater part of the submerged plat- 
form outside Aloskenesoi with its surface at depths between 30 and 40 
metres and even between 40 and 50 metres, can actuallv, in its present 
shape, be considered as a submerged strandfiat. It seems hardly probable 
that it can have been formed at a level so deep below present sea-level. 

Th. \ ogt may be right in thinking that the reason why there is less 
of a strandfiat round Aloskenesøi than on the \^æroi and Røst Plateaus, 
may be that there has been a more vigorous glacial erosion under the high 
mountains of this island. It seems to me probable that it is the glacial 
erosion which has lowered the level of the submerged platform along the 
outer coast of Moskenesoi, although the surface of this platform is more 
even and less deeply dissected by channels or hollows than is generally 
the case where there has been an effective erosion by glaciers. It seems 
also difficult to understand why there is no submerged strandfiat along 
the inner coast of Moskenesoi. The less effective marine erosion on the 
inner, less exposed coast is not sufficient to account for this fact, for 
we would at any rate expect some indications of a submerged platform, 
even though narrow, especially as there actually is an emerged strandfiat, 
and there are parts of a submerged one along the coast further to 
the north-east in \'est Fjord, and a well developed submerged strandfiat 
often occurs along more sheltered coasts. 

Is it perhaps possibile that the big glacier which deepen?d the 
\'est Fjord, has cut away the submerged strandfiat? 

It is, however, also obvious that the nature of the rocks has some 
connection with these differences in the development of the submerged 
strandfiat, for here again we find the same feature, that the submerged 
platforms cut in more resistant rocks have greater depths and more sloping 
surfaces than those cut in weaker rocks. The gneiss of the Værøi and 
Rost Plateaus is certainly considerably less resistant to glacial erosion 
as well as to shore erosion than the rocks of Moskenesoi. 

It seems probable that these platforms cut in very resistant rocks 
where the initial land was comparatively high, have required a verv long 
time for their formation. They are therefore comparativelv old, and have 
been exposed to much glacial erosion during several glacial periods. The 

Vid.-Selsk. .Skrifter. I. M.-N. Kl. 1921. \o. 11. 11 



102 Fl<l\>l\]Oy NANSKN. M.-X. Kl. 

rfsull is lliat i1k\ liavc been more or less lowered, so that their rlepths 
lia\c hccii imicli iiicrcascfl, llicir surfaces have become more or less 
(iiit\\anl-sl()|jin,L;, and tluir outer clj^-e has been rounflerl off anrl is less 
sliarpK marked. As was pointcil out abo\c, it is, however, stran/:(e that 
the glacial erosion has not sculjjlured deejjer valle\s and channels on this 
platform, similar to those that occur on land. 

As has been pointed out b\- Idi. X'o^^t^ some mou^tai^^ and nîountain 
ridg'es on Moskenesoi li<a\c' flat to])s formed by the initial, Palæic surface 
of the land, which still exists more or less intact, extending with its 
undulating level surface to the very sharply defined edges over the pre- 
cipitous side walls of the many cir([ue \alleys. 

This prox'es that in this region during the period when these valleys 
were finally formed, the effect of the subaërial denudation lias been very 
insignificant as compared to that of the erosion of the cirque glaciers, 
which works to some extent in a manner similar to the shore erosion by 
frost, cutting back the sides of the cirque valleys into the mountain block 
with the initial fairly flat surface on top. The sharp edges between this 
surface and the mountain walls have thus arisen. 

Along the south-eastern coasts of r\'st-l\:goi and Öst-J'ügoi (north- 
east of Moskenesoi) and the south coast of Ilinnoi, bounding A'est Fjord 
to the north-west, there is a distinctly developed, although narrow strand- 
flat, consisting to a large extent of a platform with numerous small is- 
lands and skerries at the foot of the steep mountains (see Figs, i and 2). 
The surface of the submerged platform between the emerged islands and 
skerries is uneven and irregular with depths varying between 10 and 50 
metres or even more, as they might be expected to be where the platform 
has been much exposed to glacial erosion. 

The rocks of this region are gabbros and granites. 

Along the outer, north-western coast of Flagstadai, Vcst-Vågøi, 
Grijiisoi, and Ost-Vågoi there is a narrow submerged platform. In some 
places, c. g. north of Flagstadoi, it is flat and fairly even, at depths of 
about 12 to 16 metres below sea-level, having a fairly well marked edge 
at this level up to 7 kilometres from the coast. In other places it is more 
irregular. North of Grimsoi there is a 5 kilometres broad and very even 
platform with depths less than 10 metres, to a large extent between i and 
6 metres below sea-level. To Avhat extent these plateaus are levelled by 
recent sediment cannot be decided. 

Along the outer coast of Vesterålen, further north, there is an ir- 
regular submerged platform similar to that along the outer coast of 
Lofoten. The islands of this region are likewise built up of gabbros, mon- 
zonites, and granites. There is also a well-marked emerged strandfiat 
along the coasts of the islands [cf. Th. \ ogt; 19 10, photographs Pis. II 
and III]. 



I92I. No. II. THE STR.ANDFL.\T AND ISOST.ASY. 1 63 




Fig. 122. Strandtlat west of Komag Fjord on the southern side of Soroi, view towards 
Kobbe Fjord and Oi Fjord. July 6, 191 2. 

Senjen to Ringvasøi. 

As there are few detailed charts of the coast north of Vesterålen, it is 
difficult to study the submerged strandfiat in this region, but on the 
whole it seems to have very much the same character as along the outer 
coast of Lofoten and Vesterålen. Outside the great island of Senjen 
(where there is a detailed chart, Xo. Sj) the submerged platform is in 
places as much as 9 kilometres broad, but has a very irregular much 
dissected surface, with varying depth. The outer coast of this island is 
also dissected by long, deep fjords. The rocks are to a great extent 
granites. 

A submerged strandfiat of very much the same character as the one 
just described to the south-west, extends along the outer coast north- 
eastwards as far as Rebbenesoi, Grotoi, and northern Kvaloi (70° 16' 
N. Lat.). In some places, e. g. outside southern Kvaløi, it attains a width 
of 18 kilometres (in about 69" 55' X. Lat.). It is much dissected, and 
carries a great number of scattered islets, skerries, and shoals. The coast 
is supposed to be built up of igneous rocks similar to those of the coast 
to the south-west. 



Finmarken. 

There are only few indications of a strandfiat. above or below 
present sea-level, along the coasts east of the region of Ringvasøi, 
northern Kvaloi, and \'annoi. This sudden disappearance of a fairlv 
well developed strandfiat coincides in a striking manner with the sudden 
change in the geological structure of the coast, as I have already mentioned 
on pp. 50 f., where this fact has been discussed at some length. The lack 
of detailed charts of this coast with sufficiently numerous soundings 
prevents us from studying the detailed topography of its sea-bottom, and 
of possible submerged platforms at deeper levels. 

In some places, e. g. on Sørøi (Fig. 122) where there are igneous 
rocks (gabbro) I have observed indications of what mav be considered 



164 FRIDTJOF NANSEN. M.-X. Kl. 

to l)c an cnicrj^cd strand flal . I'ut on tlie wliolc there are very few such 
formations alone; the coasts 01" !• ininarken, whicli as a rule is extremely 
steep on the outer, seaward side. It is, however, possible that the low 
land extending along the coast t'roni \ ardo to \'adsø and along the north 
side of Varanger l-'jord may be considered as a strandflat, as Reusch has 
indicated on his map [1894]. As I have pointed out [1904, p. 119] it is 
also possible that there are indications of an emergerl stranrlflat in the 
inner parts of the I'inmark h'jorils, e. [^. in the inner end (jf Porsanger 
h'jord. 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASY. 



l6: 




Fig. 123. Norwegian Harbour (Norske-havn) and Mount Misery, seen from the south. 
July II, 1912. [From Nansen 1920]. 

XI. THE STRANDFLAT OF BEAR ISLAND. 



The northern part of Bear Island, comprising nearly two thirds of 
its whole surface, forms a very flat and low plain (cf. Fig. 124, "Lavt, 
flatt slettelann"). cutting horizontally through the various geological 
formation and rocks, and through several faults (cf. Fig. 125), without 
any appreciable difference in height. Its flat surface is composed of 
sandstones, shales, limestones, and conglomerates of the Carboniferous 
and Devonian systems [cf. J. G. Andersson, 1900 a, and Olaf Holtedahl, 
1919]. 

This plain is about 13 kilometres broad, and about 9 kilometres 
wide from the north coast to the foot of the mountains which rise steeply 
from the plain in the southern part of the island (cf. Figs. 124 and 125). 

Seen from the sea. this plain gives the impression of being perfectly 
level (cf. Fig. 127). It is bounded along the shore on all sides by a vertical 
cliff (Figs. II and 128"), which, according to Joh. Gunnar Andersson, is be- 
tween 25 and 30 metres high above the sea. Only at some few isolated 
places is there a gentler slope from the plain down to the shore. 

Professor Holtedahl informs me that according to his observations 
the height of the shore cliff mav exceed 30 metres and even approach 
40 metres on the north-west coast of the island, while it may be but little 
more than 20 metres above sea-level on the north-east coast in Coal Bay 
( "Kulbukten"') near the Norwegian coal station (see Fig. 124). Perso- 
nally I have not been ashore in this northern part of the island. 

Joh. Gunnar Andersson [1900 a. p. 248] says that "from the said 
height of the cliff (25 — 30 m.) at the shore the plain rises uniformly 
by almost imperceptible degrees landwards to the region between the 



1 66 



FRIDTJOF NANSEN. 



M.-X. Kl. 




vm. 



Stalle,,^!, i ,, 



19° ÖL. 



Fig. 124. Map ot" Bear Island, according to the survey of C. J. O. Kjellstrom and A. Ham- 
berg in 1898, with additions by J. Kessler 1899, and O. Holtedahl 191 8. 
I Ella Lake. 2 Alfred Mountain. 3 Ymer Valley. 4 Antarctic Mountain. 5 Hamberg 
Mountain (424 metres). 6 Fugle Fjell (Bird Mountain). 7, 8 and 9 Three peaks Urd (539 
metres), Verdandi (465 metres) and Skuld (464 metres) on Mount Misery. Lavt ßait Slettclami 
is the low, flat plain of the northern part of the island. [From Nansen 1920]. 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASY. 



167 



Xo^^l^a^pf; 



fCaafi 



Kf/eintiu. 



Xopsfci hÆ-onJitr- 




y no uin'tc Kail -Cid 

Pld'tC ,(HMc/CuU-tui 
]TuiicU.yuc kcUk&Uyi.) 

qui Aam^t£7t mtxiko-nûlo- 
'm eA-aU-fi^, ^Q/ri^nl -nUl/ie. /Cull. 



[Miltc 



chue cUZ, mtôl kcUkått-n 




\KuU-t. i^ai-^ic ciU -KLeit tod baniteii 

\ {zia^iit. iikkc. Scon-iten cfk ifkfälc ndr 
j ) med kuU-lo^ tn-evL ^ki/b i "^ 

Lofi^tie. Ouùn^<cjKc I ncKLu -Ituk. 



Hg. 125. Geological map of Bear Island by O. Holtedahl, based on his investigations in 

1918, and those of J. G. Andersson in 1898. [Holtedahl 1919]. 
Translation of explanation, to the left : i Trias. 2 Younger Upper Carboniferous (Spirifer- 
Cora-coral limestone). 3 Older Upper Carboniferous (Fusulina limestone, &c.). 4 Yellowish 
sandstone with conglomerates, probably Middle Carboniferous. 5 & 6 Middle Carboniferous. 
5 Upper part chiefly limestones with limestone conglomerate at the top. 6 Lower part 
chiefly red sandstone and conglomerate. 7 & 8 Lower Carboniferous and Upper Devonian 
sandstone series with coal seams. 7 Chiefly sandstone. 8 Sandstone often interbedded with 
shale. 9 Tetradium limestone. Middle Ordovician. 10 Younger dolomite series, Lower Ordo- 
vician. 11 Slate-quartzite series. 12 Older dolomite series, presumably Ozarkian. 

9 — 12 Heclahook System. 

Explanation to the right : i Faults, probably older than the Spirifer limestone. 2 Probable 

faults of the same age. 3 Faults younger than the Spirifer limestone. 



1 68 



FRIDTJOF NANSEN. 



M.-X. Kl. 



iiorlluTiiniosI i>art of ()s\\aWl I 'rdiiioiilory (C JswaM's \-(>r])cr^, l'ij^'. 124) 
and the western corner of Mouiil Misery, where its altitu'le may amount 
to about 100 metres". 

Judging from m\' (jl)serA'ations mafic from the sea, I unrlerstand that 
it is only in the southern central part of the plain that its surface attains 
altitudes as high as this estimate. On the .south-eastern side of the island, 
south and south-west of Mount Misery, inside Russian Harbour and Nor- 
wegian Harbour ("Russe-havn" and "N(;rske-havn" Fig. 124, see also 
Fig. 125, D) there is a low kuul rising gently inland with somewhat 
similar heights (cf. Fig. 123). 

During the Isachsen Expedition to .Spitsbergen in 19 10, Adolf Hoel 
visited the northern plain on June 24th and kindly gives me the following 
extract of his diary: 

"24/6, 19 10. Together with Koller, Håvimb, and Malme I went out 
to take the levels of the plane of abrasion north of Mount Alisery. A pro- 
file southwards from the north point of Bear Island looks like this: 




P/ane./Mu.SÙ.n '^^^^f f^' 



(UoüU/SOrrv. 
atovc sea 



Fig. 126. 



The plane of abrasion extends (from the coast) to a point a good 
distance north of Mount Misery. South of that point a low, but more 
broken land begins, which is not a plane of abrasion at all." 

Hoel adds that unfortunately the time was too short for carrying out 
the intended measurements by levelling, the heights given in the above 
profile Fig. 126 are, therefore, based on aneroid readings, and are not 
accurate. The altitudes 50 metres and especially 150 metres may be 
too high. 

Holtedahl also informs me that in its central southern parts the plain 
is somewhat less even than further north, and that it rises gently in low- 
ridges towards the higher mountain slopes to the south (cf. Fig. 129). 

The general level of the very flat northern part of the plain, and of 
its marginal parts east and west, is betw-een 30 and 50 metres above the 
sea. On the north side of Mount Misery there is at this level a sharply 
defined boundary between the level plain and the steeply ascending 
mountain side (see Fig. 127). I estimate the height above the sea of this 
demarkation line at the foot of the mountain slope, to be about 30 or 
35 metres on the east coast near Cape Levin, and it rises gently with the 
plain inland. 



1921. No. II. 



THE STRANDFLAT AND ISOSTASV 



169 




-o c 

c « 

fc S 

.-H ^i^ 



lyo 



KklDI JOK NANSKN. 



M.-N. Kl. 



-jy^T,— «— w^lHirh 




Fig. 128. West coast of Bear Island, showing the horizontal plane of the strandflat, and 
the hills rising steeply above it. (Photograph by O. Holtedahl 19181. 



J. G. Andersson says that in its western part the plain is perfectly 
level. The surface consists there of Carboniferous limestone and sand- 
stone which by weathering form a boggy argilaceous soil. The region of 
Spirifer limestone east of North Harbour (Nord-havn) is somewhat 
undulating, and so is also the extensive area of Devonian sandstones. 
Over the whole plain there are numerous shallow depressions, with 
hundreds of small lakes, but no indications of any valleys. 

Most part of the plain is covered by débris, big, and small stones, 
obviously formed /;; situ by frost disintegration, but in some places, 
especially in its inner higher part, low flat ridges of bare rock are seen 
(Fig. 129), as is mentioned by Holtedahl. 

Andersson considers this remarkable ])lain to have been formed in 
preglacial, and posttriassic. time by marine abrasion (i. c. wave erosion), 
in the manner suggested by Richthofen. "In the weak and easily dis- 
integrated Devonian and Carboniferous layers which build up this low 
land, the abrasion has advanced relatively rapidly, Avhile the Heclahook- 
massive to the south has offered much greater resistance" [1900 a, p. 278]. 

I think there can be no doubt that Andersson is right in assuming 
that this is actually a plane of marine denudation, at any rate the lower 
anrl flattest part of it, as there is no other process that can cut an extensive 
level surface like this in solid rock with no valleys and no actual drainage 
system, only numerous shallow lakes with accidental outlets. According 
to my view, however, it is not the wave erosion, but the shore erosion by 
frost that has been the chief agent for the development of this plane as 
w^ell as that of the Norwegian strandflat. The wave action has been of 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY 




Fig. 129. From the sandstone area in the central part of Bear Island. lo the right where 

the ground is darker is seen a part of a small isolated ridge of Spirifer limestone north of 

Alfred Mountain. (Photograph by O. Holtedahl 19 181. 

great importance mainly by washing away the debris of the shore erosion, 
and also by wearing down the outer parts of the plain. 

We can at present observe this denudation process going on along 
the shores of Bear Island, wliere it has formed the vertical shore cliffs, 
with many small accumulations of ice and snow lying on the beach at their 
foot during a great part of the summer (see Fig. 11). 

That the wave erosion also has much direct effect along this weather- 
l:)eaten coast is demonstrated by the many caves formed in tlie present-day 
shore-line [see Xansen, 1920, PI. III]. I have observed such caves in 
South Harbour ("Sorhavn", Fig. 124), in Walrus Harbour ("Hvalross- 
havn", Fig. 124), and in Norwegian Harbour ("Xorske-havn", Fig. 124). 
Holtedahl has found deep caves penetrating far inland on the south-west 
coast of the island. The caves are generally formed along lines of frac- 
ture in the rock. 

As I have pointed out before (p. 144) the disintegration by frost is, 
in my opinion, of much importance also for the formation of caves in this 
cold climate. Where the average temperature of the rock is below freezing 
point of water, there will easily be alternate thaw and frost according as 
the rock surface be washed by the sea or left dry. As far as I have seen, 
it is on the whole striking to what a small extent the walls of these caves 
exhibit traces of direct wa\e erosion, /. c. forms rounded by the waves. 
The surfaces are generally rough like that of rock exposed to disinte- 
gration by frost, and the edges are only to some small extent water-worn. 

Where there are limestones or dolomites, or where there are fractures 
in the rock filled with limestone, the lime may be dissolved by the sea- 
water, or also by fresh water, provided that the low temperature of the 
rock does not prevent liquid water from percolating through the fractures 



172 



FRIDTJOF NANSFN. 



M.-N. Kl. 




Fig. 130. The vertical cliff and the older, undulating mountain surface, on the west side of 
.South Harbour („Sorhavn"). July 10, 1912. [From Nansen 1920]. 



The efficacy of the shore erosion on Bear Island is clearly demon- 
strated by the vertical cliffs, often some hundred metres high, along the 
southern coast. Tn most places they bound the older undulating mountain 
surface of the island with sharply defined edges (Fig. 130). This mountain 
surface has probably been developed partly by the subaërial denudation 
in the length of time and partly by glacial erosion. Although the sub- 
aerial denudation is obviously very effective in this region where the 
rocks have so little power of resistance, and where the disintegration by 
frost is so very active, it has not been able to keep pace with the shore 
erosion and round off the edges of the cliff to any appreciable degree, 
along the exposed parts of the coast. 

Traces of a strandfiat at the same lieight as the outer parts of the 
northern plain of the island may also be found along its southern coast, 
where it forms parts of a horizontal platform at the foot of the steep 
mountain slope (see Fig. 131). Along this part of the coast it has. how- 
ever, to a great extent been cut away by the effective marine denudation 
of more modern time. 

It is a striking fact that there are very few traces to be found of a 
submerged part of the strandfiat near present sea-level round the coasts 
of Bear Island. In some few places there are narrow submerged ledges, 
with depths less than 20 metres and with some few rocks emerging above 
water here and there, but they are hardly more than a few hundred metres 
broad, and as a rule the sea-bottom near the shore sinks steeplv down 
to depths of 30 and 40 metres. At about this level there is, however, a 



I92I, 



No. 



THE STRAXDFLAT AND ISOSTASY. 



173 




Fig. 131. Fugle Fjell (Bird Mountain) on the south side of Bear Island. 
After a photograph by A. Hamberg 1898. [From Nansen 1920]. 

submarine platform 9 to 20 kilometres broad, surrounding Bear Island. 
A similar submarine platform evidently surrounds Hope Island to the 
north-east, and on the submarine ridge connecting these two islands the 
soundings indicate three flat banks at the same level, with depths about 
30 to 40 metres below the sea surface. 

The question is whether these platforms ought not to be considered 
as parts of the strandfiat which during times of emergence have been 
cut down to somewhat lower levels than the Norwegian strandfiat. The 
probability is that they are built up of rocks with relatively little power 
of resistance to erosion, like those of Bear Island and Hope Island, and 
they have, therefore, easily been planed off during periods of emergence, 
even though these may have been relatively short. 

In my opinion it is impossible that the emerged strandfiat of Bear 
Island (the northern plain) can have been developed to its present shape 
during preglacial time, neither by marine denudation nor by subaërial 
denudation. In the latter case a drainage system would have been 
developed on the emerged, gently sloping plain. It might be objected 
that all traces of the broad, shallow valleys of this system have b?en 
obliterated by later glacial erosion. But if so, this glacial erosion would 
naturally have dissected the plain and made it more uneven. By whatever 
process this strandfiat was formed, it is obvious that no preglacial plain. 
as level as this and cut in rocks with so little power of resistance, could 
have survived the destructive erosion of the glaciers of the great Ice Age. 

It seems to me probable that this emerged strandfiat was planed off 
to its present level surface by the shore erosion, as I have previously 
described, during periods when the island was standing somewhat lower 
(at least 30 to 50 metres) than now. Before that time the island had been 



174 FRIDTJOF NANSFN. M.-N. Kl. 

iinu-|i (Iciiiulcil 1)\- siilj.'K-rial erosion, iiilciisif icil hy ili(- frost in a climate for 
the most part severe, aii'l •Inriii;,'- iiiiMcr climatic periods by fluvial erosion 
which rk'\clopcfl a r|raiiia<;c s\>l(i)) of hroa^l flat valleys, which, however, 
have \)vv\] oMiicraldl 1)\- the sul)sei|uci)t j^lacial erosion anrl shore erosion. 

1 )urinj4- the ice Aj^'-es the lanrl has probably been much fjenufler] In- 
glacial erosion, which, howe\'er, has not cut \-ery 'leep \alleys and fjfjrds 
in this region of soft rocks. 

Another pr(icess which has been of importance for the surface forms 
and the planing of the land in this r(,'gi(;n, is the flowing soil ("soli- 
fluction") Icf. ( i. M(jlmsen, 1915], which J. (i. Andersson [1900] has studied 
on J'ear Island. 

I)y the disintegration of the frost the soft rocks are in many places 
transformed into an argilaceous mud which, when soaked with water by 
the melting of snow and ice in the spring- and summer, becomes semifluid 
and slowlv flows down the slo])es even though they be very g-entle. The 
freezing of the water in tliis boggy soil anrl the melting- of it again, 
greatlv helps this movement. Tn this mantier the moving soil has a 
tendencv to flow rlown into the small valleys and depressions and to some 
extent i)rotect them against the frost erosion, while the higher sloping 
parts are more or less deprived of their coating of soil and their solid 
rock is more exposed to the disintegration by frost. This process mav 
have some effect towards planing the land surface, and may help to 
obliterate the valleys [cf. Nansen, 1920 & 1921, Chap. TI]. The northern 
plain is, however, to a very great extent covered by stones formed by 
the disintegration of the sandstone, and the above process cannot have 
been of so much importance in those regions. 

After its final formation the strandfiat has been covered by a local 
glacier during the last glacial period of the island. This glacier has not. 
however, eroded the surface of the strand flat to anv appreciable extent, 
but on the surface of the rocks in some places, striæ still occur, radiating 
from the central part of the island towards the coast, as was first dis- 
covered by Nathorst [cf. J. G. Andersson, 1900, p. 438]. It is possible 
that the A'ery shallow depressions now forming the many lakes, are due 
to this glacial erosion. 

After the last glacial covering the bare rock-surface has been much 
disintegrated by frost, but as there has been very little transport on this 
flat plane the débris has to a great extent remained in siiit. 

It might l)e more difficult to decide the age of the now submerged 
strandfiat of Bear Island — and between that island and Hope Island — 
now standing at a level of about 30 to 40 metres below the sea-surface. 
The probability seems to me to be that this part of the strandfiat has 
been developed during the glacial periods. 

It is a striking fact that no traces of a postglacial elevation of the 
land have been discovered on Bear Island, as has been mentioned bv 



I92I. No. II. THE STRANDFLATAND ISOSTASY. I75 

Nathorst and J. G. Andersson [1900, p. 439]. As the latter points out it is 
possible that such marks might have been obliterated along these shores 
where the rocks have but little power of resistance and the postglacial 
erosion, especially by frost, has been considerable. But if there had been 
shore-lines or shore-cliffs on the strandfiat further inland, it seems none 
the less probable that some traces of them might have been found some- 
where. 

As long as there is no evidence to the contrary we may therefore 
assume as most probable that this island has had no postglacial elevation, 
as seems also to be the case on other small island plateaus like that of the 
Faeroes, Shetland, and possibly Jan Alayen. 

It might be more probable that there has been a positive movement 
of the shore-line in postglacial time, otherwise it may, for instance, be 
difficult to explain why there is not a broader submerged shore bench 
near present sea-level round the island, if the present-day shore was ex- 
posed during the whole of postglacial time to the very effective shore- 
erosion now going on. In the few places where a quite narrow shore bench 
with shallow water has been developed, c. g. Walrus Harbour, Norwegian 
Harbour, Herwig Harbour, North Harbour, and west of that region, the 
shore is built up of rocks, with very little power of resistance to erosion, 
and where it cannot have taken a long time to erode the existing sub- 
merged shore benches. 

In spite of the considerable postglacial elevation of the land that has 
taken place in Scandinavia to the south and on Spitsbergen to the north, 
and also on Franz Joseph Land, the disappearance of the ice cap has 
probably caused no appreciable postglacial elevation of the crust in this 
region, for during the last glacial period the island was probably covered 
by a small local glacier only, the weight of which depressed the crust less 
than the sea-level was temporarily lowered by the reduction of the volume 
of the Ocean owing to the accumulation of frozen water on land. 

But of more importance in this respect is probably the considerable 
displacement of the semi-plastic 'magma' underlying the rigid Earth's 
crust, which must have taken place when by the weight of the ice caps 
Scandinavia to the south was depressed 300 metres, or more, in its 
central parts and Spitsbergen some hundred metres to the north. By 
the displacement of the 'magma', the intervening region of Bear Island 
has probably been raised, and when, after the disappearance of the ice- 
caps, the displaced 'magma' again 'flowed' slowly back to its former 
position more or less, and the previous equilibrium was restored, there 
would be a slow subsidence of the crust in the region of Bear Island. 

If in this manner, the shore-line in the Bear Island region stood 
during the glacial time about 30 or 40 metres lower than now, the plat- 
forms may have been formed which now are submerged about 30 and 
40 metres below present-day sea-level. 



t yô I- KIDTJOF NANSEN. M.-N. Kl. 

'I1ic \crtical raii^c hclwcin tlic iippc iinost level ahf>vo ihe sca and 
the lowest suhnicrj^-ed level <>\ llie api)areiit strandflat is thus coiisiflcrahly 
f,^rcater on Px-ar Tslainl llian is L^-enerally ihc case along the Xorwegian 
coast (with the exceiilnni of l''inniark perhaps). This may be due to the 
fact that the rocks of the l^ear island region have so little power of 
resistance that the vigorous shore erosion may have managed to plane 
down fairlv hroail platforms rjuring relatively short periods. The planes 
cut at tenii)()rar\ high or low levels at wliicli the shore-line may have 
stood during vertical movements of the land crust or of the sea-level, may 
therefore in this region appear as though they l)elonged to tlie strandfiat. 
In Norway, however, a much longer time has generally ])C(tn required for 
the development of broad shore benches. The Norwegian strandfiat was 
therefore formed during long periods when the land crust stood at its 
normal level of equilibrium (cf. p. 42), while the higher shore levels lasting 
for shorter periods liave left few conspicuous marks only in the shape of 
old raised beaclies and shore-lines. 

While there can hardly be any doubt that the outer flattest parts of 
the plain of Bear Island have been levelled by shore erosion, it is very 
difficult to decide how its inner part, higher than 50 metres and rising 
to 100 or, according to Hoel, even to 150 metres above the sea, has been 
formed. We have seen that Hoel does not consider it to be a plane of 
marine abrasion because it is more uneven than the outer very level plain. 
On the other hand there is as a rule a fairly well marked boundary between 
this higher plain and the more steeply ascending mountain sides along its 
southern margin, especially at Mount IVIisery. Although there may b2 
some difference in tlie power of resistance of the rocks of the mountains 
and of the plain, it is hardly sufficient to account for the differencs in 
slope, and it seems to me to be possible that also the inner, higher part 
of the plain may have been formed by shore erosion, but during a more 
remote period than tlie lower flatter part, and by exposure to later erosion 
it has become more uneven. On the other hand it is difficult to believe 
that, during some comparatively recent' period, the land has been sub- 
merged up to these higher levels for sufficiently long time to have an 
extensive plain like this cut by shore erosion. 

This higher plain of Bear Island has a certain resemblance to the 
plain across Hitteren in Norw-ay, previously described (p. 123), rising 
gradually above 60 metres in the middle of the island (cf. map Fig. 107). 
It has likewise a width of 13 kilometres. On Alsten Island at the foot 
of the "Seven Sisters" there is a similar plain (about 8 kilometres across) 
rising above 60 metres in the midflle of the island. It is probably cut in 
mica-schist although to a great extent it is covered by quaternary marine 
deposits and moraines. 

If it were not for the small areas of these islands one might be 
tempted to assume that these plains have been gradually raised by iso- 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. I77 

Static movements during and after the formation of the strandfiat. We 
do not know yet within how small areas the isostacy may produce more 
or less local uplifts; but the gradient of lateglacial and postglacial upheaval 
sometime^ seem to vary appreciabl\' within comparatively short distances. 
Besides the rigidity of the crust has not prevented the formation by 
mountain folding of ridges no broader than these islands. 

On Bear Island as well as on Hitteren, and probably also on Alsten, 
considerable quantities of rock have been removed, which originally 
covered the areas of the plains. The mountains on the sides of the plains 
may give some indication of the initial height of the land. Where the 
rocks are so extremely weak as those forming the top of Mount Misery 
(539 metres above the sea) it is also obvious that a great thickness of 
rock has been removed from the top of the mountains during the long 
time that has elapsed since the development of the present strandfiat 
commenced. 

The removal of the weight of hundreds of metres of rock from above 
the present plains has caused a slow elevation of the crust in the region 
of the islands. During this continuous elevation of the original strandfiat 
its outer parts have been gradually lowered by the shore erosion, and 
thus a gentle outward slope from the inner parts of the plains has been 
formed. As the elevation was much slower during its latest stage, the 
outer part of the plain became most horizontal. 

It might be objected that, if this explanation be correct, the extremely 
level plain of Smolen (cf. pp. 120 f.) ought to have been elevated in a similar 
manner. The probability is, however, that the initial land of Smolen was 
much lower, the isostatic elevation was therefore much smaller and much 
slower, and while the shore erosion was still most vigorous, it had time 
to plane down the whole plain, and make it almost horizontal. 

This might seem a rather bold hypothesis, and it has to be admitted 
that weighty objections may be raised. If the isostatic compensation 
can assert itself, to some extent, within such small areas, it might be 
difficult to understand why the strandfiat now stands at levels which are 
very similar over extensive regions although the quantities of rock re- 
moved may differ greatly. These problems will be discussed m a later 
chapter. 



Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 11. 12 



,78 



IKIDIIOF NANSKN. 



M.-N. Kl. 




Fig. 132. Vogel Hook, the northern point ol Prince Charles Foreland, seen trom the north 

(July 27th, 1912). The difference between the steeper west side of the island, and the 

more sloping east side with a broad strandilat is noteworthy [from Nansen 1920]. 



XII. THE STRANDFLAT OF SPITSBERGEN. 

The etnerged strandfiat is a very conspicuous feature in the land- 
scape along- the coasts of Spitsbergen. It is very different from the 
Norwegian strandfiat as it forms more or less continuous low plains in 
front of the often oversteepened mountain sides, and it is not split up 
into numerous islands and skerries. 

Literature. 

A d o 1 f H o e 1 [1909, 1914] has subjected it to special investigations. 
He says [19 14, pp. 25 ff.] that "along the whole of the Spitsbergen coast 
which he had an opportunity of examining, frotii Ice Fjord to Wood Bay. 
the coast, between the sea and the mountains, is surrounded by a border 
of low land from which the mountains rise abruptly and partly vertically 
forming a rather conspicuous precipice". 

"The boundary between the plain and the mountain is often covered 
by quaternary marine terraces, or by a talus (scree). But it also often 
appears very sharply marked, and at a certain distance one always gets 
the impression of a well defined line of demarkation." 

"These coastal plains are missing only in the inner parts of the fjords. 
c. g. in the inner part of Cross Bay. They are especially well developed 
along straits, thus along Foreland Sound 10 to 20 kilometres wid;. they 
attain a considerable width on both sides of this sound. Thev reach their 
maximum development at prominent points, c. g. at the north-east corner 
of Prince Charles Foreland, at Ouade Hook, at Cape Guissez, and Cape 
Mitre." 

"Under otherwise equal conditions, thev are wider where tne rocks 
are less resistant, c. g. Carboniferous limestones and schists of the Hecla 
Hook system, and are narrower where more resistant rocks predominate." 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. I79 

Hoel States that, in the region between the southern end of Prince 
Charles Foreland and Cross Bay, the inner margin of the strandfiat, at 
the foot of the steep mountain-slopes, is about 25 and 30 metres above 
sea-level. 

He thinks that this strandfiat is a plane of marine abrasion, formed 
by wave erosion in the manner suggested by Richthofen. It is a plane 
of perfect regular it}-, much more level than the Norwegian strandfiat is 
generally. 

He agrees with my views [1904] as to the age and formation of 
similar strandfiats, and does not consider it possible that this strandfiat 
can be of preglacial age, but assumes it to be "more recent than the time 
of maximum development of the ice covering: it seems impossible that 
a plain could preserve so perfect an evenness as it has here, if it had been 
traversed by inland-ice". He also considers it to have been formed after 
the excavation of the fjords by the glaciers; and gives several convincing 
pieces of evidence [19 14. p. 2~'\ to prove that such has been the case. 

On the other hand, he thinks that, "after the development of the 
strandfiat there has been a period wlien the glaciation was more con- 
siderable than at present. At that time erratic blocks and moraines were 
left in several places, c. g. on the south-east extremity of Danes Island." 

"The geological conditions as regards this coastal plain prove that 
the glacial covering has varied considerably on Spitsbergen, and that the 
plain has been formed by marine abrasion at a time when the glaciation 
was relatively less considerable." 

Adolf Hoel maintains that the sea must have remained at its present 
level for .a long time along the coast of Spitsbergen, for the wave erosion 
has had time to form typical shore cliffs in numerous places along the 
open coast and in the great fjords. At the foot of these cliffs there is 
generally a sandy beach, dry at low water. This beach, is often continued 
in a narrow submerged platform sloping gently seawards, till the bottom 
suddenly descends by a steep escarpment to depths of 200 and 300 metres. 
This submerged platform must be a plane of marine abrasion, which he 
has especially observed in Cross Bay. 

Gerard de Geer has also discussed the strandfiat of Spits- 
bergen! but tries to explain its formation by his theory of dislocations 
(faults), and considers it to be thus preserved remnants of a peneplain, 
an explanation which, according to my view, is especially improbable on 
Spitsbergen, for several weighty reasons. First there arc strandfiats 
backed by steep mountain sides in many places where there are no traces 
whatever of any faults or dislocations, as proved by the investigations 
of Hoel and Holtedahl. 

Secondly the surface of the strandfiat of Spitsbergen is decidedly 
not that of a regidar peneplain. As Hoel has pointed out, it is often 




Fig. 133- 



7é 



30^ 



ri 



I Land lowet than. 50 md/us 



=Hl Glaci£/t 



GLacLCt coveti-tiq the 
Sluuiäflat. 




Submetged Sltandflat kss than. 
30 mettes below sea,-Uvel. 



Map showing the Emerged and Subtmuj^d 
Sttandfiül of West Spitsbetgen 

Based chiefly on Utt Sutveys by the. Notweqlcuv 

Spitsbe tgeru-[lxpulitu)nsl906-l9H 

The Contout of Ou SuinteiçcU St^xuuifLat behwetv Vojel Hook, 
Ptinct Chatlis retilanri, atui South. CaJat is btvsui on. tke. 
rj^'yt SouHtii.n^s mcuk. bif the HoeL-Expeclitions I9I3~/3ZI- 



SOUTH CAPE> 



ir 



15° 



t'ig- 134- 

'gs. 133 & 134. Map of the North and West coast of Spitsbergen, showing the distribution of the Strandfiat. 

Fig. 134 is a direct continuation of Fig. 133. 



182 IKlbl JOK NANSKN. M.-N. Kl. 

pcrfecth' even, ;iiiil 1I sliows 110 iii'Iications of any drainage system, which 
iinisl l)e tlic l\|)ical feature of a j>encijlaiii. There- may be a good many 
shallow lakes on lliis surface, drainiuj^' tlirouj^li more or less acciflental 
ch.-mnels of pos1,t;lacial ori^ni), hut there are no inrlications of older valleys 
on tlie flat i)lanes. 

P) e r t i 1 lio_i;honi has not studied the strandfiat of Spitsbergen, 
hut like lloel he points out [1914, P- 294] tliat shfjre-chff.^ are charac- 
teristic of tlie Spitsbergen fjords, and are well developed even in places 
where the wave action can <jnly ha\e been quite insignificant. 

lie maintains that the surf anrl the ice flrift are essential for the 
transport of the waste, and arc therefore necessary for the formation of 
a cliff-shore, but their direct power of erosion may be of minor im- 
portance. The destruction of the rocks by the frost must be much more 
effective. The fact that accumulations of snow ("snow-foot") frequently 
remain on the beach at the base of the small cliffs during a great part 
of the summer without being washed away by the breakers, also dis- 
proves the direct importance of the waves for the development of the cliff. 

He thinks that by keeping the cliffs always wet, the freezing sea- 
water must have much disintegrating power, especially as even open 
crevices are kept full of water. The sea-w^ater will furthermore thaw the 
frozen rock, and by alternate high- and low-tide a regular "regelation" 
may thus be produced. 

Where the coast is relatively much exposed, and the surf can assist 
the frost in breaking loose the material and carrying it quickly away, a 
considerable abrasion may be produced, and he assumes that the sub- 
merged flat platforms, i kilometre broad, outside more exposed coasts 
in Ice Fjord — c. g. between Cole Bay and Advent Bay and at Cape 
Thordsen — have been thus formed. 

As may be seen, the views of the present writer coincide in several 
respects with those of B. Hogbom. He justly pcnnts out that, although 
a great dislocation may probably have occurred along Tee Fjord, as sug- 
gested by Gerard de Geer, this cannot explain the formation of the sub- 
merged platform, intervening between the deep channel of the fjord and 
the steep mountain side. 

B. Hogbom does not think it is possible to decide finallv at present 
whether or to what extent, these platforms with their shore-cliffs have 
been formed in postglacial time. 

His view obviousi}' is [cf. 10 14, p. 295] that the Norwegian strand- 
flat has been formed in a similar manner. 

Angus AFE w e n Peach [1916] describes the strandfiat (which 
he calls the Preglacial Platform) of Prince Charles Foreland, and also 
along various parts of the coast of the mainland- of Spitsbergen. He 




Ordoi'icisk tiler Vmur- 
SiMir(HeJcl<i-huk Formasjon) 



^T-A Granlit og Gntis-grcinlttj 
y+^ sansifnliß Fra. SiLvr(eJïir/^ocl) 



so 
' ' I ' ' "^ 



100 km. 



56 O 



Sifclkjxpp 



Fig. 135. Geological Map of Spitsbergen, chiefly after A. G. Nathorst, based on the Swedish, Rus- 
sian, Scottish, and Norwegian researches before 1910. The broken and the dotted lines indicate 
the course of the "Veslemoy" in 1912, with the océanographie stations. [From Nansen 1920]. 
Translation of the explanation in the lower left corner, i Tertiary. 2 Diabase. 3 Jurassic und 
Cretaceous (Neocomianl. 4 Trias. 5 Carboniferous and Permian. 6 Devonian. 7 Ordovician 
or Under-Silurian (Hecla-Hook Formationl. 8 Granite and gneiss-granite, probably Post-Tertiary 

(according to Hoel). 



184 IKlDljOK NANSF.N. M.-X. Kl. 

States tli.'il "tilis slicif (011 l'rinci- ' liarics l'^rclaivJ) has in general a 
rocky surface, planed (jr carxed partis out of nearl\- vertical members 
of the Hcckla Hook series. It rises gently from near sea-level towards 
the slopes of the central hills, terminating inland at an altitude of about 
150 feet (46 metres). 'J'he constancy of level maintained by this inner 
margin and the fact that it is often marked by lofty cliffs or at least 
precipitous shapes, indicate very clearly that the i)latform is the result 
of marine erosion." . . . "Jjeach deposits occur here and there on the plat- 
form, but it is clear that they have no genetic connection with it. They 
are obviously of postglacial age, while the rock platform itself is ob- 
viously preglacial." 

His reasons for the latter conclusion are i) that the surface of the 
platform "has been shown to be striated in several places", 2) "it is 
fre(|uently strewn with iccborne erratics from the mainland of Spits- 
bergen, while patches of drift containing foreign boulders have been 
found here and there covering its surface." 

The probability that these striæ, erratics, &:c. may be due to the last 
advance of the glacial covering is, however, not considered, nor is it 
explainer] how this remarkably level platform might have been able to 
preserve its evenness if it had been exposed to the erosion of the big 
glaciers of the Great Ice Age. 

The platform is in several places on Prince Charles Foreland and 
on the mainland of Spitsbergen described as rising from the shore to a 
general level of between 100 and 150 feet (30 and 46 metres); while it 
is said to rise to about 200 feet (61 metres) at the base of the over- 
steepened hills, in several places on the mainland of Spitsbergen, c. g. at 
Ouade Hook, at the mouth of St. John's Bay in Foreland Sound, in Lowe 
Sound (\'an IMijen's Bay), in Recherche Bay, on the south-west side of 
Advent Bay near Advent Point, south of Sassen Bay. 

Peach considers it to be beyond a doubt that the cutting of this plat- 
form was a later event than the partial submergence of the old river- 
valley system, by which according to his view the fjords originated, 
consequently in preglacial time. 

His view is that "the Spitsbergen rock-platform is clearly the same 
as that of the coast of Norway", and a platform of similar type "is found 
also on the west coast of Scotland, at a height of 100 to 140 fe?t (30 to 
43 metres) above sea-level. There is also a platform of marine erosion 
of preglacial age a short distance above sea-level in England and in the 
south of Ireland." 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASY. 




Fig. 136. The Saddle Mountain, with the strandflat on both sides, on the southqfn end of 
Prince Charles Foreland, seen from Foreland Sound on July i6th, 1912. [From Nansen 1920]. 

Relation between the Development of the Strandflat and the 
Geological Structure of the Land. 

The strandflat of Spitsbergen is to a great extent cut ni rocks with 
relatively little power of resistance to the frost erosion. It has therefore 
easily been levelled to a fairly regular plane in most places, rising gently 
inland from the shore. Where, however, the rock is more resistant, the 
strandflat is much narrower, or is poorly developed, with a more uneven 
surface. This is. for instance, the case in the region of the north-west 
corner of Spitsbergen, where there are gneiss-granites and granites (see 
Fig. 145). The strandflat is here poorly developed, and its surface is un- 
even like that of the Norwegian strandflat cut in resistant rocks. 

On the east side of Wijde Bay the land is built up of gneiss-granites, 
and in its northern parts of mica-schists and other crystalline schists. 
There the strandflat is well developed, with a flatter, more regular sur- 
face, obviously because the rocks are less resistant. 

A comparison between the geological map Fig. 135 and the map 
Fig. 133 and 134 may give some idea of the relation between the distri- 
bution of the strandflat and the geological structure of the land. 



Prince Charles Foreland. 

The emerged strandflat is well developed on Prince Charles Fore- 
land, and it is a noteworthy fact that in the northern part of the island 
it is wider along the relatively sheltered east coast than along the west 
coast exposed to the violent wave action of the open ocean, and where 
we might expect to find it especially well developed. 

The explanation may to some extent be that the strandflat has chiefly 
been finally planed by the shore erosion by frost after the land had been 
much denuded especially by the local glacial cirque erosion, while the main 
importance of the wave action has been to carry away the débris formed 
by the sliore erosion. For this purpose there has been sufficient wave 
action in the Foreland Sound, now 10 to 20 kilometres broad, and 



i86 



i KIUI |OF NANSEN. 



M.-N. Kl. 




\"\ii. 137. Kast coast of Prince Charles Fonrland northwards from the Foreland Plain 



formerly, when the shore-line stood hii^iur 'lurin,^- the plaiiiiif^ of the 
strandflat, even broader. 

Tt is, however, strikini,'' that the hi,L;hest mountain ridf,a' of the nor- 
thern part of the island is much nearer its west coast than its east coast 
and is steeper and less dissected by valleys along its west side than along 
its east side (cf. Fig. 132), where there is als(j a greater accumulation of 
glaciers. These are features somewhat similar to those previously men- 
tioned on Aloskenesoi, and also on \'æroi and Rost, in Lofoten (see p. 160). 
This similarity may be accidental, anrl may be more or less due to the 
geological structure of the lanrl. l'ut it is also possible that the greater 
accumulation of glaciers on the eastern side of Prince Charles Foreland 
is due to the meteorological conditions, and that, therefore, there has been 
a more active local erosion by glaciers (cirque glaciers) on the eastern side 
of the island than on the western side, cutting back the mountain slope. 

It might seem most natural to explain the difference in steepness 
between the eastern and western sides of the mountains as an effect of a 
more vigorous marine erosion along the more exposed west coast. But 
in that case there should be a broader strandflat along the west coast. 
As will be mentioned later, there is a submerged platform along this 
coast, but not very broad, and the width of the submerged and emerged 
strandflat is not broader along the west coast than along the east coast 
of the island. 

In its southern part the strandflat extends across the island, forming 
the Foreland Plain (or Foreland Laich), 4.5 to 9 kilometres Inroad from 
coast to coast, and 15 kilometres long between the slopes of Alcunt Methuen 
to the north (Fig. 137) and of Persis Crest (Fig. 140) of the Saddle 
Mountain to the south (Fig. 136). 

According to Hoel [1914, pp. 25 f.] Mr. Koller of the Isachsen Ex- 
pedition took the levels of this plain in 1909. It rises very gently land- 
wards from both shores, but the highest point along his line of levelling. 
16.3 metres above the sea, was 4.6 kilometres from the west coast and 
only 1.95 kilometres from the east coast, which may indicate that the 
western part of this plain has been slightlv more denuded bv the wave 
erosion. The width of the strandflat mav otherwise varv between i and 



1 9? I. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY 



187 




and Mount Methuen, seen from Poole Point on July 23rd, 1912 [from Nansen 1920] 



5 kilometres along the east coast, while along- the west coast it is as a 
rule less than 2 kilometres and often less than i kilometre. 

The strandflat of Prince Charles Foreland is cut in shales and lime- 
stones of the Hecla Hook system, which are easily disintegrated bv the 
frost. In some places there arc lenses and layers of quartzites forming 
quite low ridges, as pointed out bv Hoel. 

The bare rock is seen everywhere in the surface of this strandflat, 
or is covered by a thin layer of débris evidently disintegrated bv frost, 
and in many places transformed into an argilaceous or muddy soil, with 
tlie characteristic network of stone-rings, which, according to mv view, 
is due to the effect of the frost and thaw causing expansion and con- 
traction of the water in the wet soil and in the patches of snow when 
freezing and melting [cf. Xansen, 1920 and 1921, Chap. \'ni]. 

Wherever the rock is bare near the shore, there is a vertical shore 
cliff 4 to 8 metres high. At many places there is, however, a flat shore 
of sand, gravel, or pebbles, evidently to some extent consisting of water- 
worn moraine material (Fig. 138). In some places in the shore I noticed 
that this drift was stratified. It lies nearly at the same height as the real 
strandflat cut in solid rock, and rises to a few metres above high water 
level. 

The wave-action, and perhaps also the sea-ice, have built up ridges 
which may often extend considerable distances along the shore, forming 
lagoons, which are common along the east coast of this island. 

On July 25th, 1912, I observed that on the north side of Poole Point 
the storm of the previous day and night had entirely filled up and ob- 
literated the entrance to the lagoon. Only two days earlier this entrance 
had been so deep that I could not easily wade across it. 

The strandflat rises gently from the shore, or the shore cliff, to a 
general level, the altitude of which I estimated to be about 6. to 10 metres 
in the region of "Sandbukta". near the northern part of Saddle Mountain, 
south of the Foreland Plain. 

Further inland there are a good many ridges rising to a higher and 
fairly uniform level extending to the foot of the mountain, which rises 
abruptly frc^m the plane (cf. Figs. 139 and 140). I estimated the height 



1 88 



FRIDTJOF NANSEN. 



M.-N. Kl. 



of tilis k'\cl to l)c 20 to 30 metres above ihe sea, hut 1 lia'l no oportuuity 
of ineasuriu;^'^ it 1)\' levellinj^-. My cstiiTiate aj^recs, however, with Hoel's 
statement that the inner marj^nn of the j)lain is 25 to 30 metres alxn-e 
sea-level. I'each's estimate of tlie hcij^ht <>ï the inner marj^-in of the 
stranrlflat of tlie I'oreland, 150 feet (or 4O metrc-s), apj^ears to he too hij:(h. 
At the foot of the mountain, I'ersis Crest, there is a conspicuous 
leflge, to a j:]^rcat extent huill u|) of loose stones. I observed indications 
of similar ledj^-es at the foot of the mountain, on the srjuth-eastern side 
of .Sarldle Mountain, and also on the south-east sirle of Mount Methuen 
(cf. Fig. 137). 1 found a similar conspicuous ledge in Lake Valley on the 
west side of Wijdc Bay (see later). 




Fig. 138. The Sand Bay ("Sandbukta") with the southern part of Saddle Mountain and the 
strandflat on Prince Charles Foreland. July 22nd, 19 12 [from Nansen 1920]. 

Hoel assumes that these stone-ledges have been formed by stones 
that have fallen down from above on to the surface of small glaciers 
along the mountain sides, and have gathered at the foot of these glaciers. 
But considering the evenness of their upper surface, situated very nearly 
in the same level as far as I could see from the distance. — I think that 
thev indicate a raised shore-line, and they may have been form;d to a 
great extent bv stones that, owing to the regular shore-erosion, described 
pp. 30 ff., have continually been falling down from the mountain side above 
on to the accumulations of snow on the shore ledges, and have gathered 
on the beach and on the relatively steeply sloping sea-bottom outside the 
foot of these shore-accumulations of snow and ice. The stones that 
remained above water were soon disintegrated by the frost, while those 
below water were more protected. 

The fact that the stones on these ledges are sharp-edged as* a rule, 
and show no traces of ha vine: been worn l)v the surf, mav be accounted 



I92I. No. II. THE STRANDFLAT AND ISOSTASV. 



189 




Fig. 1 30. 



01 tiic stranaiiat witti stone-nngs , on the east side ot Saddle .Mountain. 
July 22nd, 1912 [from Xansen 1920]. 




Fig. 140. The strandflat at the foot of Persis Crest, on the north-eastern side ol ^addle 
Mountain. July 2nd, 1912 [from Nansen 1920]. 



IÇO IKIDIJOI' NAN.SKN. M.-N. Kl. 

for as an effect of tlie frost action, in a similar way as th • fact tliat the 
rocks of the shore-lerlj^'es cut in solid rock are also shariJ-erlj:(ed, with very 
few traces of wave-erosion. .Some stones have obviously also fallen down 
on these ledj^es after tliey were raised ahoxc sea-level. 

It is striking;- that these led^a's, wherever I have observed them, 
always occurred at the foot of very steep mountain sides, where stones 
may be expected to liaxc been perpetually falling;, especially when the 
shore erosion worked at the foot of the mountain. The probability is. 
how-ever, that the inner part of these ledt^es are more or less cut in .solid 
rock, but this is often difficult to examine, as their inner margin is 
generally covered ])\ a talus (scree). 

I had no opportunitx' of measuring tlie heiglit of these stone-ledges 
on Prince Charles Forelanrl, but estimate it to be about 50 metres above 
sea-level. 

At the foot of the ledges accumulations of snow are now eroding 
small cir(|ues in their steep slope, oversteepening them (cf. Fig. 140). 

Along a part of the east coast of Prince Charles Foreland, from 
J. IVIurray Point southwards, the strandfiat is now covered by flat gla- 
ciers, Murray Glacier and Buchanan Glaciers (Fig. 9 and 10). The rocks 
of the strandfiat project in several places along the outer edge of the 
glacier. I agree with Hoel, that the strandfiat was obviously formed 
during periods wdien the glaciers had less extent than at present. They 
have afterwards extended over this flat plane, and form a nearly hori- 
zontal ice-sheet, to some small extent fed by the small glaciers of the 
mountain slopes behind. 

The glaciers to the south, Geikie Glaciers, form thin sheets extending 
from the mountain slopes over the undulating lower land (Fig. 137). 

The strandfiat of Prince Charles Foreland is continued outside the 
shore in a siihiiicrgcd strand flat (see Fig. 133) the contours of wdiich 
may be approximately traced along the east coast by the soundings of the 
Isachsen Expeditions of 1909— 1910 in the Foreland Sound [cf. Isachsen, 
19 1 2, chart], and along the west coast by the soundings taken by the 
Hoel Expeditions in the years 19 13 to 1921. By Hoel's kind permission 
I have been able to study the detailed charts with these soundings. In the 
map Figs. 133 & 134 is drawn the isobath for 30 metres below sea-level. 
The edge of the submerged strand flat seems to be very near this isobath 
at depths between 20 and 30 metres or often near 20 metres. In some 
places the edge seems to be quite sharply defined, whilst in other places 
the side slope of the submerged plateau is more gentle, making the edge 
less distinctly marked. 

Along the northern part of the west coast of the island, north of 
Cape Cold, the submerged strandfiat is mostly about 2 kilometres broad, 
the width decreasing north of Cape Cold to about i kilometre. Off Cape 



1921. No. II. THE STRANDFLAT AND ISOSTASY. I9I 

Cold it is broader, nearly 5 kilometres, and along the coast to the south 
it is between 2 and 3 and 4 kilometres broad. 

Along the east coast of the island the width of the submerged strand- 
flat varies much, being mostly between i anrl 2 kilometres, but off the 
north-east coast it has a wide extent, 7 to 8 kilometres. The surface of 
this broad platform slopes gently from a depth of 10 metres near the 
shore, to about 27 to 30 metres near the edge, which seems to be very 
sharply defined towards the east and north-east, where the depths suddenly 
increase to 126 and 178 metres, while towards the north and north-west 
of Vogel Hook tlie sea-bottom slopes more gently outwards. 

It is a striking contrast between this wirle submerged plateau and 
the total absence of a su])merged strandfiat along the west coast of 
Brogger Peninsula on the opposite side of Foreland Sound. As the north- 
east coast of Prince Charles Foreland is to some extent built up of 
Tertiary sandstone and conglomerate, the probability is that the wide 
submerged strandfiat, being a continuation of the very broad emerged 
strandfiat, has, to some extent at least, been cut in rocks of this system 
which have offered relatively little resistance to the shore erosion. On 
the opposite side of Foreland Sound the strandfiat may, to some extent, 
have been cut away by the glacier formerly extending far towards the 
north-west from Comfortless Glacier. At the north-western end of 
Brogger Peninsula, off Ouade Hook, however, there is a submerged 
platform, which has not been cut away by the glaciers of Foreland Sound 
or King Bay. 

Further south there is a Barrier ("Revet") across Foreland Sound, 
from John Murray Point to Michael Sars Point, which, however, seems 
to be built up of loose material, mostly sand, like the sand spit of Michael 
Sars Point. 

Between this region and St. John Bay there is a submerged strandfiat 
2 to 3 kilometres broad. The rock of the shore is here Tertiary and it is 
the same on the opposite side of the sound on both sides of Ferrier Haven 
and at Poole Point where the submerged strandfiat is also fairly broad. 

Considering that the emerged strandfiat is cut in solid rock, it seems 
probable that the rock surface continues more or less in the platforms 
under the sea surface. It is a striking fact that these platforms are 
especially broad wdiere the shore is built up of Tertiary rocks, vis. off 
the north-east coast of Prince Charles Foreland, north and south of 
Michael Sars Point, at Ferrier Harbour, and at Poole Point. This seems 
to indicate that the platforms are not built up of loose material to any 
great extent. The Tertiary rocks have obviously been less resistant to 
the shore erosion than the rocks of the Hecla Hook system. 



192 



IRIDTIOF NANSEN. M.-N. Kl. 



m 



l'ig. 141. Axel l^lanil. i\u^Li^)l 31, 191;^. 

West Coast of Spitsbergen south of Ice Fjord. 

Aloni^' the west coast between Ice l-'jord and Ik-ll Souivl, there is 
a well (levelopefl enierj^ed stranrlflat cut in rocks of the Hecla Hook 
svstem. The suhmerg-ed strandfiat is also broad in this rej,^ion, as much 
as I 1 kilometres, with rlcpths between lo and 20 metres and the sharply 
defined ed^e not much deejjcr than 20 metres. The breadth of the 
emerj^ed and submerj^ed strandfiat t(,\^ether is alx.ait 14 kilomètres. 16 to 
25 kilometres outside this coast is the Sentinelle I'ank, the hii^diest part 
of which has depths less than 30 metres. 

Axel Island in Bell Sounrl has a sharply defined level surface, less 
than 20 metres high, cut in rocks of the Carboniferous system (J'^ig". 141). 
On the north side of Lowe Sound there is an extensive low plain which 
to some extent is a real strandfiat cut in solid rock, but a part of it is 
the floor of the wifle (JndiejK^ \'alle}' which has been filled u;) with 
sediment. 

Along the coast southwards from Cape Lyell, in Bell .Sound, there is 
a well developed emerg-ed as well as a submerged strandfiat. The latter is 
3 to 5 kilometres broad, and off Torell Glacier even 9 kilometres brc ad. 
It is mostly less than 20 metres below sea-level. 

On the north side of the entrance to Horn Sound there is practically 
no emerg-ed strandfiat because, as W. Werenskiold tells me. the coast is 
here built up of gabbro. 

South of Horn Sound there is a well developed emerg-ed strandfiat 
which has been surveyed by A. Hoel and W. Werenskiold during their 
expeditions of recent years. This strandfiat is especially extensive at the 
southern end of the land. 

As the soundings taken by the Hoel Expeditions during the last 
years have shown, there is also a broad flat submerged strandfiat outside 
the coast in this region. Tt is near sea-level with depths to a great 
extent less than 12 and 15 metres and carries many shoals and rocks. 
It is 9 to 10 kilometres broad, and at South Cape and east of it even 
12 kilometres broad. The sea outside is not very deep and the sea-bottom 
slopes fairly gently, especially on the south-western side, to depths of 70 
to 120 metres without any very distinct erlge. On th? south-eastern side 
the slope is somewhat steeper. 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. I93 




Fig. 142. Shore clirt" and strandtlat at Cape Guissez, with Mount Grimaldi. August 28, 1912. 

Cape Guissez in Cross Bay. 

Along- the shore east of Mount Grimaldi, on the east side of the 
entrance to Cross Bay, northwards from Cape Guissez, there is a vertical 
cliff of solid rock (see Fig. 142) which, according to Hoel's statement 
[1914, p. 26], is about 20 metres high. The plain rises gently inland from 
the edge of this cliff to the foot of the mountain where, according to 
Hoel, it attains an altitude of about 35 metres above the sea. It is here 
covered by more recent terraces of gravel, and the surface of the strand- 
flat, cut in solid rock, is lower. Hoel estimates it to be about 25 metres 
above the sea. 

Outside this shore, north and south of Cape Guissez, there is a 
submerged platform, half a kilometre to one kilometre broad, or even 
more, with depths less than 20 metres [cf. Hoel, 19 14, p. 40, Pi. XX\', il. 



Cape Mitre Peninsula and the North-western Part of Spitsbergen. 

On the south and west side of Cape Mitre Peninsula, west of Cross 
Bay, there is a very flat and well developed emerged strandfiat. the 
Dieset Plain ("Dieset Sletten", map Fig. 143), as has already been pointed 
out by Hoel. It is 2 to 4 kilometres broad, and rises very gently from 
the low shore to the base of the abruptly and steeply ascending mountains 
(Fig. 144), at about 25 to 30 metres above sea-level. There are a good 
many shallow lakes on this strandfiat. Towards the south it is continued 
as a submerged platform, 3 to 4 kilometres broad and with depths less 
than 20 metres. As regards the occurrence of the emerged strand flat 
there is a striking difference between this peninsula and Prince Charles 
Foreland. Here the strandfiat is well developed along the west coast, 
while there is no strandfiat along the east coast of the peninsula, in Cross 
Bay, except near Ebeltoft Harbour ("Ebeltoft Havn", Fig. 143). The 
reason why there is no strandfiat along the coast north of Ebeltoft Harbour 
is obviously that after the development of the strandfiat, Lilliehöök Fjord 

Vid.-Selsk Skrifter. I. M.-N. Kl. 1921. No. 11. 13 



T94 



FKIirrjDF NANSF.N. 



M.-N. Kl. 




Fig. 143. Cross Bay and the Cape Mitre Peninsula, based on the maps by Gunnar Isachsen. 

The horizontal liatcliiiig indicates the strandflat. Isobaths are drawn for 50, 100, 200, and 

300 metres of depth. The contours on land for every 50 metres. [From Nansen 1920]. 



and Cross Bay ("Kross Fjorfl") have been much deepened and excavated 
bv glacial erosion, which has cut away the strandflat and steepened the 
mountain sides. It is. however, noteworthy tliat here too there may b» 
observed a certain tendency towards formation of cir(|uc glaciers on the 
eastern side of the mountain ridges which is especially conspicuous on 
King Håkon Peninsula ("Kong Håkon's Haho". !-~ig. 143). 

The strandflat of the Cape IMitre Peninsula is cut in schists of the 
Hecla Hook system, which are fairlv easily rlisintegrated by frost. But to 
the north of this peninsula the coast is l)uilt up of very resistant granites 



1 92 1. No. II. 



THE STRANDFLAT AND ISOSTASV. 



195 




Fig. 144. Strandtlat on the west side of the Cape Mitre Peninsula. The Dieset Plain 
iDieset Sletten 1 and Dieset Lake (Dieset Sjol. [From Nansen 1920]. 

and gneiss-granites which fact causes an abrupt and striking change in 
the development of the strandfiat, as has been pointerl out by Hoel. 
In this granite region the emerged strandfiat is, as a rule, hardly more 
than a hundred or a few hundred metres broad. On both sides of Ham- 
l)urg Bav, south of Magdalena Bay, it is 300 to 400 metres broad. Tn 
some places in this region the emerged strandfiat is missing altogether. 
The submerged strandfiat also seems to be poorly developed along a great 
part of this coast, but there are too few soundings to trace its extent. 

On Danes Island and Amsterdam Island, built up of granite, the 
emerged strandfiat is also narrow or missing. P^ig. 145 shows the strand- 
flat along the north-east coast of Danes Island. It is in this region less 
even than in the region of the Hecla Hook rocks to the south, and it is 
more like the strandfiat in the granite regions of Norway. 

The low level plain, the Hollaender Ness, on the east side of Amster- 
dam Island (Fig. 146) is chiefly formed of loose material (sand and 
moraine material), but in its inner part near the foot of the mountain 
there is obviously rocky ground near the surface (Fig. 147). 

Off these islands and also north of the islands X'ogelsang, Cloven 
Cliff, and Norway Islands, there is a submerged strandfiat with skerries 
and rocks and with depths of less than 30 metres. Its extent cannot be 
traced, as the soundings are too few. North of Norwav Islands it seems 




Fig. 145. The coast east of Virgo Harbour on north-eastern side of Danes Island. 

August 23, 1912. 



196 KKIDTJOF NAN.SKX. M.-N. Kl. 






i»^^cSi^:-.^Äu.:-^. 




Fig. 146. The east side of Amsterdam Island, with the low Holiaender Ness. 
August I, 1912. [From Nansen 1920]. 

to have a width of 9 kilometres. On the whole the sea-bottom seems 
to he uneven in this region, as we have generally found it in regions of 
irranitc and resistant igneous rocks. 



Reindeer Land. 

Almost the wdiole of Reindeer Land forms a very level strandfiat 
which is about 18 kilometres across (Fig. 148). I have only been ashore 
on its east coast 5 or 6 kilometres north of its south-eastern corner. 
There were in some places low vertical shore cliffs, with a flat beach in 
front, while in other places the rocks sloped more gently down to the shore 
(see Fig. 149). 

The general level of the flat gently undulating surface of the land 
(Figs. 150 and 151 ) is between 15 and 20 metres above the sea. Low 
lidges rise above this level to about 26 metres. The highest ridge inside 
our anchorage on the east coast, was 27.5 metres above sea-level. 
Most ridges seen inland had aliout the same height, but some ridges 
further in to the north of us, were higher, and may have attained altitudes 
of about 50 metres. Bare rock was seen in the surface in many places 
especially on the ridges. 

Near the northern point of the peninsula, near Welcome Point, there 
is a solitary hill rising to 98 metres above the sea, as determined by the 
Isachsen Expedition 1909 — 1910. 

The whole peninsula is built up of Devonian schists. \\\ its south- 
western part there are higher mountains consisting of the same kind of 
rocks. According to what I was told by Hoel, the plain extends almost 
horizontally to the foot of these mountains, which rise abruptly, with a 
sharply defined line of demarkation between the plain and the mountain 
sides. 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY. 



197 




Fi£ 



147- 



The Hollaender Ness on the south-eastern side of Amsterdam Island. 
August 22, 1912. iFrom Nansen 1920]. 



The Peninsula between Liefde Bay and Wijde Bay. 

On the east side of Liefde Bay, where 1 was ashore, perhaps 12 kilo- 
metres south of Grey Hook, there is a sk)ping platform cut in solid rock 
(see Fig-. 152), l)Ut its upper level is very much higher than the strandfiat 
of Reindeer Land. PVom a vertical shore-cliff, about 10 or 11 metres 
high, backing a flat l)cach, there is a relatively steep ascent to a sharply 
marked shore-line cut in solid rock 41 metres above sea-level. It is 
marked by a horizontal series of small accumulations of snow (see Fig. 152) 
and is backed by a vertical shore-cliff about 6 metres high. From the 
edge of this cliff, 47.25 metres above sea-level, there is agam an ascent 
to 62.25 nietres above the sea, where a relatively broad plateau extends 
southwards along the foot of the steep mountain side (see Fig. 153). The 
foot of the steep scree along the inner margin of this plateau stands about 
4 to 8 metres higher, i. c. 66 to 70 metres above sea-level. On the plateau 
ihere were indications of a shore-line marked by a series of snow- 
accumulations (see Fig. 153, along the middle of the plain, towards the 
mountain slope) . 

The surface of this plain consisted of loose material. The highest 
level at which solid rock was observed was that of the shore-cliff at about 
47 metres above the sea. This may be the actual height of the rock plateau, 
which may, however, rise gentlv to a few feet more under the layer of 
loose material at the foot of the mountain side. 

To the north of this place, where we made our measurements, there 
was a lake (see Fig. 154) lying at a lower level. On the north side of 



lyB IKlI/ljOF NANSrCN. M.-N. Kl. 



Fig. 148. l'anoramic view of the cast coast of Reindeer 

lliis lake ami ils i-utlcl, the jilain under tlie stee]) nuniiilain-side, ascenfler] 
lo a liii^lier le\i'l iliaii tlial of our ])lateau, perhaps 30 metres liigher, givirg 
an altitude of al)out y)u metres above sea-level (see Fig. 154 to the rightj. 

The ri\er runninj,'- out of the lake has formed a canyon through the 
broad moraine terrace which has daimiud up the lake (see Fig. 154). This 
moraine has a considerable thickness as is pro\ed by this canyon, and is 
built up of coarse material, with numerous boulders embedded in its mass, 
as is seen in the sides of the canyon. The surface of the moraine is levelled 
by the sea to form a horizontal terrace at about !("> metres above sea-level. 

The well-marked coast-ledge of this region extends along the coast 
northwards to Grey Hook where there is a well-developed strandfiat in 
front of the very steep mountain side (see Fig. 150). The coast-ledge 
continues from Grey Hook south-eastwards along the west coast of 
Wijde Bay. 

I was ashore in Lake Valley, on the west side of Wijde Bay. The 
lake in this valley is a big lagoon to which a narrow channel leads from 
the sea through the shore terrace built up of drift material. The channel 
leads first into a smaller lagoon, and then a longer distance into the big 
lagoon. Strong tidal currents run through this narrow channel like a river. 

I took the height of the strandfiat by levelling. 

The ground rises from the plain near the shore (see Fig. 155) inland 
to a height of 46.8 metres where there is a well-marked shore-line partly 
cut in solid rock. Its exact level was difficult to determine as it was 
covered by too much gravel and loose niaterial. This shore-line forms a 
conspicuous horizontal line, marked bv accumulations of snow, along the 
coast southwards. 

From this shore-line an extensive plain rises gently inland towards 
a terrace built up of stones at the foot of the much oversteepened mountain 
side (see Fig. 156). The base of the terrace is about 55.8 metres above 
the sea, and its upper level surface about 64.8 metres (Fig. 157). This 
stone ledge is ol)viously a formation similar to those observed on Prince 
Charles Foreland, and has been formed in the same manner. Similar 
terraces were seen northwards along the coast towards Grey Hook, ap- 
proximately at the same height at the foot of the mountains, along the 
inner margin of the coast platform. 

On the north side of the big lagoon of Lake \^alley, the platform 
appeared to lie, to a great extent, at about the same level as the well- 





iq2l. 


No. I 1 . 




THE STRAXDFLAT AXD ISOSTASY. 




199 














1 


^^■1 




^■■1 


■1 




W^^^^^^^^^M 


■■■ 


^ "* ^ 








JL— — '. iU-LLil . . 1^:^ . ,„.- , . _jr 











from our anchorage in Liefde Ray. August 8, 1912. 

markcil slure-line (lcscril)C(l above, at al)out 47 metres above the sea. It 
rises i^ently inland towards the foot of the mountain. 

Hence we may assume that the upp^r level of the platform cut in 
solid rock lies about 47 metres above the sea in the reijion of Lake \'alley, 
and is probablv a few metres higher at the foot of the mountain-side 

This is exactly the height of the coast platform, which we found on 
the west side of this peninsula, on the east coast of Liefde Bay. But 
it is considerably higher than the observed heights of the strandfiat on 
Reindeer Land, and on Prince Charles Forelaufl. 

The ([uestion is then whether this coast platform may be considered 
as actualls' belonging to the strandfiat, or whether it is rather a shore- 
ledge similar to the raised beaches and shore-lines, formed during some 
more temporary depression of the land. On the one hanrl it seems to be 
too high to be a part of the regular strandflat, on the other hand 
it seems difficult to assume that a platform as broad as this has been 
formed f luring temporary depressions of tlie lanrl. Lately Hoel [1922] 
and \\'erenskiold [1920 , however, have found some most remarkable 
broad ledges or plateaus, cut in solid rock, on the west coast near the 




Fig. 149. East coast of Reindeer Land. August 8, 191 2. [From Nansen, 1920I 



KU IDT JOK NANSKN. 



M.-X. Kl. 




Fig. 150. On the plain near the east coast of Reindeer Land, with the "Veslemoy" at 
anchor. The crows nest of the latter is 13 metres above the water. As it is very nearly 
level with the horizon, the plain at this place is about that height above sea-level. The 
land at Grey Hook with the strandOat outside is seen in the back ground. Augusts, 1912. 

southern end of Spitsbergen. They must have been former! during periods 
of temporar}' sul)mergence of the land, for they are at various heights, 
the highest plateau being as much as 340 metres above sea-level. They are 
as much as some hundred metres broad, and must have been formed after 
the land was covered by glaciers at that place, for there were many pebbles 
and round shore boulders on the highest plateau. 

This seems to prove that, owing to the vigorous shore-erosion in this 
region, fairly broad shore benches or shelves may be cut in relatively short 
time where the rock is not verv resistant. 

There is thus a possibility that these high platforms on the west and 
east coast of the peninsula between Lief de Bay and Wijde Bay, are raised 
coast platforms of the same nature as the raised beaches, and are not a 
part of the regular strandfiat. 

If this be correct, we may assume that the real strandfiat on both 
sides of this peninsula has been more or less cut awav bv the big glaciers 
which filled Wijde Bay and Liefde Bay (and Wood Bavi and have 
deepened and excavated these fjords after the strandfiat had been 
developed. 



East Coast of Wijde Bay, and Verleegen Hook. 

There is a well developed strandfiat along the east coa.^t of Wijde 
Bay, which was observed from the region of Aldert Dirkse's Bay and 
northward to \'erleegen Hook (Figs. 158 and 160). I did not land on 
this coast, but as far as I could see from the sea, the strandfiat is cut in 



1921. No. 1 1. 



THF STRAXOFLAT AND ISOSTASV 




Fig. 151. The plain of Reindeer Land some kilometres from its east coast. August 8, 1912. 

.solid rock, and is to no considerable extent covered by drift or moraines. 
In many places the surface of the bare rock slopes gently down to the 
shore, while in other places there is a low shore-cliff, with accumulations 
of snow on the beach at its foot. 

The rock consists here chiefly of crystalline schists (mica-schists, 
hornblende-schists, gneisses) and granites, and the strandfiat once formed, 
has been relatively well preserved. The land on the west side of Wijde 
Bay is built up of sedimentary rocks of the Devonian system which have 




Fig. 152. Platform on the east sitle of Liefde Bay. August 9, 19 12. [From Nansen, 1920]. 



FKIOTJOK NAN.SFvN. 



M.-N. Kl. 




i' 'o- 1 53- ihe level upper suiiace ol 11),- jjlaliurni of the east coast ol" Lic-ide Hay. 

August 9, 1 9 12. 

obviously less power of resistance to the frost erosion tlian the crystalline 
rocks of the east side of the fjord. 

'^^rhe onl\' niamier in w liicli 1 can explain wliy. in spite of this, the 
strandfiat is so much broader aufl lower on the east side, is that the 
glaciers have cut away the Devonian rocks more easily than the crystalline 
schists and granites, and that during the deepening of the fjord during 
late glacial periods, after the strandfiat was formed more or less, the 
western side of the fjord has been more excavated than its eastern side. 

I did not land at Grey Hook, but, seen from the sea, it looked as 
if the strandfiat is wide and fairly low in that region, which has been 
less exposed to the erosion of the glaciers of \\ ijde Bay as well as 
of Liefde Bay. 

On ]\'iiccgcn Hook 1 was ashore. The strandfiat is very wide and 
low, cut in mica-schist anrl hornblende-schist, its surface being to a great 
extent formed of bare rock which is in manv places scoured anrl rounded 
by glacial erosion (Fig. 159). At the outer edge of this plane a cliff much 
disintegrated by frost was observed inside a flat somewhat raised beach, 
near the present shore (see Fig. 15). 




1 i.y. 15^. I'anorain 



ic view 



uf the sjreat terraces of moraine material outside the 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 203 

The low rocky ridges of this flat plain attained an altitude of i6 
metres above sea-level. Further inland there was a slightly lower plain 
covered with gravel and pebbles. 

Near the shore, at an altitude of about 9 metres, was a broad flat 
plane in the rock, a raised shore-line several hundred metres long. 

Small shore ridges of wave-washed pebbles were seen in several 
places up to the highest level of 16 metres. 

The strandfiat extended inland, rising slightly, to the foot of the 
oversteepened mountain sides (Fig. 160). 

About 3 kilometres north of \'erleegen Hook I found the depth 
to be only 30 metres, indicating that the strandfiat continues as a sub- 
merged platform some distance seawards. About 5 kilometres north of 
the Hook I found a depth of 48 metres, but then it suddenly dropped to 
152 metres and deeper further north. 

North-west Coast of North East Land. 

Great Stone Island, and Low Island outside, on the north-west side 
of North East Land, are low and seem to form a well-developed strand- 
flat cut in rocks of the Hecla Hook system. I have not. however, landed 
in this region. From this coast a wide submerged platform extends about 
27 kilometres (15 naut. miles) north-westwards, with depths chiefly about 
15 to 25 metres. The outer edge may possibly be at a depth of al)out 25 
or 30 metres, where the bottom begins to slope more steeply towards 
depths greater than 100 metres. In about So*' 25' N. Lat.. at a distance 
of about 35 kilometres from North East Land, I sounded 30 metres. 
This wide platform mav probablv be considered as a part of the strandfiat. 

The Fjords. 

In the inner parts of the fjords of Spitsbergen, the strandfiat is not 
so well developed as along the outer coasts, but there are nevertheless 
traces of a strandfiat in many places, indicating that the shore erosion 




le east coast of Liefde Ba\\ The lake is seen to the right. August 9, 19 12. 



204 



F RI DT I OK NANSKN. 



M.-N. Kl. 




^'S- 155- Strandllat in Lake Valley, view southwards along the west coast of Wijde Bay. 

August lo, 1912. 

has been at work there too, although not so effectively and perhaps during 
a shorter time than along the outer and more exposed coast. 

On the north side of Lowe Sound (Nan Mijen's Bay) there is an ex- 
tensive flat plain corresponding to the strand flat as was mentioned above 
(p. 192). 

In Ice Fjord there are ol)vious strandfiats in a good manv places, 
c. g. Cape Erdmann, Cape Boheman, Cape Wocrn, Cape Thordsen, Bjona 
Haven in Sassen Bay (according to Peach), Advent Point west of Advent 
Bay, &c. 

As was pointed out by Bertil Hogbom [19 14] there are also narrow 
submarine platforms outside the shore in several places in Ice Fjord, and 
they seem to be best developed where the coast is most exposed, c. g. west 
of Advent Bay, and off Cape Thordsen. \\'est of Cape Boheman there is 
also a submarine platform with several rocks, the Twin Rocks Szc. 



The Age and Genesis of the Strandflat of Spitsbergen. 

The strandflat of Spitsbergen is obviously a relatively young form- 
ation. As was previously mentioned (p. 179), and was already pointed out 
by Hoel [1914, p. 2/], its present very level surface has not been exposed 
to the erosion of the glaciers of the Great Ice Age, or Ages, which covered 
the whole land and have deepened and excavated the fjords, and have 
sculptured the land surface. The present plane or planes of the strandflat 
have, therefore, been developed after that time. 

On the other hand there has been a considerable advance of the 
glaciers after their formation, as is proved by the erratic blocks, mentioned 
bv Hoel and Peach, and bv the moraine material found on the surface 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY. 




Fig. 156. The level surface of the platform in Lake Valley, with a terrace of big stones at 
the foot of the steep mountain. August 10, 1912. [From Nansen, 1920]. 



of the Strandfiat. Its rocks are also scoured and rounded, e. g. at \"er- 
leegen Hook, and striæ have been observed, as mentioned bv Peach. 

This strandfiat niust, therefore, have been planed to its present levels 
after the Great Ice Age, or Ages, and before the last great advance of 
the glaciers of Spitsbergen. 

In Chapt. \'I (pp. 47f.) I have pointed out that the strandfiat of 
Norway has obviously been developed during periods when the land crust 
had attained its natural level of ei|uilibriuni, anrl the shore-line remained 
stable during a long time. i. c. during periods when the land was not 
depressed by the weight of any ice-caps. At the same time there must 
have been a severe climate favouring the shore erosion bv frost. I there- 
fore consider it probable that the Norwegian strandfiat has to a great 
extent been formed during periods with a severe climate preceding each 
glacial period, and before the land had been depressed by the inland ice. 

In Spitsbergen the conditions are different. The land is still covered 
with glaciers to a very great extent, and we cannot say what the natural 
level of ef|uilil:)rium of the crust may actually be. At the same time the 
climate of that region is so severe that probably, even during warmer 
interglacial periods, it was sufficiently cold for an active shore erosion 
by frost. 

Nevertheless it is a striking fact that the emerged as well as the 
submerged strandfiat of Spitsbergen have levels that are very similar 
to those of the Norwegian strandfiat, and we may assume that they 
indicate levels of an approximate natural equilibrium of the crust in 
this region. 

I imagine that the Spitsbergen strandfiat was planed to its present 
levels during interglacial periods when the shores were not covered by 
glaciers and the ice covering of the land was similar to what it is now. 



2o6 



FRIDTJOF NANSFN. 



M.-N. Kl. 








I 



1 i,<- ',-1 /• 



moramic \ii., .. ,,,c coast platform from the top of the stone t<: 



^A ' .^^ — isr- 




Fie;. 1^8. Panoramic view of the strandflat along the coast south ol Mossel B: 



During some part of tlie time when the strandflat was formed th2 
glaciers had less extent than at present. As pointed out by Hoel [iÇiJ, 
p. 2/] this is especially proved l)y the Buchanan (Placiers and the Murray 
Glacier now extending over the strandflat on the east coast of Prince 
Charles Foreland. 

It has been already mentioned, that two levels of the strandflat were 
probablv observed on Prince Charles Foreland, one at about 6 to lo 
metres and one at about 20 to 30 metres above the sea. 

The lower level is probably the same as that of the low strandflat 
at A>rleegen Hook, where it was between 10 and t6 metres above the sea. 

The higher level of 20 to 30 metres is generally found on the strand- 
flat along the whole west coast of Spitsbergen, and also predominates 
on Reindeer Land. 

It was pointed out before (p. 200) that the higher shelves, at about 
47 metres or more above the sea, observed on the west and east side of 
the peninsula between Lief de Bay and Wijde Bay cannot really belong- 
to the strandflat, but must have been formed during some more temporary 
submergence of the coast. According to Peach the altitude of the rocky 
surface of the strandflat at the base of the mountains should be 150 feet 
(46 metres) above the sea, and he states it to be even 200 feet (61 metres) 
in some places (see above p. 184). His statements seem, however, to be 
rough estimates, and not based upon very accurate investigations. 



1 92 T. No. I T, 



THE STRANDFLAT AND ISOSTASY. 



207 




•e Fig. 150I at the toot of the mountain side in Lake \'^lk;, . August 10, ign 




e land near Verleegen Hook is seen to the left. August 10, 191 2. 

Raised Shore-lines. 

Raised shore-lines and beaches fretiuently occur along the coasts 
of Spitsbergen, and have been described especially by A. Hoel [1914, 
pp. 28 ff.j and Peach [19 16, pp. 297 ff.l. They have obviously Ijeen formed 
after the strandflat and are relatively recent formations. 

At the place where I landed on the east side of the entrance to 
Liefde Fjord (sec al)ove p. 197) I found shore-lines cut in solid rock at 
the follmv in"" liei^hts abf>ve sea-level: 




iig. 159. .SlraiiiiilaL cut m lioi uulciiiic-bcliibL ami iiuca-scliisl oil Vciiccgcn llouk. 

August 12, 19:2. 



208 



IKIDTJOF NANSrCN. 



M.-X. Kl. 











Fig-. 160. Th.- Iruid north of Mossil Bay to Vcrlcegcn Hook, will) tlic stramJtlat along the shore and 




Fig. 161. The north side of Brogger Peninsula, with the conspicuous flat platform (about 200 met: 

right, with the mountain on Prince Charles Forelå 

1 1 metres. Tliis shore-line continued northwards along the slope of 
the big moraine terrace in front of the lake mentioned p. 198. 

27 metres. A fairly distinct shore-line in solid rock. 

41 metres. A sharply defined shore-line marked by a horizontal 
row of small accumulations of snow, and backed by a vertical cliff. 

47.25 metres. The upper horizontal edge of the above mentioned cliff. 

In Lake Valley, on the west side of Wijde Bay, shore-lines were ob- 
served at the following heights: 

II metres — quite distinct. 

24 metres — less distinct. 

30.8 metres — sharply marked. 

46.8 metres — partly in solid rock, very conspicuous along the coast 
as a horizontal line marked by accumulations of snow. 

At Vcrlccgcn Hook a shore-line was observed at about 10 metres 
above sea-level. 

Platforms at high Levels. 

In several places along the coasts of Spitsbergen one may observe 
traces of horizontal planes at relatively high levels. Especially conspicuous 
is the remarkable, broad plane on the Brogger Peninsula south-east of 
Ouade Hook (Fig.i6T). It extends north-westwards from Mount Schetelig, 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY. 



2og 




fill III ' -"'''I' I \h '"''^' if/ '" ^iiin 



]f''miip'' 



the old flat surface ipeneplain» at the top of the mountains. August lo, 1912. [From Nansen, 1920]. 




above sea-level) in front of ScheteUg Mountain. The low strandfiat at Quade Hook is .seen to the 
behind. July 31, 1912. [From Nansen 1920]. 



is nearly 2 kilometres wide and is bounded along its outer edge by a vertical 
cliff more than 150 metres high. According to Hoel's statements and the 
measurements of the Isachsen Expedition 1909 — 1910, this remarkably 
flat plane rises gently inland from the top of the cliff, at a height of 
about 200 metres above the sea. or somewhat less, and the plane attains 
a height of about 240 or 250 metres at the foot of the mountain, where 
the boundary between the plane and the steep mountain slope is to a 
great extent covered by a glacier. 

As proved by O. Holtedahl's and A. Hoel's investigations this plane 
cuts the strata of the Carboniferous system of which the land is composed, 
and, as Hoel maintains, it is obviously a plane of marine denudation. 
It has been formerl in a similar manner as the strandfiat at lower levels, 
chiefly by shore-erosion by frost, assisted by the wave-action. It is hardly 
conceivable that this very level plane, giving the impression of being a 
quite recent formation, can have been exposed to any considerable glacial 
erosion, for in that case it could not have preserved its level surface. 
The probability is, therefore, that it has been formed at some period after 
the time when this region was covered by the inland-ice of the Great Ice 
Age. It has, however, been formed before the lower strandfiat of between 
10 and 30 metres above the sea. This younger strandfiat has a fairly great 
width at Ouade Hook and has obviously been cut in under the older and 

Vid.-Selsk. Skrifter. I. M.-X. Kl. 1921. No. 11. 14 



2TO 



r-RINTIOK NANSF.N. M.-N. Kl. 



liij^licr plalforni. Tims tlic lii^-'li \crtical cliff nt llie c(]^t of the latter was 
formell. 

Traces of a similar plain of marine ^emulation liave also been ob- 
ser\(.'<l hy I loel north of l'"J)eltofl llarhour in Cros.s P^ay, hut it was there 
hint;- at a lower level of about 150 metres above the sea. 

( )n the east side of Amsterdam Island I observed indications of what 
may be an ancient platform which 1 estimated to lie at a level somewhat 
hit^hcr than 150 metres (Vi^. i4^>). On the southern Norway Island there 
is also a plateau in a similar height (see Fig-. 8). But as these platforms 
are cut in more resistant rocks (granite and gneiss-granite) they may 
be older. 

Inside the low and level strandfiat at \'erleegen Hook the mountain 
rises to a very flat plane with a similar altitude. As, however, the inner 
margin of this plane is not marked by any rleclivity, and as there is a quite 
gradual transition from this plane to the general mountain plateau inland, 
without any marked difference, it may here most probably be a part of 
the raised ancient level mountain surface, that extends southward over 
all the land east of Wijdc Bay. 

As was mentioned above (p. 199) A. Hoel [1922] and W.Werenskiold 
[1920] have found several horizontal shelves or plateaus at various high 
levels along the coast in southern Spitsbergen near South Cape. There is 
one shelf at about 40 metres above sea-level, another at about 80 to 100 
metres. This platform is especially well developed. It is in some places 
several hundred metres broad, and is widely extended along the south coast 
of the land, inside the low plain of the strandfiat, which is there very broad 
(see Fig. 134). Farther north towards Horn Sound this platform is per- 
haps somewhat lower, about 70 metres. 

There is one shelf, somewhat imeven, at 220 metres, and then a 
fairly broad and conspicuous one at 340 metres above the sea, on which 
Hoel observed many pebbles and shore boulders. 

It is indeed strange that shore formation of so recent date, that the 
pebbles have not even been removed by glaciers, occur at such high alti- 
tudes so near the southern end of the land. 

It shows on the one hand that there has been very great vertical 
movements of the crust in this region relatively recently, and on the other 
hand that the shore erosion must be so effective in this region, that in 
the rocks of little resistance broad shore platforms can be cut in relatively 
short time during temporary depressions of the land. 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 211 



XIII. THE STRANDFLAT ALONG COASTS INSIDE 
AND OUTSIDE THE ARCTIC REGIONS. 

The Coasts of Siberia, Greenland, and Alaska. 

Along the north coast of Siberia there is a well-developed emerged 
strandfiat, which I have previously described [1904, pp. 20 ff., PI. III]. 
On the Eastern Taimur Peninsula and the Chelyuskin Peninsula, between 
the mouth of Taimur River and Chatanga Bay, it forms a very conspicuous 
level foreland backed by mountains rising abruptly above its plane [see 
1904, PI. Ill, Figs. I — 4]. Its height, near the coast, was estimated to be 
less than 30 metres above the sea, it has a considerable width, and forms 
a continuous level plain along most part of the coast, extending hori- 
zontally to the foot of the mountains. It is cut in solid rock, but is to 
a great extent covered with drift material, which may in some places 
attain considerable thickness. The coast between Taimur Bay (the Norden- 
skiöld Archipelago) and the mouth of Yenisei River is low, and as there 
are few mountains or hills near the sea, it is difficult to say whether 
the whole of this low land is actually a strandfiat which has been cut 
horizontally by shore erosion. We must in that case assume that most 
initial hills surmounting the plane of this strandfiat have been planed 
down. As this coast land and the islands outside are so very flat [cf. 1904, 
Pi. Ill, Figs. 7 — 10, PI. I\', Fig. i], and considering that there is a quite 
similar low coastland with low islands to the north forming a foreland 
or regular strandfiat in front of the steep mountain sides, it seems to me 
to be probable that this land too has been levelled by shore erosion, and 
that it is actually a strandfiat. 

The sea along the north coast of Siberia is very shallow; often 30 or 
40 kilometres or more from the coast, it has depths less than 40 and 50 
metres. The soundings taken in this region are, however, much too few 
to make any study of the topography of the sea-bottom possible. It has 
to be considered that there is an exceptionally broad continental shelf, 
with depths less than 100 metres, extending a great distance north from 
the Siberian coast, in the region of the New Siberian Islands even as much 
as 600 or 700 kilometres, and that the whole of this sea is therefore very 
shallow. As it is especially shallow outside the mouths of the great rivers. 



FRIDTJOF NANSEN. M.-X. Kl. 



Yenisei, Lena, ^c, it seems innltaMc lliat it is to some ^reat extent fillef] 
up witli ri\(T sediment, l-ul on the oiIki- lianrl, as small rocky skerries 
rise ab(j\e the sea in many |)laces aUm^ the coast, and as I also ol^served 
sunken rocks near or in the water-surface, it is obvicjus that the sea- 
l)(^tt(jm consists of solid rock to a j4"reat extent, and it seems probaljle that 
there is a siih)iicr^^i'ü rocky straiulflat which is very widely extended in 
some places, and the plane of which \\c> at levels less than 30 metres 
below the sea-surface. 

In the region of Yiigor Strait anrl l^ii^^arh the land is low and flat. 
The plain is to a great extent cut in solid rock as is especially clearly seen 
along the shores of the strait and on X'aigach [cf. Nansen, 1904, p. 22]. 

Along the coasts of Novaya Zemlya tliere is in many places an un- 
usually well-developed strandfiat forming a flat foreland in front of the 
steep mountain sides ascending al)ruptly above its plane. The low anrl 
flat Goose Land seems, for instance, to be a broarl stranrlflat. On the 
Holtedahl Expedition to Novaya Zemlya in 192 1 the emerged strandfiat 
was studied and its levels taken in several places, and we may look for- 
ward to an interesting report on these investigations. 

The soundings taken along the coasts of Novaya Zemlya indicate 
that there is a submerged straiulflat. but they are too few to tell much 
about its topography and extent. 

On the whole it is striking that along all these Arctic coasts, we 
find a well developed strandfiat, which often, especially along the north 
coast of Siberia, forms extensive plains. It indicates that in these regions 
there has been a verv vigorous shore erosion, obviously due to the cold 
climate in these northern latitudes, which has caused an active dis- 
integration by frost of the rocks of the shores even during the warm 
interglacial periods. 

Well developed strandfiats obviously also occur along other coasts 
of the Arctic region. 

Along the west coast of (ircoiland there is a strand flat very similar 
to the Norwegian one, with a belt outside the coast of numerous low 
islands and skerries [cf. Nansen, 1904, pp. 90 f.j. It does not seem, 
however, to be as well developed as along the coast of Norway. 

Along the east coast of (Greenland I have found no certain evidence 
of the existence of a strand flat near present sea-level. This is what 
might be expected, as Greenland is still covered by an ice-cap which 
extends to the outer coast along most part of the east coast south of 
68" N. Lat. The jirobability is. therefore, that this coast is still much 
submerged by the weight of this ice-cap. If there is a strandfiat it may 
be at some depth below sea-level. 

Along the west coast of Greenland south of 68** N. Lat. the conditions 
are different. The margin of the inland ice is a great distance from the 
outer coast, and there is a broad coast land which is not covered bv the 



1 92 1. No. II. THE STRAXDFLAT AND ISOSTASY. 21 3 

ice-cap. It is therefore possible that tlie outer part of this coast land 
mav have risen towards its natural level of equilibrium, and even that, 
by the pressure of the ice-cap over the inner land, the outer coast may 
have been upheaved above its natural isostatic level, although this is 
hardly probable. 

These are interesting questions which have to be settled by special 
investigations on the spot. 

On the islands of the Arctic Arcliipclago north of Canada, there is 
obviously also a strandfiat. 

On the islands along the coast of Alaska, there is a strand flat 
forming a flat foreland between the base of the steep mountains and 
the descent to deep water. 

Gilbert [1904, pp. 130 ff. and 17g] mentions this low foreland in 
several places and gives some very illustrative pictures of it (see his 
Figs. 64, 65, 85, and his Pl.XMI) showing that it is obviously of the 
same type as the Norwegian strandfiat. Gilbert, however, explains 
it as a preglacial peneplain, or base-levelled plain, in the same manner 
as Ahlmann has lately adopted. This peneplain formed originally a more 
continuous foreland, which has afterwards been dissected by erosion, 
forming deep channels now separating the islands. These channels have 
to a great extent been formed even by fluvial erosion "at least 500 feet 
(150 metres), and probably 1,000 feet (300 metres) or more, below the 
present sea-surface" [1904, p. 136I. It seems to me difficult to under- 
stand how a plain of this kind could possibly have survived a fluvial and 
glacial erosion which has cut channels through it, now 1,500 to 1,700 feet 
(450 to 520 metres) deep. Gilbert has also to assume that "the glacial 
degradation must have been locally quite moderate, or the general plain 
character would not have survived" [1904, p. 131]. 

In my opinion there can be little doubt but that this low foreland 
in Alaska is a formation of the same nature as the Norwegian strandfiat, 
and has been formed in the same manner by shore erosion, during cold 
interglacial periods, and during the beginning of glacial periods, before 
the land began to sink, and before the coasts were covered bv glaciers. 

The heights of the Alaskan strandfiat seem to be similar to those of 
the Norwegian one. Gilbert states that on the islands along the coast 
of Kadiak "the height ranges from about 100 feet (30 metres) to sea- 
level" [1904, p. 179'. On the Annette Island, in the Alexander Archi- 
pelago, he says, however, that "there is a general and gradual ascent 
from the sea front to the mountain base, where the altitude may be 
three or four hundred feet" (90 or 120 metres'). This description does 
not, however, agree with the impression given by his illustrations, which 
show a low fairly level plane, and I think, therefore, that this high alti- 
tude of the mountain base is not the general one. 



214 



)• KID I lot' NANSKN. 



M.-N. Kl. 




Fig. 162. Low land along the east coast ot' the southern portion of Mainland. 
I The coast northwards from Sumburgh Head, seen from the north-weast. II Southward 
view along the coast from the sea off Mousa. Ill View northwards along the coast from 
the sea off Mousa. IV View southwards along the coast from the sea off Helli Ness. 

V View northwards through Bressay Sound from Lerwick. August 23, 191 1. 



Shetland Islands. 

As I considered it to be of interest to study the strandfiat on an ex- 
posed island group far out in the ocean, I sailed across to Shetland for a 
few days in August 191 1. The stay was, however, too short for a real 
survey, and I could only expect to get a general impression of the topo- 
graphy of the coast. I sailed along the whole east coast of Mainland and 
nearly to its north end, and I landed at several places. I also crossed the 
island from Lerwick to Weisdale Voe on the west coast. 

The Shetland Islands have a very complicated geological structure. 
In the southern portion of Mainland the clay-slate series, with associated 
limestones and quartzites, prevails. Its eastern sea-board from Lerwick 
southwards is skirted by the lower Old Red Sandstone (Devonian). The 
north-western half of Mainland, as well as the islands ^^'halsev. Yell, and 



1 92 1. No. II. 



THE STR.WDFLAT AND ISOSTASY. 




Fig. 163. The Nab south of Lerwick, the west side of Bressay iti the background. 



the western sea-board of Unst consist chiefly of micaceous and horn- 
blendic gneiss with limestones and quartzites. In Unst and Fetlar. the 
north-eastern islands, there are large masses of serpentine and gabbro. 
In Delting and Northmavine, in the north-western portion of Mainland 
a large area is occupied by diorite. The islands Foula, far out to sea 
toward the south-west, and Bressay, east of Lerwick, are built up of the 
lower Old Red Sandstone, and so is the greater part of Walls, in the 
western portion of Mainland. 

As these rocks, building up the islands, ha\'e to a great extent con- 
siderably less power of resistance to erosion than the rocks of the west 
coast of Norway in the same latitude, it is natural that the islands have 
an aspect very different from the Norwegian coast. They are on the whole 
low with rounded forms, while the shores are cut back by recent marine 
erosion, forming cliffs and many isolated vertical rocks, the so-called 
"drongs", in the sea outside. 

But in spite of these conspicuous marks of a recent shore erosion, 
no ([uite convincing evidence of the existence of a real emerged strandfiat 
could be discovered along the shores. Many low islands occur, and low 
flat plains extend inland from the shore in many places in Mainland. 
But nowhere did I ol)serve a sharplv defined boundary between the coastal 
plains and the hills rising above them. 

It seems to me, however, to be probable that the low land, about 
20 to 30 metres above sea-level, extending along the southern east coast 
of Mainland, northwards from Sumburgh Head (see Fig. 162, I — IV) is 
actually an old strandfiat. Inside this low land the hills rise more steeply, 
but as a rule not abruptly with a sharply marked boundary. In some 
places especially on the points, there may be a more marked difference 
between the flat shore land and the rising hill side (cf. Fig. 161, \s the 
distant low point on the west side of the sound north of Lerwick, see also 



2i6 KKiD'i.ioi- N.\.\sr:,\. M.-N. Kl. 



^^^^^^ 










■■ 














J 



Fig. 164. 1 Lunna Holm and the Islands west of Lunna Ness. II Out Skerries with the 
Lighthouse. The height of tiie highest islands is 52 metres above the sea. August 26, 191 1. 

Fig. I (^"3. the west side of Bressay in the background). On some of the 
islands similar formations are. noticeable, a low outer border and hij^^her 
ridges inside (see Fig. 164). 

In Yell Sound at Ollaberry Bay, on the east coast of Xorthmavine 
(in the northernmost portion of Mainland), I found, by levelling, the plane 
of the supposed strandfiat to be about 33 metres above the sea. The is- 
lands Bigga and Samphrey in the middle of the sound are 33 metres and 
30 metres above sea-level. 

If this low- land may actually be considered to be a regular strandfiat 
of the same kind as the Norwegian strandfiat, it seems probable that the 
slopes of the hills inside have been much changed by subacrial denudation 
and perhaps by glacial erosion after the formation of the strandfiat. 
During the periods when the strandfiat was developed, the shore erosion 
by frost was probably not very effective in this region, so near the warm 
Atlantic Current, where the climate was probably not very cold. Hence 
the strandfiat was not cut very broad, and as its formation took a long 
time, the surface of the land was at the same time to a comparatively 
large extent affected by the subaërial denudation. 

There is no well-developed submerged sirandflat along the shores of 
the Shetland Islands. Along a great part of the coast the sea-bottom 
slopes gradually from the shore down to depths of 100 metres and more, 
without any distinct break or well marked edge. At many places there 
are, however, indications of a submerged narrow platform, less than 15 
or 20 metres below sea-level. They may be as much as i kilometre broad 
and are bounded by a somewhat steeper slope outside. 

Considering the exposed situation of these islands to a stormy sea, 
and the comparatively small power of resistance of their rocks to erosion, 
it may seem remarkable that there is not a better developed emerged or 
submerged strandfiat along their coasts. 

It proves: — on the one hand, that the wave action alone has not much 
power to form a strandfiat. and on the other hand, th.at there cannot 
have been much shore erosion bv frost for its formation in this region. 



I92I.X0. II. THE STRANDFLAT AND ISOSTASY. 217 



XIV. THE LEVELS OF THE STRANDFLAT AND THEIR 

DEVELOPEMENT. 

We now propose to discuss the possible changes in the level of the 
shore-line and the relation between the two similar formations, the strand- 
flat standing- at levels near or slightly above present sea-level, and the 
much broader continental shelf standing at levels one or two hundred 
metres below present sea-level. Let us first, however, summarize the 
conclusions on the nature of the strandfiat to which our investigations 
have led us. 

Summary of the Results of our Investigations of the Strandflat. 

W'e have found: 

On the one hand that, during the long periods required for the form- 
ation of the strandflat, the shore-line has maintained very uniform levels 
along the extensive coasts where a strandflat occurs, /. c. especially in 
Arctic and Subarctic regions. 

On the other hand that after the formation of the now emerged 
strandflat began, there has been a negative shift of the shore-line caused 
either by an upheaval of the coasts or by a lowering of the sea-level. 
These changes of level have been very uniform over extensive areas. 

On the whole, the strandflat creates the impression of having been 
formed during long periods when the earth's crust in those regions stood 
at its natural level of equilil)rium; and since then there has been no ap- 
preciable tilting of this level on the whole, although there may have been 
small local changes in the course of time. 

Our investigations show that there are probably several levels of the 
strandflat: the emerged strandflat has probably in most places two levels, 
and the submerged strandflat represents at least one level. 

The results of our investigations as regards these levels may be sum- 
marized as follows: 

The emerged Strandflat. 

Along the west coast of southern Norway, in the regions of Stav- 
anger, Karmoi, Hardanger Fjord, Sotra, Radoi, Lindås, and Sogne Fjord, 
the emerged strandflat seems to have two levels: An upper level at be- 



2l8 FRIDTJOF NANSEN. M.-X. KI. 

twccn 30 and 40 metres, or in s(jine places {perhaps between 26 anrl 35 
metres ahfne the sea — and a Unver level at about 15 to 18 metres above 
the sea. In most places in these regions the upper level comprises the 
greater part of the area of the strandfiat. The lower level extends near 
the shore, is more even, and is obviously of younger age than the upper 
level. Jn the region of Stavanger, this lower level is of comparatively 
wide extent, especially where the rocks are phyllite and chlorite schists, &c. 
which have comparatively little power of resistance to erosion. 

In tlic region of Smolen, IJitteren, and Froia (Fig. 107) the emerged 
strandfiat has probably also two levels: an upper level about or higher 
than 30 metres above the sea, and a lower level lower than 20 metres 
above the sea. The lower level is here of a comparatively greater extent 
llian in tlic above mentioned region of the west coast. It compri.ses the 
most part of Smolen, and a great part of Froia, and is also widely extended 
along the north coast of Hitteren. 

Along the coast of Helgeland the lower level of the emerged strand- 
flat comprises the greater part of its area, especially in regions where 
rocks with comparatively little power of resistance to erosion prevail. 
On Donna and Heroi it forms very even planes at altitudes of 8 to 10 
meters above the sea, and in some parts of the islands even lower. This 
low plane of about 10 metres seems to be of very wide extent on the 
many islands of Helgeland. 

The upper level at about 30 metres, or between 30 and 40 if it exists, 
is at any rate of very small extent in the region of Helgeland (cf. the maps 
Figs. 1 13 and 114). 

On Lofoten and Vesterålen the emerged strandfiat has probably also 
two levels, and the upper level, perhaps at about 30 metres, seems to be 
of a greater extent, c. g. on Hasseloi, than in Helgeland. The lower level 
is well marked on the small islands along the coast (see Figs, i and 2). 



The siibiiierged Strandfiat of Norzvay. 

Along the west coast of southern Norway south of Sogne Fjord 
there is not much of a submerged strandfiat. There are numerous shoals 
sunken rocks here and there, at somewhat varying depth, but there are no 
well defined platforms with horizontal planes. 

Along the coast between Sogne Fjord and Stat, there are more indi- 
cations of a submerged strandfiat in the shape of small submerged plat- 
forms and groups of shoals and sunken rocks often far out to sea. The 
soundings given in the charts of this coast are not sufficiently numerous 
for a study of the shape and topography of these platforms. Their depths 
seem to vary somewhat, but are to a great extent less than 10 metres 
below sea-level. 



IQ2I. No. II. THE STRAVDFLAT AND I^OSTASV. 



219 



Along the Sondmor and Romsdal coast the submerged strandfiat is 
considerably wider and more developed, forming platforms extending 
far out to sea. Outside the islands Fjertoft and Haroi. north-east of Åle- 
sund, these platforms extend as much as 8 or 10 kilometres seawards from 
the outer coast of the islands, and have depths to a great extent less than 
10 metres below sea-level. Their surface is somewhat uneven and dis- 
sected by channels and depressions, and their edges do not seem to be 
very sharply defined. 

The submerged strandfiat is similar outside the coast at Hustad. 
It is especially well developed in the region of Smolen. Froia. and the 
Froan Islands where it extends 12 to 20 kilometres, or more, seawards 
from the outer coast of the big islands, and has a level surface, less than 
10 metres below sea-level, and well defined edges. It is. however, dis- 
sected by numerous channels. This strandfiat is cut in pressed igneous 
rocks and granites, which, however, seem to have no very great power of 
resistance to erosion by frost. The submerged strandfiat of this region 
has, on the whole, a greater area than the emerged strandfiat. 

Along the coast to the north-east the submerged strandfiat is well 
developed, and its surface topography is largely similar to the one just 
described. 

Along the coast of Helgeland the submerged strandflat is of ex- 
ceptionally wide extent. The area of the emerged strandflat is quite small 
in comparison. The surface of this submerged strandflat is. as a rule, 
very level, almost perfectly horizontal, and less than 10 metres below the 
sea-surface, largely in fact between 2 and 6 metres. Its edges are to a 
great extent sharply defined with steeper side slopes. Where the rocks 
are mica-schist, limestone, and weak gneiss the surface of this submerged 
strandflat is more level and the edges more sharply defined, than where 
the rocks are granite or other fairly resistant igneous rocks. 

In Lofoten and \'esteràlen the submerged strandflat is not broad, 
obviously owing to the great power of resistance of the rocks to erosion, 
and also owing to the considerable initial height of the land. The sub- 
merged strandflat of this region has, however, as a rule a much greater 
width than the emerged strandflat. Its depth is as a rule greater than 
those of the submerged strandflat of Helgeland, its surface is less hori- 
zontal, and the edges less marked. 

Along the coast to the north-east there is also a well developed sub- 
merged strandflat at about the same depth, as far as Ringvasoi and Kvaloi. 

The three Levels of the Strandflat. 

We may thus assume that the strandflat of Norway has at least three 
different levels: An upper level which in southern Norway is mostly be- 
tween 30 and 40 metres above the sea (in some regions, as for instance 



220 FF^IDTJOF NANSEN. M.-N. Kl. 

on Storfl, ]Krlia|)s soniiuliat lii.L;licr lliaii in olhcr rej^ionsj. J ii iiortlicrn 
Norway, especially in J lelj^elaml, this level is ]K-rlia])S somewhat lower, 
and is less conspiciujus. 

A lower level which in southern Norway is about 15 to iS metres 
above the sea (in the inner j)art of Sogne i^'jrjrfJ about 10 anrl 12 metres) 
and in Plelgeland in ncjrthern Norway is somewhat lower, about 8 to 10 
metres (or even 5 metres), and is there of wide extent. 

A submerged level which is not niucli developed along the coast of 
southern Norway, is much more developed along the coast of Søndmør, 
Romsdal, and Nordmor coast, and is of a very wide extent along the coast 
of Helgeland. Wherever this level is well developed it has a depth of 
only some few metres, and less than 10 metres, l)e!ow present sea-level. 

These different levels of the strandfiat obviously indicate that the 
shore-line has stood at different levels during the long periods when the 
strandfiat was developed. We must assume either tliat the land has risen 
or that the sea-level has been lowered. 

The inner part of the emerged strandfiat (at the upper level) has 
obviously been formed at some earlier period than the lower part of the 
strandfiat (at the lower level). When the submerged level was formed 
is more difficult to decide. We have seen that in Helgeland where this 
level is especially well developed and is of very wide extent, it is only 
some few inetres below the lower level of the emerged strandfiat, and it 
might almost seem doubtful whether, in this region, it has not been 
formed more or less during the same long period as the latter. 

In Spitsbergen we have found a well developed emerged strandfiat 
with heights above sea-level very similar to those of the Norwegian strand- 
flat, and there is also a distinct submerged strandfiat with a depth of less 
than 20 metres. 

On Bear Island there seems to be a wider range between the upper 
limit of the emerged strandfiat and the level of the submerged strandfiat, 
but this may be due to special reasons as has been previously mentioned. 
In other regions, especially Arctic, where there is a well developed strand- 
flat, the investigations are not sufficiently detailed to give definite in- 
formation about the heights of the levels, but they seem to be somewhat 
similar to the heights we have found in Norway and in Spitsbergen. 

Causes of the Changes in the Level of the Strandflat. 

It is, in my opinion, probable that the different levels of the strand- 
flat in Norway indicate different interglacial periods of its formation, 
and mark the levels of equilibrium of the land crust during each of these 
periods. 

Owing to the great quantities of rock and the débris of the inter- 
glacial erosion carried away from the land-surface into the sea by the big 



1 92 1. No. II. THE STRANDFLAT A\D ISOSTASY. 221 

glaciers, the land has been somewhat raised by isostatic movement to the 
new level of equilibrium after the disappearance of the ice cap of each 
glacial period. 

It is not known how many Pleistocene glacial periods there mav 
have been in Norway. But considering that at least four different glacial 
periods are now established for Central Europe, it seems hardly probable 
that there should have been less in Scandinavia. 

If the strandfiat has two distinct levels it seems to indicate that there 
have been at least three glacial periods in Norway as was already assumed 
by Oxaal [1914, pp.42 f., cf. above p. 48]. If there are actually three 
different levels, it may indicate four glacial periods. 

If we assume that the changes of level are solely due to isostatic 
movements of the earth's crust, and in no degree to changes in the sea- 
level (caused by changes in the volume of the Ocean), it may seem diffi- 
cult to understand why the heights of the raised strandfiat are so very 
similar along the Norwegian coast and even on Spitsbergen, although the 
quantities of rock removed may probably have varied much in the divers 
regions of the coast. 

If we assume that the upper limit of the strandfiat is universallv at 
some height between 30 and 40 metres above the sea, we would then have 
to assume that the quantity of rock material removed from the surface 
of the coast land corresponds to a continuous layer of rock with an aver- 
age thickness of at least 36 to 48 metres, which has been removed after 
the upper level of the strandfiat was first developed. This seems a great 
deal. Although much thicker layers of rock have obviously been excavated 
from the deepened valleys and fjords during this long period, the thickness 
of rock removed from the high land surface, between the deep vallevs and 
fjords, has probably not been very consideraljle. It is also striking that 
the level surface of the strandfiat itself cannot have been much denuded. 
On the other hand, if there have been at least two glacial periods, 
and probably three, after the upper level of the emerged strandfiat was 
developed, it is not inconceivable that the weight of the rock material 
removed by the glacial erosion of the coast land during these periods 
may go some way towards explaining the elevation of the strandfiat above 
present sea-level. 

The submerged strandfiat offers, however, another difficulty to this 
assumption. If, during some period before the last glacial epoch, this 
part of the strandfiat was cut at a level actually lower, though slightly 
lower, than the present sea-level, the level of the land cannot have been 
permanently raised after that time, unless the general sea-level has also 
been raised. 

To me it seems probable that, in addition to the isostatic upheaval 
of the land, there have been chancres in the level of the sea. 



222 FKIDIjOK NANSEN. M.-X. Kl. 

There arc several processes conlinually at work causiiij^- changes in 
the Y(jlume of the Ocean. The continual emission of water vapour from 
the volcanos a^lds new water to the Ocean, while on the other hand water 
is substracted from it I)y the alteration of rocks which hinds water chemi- 
cally. These processes are slow, however, and the one more or less checks 
the fjther. Tt is, therefore, harrlly probable that, during- the periods we 
are here consiflerinJ^^ they have prcjducerl changes of sufficient importance 
to account for the (jl)served shifts of the shore-line after the formation 
of the strandfiat. 

The displacement of sea-water caused by a volcanic eruption on tlie 
sea-floor would probably only cause a temporary rise of the general sea- 
level, as the effect would ])robably be more or less readjusted bv the 
isostatic movements of the crust under the Ocean. Crustal movements 
changing the size or depth of the Ocean basin would naturally also cause 
changes in the general sea-level, Ijut as these movements would be finally 
controled by isostasy, it is hardly probable that they can have produced 
appreciable changes during the periods we are discussing. The deposition 
of sediment on the sea-floor may gradually raise tlie general sea-level, as 
will be mentioned later; but this is also a slow process and is partly 
checked by crustal sinking. 

Changes in the position of the poles would cause changes in sea-level. 
But if the lowering of sea-level since the first formation of the strandfiat 
should be thus explained, we would have to assume, that the pole has come 
nearer to the regions of the strandfiat, which would involve the probability 
that the climate has become more arctic since that time, which is exactly 
contrary to what might be expected. Changes in the earth's centre of 
gravity might also change the sea-level, but is probably so slow a process 
that it would not help much to explain the changes under consideration. 
The same is also the case with possible changes in the earth's rotation. 

The subtraction of water from, and the addition of water to, the 
Ocean caused by the formation and melting of the great present and 
Pleistocene ice-caps have probably produced the most considerable changes 
in sea-level during comparatively short periods. 

Much water is now^ accumulated in the extensive ice-caps of the 
Antarctic and of Greenland. 

According to Hess, the area of the Antarctic ice-cap is about 13 mil- 
lions square kilometres and that of Greenland 1.9 million sq. km. If we 
add to this the glaciers of the rest of the world, we get an ice-covered 
area of about 15,156,000 sq. km. altogether. We do not know what the 
average thickness of these ice-caps may be. The ice-caps of Greenland 
and the Antarctic with their general level surfaces standing at altitudes of 
2,000 and 3,000 metres and more above the sea, must obviously have quite 
considerable thicknesses. If we assume the average thickness of the ice- 
caps to be 600 metres, the melting of them would increase the average 



ICSI. No. I]. THE STRANDFLAT AND ISOSTASY. 223 

, 15.156 

depth of the Ocean by an amount ot — ■ ooo >< 0.9 = 22.7 metres, 

361.1 

the area of the Ocean being about 361,100,000 sq. km. and the specific 

gravity of the glacial ice about 0.9. 

If the average thickness of the ice-caps be 1,000 metres, the increase 
of the Ocean depth caused by their melting would be 37.8 metres. 

The final rise of the general sea-level caused by this increase of the 
average depth of the Ocean, would be somewhat reduced by the gradual 
crustal sinking of the sea-floor under the increased weight of the oceanic 
water masses. If the specific gravity of the plastic rock or magma 
underlying the crust be 3, the depression (it) of the sea-floor will be 

d 

It ^ — 

3 

where d is the increase of the average depth of the Ocean. Hence, if the 
increase of the Ocean depth be 22.7 metres, the rise of sea-level will be 
1 5. 1 metres. It will be 25.2 metres if the increase of the Ocean depth be 
37.8 metres. This is provided that the crustal sinking of the sea-floor is 
entirely compensated for by the upheaval of the areas which were covered 
by the ice-caps. If not, as will generally be the case, the sinking of the sea- 
floor will be somewhat reduced, as it will then have to be compensated for 
by the upheaval of the continents surrounding the Ocean (cf. pp. 235 ff.). 

We see that the probable changes in the general sea-level caused by 
the reduction of the existing ice-caps would be quite sufficient to explain 
the negative shift of the shore-line after the first formation of the strand- 
flat, if we add the probable isostatic upheaval of the land caused by the 
removal from its surface of glacial and interglacial waste. 

The question is, however, whether we can assume that all ice-caps 
had disappeared when the strandfiat was formed. The probability is that 
during long preglacial periods, there have been no ice-caps of significance, 
and that during those periods the Ocean had what we might call its normal 
average depth, which was somewhat greater than the present one. 

During the warm interglacial periods it is also probable that the ice- 
caps of the earth may have been essentially smaller than they are now. 
The difficulty is. however, that the strandfiat has obviously not been 
developed during these warm interglacial periods, but during periods with 
a cold climate favouring a vigorous shore erosion by frost. And these 
periods may have preceded the glacial periods, or have formed their be- 
ginning, before the crust was pressed down by the weight of the accu- 
mulating ice. 

The development may also have continued some time after this sub- 
sidence of the land began, as the sea-level sank simultaneously because of 
the subtraction of water from the Ocean by the increasing accumulation 
of water in the ice-caps on land. But it should be noticed that this sub- 



224 FRIDTJOF NANSEN. M.-N. Kl. 

sidence was not unifonn, it Ix-iii^' ^^rcatcr iiilaml than iicar<-r ilie coast, 
and in \]]v outer ccjastal rej^'icjn thcrf nia\ jirobably have been an upheaval 
uf the land flurinj^'' the first part of the suhsirlcnce of the inner land. 

Hence it is not prohahic that tlie strand flat was developed to any 
larj^e extent during this pcriot] wlicii the laiul surface was warped, a 
characteristic feature of the strandfiat heinj^- that its levels are practically 
horizontal. 

Howe\'er this may be, it seems i)rol)al)le tliat tlie j^eneral sea-level 
has sunk somewhat since the strandfiat beji^an to he devehjped, and it is 
possible that changes in the thickness and extent of existing ice-caps may 
at least to a large extent have caused these changes in the sea-level. 

As long as there are ice-caps which during some periods increase 
and during others decrease, the average depth of the Ocean and the general 
sea-level will not remain c|uite stable. 

It might then be |)ossil)le that the jMX'sent submerged strandfiat was 
planed during some long period when the ice-caps of the earth were a 
good deal greater than now, and that although there has been an isostatic 
upheaval of the land since that time, the level of this lowest part of the 
strandfiat has not been raised above the present level of the sea surface. 

As will be mentioned in the next chapter there is, however, weighty 
evidence to show that the lower level of the emerged strandfiat was formed 
just before the last glacial period. If therefore the plane of the submerged 
strandfiat represents a different stage, we would in that case have to 
assume that it was formed before the lower level of the emerged strand- 
flat, although this would be somewhat difficult to understand. 

By the assumption that, during the long time which has elapsed since 
the strandfiat began to be formed, the shore-line has been shifted partly 
l)y isostatic upheaval of the land and partly l)v changes in the general level 
of the sea, we obtain a simple explanation of the reason for there not 
being greater differences in the heights of the strandfiat in the divers 
regions. 

The explanation of the fact that the lower level of the emerged 
strandfiat seems to stand somewhat lower along the coast of Helgeland 
and in the inner part of Sogne Fjord than along the outer west coast of 
southern Norway, in the region of Sogne Fjord and south of it. may 
then be, either that a smaller quantity of rock have been removed from 
the land surface in these regions after the formation of this level, — or 
that the isostatic upheaval of the land has not yet been quite completed 
in these regions. 

The former explanation seems hardly probable, considering that the 
rocks of Helgeland are not on the whole very resistant. It seems less 
improbable that the upheaval of the land after the last glacial period is 
not yet finished in these regions which were considerably depressed 
during the last glacial epoch. The inner region of Sogne Fjord was de- 



igsi. No. II. THE STRANDFLAT AND ISOSTASV. 225 

pressed more than loo metres and was probably covered b}' big glaciers 
comparatively recently, and the extensive Justedal glacier still remains, 
weighing down the land to some extent. In the region of Helgeland 
where the outer coast was depressed more than 75 metres, there mav 
also have been a glacial covering comparatively recently, remnants of 
which still exist in the big "Svartisen" glacier. 

Another fact which nia\- also have retarded the upheaval of the crust 
in this region, is that the extensive submerged strandfiat was covered 
during the depression by a layer of water as much as 75 metres deep or 
even more near its inner margin, and the weight of this water made the 
upheaval of the land slower. It may, therefore, not seem improbable that 
in Helgeland, as well as in inner Sogne Fjord, the upheaval of the land 
may still be a few metres short of its full coniylciion, while it is long ago 
completed along the outer west coa-f ." -nuthern Norway as well as in 
Lofoten and \'esterålen where the late-glacial depression was verv small 
or nothing at all, and where there was no extensive area of submerged 
oea-floor outside that should be elevated. 

The Nature of the Rocks and the Topography of the Strandflat 
and the Continental Shelf. 

Our investigations have shown that the .shape and development of 
the strandflat vary to some extent with the nature of the rocks. This is 
especially conspicuous in the topographv of the submerged strandflat of 
Norway. 

\\'here the rocks possess comparatively less power of resistance to 
erosion, the submerged strandflat has a very level surface forming a 
nearly horizontal plane only some few metres below present sea-level, 
and its edges are sharply defined at about the same depth. Its side slopes 
are often steep descending to considerable depths on the continental shelf 
outside. 

Where the rocks possess a comparatively great power of resistance 
to erosion, the surface of the submerged strandflat is considerably more 
uneven and irregular. It is generally sloping seawards, its edges are less 
sharply defined, and its depth may vary a great deal. Its outer side slopes 
are as a rule not steep, and the depth of the continental shelf outside not 
considerable. 

If this be compared with the surface of the continental shelf of 
Norway outside the strandflat, we finrl a striking difference. 

In regions of weaker rocks the continental shelf is broad, but is 
comparatively uneven and its ^^urface is to a great extent lying at a con- 
sideral)le depth below present sea-level, while, as we have seen, the sub- 
merged strandflat is extremely level in those regions, and is standing 
comparatively high, being near present sea-level. 

Vid.-Selsk. Skrifter. I. M.-X. Kl. 1921. No. 11. 15 



226 t KIDTJOF NANSKN. M.-X. Kl. 

hl llic rf,ij;-i()iis uf niorc rcsisiaiil rocks, c. g. outside llie coasts of 
Soiifliiior, konisdal, Nordmor, Lofoten, X'esteràlen, Senjen, &c., the con- 
liiieiilal shelf is comparati\ely narrow, its surface fairly level, and its 
depth below present sea-level to a j^rcat extent less than 150 metres, and 
in some reg'ions even less than io(j metres, while the siiitmerj^a-d strand- 
flat, as we liavc seen, is also narrow, hut is comparatively uneven and 
irrcg-ular, slopinj^ scawarrls, and lyinj^^ deeper helmv present sea-level than 
in the reg'ions of weaker rocks. There is often n<j very sharply defined 
boundary between the low strand flat anrl the hij^di inner jtart of the con- 
tinental shelf. 

Althouj^h tlie submerged strand flat is very broad outside Helgeland, 
its outer side slopes seem to have been cut back more and to have been 
made steeper in the regions of weaker rocks than in those of more 
resistant ones. Its very level surface in the former regions seems there- 
fore largely to be of a younger age than the more uneven surface of the 
latter regions. 

Let us now leave the strandfiat for a while anrl consider the form- 
ation of the continental shelf. 



1921. No. II. THE STR.AXDFLAT AXD ISOSTASY. 227 



XV. THE CONTINENTAL SHELF AND ITS 
FORMATION. 

In a previous treatise on the bathymetrical features of the northern 
seas [1904J I have described the continental shelf of Norway and of other 
northern regions and discussed its formation. There is this resemblance 
between the continental shelf and the strandfiat, that both of them form 
fairly horizontal planes, in front of the coast which in Norway and several 
other countries is steep and ascends abruptly from the inner margin of 
the planes. Both formations as far as they are cut in solid rock, may 
most naturally be assumed to have been planed largely by marine de- 
nudation of some kind. There is, however, this striking difference be- 
tween them, that the typical strandfiat is largely confined to cold regions, 
where there has been severe climates favouring shore erosion by frost, 
and it is only exceptional that strandfiats occur outside such regions, in 
places where there have been special climatic conditions, as for instance 
along the south-east coast of India — while the continental shelf has a 
universal distribution along most old coasts in all latitudes. 

The continental shelf is obviously to a large extent built up of waste 
from the land carried into the sea during long geological periods and 
deposited outside the coasts. Rut as I previously pointed out [1904] 
it is also to some extent cut in solid rock. It would, for instance, be 
extremely difficult to explain in any other way the striking fact that 
outside the Norwegian coast its shape and the height of its level is ob- 
viously to some extent dependent on the nature of the rocks of the coast. 
It is especially narrow and high where the rocks possess a comparatively 
great power of resistance to erosion, as along the Romsdal, Sondmor coast 
and the coast of Lofoten, A'esterålen and Senjen, while it is very broad 
and lies comparatively deep below sea-level outside Helgeland where the 
rocks on the whole possess much less power of resistance to erosion. 

The submerged fjords on the continental shelf, c. g. outside Søndmør, 
and outside Helgeland, and the coast north of Trondhjem, also indicate 
the rocky nature of the shelf. These fjords, which to a large extent follow 
the same direction as the valleys of the great Caledonian mountain folding 
on land, are obviously sculptured in stjlid rock [cf. Nansen, 1904, p. 151]. 



FRID'I.IOF NANSKN. M.-N. Kl. 



y\s a whole llie coni iiiciital sliclf is ohxiously a verv >>](] formation 
(Icxclopc'il during; loiij^' j^coloj^ical |KTio(ls, to a ^reat extent prej^'lacial, 
wliilc llic sliamlftat has been fIe\'eloped atter tlie first TJreat Ice Ag'e. 

/\ few points nia\ ]>v specially mentionerh 

Onr in\csli,i;ations of the Xorwej^-ian stranfjflat anr] the late-j^'-jacial 
sLibmcrj^ence of ihe laml (see nt-xl chapter) pro\e the almost p^-rfect 
isostacy (;f the crust. 

It seems then improbable that the initial laii'l in the region of the 
continental shelf can on tlie averag'c have been ^■e^y high. 

( )ntsi(le I lel,<4e!an(l tliis shelf has a width of 250 kihjmetrcs, vcrv nearlv 
e(|ual to thi- distance from the coast castwarrls to the central region of 
Fenno-.Scandia wliere the land had its greatest late-glacial submergence 
of 300 metres or more. Idiis shelf is consequently so broad that the crust 
in that region must stand very nearly at its level of equilibrium. 

If the average specific gravity at the crorled surface rocks be 2.6, 
that of tlie underlying ])lastic rock or magma, displacing the uplieaval {11) 
of the crust, be 3, tlie tliickness of rock eroded be //, and the permanent 
lowering by ere)sion of the level of the land surface be /, we have: 



MX3 = /1X2.6 
/ = h — n 



2.6 



/0 



7 



Hence m order to lower the average level of the land surface 100 
metres it would be necessary to remove a layer of rock having an average 
thickness of 750 metres. If the specific gravity of the underlying magma 
be higher, the thickness of the layer of reck which had to be removed 
would be smaller, c. g. with a specific gravity of the magma of 3.5, it 
would be only 389 metres. 

As the planing of a shelf in solid rock by the joint action of subaërial 
denudation and marine denudation is obviously a very slow process, 
especially if it is not vigorously assisted by frost disintegration, and as 
the continental shelf of Norway cannot be assumed to be of sufficiently 
old age to make a vertical denudation of very great dimensions probable, 
we may assume that the initial land in the region of the present continental 
shelf was fairly low. 

It may, however, have been much more uneven, and there may have 
been isolated parts or mountains rising to greater heights. The sub- 
merged lower parts, the depressions and valleys, of this initial land, may 
then in the course of time have been filled up with waste from the land, 
and the crust mav thus have been depressed to a certain degree. 

There is also the possibility that in the outer part of the shelf there 
is only a small amount of rocky ground, and that this part is chiefly 



I92I. No. II. THE STRAXDFLAT AND ISOSTASY. 229 

built up of waste. This accumulation of waste will have pressed the crust 
down to some great extent. On the deeper sea-floor outside there may 
probably also haye been formed thick layers of deposits, weighing down 
the crust. But why is there such a comparatively steep and sudden 
descent from the outer edge of the shelf on to the flatter sea-floor if the 
shelf and the sea-floor are Ijuilt up of the same kind of material? 

There is also another difficulty with which we are faced. The level 
surface of the continental shelf, whether cut in solid rock or built up of 
loose material, indicates an earlier sea-level or rather sea-levels, when 
either the land stood so much higher than now or the sea-level so much 
lower. 

In Xorway there is furthermore this difficulty that the height of 
tlie level of the shelf varies a great deal as was previously mentioned. 
Outside Romsdal and Sondmor as well as outside Lofoten the level of the 
continental shelf stands at a depth of lOO to 150 metres below sea-level, 
and outside A'esterålen and Sen j en the depth below sea-level of the very 
flat shelf is even less than 100 metres. Outside Helgeland the surface of 
the continental shelf sinks down to 300 and 400 metres below sea-level. 

Considering the nearly perfect isostasy of the earth's crust, which 
the crustal movements of Fenno-Scandia indicate, it seems extremely 
difficult to understand that, rluring the long period when the shelf was 
formed, the land as a whole can have stood so much higher as the depth 
below sea-leve! of the continental shelf: and when we consider that the 
other coasts of the globe must in that case also have been similarly 
elevated, the idea becomes quite impossil)le. 

If the continental shelves of the world actually indicate one or more 
earlier sea-levels we would therefore be bound to assume that there have 
l:)een considerable fluctuations in the level of the Ocean, and that during 
the periods when the continental shelves were developed it stood perhaps 
100 to 150 metres lower than now [cf. Xansen, 1904, pp. 200, 211 f.]. 



Fluctuations in Sea-Level caused by the Formation of Ice-Caps. 

I'uring the Ice Ages great ((uantitics of water were accumulated in 
the extensive ice-caps of the northern and southern hemispheres. Penck 
[i88r. p. 76, 1894. II pp. 528. 660] estimated that the abstraction of water 
from the Ocean caused by the glaciation of the northern hemisphere alone 
sank the general sea surface 67 to 71 metres. Assuming that the glaciations 
of the northern and southern hemispheres were simviltaneous, Drygalski 
[1887, p. 274^ calculated that the lowering of the general sea-level thus 
caused was about 150 metres. The present writer pointed out [1904 
pp. 211 f.j that the ice-cajxs of the northern hemisphere may probably 
during their widest extent have had a considerably greater area than 



230 KKIDTJOF XAN.SF.N. M.-N. Kl. 

assumed hy I't-nck ;iiii| tlic i^ciicral sca-lc\cl iiia\- |)ossiij|\- have been sunk 
as mucli as kk) niclrcs hy llicir format i'^ii. \\ il bf assume! that 
simultaneously there was a ^Teat cxlension of ice-caps in the southern 
bemispliere, aiul tlie antarctic ice-cap liar] a f^reater thickness, T thouj^-ht 
it "concci\ablc that tlie le\el of the Ocean was at times lowered as much 
as 200 metres or even more". 

Ren;-inald A. Daly [1910, 1915, \). 173] has estimated that by the 
nultinj;- of the I'leistocene ice-caps since their maximum development 
"the general sea-le\'el lias been raiserl by an amount ran^-ini,'' betwe.-n 
23 and 129 metres". He thinks that this minimum estimate is likely 
to be too small, while the maximum estimate is too large. He thinks 
a rise of sea-level of the order of 50 to 60 metres to be most probable. 

I think Daly's estimate is too low. It seems to me to be probable 
that during their maximum development the T'leistocene ice-caps had a 
greater average thickness and a greater extent than assumed even by 
Drygalski and Penck. 

The Thickness of the lee-eaps. 

During the last glacial epoch Fenno-Scandia was depressed l)y the 
weight of the ice-cap iirobably about 350 metres in its central region 
about Bottenviken (the Gulf of Bothnia). If the specific gravity of the 
plastic magma underlying the rigid crust be 3, the specific gravity of 
the ice 0.9, and the thickness of the ice-cap /i, we have: 

//X0.9 = 350 3 -^ M 
where M is a certain quantity due to the pressure counteracting the de- 
pression created by the upheaval of the peripheral areas surrounding the 
ice-caps, as will be mentionerl below. If for the moment we leave this 
quantity out of consideration we have: 

H = ^^ ^ = 1 167 metres. 

0.9 

This is, however, a minimum value. First, it is improbable that the 
land had been depressed to the full extent corresponding to the weight 
of the ice-cap before the latter began to decrease. 

Secondly, there is considerable evidence to prove that the depression 
of the areas covered by the ice-caps causeri an upheaval of the regions 
surrounding the ice-caps. This upheaval will gradually spread outwards 
from the ice-caps over wider and wider areas, and the elevation will thus 
be reduced again very slowly. It is, however, obvious that as long as it 
lasts, the weight of the elevated magma under these upheaved regions 
will counteract to some extent the depression of the ice-caps, and will 
reduce accordingly the depth to which the underlying crust can b? de 
pressed bv their weight. The wave of upheaval surrounding a depressed 
area will as it were temporarily raise the level of e(|uilibrium of this area 



IQ2I. No. II. THE STR.WDFLAT AND ISOSTASV. 23I 



by the height of the wave. It is, however, extremely difficult to determine 
what the height of this upheaval may have been at the time when the ice- 
caps began to decrease, or, more correctly, when the sinking of the de- 
pressed areas ceased. A great part of these surrounding areas are now 
under water. The probability is, however, that owing to the upheaval of 
these areas we have to add a considerable amount to the thickness of the 
ice-cap computed from the late-glacial submergence of the land below its 
present level. 

If the volume of the underlying magma be not altered by any chemical 
changes, it is cbvicus that the total volume of upheaval must l)e equal to 
the total volume of depression, without taking into consideration the 
elastic compression which in this connection is insignificant. If we 
assume that the area of the depressed land under an ice-cap is equal to 
one fourth of the area of the upheaved region surrounding the ics-cap, 
the average height (//) of upheaval must therefore be equal to one fourth 
of the average height (/;) of depression. As, in the case of equilibrium 
inside these regions, the pressure should be uniform at a certain level 
below the earth's surface, this means that 

?/><3 = d: o.g — /1x3 

where d is the average thickness of the ice-cap, 3 the specific gravity of 
the magma displaced by the depression, and 0.9 the specific gravity of 
the ice. As 

u = 0.25x7? 
we have 

0.9 

If for instance the average height (//» of the depression in the whole 
depressed area were 200 metres, the average thickness of the ice-cap would 
be 834 metres. If, however, the upheaved area were greater in proportion 
to the depressed area the thickness of the ice-cap would be less. 

Thirdly, there is a probability that the specific gravity of the rock 
or magma underlying the rigid crust, at depths of perhaps 120 kilometres 
below the earth's surface, is somewhat higher than 3. In that case the 
thickness of an ice-cap computed from its depression of the crust would 
have to be increased accordingly. 

According to what has been just pointed out, the probability is there- 
fore that the thickness of the ice-cap was a good deal greater than the 
amount computed from the submergence of the land. 

In the Christiania region, with a lateglacial submergence of 218 
metres, the corresponding minimum thickness of the ice-cap, computed 
as on p. 230. would be 727 metres, and in the Trondhjem region, with a 
submergence of about 200 metres it would be 667 metres. 



232 FRIDTJOF NANSFN. M.-.\. Kl. 

Tlic Tromlli jcm rcj^-^ion was very near lo the outer üiar^'in of the 
Fenno-Scandian ice-cap. In its eastern, south-eastern, and southern 
portions the ice-cap prohahly thinnerl off niucli more {.(radually. In the 
region of (iniland its averaj^^e thickness may lia\e Ijeen about 250 inetres 
as a niininiuni, l)ut was ])rohaIjly much more. 

If we estimate the averaj^fc tliickness of the Fenno-.Scandian ice-cap 
at its maximum rlevelfjpment durinj:;^ the last glacial period to have been 
at least ^«0 metres, we are certainly on tlie safe side. Probably it was 
a good rlcal more. 

This icc-caj) was, however, very small when compared with the north 
European ice-cap of the Great Ice Age during its maximum development, 
and the average thickness of that ice-ca[) was probably much greater. 

Near South Cape on Spitsbergen A. Hoel and \V. Werenskiold have 
recently found raised strand-lines and beaches at a height of as much as 
340 metres above present sea-level. A submergence of this magnitude 
would corresi)onfl to the weight of an ice-cap which was at least 11 30 
metres thicker than the present inland-ice of Spitsbergen and probably 
much more. The formula p. 231 gives 1418 metres. As this raised beach 
was observed near the southern end of the land, it means an enormous 
increase of its glaciation. 

On the top of Cape Flora, on the south coast of Franz Josef Land, 
at a height of about 330 metres above the sea. Dr. Reginald Koettlitz 
[1898] found water-worn pebbles which seemed to indicate a beach. 
A submergence of this height might indicate that this part of Franz Josef 
Land has recently been covered by an ice-cap which was at least iioo 
metres (or 1376 metres according to the formula p. 231) thicker than its 
present glacial covering. 

In Greenland and in Grinnell and Grant Land there are also numerous 
raised beaches and strand-lines proving a recent upheaval of the land and 
indicating that during the last glacial period the ice-cap has been con- 
siderably thicker than now; and it was more widely extended over the 
whole now bare land of the west coast of Greenland, as well as of its 
northern east coast. 

During the maximum development of the ice-caps of the Great 
Ice Age, they were much more widely extended in Europe as well as in 
AiTierica and Greenland, &;c., and also covered the broad continental 
shelves surrounding the coasts. For instance the whole of the shallow 
Barents Sea between Norway, Russia, Novaya Zemlya, Franz Josef Land, 
and Spitsbergen was probably covered by a thick ice-cap. The mean depth 
of this sea is perhaps 200 metres. I have moreover found indications of 
a glaciation in northern Siberia, and have also observed raised strand- 
lines on islands outside that coast, probably indicating a considerable 
lateglacial submergence of the coast land of those regions. 



1921. Xo. II. THE STRAXDFLAT AND ISOSTASY. 233 

In Xurth America the lateglacial submergence was at least 500 
metres, probably a great deal more. Computed in the same manner as 
above (by the formula p. 231) this would correspond to a thickness of 
the ice-cap of about 2085 metres. If, however, we do not take into 
account the upheaval of the zone peripheral to the ice-cap, we find ac- 
cording to the formula p. 230 that the thickness of the ice-cap mav have 
been 1667 metres only. But this is a minimum value, and it was probablv 
a great deal thicker. 

If we assume that the area of all Pleistocene ice-caps in the northern 
and southern hemisphere, during their maximum development was at least 
50 millions square kilometres, that their average thickness above sea-level 
was 1000 metres, and that the average thickness of the extended Antarctic 
and Greenland ice-caps were increased by a similar amount: if we further- 
more assume that the area of the Ocean be reduced by 4 per cent bv the 
ice-caps covering some parts of the continental shelves, and by the sinking 
of sea-level, we find that, by the formation of the Pleistocene ice-caps, 
the average depth of the Ocean was decreased by an amount of abcut 

50 

— -xiooo -c.g = 130 metres. 
346 

I consider this to be a minimum value, for the average thickness of 

the greater part of the ice-caps during their maximum development was 

probably more than 1000 metres. It is also probable that the total area 

of all Pleistocene ice-caps during their maximum development was more 

than 50 millions sq. km. Drygalsky [1887^ and Penck '1894 II, pp.528 

and 660^ have estimated the reduction of the Ocean by the formation of 

the Pleistocene ice-caps to have been equal to a water layer 150 metres 

thick. 

Shifts of Sca-lez'cl and Shore-line caused by Changes in the Volume 

of the Ocean. 

\\ e thus see that there must have been an appreciable reduction of 
the volume of the Ocean during the maximum development of the Pleisto- 
cene ice-caps, which would cause an immediate sinking of the general 
sea-level. 

By this shifting of water-masses, however, the equilibrium of the 
earth's crust is disturbed. What will be the result? In the course of time 
there will be a gradual and very slow readjustment. If perfect equilibrium 
be restored, the crust under the ice-caps will be depressed to an extent 
corresponding to the weight of the ice-caps, and the floor of the Ocean 
will be upheaved to an extent corresponding to the weight of the re- 
duction of the volume of the Ocean. 

If the specific gravity of the plastic rock or magma underlying 
the rigid crust be 3, the upheaval (//) of the general sea-flcor, caused by 
a reduction {d) of the average depth of the Ocean, would be: 



234 FRIDTJOF NANSEN. M.-X. Kl. 

(i 

U ~ — 

3 

'l'lic final lowerinj^ of the ji^eneral sea-level would in that case be 
al)(Ail iwo tliirds of the aiiKiunt hy which the averaj^e depth of the Ocean 
was reduced. 

The al)o\'e formula for // will only be valid if perfect equilibrium 
(jf the entire crust of the earth br restored. In that case the depression 
of the crust in the regions of the ice-caps would be entirely compensated 
for by the ui)lica\al of the general sea-floor, and no permanent changes 
in the general level (jf the continents w(juld then be caused. 

As, however, the shifting of the plastic rock or magma underlying 
the rigid crust is an extremely slow process, this state of perfect 
c(|uilibrium is hardl\- ever attainerl. \Vhat wouUl actuall\- liappen may, 
perhaps, be as follows: 

The immediate effect of the accumulation of water in the ice-caps 
will be a sinking of the general sea-level to the full amount by which the 
average depth of the Ocean was reduced. Then the isostatic readjustment 
of tlie crust would gradually ])egin. 

The crust under the ice-caps would be depressed by the increase of 
the weight of the ice-caps. This depression would cause a corresponding 
upheaval of the crust in the areas surrounrling the ice-caps. If these 
areas consist more or less of sea-floor, their upheaval will reduce the 
sinking of the general sea-level accordingly. 

The reduction of the depth of the water in the other parts of the 
Ocean will gradually cause an upheaval (//) of the sea-floor. This up- 
heaval, wdiich is not compensated for l)y the depression of the crust in the 
regions of the ice-caps, has to be compensated for l:)y the sinking of the 
continents surrounding the Ocean. The upheaval ;/ of the general sea- 
floor will then be: 

// = M 

3 

where M is a quantity determineii by the reduction of pressure at a 
certain level under the continents due to their sinking. 

The following consideration may help to make this clear. 

If the globe were covered entirely by an Ocean, the increase or de- 
crease of the volume (/. c. the depth) of this Ocean would have no effect 
upon the level of the sea-floor. If, however, continents emerge above the 
surface of the Ocean, the conditions are entirely altered. An increase of 
the volume of the Ocean will then press down the Ocean floor and press 
up the continents accordingly. 

If the average depth of the depression of the Ocean floor be ;/, the 
average height of the upheaval of the continents be //. the area of the 



ig2r. No. II. THE STRAXDFLAT AND ISOSTASV. 235 

Ocean being a])out 36 i.i millions square kilometres, and the area of the 
land surface of the earth being- about 148.8 millions sq. km., we have: 

// -36 1. 1 = // 148.8 
// = 2.43 II. 
To give room for a universal depression of the Ocean floor the 
continents have consequently to be elevated on the average 2.4 times the 
average amount of this depression. This reduces substantially the depth 
to wliich the Ocean floor can be depressed by a certain additional weight. 
If perfect equilibrium be established, the pressure at a certain level below 
the earth's surface should be uniform in all regions. Hence the increase 
of pressure at this level created !)y the increase of the depth of the Ocean 
must be equal to the increase of pressure caused by the upheaval of the 
continents. Hence 

dxi — WX3 = h>^2> = 2.43 >'//>' 3 

II ^= ^ a ■ o.og-] 

3 - 2.43 3 

where d is the increase of the depth of the Ocean. If this increase be 
100 metres, we find that the depression of the Ocean floor thereby caused 
can be no more than 9.7 metres, while the average upheaval of the con- 
tinents would be 23.6 metres. 

To simplify our computation we have here assumed that the volume 
of the Ocean was increased by water coming from outside the globe. In 
reality the water is removed from the continents to the Ocean, which 
complicates the matter. This removal causes an upheaval of the land in 
the region of the ice-caps, and a sinking of the crust in the surrounding 
regions, which will gradually extend over wider areas, but will only reach 
a certain part of the Ocean floor. 

In this manner the universal upheaval of the continents caused by 
the shift of the water from the ice-caps to the Ocean will be reduced, 
and the possible depression of the Ocean floor may be increased 
accordingly. 

On the other hand it would take an extremely long time before the 
depression of the Ocean floor could effect a universal and equal upheaval 
of the whole area of the continents. It would begin by pressing up th.e 
coasts surrounding the Ocean, and the upheaval would only very slowly 
extend landwards over the continents. Thus the depression of the sea- 
floor would also be much checked. 

The whole process is thus seen to be a very complicated one, and 
as so many important factors are unknown, it would be useless to try 
to form any accurate estimate of the depression of the Ocean floor and 
of the upheaval of the continental coasts. 

\\'e have here considered the case where there was an addition of 
water to the sea l\v the melting of the ice-caps on land. In the opposite 



236 FRIDTJOF NANSEN. M.-N, Kl. 

case of an ahstractinn nf water frum llic sca 1j\' llic fonnatioti of ice-caps 
oil land, the same compulation can he mafic, only in the inverse direction. 

It lias to he noticed tliat as the continental coasts are to some extent 
raised or sunk hy the crustal movements, the iiej^^ative or positive shifts of 
shore-hne will not corresj)ond to the clianges of the j^cneral sea-level, hut 
will as a rule he somewhat smaller. 

For instance, hy a rlecrease of the avera.i^'e dejjth of the Ocean, the 
sea-floor will he raised, whicli will reduce the sinking- of the general sea- 
level accordiuLjl}-; hut the continental coasts will sink somewhat, which 
will reduce the neg"ati\e sliift of tlie shore-line causeri hv the sinking of 
the general sea-level. 

These reductions, however, will not he considerahle. 

T.et us take an example; let us assume that the thickness of the Ant- 
arctic ice-cap is increaserl on the average 1000 metres. As the area of 
the ice-cap is ahout 13 millions square kilometres, this would mean a 

13 

reduction of the average depth of the Ocean by — — xiooo = 36 metres. 

36 1 

Let us furthermore assume that the depression of the crust caused 

hy this increase of the ice-cap is compensated for by an upheaval of the 

sea-floor in a surrounding area which is four times the depressed area 

under the ice-cap. According to the formula given on p. 231 we have then 

the average upheaval (U) of the sea-floor in this area 

^, 1000 -0.9 900 

L -— — = = 60 metres. 

3 + 3/0-25 15 

As the area of upheaval is 52 millions square kilometres, this would 
mean a reduction (r) of the sinking of the general sea-level by 

52 

r = x6o = 8.S metres. 

361 ^ 

Hence the general sea-level would be lowered 36 — 8.6 = 27.4 metres. 
This reduction of the depth of the water of the rest of the Ocean, 
with an area of 361 — 52 = 309 millions square kilometres, would cause 
an upheaval of the sea-floor and a sinking of the continents. In order 
to simplify our computation we will now leave the upheaved part of the 
Ocean round the Antarctic out of consideration as we well may in this 
rough estimate. If we assume that the area of the lowered continents be 
J49 — 13= 136 millions square kilometres the upheaval (//) of the sea- 
floor would be 

27.4 



3 + 3£^ X 3 
•^ 136 



2.85 metres 



and the average subsidence (//) of the continents would be 

3p9 
136 



/; = 3p9_^^2_g ^ g metres. 



I92I. No. II. THE STRANDFLAT AND ISOSTASV. 237 

Hence the general sea-level would be lowered about 24.5 metres in- 
stead of 27.4 metres and the negative shift of the continental shore-line 
would be 18 metres, provided that the coast would sink as much as the 
average subsidence of the continents, which, however, is not probable. 

The values of tlie upheaval of the sea-floor and sinking of the con- 
tinents found above are. however, maximum values. It would probablv 
be a verv long time before the whole area of the continents were sunk. 
For all practical purposes we may assume that only some parts of them 
near the coasts would sink and this would substantially reduce the amounts 
of the upheaval of the sea-floor and the subsidence of the continental 
coasts. 

We may assume that the sinking of the general sea-level caused bv 
the formations of ice-caps will at least be 70 per cent of the amount by 
which the average depth of the Ocean is thus reduced, and the negative 
shift of the general continental shore-line, outside the regions depressed 
and raised by the weight of the ice-caps, will be at least 53 per cent of 
the amount by which the Ocean depth is reduced, but as rule much more. 

The preceding computations are only rough estimates. They prove, 
however, on the one hand that quite considerable changes in the average 
depth of tlie Ocean mav have been caused by the formation of the Pleisto- 
cene ice-caps and bv their melting, and on the other hand that the shifts 
of the general sea-level and the general continental shore-line have been 
at least 70 and 53 per cent of the amounts by which the depth of the Ocean 
was changed. 

It is quite possible that during the maximum development of the 
Pleistocene ice-caps the general depth of the Ocean was reduced by 150 
metres. In that case the general sea-level would sink at least 105 metres, 
and the nes:ative shift of the general shore-line would be at least 80 metres. 



The Lowering of Sea-Level during the Glacial Periods and the 
Surface of the Continental Shelf. 

The (juestion is now, whether this lowering of the general sea-level 
during the glacial periods can help us to explain the formation of the 
continental shelves? My view is that, although the great glacial periods 
may have lasted a very long time, the time would nevertheless not be 
sufficient for the development of such broad formations as the floor of 
the North Sea. the continental shelf west and south of Great Britain and 
Ireland, Sec, not to speak of the continental shelf of Norway, Iceland. 
Greenland, and the coasts of the North Polar Sea. It must also be 
remembered that during the maximum development of the Pleistocene 
ice-caps these northern shelves were to a great extent covered by ice, and 
were more or less depressed. 



238 FRIDTJOF NANSEN. M.-N. Kl. 

The sinkinf4' of sc'a-lc\cl iluiiii^ tlic ^^^lacial pcriofls may, however, 
hel]) lo explain ccrlaiii features of the continental shelves which seem 
to iiKlicaie lliai tluy lia\c hci'u receiitl\- enier^cfj, c. ^. the occurrence at 
dc'ptlis of 100 I0 150 metres of numerous water-worn shore fx;bbles ard 
shells of littoral molluscs, Ixin^»- on or just hcneatli llie surface of the 
continental shelf and harrlly covered hy later deposits. A great many river 
channels and \alU'\s traversing the cont iiuiital slielves also indicate that 
their surl'accs must luaxe heen emerged during sfjine recent perio'l, when 
these channels were either formed or recjpeneil. 

In regions where the lanrl was covered by I'leislocene ice-caps the 
continental shelves outside the coasts may also have been upheaved to 
some limited extent wlien the land insifle was flepressed bv the weight 
of the ice-caps Icf. H. Munthe, if;io. p. I20^)'. 

It is hardly ])ossil)le to compute the magnitude oi this periph.eral 
upheaval. It seems probable that it reached its greatest heights during 
the first period of the depression of the ice-covered region, when only a 
comparatively narrow peripheral belt was pressed tip by the underlying 
magma displaced by the depression. The height (jf this upheaved wave 
gradualh' decreased again as the wave slowly extendecl its area outwards. 
We have previously found that in the case of local equilibrium 

II >< 3 = d^o.<^ — // >■- 3 

where u is the average height of upheaval in the peripheral belt, d the 
average thickness of the ice-cap, and /; the average height of the depression 
under the ice-cap. This means that as tlie depression (li) gradually in- 
creases the height of tipheaval (//) will decrease, and when // has reached 
its full extent and 

/(X3 = c/> 0.9 

the upheaval ;/ will be zero, which will not happen before the whole crust 
of the earth has regained its level of perfect equilil)rium. In that case 
the upheaval of the crust compensating for its depression under the ice- 
caps will have been entirely transferred to the floor of the Ocean the 
volume of which was reduced by the formation of the ice-caps. 

In the case of the continental shelves of Norway and other glaciated 
regions the development may have been the following: 

Bv the reduction of the volume of the Ocean there has been an im- 
mediate sinking of the general sea-level, which during the maximum 
development of the ice-caps may have amounted to 100 metres at least, 
and perhaps to as much as 150 metres. By the gradual depression of the 
interior areas of the land under the weight of the ice-cap the outer coast 
and the region of the continental shelf have been upheaved. If this up- 
heaval was as much as 50 metres, there may have been a negative shift 
of the shore-line of as much as 150 or 200 metres. 



ICSI. No. II. THE STRANDFLAT AND ISOSTASY. 239 

This Stage may have lasted for some time and during tliat period 
tlie surface of the continental shelf was to a large extent above sea-level 
and was exposed to suhacrial denurlation, fluvial erosion, and to a vigorous 
shore erosion by frost, which ha\e left their marks on this surface. 

I have previously [cf. 1904, pp. 151 f.î pointed out that the longitu- 
dinal arrangement of submerged valleys and fjords in the inner regions 
of the continental shelf of Norway must stand "in genetic relation to the 
tectonic structure of the underlying rocks", and that this fact is "conclusive 
evidence that the underlying ground of at least the innermost part of the 
shelf is solid rock at no great depth below the surface". And I furthermore 
pointed out 'cf. 1904, pp. ^j f.] that these longitudinal valleys, following 
directions parallel to the coast and to the direction of the Caledonian 
mountain folding, "must owe their prime origin to the subaërial erosion 
of rimning water", because it is hardly conceivable that the glacial erosion 
alone would excavate such valleys at right angles to the direction of the 
general glacial movement, unless the glaciers were to some extent guided 
by initial fluvial valleys. 

This means, however, that the least the inner part of the continental 
shelf must have been above water when these initial fluvial valleys were 
developed. This may to some great extent have occurred during preglacial 
periods: but it is also probable that by the emergence of the shelf during 
the glacial periods, there has been some fluvial erosion of these valleys. 

\Mien the ice-caps increased and extended outwards, the shelves were 
graduallv more or less covered by the ice. They were then depressed in- 
stead of being upheaved, and the peripheral belt of upheaval was gradualh- 
shifted outwards. During the last glacial period this was obviously the 
case in the inner part of the broad continental shelf outside the coast of 
Helgeland and the Trondhjem region and also to some small extent out- 
side the coast of the Xordmor and Romsdal region. 

During the maximum development of the Pleistocene ice-caps the 
whole of these shelves were obviously covered by ice, and they were then 
much depressed, and were probably also much eroded by the glaciers, at 
least in some regions, c. g. outside Helgeland and the Trondhjem region, 
where the surface of the shelf was thus much lowered. 

The Deposition of Sediment has gradually raised the general 

Sea-Level. 

If the continental shelves of the world cannot have been initially 
formed and developed to their great width, and if their rocky planes 
cannot have been cut in solid rock during the comparatively short periods 
when the general sea-level was lowered by the formation of the Pleisto- 
cene ice-caps, the question still remains open how and when were they 
formed. 



24*^ IKIOTJOF NANSKN. M.-N. Kl. 

Is it prohalilc tli.'it llic f^ciicral lc\cl of the Ocean lias been as much 
lower as llu' (lc\flo])m(nl of these shelves near sea-level wouM necessitate, 
and llial lliis low level has reniainerl fliirin^ such lon^ i^re^^lacial periorl.-, 
as wonlil lie r((|uirefl for the planinj,'- of the hroad shelves, provifled that 
tlicy are fonneil chiefly hy marine anr| suharrial fknudation an'l hv sedi- 
mentation? 

The fol I o w i n,!^'' consideratifjns may perhaps help to find an explanation. 

The continual deposition of serliments on the sea bottom has a ten- 
dency to raise the g^cneral sea-level. The rise of level thus caused is to 
some extent compensated for b\ the i^radual subsidence of th? sea-floor 
undt'r the weij^ht of the sediment. If the specific ^a'avity of the sediment 
be 2.5, that of the ma_L;nia underlyinj^- the ri.^-id crust 3, the subsidence (u) 
of the sea-floor caused by a la\er of sediment with the averajij^'e thickness 
of 100 metres will be: 

ioo> 2.5 

u = ^ ^- o'i metres. 

3 

As the sediment came from the continents, these would be raised bv 
an average height (//) of (cf. p. 235 ) 

// = 2.43 - 83 = 202 metres. 

Hence, although the general sea-level would be raised 17 metres, there 
would be a negative shift of the shore-line. If we assume that the amount 
of the upheaval of the coast is as much as (^o metres or mirlwa\- between 
the upheaval of the continents inside and the subsidence of the sea-floo."" 
outside, the negative shift of the shore-line would be 43 metres. 

These calculations are very rough. On the one hand, we do not kn.ow 
the specific gravity of the plastic magma underhing the rigid crust, nor 
do we know the exact specific gravity of the sediment on the sea-floor. 
The oceanic water contained in this sediment has to be left out of our 
calculation, as it was in the Ocean before the sediment was deposited. 
What we wish to know is the specific gravitv and thickness of the sedi- 
ment without the water. 

On the other hand the processes that actually take place are of course 
much more complicated than here assumed. The deposit of scdiiuents is 
distributed very unevenly over the Ocean floor. The deposition of terri- 
genous matter is largely limited to the regions near the coasts and the 
deposition of calcareous and silicious sediments (of organic origin) is 
limited to the regions of the sea-floor which have moderate depths, while 
there is extremely little deposition going on in the deepest areas of the 
sea-floor. This means that the depression of the sea-floor caused by the 
comparatively rapid rleposition of sediment in certain areas is largely 
compensated for by the upheaval of the sea-floor in adjoining regions 
wdiere the deposition is much less. In that case the rise of the general 
sea-level caused l)y the deposition of sediment will naturally be consider 
ably luore than found by our computation above. 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 24I 

It thus seems probable that thie continual deposition of sediment on 
the sea-floor has in the course of time actually caused an appreciable rise 
of the g-eneral sea-level. At the same time there has been a gradual sub- 
sidence of the sea-floor and an upheaval of the continents. 

If we imagine that, to begin with, the outer border of the continental 
surface was low and to some extent submerged below sea-level, then the 
emerged parts of this continental border would be exposed to marine 
abrasion as well as to subaerial denudation. By the joint action of these 
two agencies the emerged portions of the land surface would after a long 
time be planed down towards sea-level, while the depressions of the sub- 
merged surface would be more or less filled up with the waste from these 
emerged parts and from the continental surface inland. At the same time 
there would be a continual deposition of sediment on the continental slope, 
outside tlie submerged outer border of the flat continental surface. 

In the course of time the result of these processes would be that the 
sea-floor of the continental slope would be gradually depressed under the 
heavy weight of the increasing layers of sediment. On the submerged 
continental border there has been a depositic i of waste from the inner 
part of the continent in addition to the Avaste formed by the denudation 
of the emerged portions of the coast. This border will consequently also 
be pressed down to some extent. The emerged continental coast land 
inside this submerged border, however, has been denuded by subaerial 
erosion and A\ill be tipheaved. At the same time the general sea-level will 
be somewhat raised by the general deposition of sediment on the Ocean 
floor, and the submerged continental border will be still more sub- 
merged. 

As the coasts would be continually exposed to the joint action of 
subaerial denudation and marine abrasion they would gradually be cut 
back; but this process would be extremely slow, as, owing to the general 
subaerial denudation of the continental surface, the denudation of the 
coastal land would be largely compensated for by its isostatic upheaval, 
and it would therefore be chiefly the marine abrasion which would cut 
T)ack the coast, as this denudation is limited to the shore and would 
therefore create no appreciable upheaval of the land. 

Although weighty objections may certainly be raised against many 
details of this theory, still I think it possible that on the whole it may 
give a fairly feasible explanation of the formation of the continental 
shelves. 

It would be easy to understand that along coasts where there was 
no initial submerged border of the continental surface, the formation of 
a continental shelf will be a difficult and extremely slow process. We 
therefore find that the continental shelf is often very narrow outside such 
high coasts, while it is, as a rule, broad outside coasts where the continental 
surface is low and slopes gently towards the sea. 

Vid.-Selsk. Skrifter. I. M.-N. Kl. 1921. No. 11. 16 



242 FRIDTJOF NANSFN. M.-N. Kl. 

As tlic sIkji'c erosion I)v frost will larj^a-ly increase the- rate of the 
marine abrasion, the continental shelf may be expecterl to be very broad 
HI the cold rcf^ions of the j,'lobe, where the climate may have favoured 
this erosion also durin/^-- preg'laci.'d periods, dlie fact is that the continental 
slielf is exceptionally well developed in the Arctic rejj^ions, and esijccially 
outside the coasts rouml the Nortl: I'olar Sea. 

It nuL^ht be objectefl that by tlie j^^reat auKnuit of waste carried from 
the land on to the surface of tlie continental shelf formed in this manner, 
its level would be built up towards sea-level faster than the general sea- 
level could be raised I)y the deposition of sediment on the floor of the 
Ocean. Owing to the considerable movements of the water, however, 
especially those caused l)y tidal-currents and wind-currents over the con- 
tinental shelf, as a rule comparatively little sediment can be deposited on 
its level parts or higher parts. The sediment is therefore chiefly deposited 
in the hollows and depressions or is gradually moved beyond the shelf, 
and is deposited on the continental slope outside. 

Hence tliere will not as a rule be much deposition of sediment on the 
continental shelf after its depressions and liollows liave been filled up, 
and it will not be much depressed by the weight of the sediment after it 
has been once fully developed. The probability is therefore that its level 
vxill remain fairly stable, while there is a slow upheaval of the emerged 
land inside to compensate for its denudation by subaërial erosion, and a 
slow subsidence of the sea-floor outside caused by the deposition of sedi- 
ment. As the general sea-level gradually rises and the depth of water 
over the continental shelf increases, it will be slightly depressed by the 
increasing weight of this water. 

If the submerged shelf of the continents has been developed in a 
manner somewdiat similar to wdiat has been indicated above, it is also 
fairly easy to understand how it is possible that there are two wideh' 
extended formations, which seem to indicate two distinctly different levels 
at wdiich the shore-line of the Ocean has remained for very long periods. 
The one formation is the low continental plains, which rise gentlv inland 
from present sea-level, and the development of which would be extremely 
difficult to understand unless we assume that the general sea-level has 
remained more or less the same as now during long geological ages. The 
other is the continental shelf which also indicates a sea-level and which 
must have required a very long time for its development. The levels of 
these two planes are as a rule distinctly different, and the difference 
between them is often marked by a comparativelv steep descent from the 
coast, or in Norway from the edge of the strandfiat, down to the inner 
surface of the continental shelf. 

Both plains have been formed at about present sea-level more or less, but 
while the continental shelf has been somewhat depressed and is more and 



I92I. No. II. THE STRAXDFLAT AND ISOSTASV. 243 

more submerged hv the slow rise of the general sea-level, the continental 
plains keep to a certain degree their levels, owing^ to the upheaval of the 
land compensating for the denudation. The denudation is, however, quite 
insignificant in the lowest parts of the plains, the outer borders of which 
will graduallv be submerg'ed by the rising sea. 

The difference between the level of the continental shelf and the 
levels of the coastal continental plains as well as of the strandfiat. may 
have been accentuated during the glacial periods when the general sea-level 
was lowered more or less by the formation of ice-caps on land. 

Bv a negative shift of the sh.ore-line to an amount of say 40 to 60 
metres, or more, tlie inner parts of the continental shelf may have been 
more or less planed down by marine abrasion (shore erosion), and the 
slope from the present coast, or from the edge of the submerged strand- 
flat, mav have been cut back, and has become steeper. 

In the manner explained in this chapter it may be possible to under- 
stand the development of the continental shc!f without being obliged to 
assvune considerable changes in the general sea-level. There are, however, 
certain bathymctrical features of the northern seas, which might seem 
to indicate much greater changes in the sea-level. I may especially mention 
the great similarity of the depths of 400 to 500 metres of the submerged 
ridges between Scotland, The Færoes, Iceland, Greenland, and Baffin Land 
Tcf. Xansen, 1904, pp. y^,, St, f., 89. 173 f.]. 

It seems somewhat difficult to assume that the uniformity of the 
lowest levels of these ridges can merely be due to an accidental coincidence. 
It may also be mentioned that the floor of the wide outer part of the 
large submerged valley of the Barents Sea, south of Bear Island, which 
[ have called the submerged Bear Island Channel [1904, p. 30], also seems 
to have a level similar to that of the above mentioned ridges. The mouths 
of several other suljmerged fjords (e. g. the A'est Fjord) also seem to have 
similar depths. 

If these ridges and mouths of submarine fjords have actuallv been 
lowered to some ancient base level, indicated by their present depth, this 
would prfjbably mean a lowering of the general sea-level, so great that it 
could hardly be explained by the shifts of sea-level discussed in this 
chapter, and it seems extremely difficult to find a satisfactory explanation 
of changes in the sea-level of this order. 

There is another possibilit\ that these depths mav indicate a certain 
lower limit of the depth to which the Pleistocene ice-caps reached down 
during their maximum development. A depth below water of 400 to 500 
metres reached by an ice-sheet would correspond to a thickness of the ice 
of about 440 to 550 metres, which does not seem excessive, even near the 
edges of the ice-caps. Besides it has to be remembered, that during the 



244 FRIDTJOF NANSEN. M.-N. Kl. 

niaxiiinim dcvelopiiunl of llic ice-caf)s llic- sca-levcl prohahly was con- 
siderably lower lliaii now. 

Hence we may well assume that the ice-caps readied down to these 
depths. The dc-])1hs may then indicate a kind of hase level to which the 
underhin«^- .ground was eroded l>y the f.,daciers. It was j^reviously Cp. 23; 
mentioned that the erosive power of a g'lacier, moving at a certain rate, 
remains the same below water as above it, as long as it rests on the ground 
anfl does not float. 



ig2I. No. II. THE STRAXDFLAT AND ISOSTASY. 245 



XVI. THE STRANDFLAT AND THE LATEGLACIAL 

AND POSTGLACIAL SUBMERGENCE 

OF FENNO-SCANDIA. 

As was pointed out already years ago by J. H. L. Vogt [1907, 
pp. 38 ff.] and by the present writer [1904, pp. 126 ff. and 200 ff.], there 
is a striking difference between the fairly uniform, nearly horizontal level 
of the Norwegian strandfiat and the varying, generally much higher levels 
of the raised beaches and shore-lines which slope seawards from the inner 
land. This relation between the strandfiat and the raised shore-lines, 
formed later, was held by the writer to be convincing evidence of the 
isostasy (jf the earth's crust. 

The important fact that the raised lateglacial and postglacial shore- 
lines slope from the inner land towards the outer coast has been established 
for Norway as well as for Sweden, Finland, and the Kola Peninsula by 
the investigations of a great many authors. The following may be 
especially mentioned: Bravais [1838], R. Chambers [1850], Gerard de Geer 
[1890, 1898], Andr. M. Hansen [1891], H. Munthe [1892, 1900], A. G. 
Hogbom [1896, 1899, 1904], Amund Heiland [1900], J. Rekstad and 
J. H. L. Vogt [Vogt 1900], W. C. Brogger [1900], W. Ramsay [1898], 
J. Rekstad [1905 — 07, 1910], V. Tanner [1906, 1907], A. Hoel [1907], 
J. H. L. \*ogt [1907], P. A. Oyen [several papers], Kaldhol [1912], 
Danielsen [1905, 1906, 1909, 1912], Gronlie [1913 — 14, 1918]. 

A great many American geologists after Gilbert [1882, 1890], Russell 
[1885], Warren Upham ^1887], and others have found that the raised 
shore-lines of North America are tilted, and that the lateglacial and post- 
glacial shore-lines have a distinct inclination outwards from the central 
areas of the regions which were covered by an inland ice during the last 
glacial period; other quite local shore-lines were also found tilting out- 
wards from the centre of a region formerly covered by the water masses 
of a great lake (Pake Bonneville). 

The postglacial upheaval of the land has also there been greatest in 
the central parts of the ice-covered regions, diminishing outwards to nil 
near its outskirts. The details, however, of this upheaval have not yet 
been studied in so much detail as in Fenno-Scandia. 



246 



FRIDTJOF NANSEN. 



M.-N. Kl. 



The Lateglacial and Postglacial Upheaval of Sogne Fjord. 

Let us roiiipari' llic heights of the Xorwej^'ian strandflat with those 
of tlic raised latc\i;lacial ami ijoslijlacial slnjrc-lines. 

In Sof^nc i'jonl tlie rollou iii;^- aliitudes of ihe lower level of the 
emerg'cd strandflat were found: 



Locality 


Distance from the outer 

skerries (Utvær outside 

Sogne Fjord) outside the 

coast 


Height of Strandflat above 
mean sea-level 


Sogndal and Norum Fjord 

Leikanger 

Tjugum and Balestrand 

Tangen opposite Høyang Fjord . . . 
Matsnes east of Eike Fjord 


136 ki!ometr<- 

T27 

115 

80 

63 

36 

28 

16 


about 10 metres 
10 to 12 „ 
about 12 „ 

17 

17 

16 


Dingenes 

Northern Lindas Peninsula 


II. 3 » 

17 







The measured altitudes of the upper level of the emerged strandflat 
may indicate a similar rise of altitude from the inner parts of Sogne Fjord 
towards the outer coast. We found it to be aljout 25 metres on \>garnes, 
26 metres on \'angsnes, on Matsnes, and possibly on Rutlctangene, 33 to 
40 metres on Radoi (according to Ahlmann), &c. 

Prov'ided that these olxservations be correct, and tliat the difference 
be not due to glacial erosion, the strandflat has consequently been some- 
what more raised near the outer coast than further inland, as was 
previously mentioned. 

The upper limit of the lateglacial and postglacial submergence (or 
upper marine limit) in the region of Sogne Fjord can only be traced by 
the highest raised terraces. J. Rekstad [1905, 1906, 1907, 1910] has found 
the following heights: 



Locality 


Distance from the outer 
skerries, Utvær 


Height above sea-level of 

upper limit of lateglacial 

submergence 




72 kilometres 

88 

112 », 
I 70 


77 metres 


Ortnevik, Fitje 

Vik 

Hovland, Ardal Fjord 


87 » 
115 

137 



Let us assume that the isohasc (i. c. line of equal postglacial upheaval) 
for I/ metres (i. c. the height above sea-level <-)f the lower level of the 
emerged strandflat near the outer coast) passes across the middle of 



I92I 



. No. 1 1 . 



THE STRAXDFLAT AND ISOSTASY. 



247 



Yttre Sulen, across the outer part of Atleoi to Rugsund near tlie mouth 
of Nord Fjord. Let us furthermore assume that the isobases for greater 
elevations further inland follow directions parallel to this line. 

In the following table are given: 1) the direct distance of each locality 
mentioned from the isobase for 17 metres, 2) the height of the upper limit 
of lateglacial submergence above the lower level of the emerged strandfiat 
(as given in the table on p. 246), 3) the gradient of the postglacial 
elevation (minus the height of the strandfiat) a) between the isobase for 
17 metres and the locality, and b^ between each locality. 



Localitv 



Distance from 

the isobase for 

1 7 metres 



Height of upper 

marine Hmit above 

lower level of 

strandtlat 



Gradient of Elevation 



from isobase for 
1 7 metres 



between the 
localities 



Vadheim 
Ortne\-ik 

Vik 

Ho viand . 



S2 kilometres 



95 

143 



00 metres 



103 

128 



1. 1 5 per mDle 
0.99 • 
1.08 » 
0.89 • 



) 0-53 per mille 
) 1-37 * 
} 0.52 ' 



The gradients between X'adheim and Ortnevik and between Ortnevik 
and \'ik differ much from each other and from the gradient between Vad- 
heim and the isobase for 17 metres. If we assume that the terrace 
measured at Ortnevik does not actually give the upper limit of sub- 
mergence, and if we take the gradient between \'adheim and \ ik, we find 
it to be i.oo per mille, which is slightly lower than the gradient 1.15 per 
mille between \'adheim and the isobase for 17 metres. 

The gradient in the inner part of the fjord, between \*ik and Hov- 
land in Ardais Fjord, is very low, 0.52 per miille. The difference between 
this gradient and the gradient in the outer part of the fjord is much too 
great to be explained by the regular decrease of the gradient of elevation 
inland towards the central region of the depressed area. Rekstad is ob- 
viously right in assuming that the highest terraces at Hovland do not 
mark the upper limit of submergence, because the inner part of the fjord 
was filled with glaciers during the time of greatest submergence, and 
when the glaciers retreated and the terraces could be deposited, the land 
had already begun to rise. 

The probability of this explanation is confirmed by Rekstad's 
measurements of the upper marine limit in the inner part of Hardanger 
Fjord and by H. Kaldhol's measurements of the raised terraces in Xord 
Fjord. 

The observations made in Xord Fjord are of special interest to us 
here in another respect. 



248 FRIDTJOF NANSEN. M.-N. Kl. 

The Lateglacial and Postglacial Upheaval in the Hegion 
of Nord Fjord. 

krkslad fouiHl Ikjo^ tlirii tlic j^n-ariicnt of |)ost^lacial upheaval is 
much smaller in tlic (jutcrmost pari of Xor'l I'^jorrl than farther in. 
Kaldhol's numerous measurements (1)\' k\cllin.L() of the upper heights of 
the raised terraces in this re.y-ion, at ahout liundrc-il different places, have 
shown this still more clearly [Kaldli<il. i';i2a . 

At t\vent\ iliffcrenl ])laces in the ouic^r coast zone. Ijetween the north 
coast of the Slat I'c-ninsula and the south coast of Bremanger Land (and 
iMoicn Island), Kaldhol's measurements of the upper limit of submergence 
^i\c heights between 15.2 and 17.7 metres above present sea-level (see 
nia]) h'ig. 165). In most cases it is between 16 and 17 metres. At two 
places, Revik on \agsoi and l'"orr]e on llremanger Lanrj, he founrl tlie 
lieight to be 14.9 and 13.8 metres, but these values seem to be too low, 
for at Vedvik, 2 kilometres north of Revik, the height was found to be 
15.8 metres, and at Tysketveit close by, also on \"ågsøi, it was i() metres 
(cf. Fig. 165). At Kalvåg on Froien Tslanrl, 10 kilometres south-west of 
Forde and more seawards, the height was UkO metres, and at .'^teinset on 
the outer side of the same island it was 17.3 metres. 

These heights about iC) ancl 17 metres and less than 17.7 metres are 
found inland as far as Rugsund, inside the mouth of Xord Fjord, and 
about 23 kilometres from the outer coast-line (cf. Fig. 165 )^ 

Kaldhol points out that the gradient of elevation is very gentle in 
this outer region. Even if the height of 13.8 metres at Forde (Bremanger- 
pollen) be used, and heights found farther seawards be considered too 
high, the gradient will be no more than 0.26 per mille. It seems to me, 
however, to be much more proba1)le, that the exceptional heights found 
at Forde and Revik are a little too low. Kaldhol's observations give then 
practically no gradient of elevation in this coastal zone. TJw land has here 
been cicz'atcd about 16 or ly metres, keeping its horisontal leveL or at 
least zi'ithout any appreciable tilting. 

It is striking what a perfect accordance there is between the height 
of this horizontal upheaval and the height of the strandfiat in the outer 
part of Sogne Fjord and in the coastal region outside, which I also found 
to be about 16 and 17 metres. It is hardly probable that such a coincidence 
is merely accidental. 

This seems in an unexpected manner to prove the correctness of our 
assumption that this level of the strandfiat represents the level which the 
shore-line had in this region before the last glacial period and before the 
last submergence of the land. 



1 If the observations of 28.9 and 23.9 m. at Sandvik and Hammersvik, in the inner 
portion of the Stat Peninsula, be correct, the border with the upper shore-line at 

heights less than 17.7 ni. may tie somewhat narrower in that region. 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASY. 



249 




..^ 



br. - tr 



o ? ON = ^ 



-^ 


- 


^ 


1 


A 


.^ 


£ 


1 


.-^ 


-r 


d 


C3 


3 




n 





5 




be 

C 


■fl 


r- 


X 


1 


"x 


u 


- 


X 


5 


•r 


•- 


T. 


o\ 


■r. 


CI 


73 


C3 


7; 




~t 


1Ö 





K 


•-£■ 


r. 


u 


~ 


- 


§- 


C 

br. 




CO 




rt 



£ 

C 







5 





bi 



'be 


be 

C 


z 


IE 




i/. 


s 


1^ 


P 


■€ 


r. 

C 


n: 


^J. 


12 


■5 


•r. 


CJ 


^ 


Lu 


Cii 


B 





^/ 


w 


— : 


bi 


•^ 


y; 


r. 


c 


K 


- 





y; 


~ 


-i 





250 KRIirrjOF NANSEN. M.-X. Kl. 

I lie 1,'iinl lias consdnuiillx nol Ixcii ilcjdcsscl 'luriiij^ llic last j^lacial 
period in lliis coastal /one, 25 l:iloin(tres broad, in the region of Nord 
J'\ioi-d, i'.rcnian,L;cr, and Stat, altliou^li ilic-re has been a postglacial change 
of \() to \y nutres in tlie le\cl of the shore-line. The depression has onlv 
occurred in tlie re.^ion inside Ku^sund, whence the njjper limit of sub- 
nierf^ence sudden!}- begins to rise inland, as is clearlv rlenioiistrated by 
Kaldhol's profile. 

The gradient of elevation is 1.20 jjer nulle betuen Rugsunrl t.Stroni- 
nien, height =■• 18.2 metres) and Eikenes in Hyen Fjord (height = 61.2 
metres), a distance of about 35.7 kilometres if measured at right angles 
to the direction of the isobases (if we take .Strommen in Hyen Fjord, 
height 59 metres, the distance is 33 kilometres and the gradient 1.24, 
cf. J^'ig. 1^5) — while it is on the average 2.00 per mille in the inner 32 
kilometres of the fjord, between Hyen Fjord (Eikenes) and Tndvik in 
Indvik l^^jord (height = 125.3 metres), provided that Kaldhol's deter- 
minations of the upper limit are correct in this region. They give gradients 
which vary much in value. Between Lote (71.7 metres, see Fig. 165) and 
Henne (95.7 metres, see Fig. 165), a distance of 6.7 kilometres, his values 
give a gradient even of 3.6 per mille. It seems to me to be probable that 
the highest terraces, found by Kaldho! in this region of the fjord, are 
not marine. 

We may, however, expect some irregularities here, as the region of 
Nord Fjord is perhaps the most prominent centre of seismic activity along 
the Norwegian coast [cf. Kolderup, 1914, map p. iiij. 

In the innermost region e^f Nord Fjord, inside Indvik, in Stryn. 
Olden, and Loen, Kaldhol's measurements of tlie heights of the highest 
terraces give suddenly much lower values than in the region west of Ind- 
vik. The explanation may he either that the formation of terraces was 
prevented by glaciers filling the valleys and innermost parts of the fjords 
during the time of the deepest submergence of the land, or, as suggested 
by Kaldhol, that a later advance of the glaciers has destroyed the highest 
terraces in that region. 

Kaldhol's detailed investigations of the lateglacial and postglacial 
terraces in the region of Nord Fjord give most valuable information 
about the suljmergence and emergence of the land during this period, and 
his height measurements support, as we have seen above, the correctness 
of my view of the relation between the horizontal level of the strandfiat — 
indicating the level of the shore-line before the last great submergence — 
and the tilted planes of the raised terraces and beaches — indicating the 
submergence and the upheaval of the land. 

There is unfortunately no other region of our coast, which has been 
subjected to similarly detailed investigations, as far as I am aware, at 
least not in the outer coastal region. But several scattered observations 
in various localities indicate that along those parts of the outer coast. 



I92I. Xn. II. THE STRAXDFLAT AND ISOSTASV. 25I 

where the upper limit of submergence lies no higher than the strandflat, 
more detailed investigations would probably demonstrate a horizontal 
upheaval of the outer coastal region, and a tilted upheaval of the land 
inside, in quite a similar manner as we have found in the region of 
Xord Fiord. 



Gradient of the Lateglacial and Postglaciel Upheaval of the West 
and North Coast of Fenno-Scandia. 

A general impression of tlie postglacial upheaval of the land may be 
obtained by comparing the gradients of this upheaval in the different 
regions of the coast of Xorway and of the Kola Peninsula as given in the 
following table. The values of the gradients are computed from the heights 
of the highest terraces and raised beaches which are supposed to indicate 
the upper limit of lateglacial submergence. The heights were measured 
by the authors given in the last column. The heights of the lower leve! 
of the strandflat have not been deducted from the figures used for the 
computation of these values. 

In several regions it is difficult to decide which raised terraces and 
beaches actually indicate the upper limit of lateglacial submergence. In 
the outer coastal regions of Finmark Tanner [1906, 1907] has found 
raised shore-lines and terraces situated considerably higher than those 
which have been used for the computation of the gradients in our table. 
Similar higher shore-lines have also been found at several levels bv 
Gronlie [19 14" in the Tromso distrikt and by W. Ramsay [1898" on the 
Kola Peninsula. These shore-lines, however, are less distinct than the 
lower shore-lines and have a much older appearance. Thev do not seem 
to correspond to the shore-lines which are supposed to mark the upper 
marine limit (the upper limit of lateglacial submergence) further south 
in Xorway. and it is difficult to understand that they can have been 
formed during the last glacial period. They may more probablv be sur- 
vivals from a previous period of submergence, as is assumed by Ramsay 
and Gronlie. 

For my computations I have therefore used the heights of the shore- 
lines and terraces belonging to the level which Tanner calls le, as this 
level seems to correspond to the upper level of the two conspicuous raised 
shore-lines of the Tromso — Hammerfest region and \'esterålen. and to 
the generally accepted upper limit of the lateglacial submergence farther 
south in Xorwav. 

These gradients given in the table differ so much locally that one 
might doubt their correctness; but on the whole there seems to be a certain 
system in their variations. If they be introduced in a map of the Xor- 
wegian coast where the submerged continental shelf is also outlined [cf. 



252 



FRIDTJOF NANSEN. 



M.-N. Kl. 



Localitv 



(i radient ol 
Elevation in 

/v V- uullr 



Observer 



Northern Östcrdal 

Christiania Fjord 

Coast between Larvik and Clu-istiansand 

Hardanger Fjord, outer |)art 

— — central |)art 

— — Halsenoi to Rosendal 

Sogne Fjord, between Vik and Vadheim 

Sund Fjortl, outer part 

— inner part 

Nord Fjord, coastal zone 25 km 

— middle 36 kilometres 

— inner 32 — 

Sondmor, Hardeidland 

Coastal region northeast of Ålesund 

Nordmor, Reinsvik (Christiansundi to Ting\-oll . 

— Reinsvik — Vågbo (Halsa Fjord I 

— Reinsvik — Bruset I Todall 

— Tingvoll — Bruset 

— Henden (Arisvik Fjordl — Valsoibotten . 
Southern Helgeland, average 

— region of Donna 

Dunderland Valley to Træna 

Mainland to Lofoten 

Ofoten to Kvæ Fjord 

Region between Andoi and Senjen 

Region of Senjen 

— between Senjen and Ringvasoi 

— — Ringvasoi and Skjervoi 

Fjord south of Hammerfest . 

Porsanger Fjord 

Lakse Fjord 

Tana Fjord 

Varanger Peninsula 

Region south of Varanger Fjord 

Region of Ribachi (Fisker) Peninsula 

— » Kildin 

— » Woronye River, Kola Renins 

Southeastern coast of Kola Penins 



0.7 1 .6 
ca. 0.7 

0.68 
ca. 0.78 

0.68 

0.94 

1 .0 

r .0 

O.Q 

0.0 

: .2 

2.0 (?i 
1 .2 
I.I 
1.48 
2.00 
1.32 
1.23 



0-73 
0.67 
0.91 
0.76 
0.71 
0.90 
1.22 
0.85 
0.90 
0.67 
0.60 
0.48 
0.58 
0.51 
0.60 

0.53 
0.65 
0.72 
0.32 



< I. Ilolmsen 1 191 7J 

Oven 

Danielsen [1912] 

Rekstad [1906] 



— U910J 

— [1906] 

Kaldhol [1912] 



Rekstad [1905] 

— [1906] 

Kaldhol [1913I 



Rekstad [Vogt 1900I 

— [1904] 

— [Vogt 1907! 
J. H. L. Vogt [1907] 
N'ogt [Rekstad 1905] 
Heiland [1900] ' 



R. Chambers [1850J 
Tanner [1907] 

— [1906] 



Ramsav [1898] 



' O. T. Gronlie gives in his text [1914, p. 224] values (between 1.35 and 1.77 per mille» 
of these gradients for the region between Andoi and Skjervoi, which are much higher 
than those computed from Helland's observations ; but as far as I can see Gronlie must 
have made some strange mistakes in his distances. His map of the isobases gives 
much the same gradients as Helland's observations. 

2 As will be mentioned later (p. 263I it seems to me that Tanner has probably mistaken 
his level h in this region, and I have used his highest terraces for the computation 
of the gradient. 



1921. No. II. THE STRANDFLAT AND ISOSTASV. 253 

Xanseii, 1904, I'l. XI . it is seen that, as a rule, the steep.^st gradients 
occur in those regions where the edge of the continental shelf is nearest 
to the outer coast of the land, c. g. in the region of Senjen and along the- 
coast from Xordmor to Xord Fjord (cf. Fig. 166). In Sond Fjord it is 
also steep (i.o per mille) although this region is farther away from the 
edge of the continental shelf. But here the deep outer part of the sub- 
meged Norwegian chaimel is very near the outer coast, while this channel 
is shallower outside the region of îiardanger Fjord where the gradient is 
less steep (0.7 — 0.8 per mille). 

The lowest gradients occur along the south-east coast of Xorwa>' 
(0.68 per mille), in Helgeland (0.67 — 0.73 per mille), in the region from 
Hammerfest to the \ aranger Peninsula (0.48 — 0.60 per mille), and on 
the Kola Peninsula. These regions are farthest away from the outer edge 
of the submerged continental shelf. 

It is also noteworthy that the gradient seems to increase somewhat 
towards regions where there are deep submerged fjords on the continental 
shelf outside the coast, c. g. outside Træna 'cf. Xansen, 1904, Pi. XI], 
where the gradient was found to be 0.9 per mille, while it was 0.67 and 
0.73 per mille^ in the region of Donna and ^^ega to the south, where the 
shelf outside is broader and less dissected. 

As A. G. Hogboni has pointed out to me in a letter, the investigations 
of several geologists obviously show that there mav be appreciable local 
variations in the upheaval of the land. 

By careful levelling of a very distmct raised l)each i from the Tapes 
period), formed of boulders, along the coast of Lake \>nern in Sweden, 
R. Sandegren [1916] has found that along a distance of about 11 kilo- 
metres, from the region of Kleven (height 65.2 metres above the sea) to 
the region north of Otterbacken (70 metres above the sea), the gradient 
of upheaval is 0.44 per mille, and along the next 19.5 kilometres to the 
XXE, between the region north of Otterbacken to the region east of A'all 
in A'isnum (height 92.1 metres above the sea), the gradient is 1.13 per 
mille. Hence, in the latter region, the land has been elevated about 13.5 
metres more than it would have been, if the gradient of upheaval had been 
uniformly 0.44 per mille along the whole distance of 30.5 kilometres. 

By his measurements of the levels of the ancient beaches left bv the 
ice-dammed lakes in Xorthern Österdal, Gunnar Holmsen [191 5. 1916. 
1917] has found that the gabbro moimtains ma}- have had an influence 
upon the postglacial upheaval of the land which has taken place in this 
region. The isobases are deflected by, and go round, the intrusive masses 
of gabbro, and it looks as if the latter have risen somewhat more than the 
surrounding regions, and the gradient in these regions may thus locallv 
be increased from 0.70 per mille to between 0.85 and 1.63 per mille. 



' J. H. L. Vogt [1907, p. 30], however, remarks that according to later investigations by 
A. Hoel, these values are probable somewhat too low. 



254 



KRIDTJOK NANSF.N. M.-X. Kl. 



(niiin.ir I lolmscii ]iiiiii1s out lliat .\. I Icllaiul's incasurcmcnls of the 
raised sliorc-lincs [Kjooj in llic 'Irom^o dislricl indicate a similar local 
difference in the poslf^dacial upheaval near intrusive masses of g-abbro, 
es]K'ciall\- consi)icu()Us in tlie rej^don of the f.yngen Peninsula. Grønlie 
[1918] also niainlains lliat there has l)een appreciable U^cal differences in 
llie uphea\al of llic land in the Tromso flistrict. 

Il was piH'\ iciusl}' nient ioiK'd that Kaldhol's measurements of the 
hei^^'hts of the "upper marine lindt" in the region of Nord Fjord seem to 
indicate considerable local \'ariations in the upheaval of the land, the 
graillent varving even as much as from 1.2 to 3.6 per mille. His in- 
vestigations in Nordnior [191^)] give similar results. As was previously 
pointed out, however, the difficulty with these determinations of the 
"upper marine limit" is first, that to a great extent they are based on 
terraces of loose material which do not give the exact level of the shore- 
line; secondly, they are marie in fiords, where ice-danuned lakes may easily 
have been formed during lateglacial time, and the terraces may thus have 
been formed at levels much higher than the sea. It is therefore of im- 
portance to have it proved that the terraces actually arc marine. As the 
same terrace cannot be continuously followed from one fjord to another, 
it is often difficult to decide the identity of the terraces. 

Although these numerous observations by Kaldhol and others in the 
fjords may, therefore, have to be carefully sifted, before conclusions of 
wide bearing are drawn from them, still it is hardly doubtful that, on the 
whole, they prove the probability of appreciable local differences in the 
upheaval of the land, and that the gradient of upheaval may change ap- 
preciably inside distances of no more than to to 20 kilometres. 

The investigations of the Swedish geologists also indicate similar 
local differences in the upheaval of the land in Sweden. 

Tvlunthe's and Sundelin's map [Sundelin 1919, PI. X] of the raised 
shore-line of the Ancylus Lake along the east coast of Sweden, in östcr- 
götland and Småland, shows that the gradient of the upheaval of this 
shore-line varies a great deal inside the region of the map. Inside a 
distance of about 90 kilometres the gradient may increase from 0.24 per 
mille in the region of A^ästervik and Oscarshamn, to about 0.50 per mille 
in the region of Söderköping, Norrköping, and Linköping. The gradient 
of the upheaval of the Littorina shore-line seems to differ even still more 
locally in this region. 

Similar local variations of the gradient of upheaval have been found 
at a good many places in Sweden, by the measurements of the heights of 
the "upper marine limit", as well as of the shore-lines of ice-dammed lakes. 

The isobases drawn in Fig. 166 are not intended to show the possible 
local variations in the gradient of upheaval of the Norwegian coast. It 
was rather attempted to give a general idea of the upheaval according 
to the most trustworthv observations. 



1 92 1. No. 1 1. 



THE STRANDFLAT AND ISOSTASY. 



9 20 21 22 zy 




iiccotcUn^ to the IcUesù ßuuAxcs 

V.TANNEJt rai FlnmaUc 
n. CHAMBERS fot Alan -Ham7m*.f€st, 
A.HLLLAND Fol t/vt Hovisö Biskict, 
VO CT y. RIKSTAD Fol NtnM^Lru^, 
H-KALDHOL Fct Notd-mct uNetd- fjotcL 
J. REKSTAD fot SÖTiä'niutjScgn, Hauùa^^ 
D.DANIELSEN fv^ SotUtietm Nctwuff, 
PA. ÖYENfcttfie Ticndhjem utiU 
CJiùstUiTzia K^icrv 

The hobatfis for HOO anä 600 

Metres metlcatc t/ie oiUeredge 
of tka CuTiUneTüal Shelf. 



Fisr. i66. 



256 IKlDljOl- NA.NSKN. M.-X. Kl. 

Relation between the Inclinations of the two Conspicuous Raised 
Shore-Lines of Northern Norway and the Kola Peninsula. 

Ill nnrilicrii Norway tlicrc arc two especially conspicuous raised 
shorc-liiKs, lo a j^rcal cxlcnl cut in solid rock. Amund Heiland [1900] 
pointed out llie fact that llie [)lanes of inclination of these two levels do 
not inlcrscct the sea-level alc^n.^' the same line, zero isobase, outside the 
coast; but in the Troniso region ("JVonis i'\vlke) the hypothetical zero 
isobase of the lower shore-line lies 8 to 17 kilometres further seawards 
than that of the upper one. 

The two planes of inclination "intersect one another along- a line in 
the air a few metres (on the average 5 metres) above the surface of the 
sea". This line lies just over the outermost islands and skerries in the 
Tromso distrikt. 

J. Rekstad states [1905, p. 21] that there is a similar relation between 
the sloping- planes of the corresponding two levels of raised beaches and 
terraces in Helgeland, and also on the Norwegian west coast (between 
60" and 62« N. L.). 

Thinking that this peculiar relation between the two shore-lines 
proves that the postglacial uplieaval did not extend so far seawards 
during its first period, before the time of the lower shore-line, as it did 
during its later period, Rekstad assumed this to be evidence against the 
probability of the isostatic nature of the postglacial upheaval of the land, 
for, he argued, as the load of the ice-sheet was first removed from the 
border regions of the depressed area, one might expect an isostatic up- 
heaval of the crust to begin in those outer regions and gradually extend 
landwards, and not the other way. 

The inference from our studies of the strandfiat and the raised 
beaches — that there may be distinguished between two movements in 
the lateglacial and postglacial upheaval, a vertical change of the hori- 
zontal level of the shore-line as indicated by the strandfiat, and a tilted 
elevation of the land as indicated by the raised beaches — gives a simple 
explanation of the above mentioned relation between the tilted planes of 
the two shore-lines. 

Let us here, in order to simplify matters, assume for a moment that 
the former movement be due to a postglacial sinking of the sea-level, 
while the latter tilting movement is due to an upheaval of the depressed 
land. Let us then take as an example a special case, c. g. Helland's 
measurements of the heights of the two raised shore-lines at Helgoi 
(70" 7' N. Lat.) and at Havnnes (69° 47' N. Lat.) in Lyngen Fjord in the 
northern Tromso district. 

The heights of the two shore-lines were found to be 17 and 8 metres 
on Helgøi and 60.9 and 22.8 metres at Havnnes. Let us assume that 
17 metres is the height of the lower level of the strandfiat in tliis region, 



1 92 1. No. 1 1. 



THE STRAXDFLAT AND ISOSTASV. 



257 




Z 30 ^ JO .0 /0 20 30 ^iokdomeues 50 G 

Fig. 167. Diagram illustrating the upheawil of the shore-lines between Helgoi (Ol and Havnnes \G 

i. c. is the height of the lategiacial and postglacial relative sinking of the 
sea-level. At Mikkelvik on Ringvasoi, about 15 kilometres south-west 
of Helgoi, or very nearly in the direction of the isobases in this region, 
Heiland found a raised shore-line at 10.7 metres above sea-level, which 
is obviousl\" the same as his lower shore-line on Helgoi, 8 metres above 
sea-level. Let us therefore take the mean between the two and make the 
height of tlie lower shore-line in this region 9.3 metres. Let us further- 
more assume that the distance between the isobases drawn through Helgoi 
and Havnnes is about 54 kilometres. 

According to our assumption there has been no postglacial tilted 
upheaval of the land on Helgoi (Fig. 167, O), the upper raised shore -line 
standing at the level of the strandfiat; there has only been a sinking of 
the sea-level. During the first period of upheaval, before the time when 
the lower shore-line was formed, the upper shore-line rose to the line Be. 
The sea-level had then sunk y.y metres to the line DEf, along which the 
lower shore-line ^^as eroded. During the last periorl of upheaval the 
upper shore-line rose to the line BC, tlie lower shore-line to Ef. and the 
sea sank to its present-day level ////. 

The result of these movements is that we now find the upper raised 
l)each along ABC, the lower raised beach along DEF, and the present 
sea-level at HH. 

If we continue the tilted lines BC and EF, they intersect one another 
at .r, () metres above present sea-level. The plane of the upper shore-line 
will intersect the present sea-level at 3', about 13 kilometres inside that 
of the lower shore-line (c). 

Fig. 167 is naturally a simplified diagram. Li reality there will 
probably be a more gradual transition from the horizontal outer parts 
AB and DE to the rising planes inside, BC and EF. 

These results are in perfect accordance with Helland's computations 
mentioned above, anfl the explanation of the facts here given may be 
considered to be fullv satisfactorv. 

On the other hanrl, the facts described above seem to support the 
probability that we are right in assuming that the lower level of the 
strandfiat (about 15 to 17 metres above the sea) actually marks the level 

Vid.-Selsk. Skrifter. I. M.-\. Kl. 1921. No. 11. 17 



258 KUIDÏJOF NANSEN. M.-N. Kl. 

(if tlic sliorf-liiic l)cf(/rc- tlic siihsiflcnce of llic laii'l 'luring the last glacial 
|)C'ri(jfl. 

Tanner's investigations [190^1, 19071 of the raised shore-lines in Fin- 
mark show a relation hetwcen inclinations of the planes of the twfj shore- 
Inies which lie calls It: ami //./, i|uile similar to what ilellanrl has foun'I 
in ihe I'romso district. As will he mentioned later, Tanner's shore-lines 
!r and //. / correspond probably t(j the two well known shore-lines of the 
'Fromso district and Alten. 

Tanner found [1906, p. 130] that the i)lanes of the two shore-lines 
intersect one another about 50 kilometres north of Kvitnes (in the mouth 
of Tana Fjord) and 80 kilometres north of Havningberg (25 kilometres 
north-west of Wardo), at heights of 8 and 7 metres above sea-level. 

The explanation is obviously the same as before. Besides the tilted 
upheaval of the land, there has also been an elevation of the land in its 
horizontal position, or a sinking of the sea-level, which has the same 
effect. And this change of level has probably lieen of about the same 
magnitude as in the Tromsø district and in the region of Nord Fjord and 
Sogne Fjord, or perhaps a trifle higher if we may judge from the height 
of intersection of the two planes. 

In the region, however, of the Arranger Peninsula the plane of the 
upper shore-line {le) will intersect the horizontal plane of elevation (the 
plane of the lower level of the strandfiat) far outside the coast. If we 
assume the height of the latter plane to be 18 metres the distance will be 
about 25 kilometres north of Kvitnes (in the mouth of Tana Fjord) and 
54 kilometres north-north-east of Flavningberg. This might seem to 
indicate that the inland-ice, which pressed the land down, extended far 
seawards outside the coast in this region. 

Ramsay's meastirements show a similar relation between the in- 
clinations of the two planes of shore-lines on the Ribachi Peninsula, on 
Kildin Island, and along the IMurmanski Coast, as well as along the south 
coast of the Kola Peninsula. Along the east coast of this peninsula, how- 
ever, in the region of the mouth of Ponoi River, there seems to have been 
no postglacial upheaval of the land. 

Fig. 168 gives a profile of the heights of the two shore-lines along 
a line at right angles to the approximate direction of the isobases in this 
region according to Ramsay's map [1898, p. 132]. The gradients of both 
shore-lines decrea-se outwards in a similar manner as the gradient of 
Kaldhol's upper limit of submergence decreases seawards in the region 
of Nord Fjord. 

According to recent statements by Ramsay [cf. A. G. Hogbom, 1921, 
p. 137], however, the land at the entrance to the White Sea. should have 
been so much higher than now during the lateglacial period when the ice- 
sheet retreated from that region, that the \\'hite Sea mav have been cut 
off from the Ocean, and transformed into a fresh-water lake, similar to 



I92I. No. II. THE STRANDFLAT AND ISOSTASY. 259 

the Ancylus Sea of the Baltic region. Hence the heights found for the 
upper limit of submergence along the south coast of the Kola Peninsula 
may possibly have to be somewhat corrected. 



Relation between the Altitudes of the two conspicuous Raised 
Shore-lines in the Tromsø — Hammerfest Districts. 

Let us new stufly the relation between the heights of the two levels 
represented by the two conspicuous raised shore-lines in the different 
regions of northern Xorway. 

Tanner [1906, 1907] and Gronlie [1914] are probably right in as- 
suming that the upper raised shore-line (of Heiland) in the Tromsø and 
Hammerfest districts corresponds to the level of the Finmark shore-lines 
which Tanner has called le. 

It seems to me to be probable that this shore-line marks the upper 
limit of lateglacial submergence in these northern regions. If so, it 
probably also corresponds to the level of the upper limit of submergence 
(tlie upper marine limit) observed in the different parts of southern 
Norway. 

It was previously mentioned (p. 251) that, the old raised shore-lines 
and terraces, found l)y Ramsay, Tanner, and Grønlie above the level 
corresponding to Tanner's le, may probably be survivals from a previous 
submergence. 

The lower raised shore-line in the Tromso and Hammerfest districts 
obviously represents the same level as Tanner's line IIA in Finmark. 
There seems to be good reason for accepting Gronlie's and Tanner's 
assumption that this shore-line corresponds to the so-called Tapes-level 
in southern Norway. Grønlie therefore calls this shore-line the Tapes-line. 

Probably owing to the mildness of the Tapes climate, which has not 
favoured the shore erosion by frost, this level is not marked by shore- 
lines cut in solid rock south of the Tromso district (68th parallel), but 
as a rule merely by marine terraces and wave-built shore ridges, which 
are less reliable marks of the actual level of the shore-line existing at 
the time of their formation. The surface of a terrace may have been 
somewhat lower than the sea-level, and a wave-built ridge may have been 
thrown up above high-tide level during storms. 

According to the figures given above, the height of the lower raised 
shore-line (the Tapes level) is 37.4 per cent of the height of the upper 
raised shore-line (the upper limit of submergence) at Havnnes, while this 
percentage is 54.7 on Helgoi if the height of the Tapes-line be 9.3 metres. 
According to what has been said above, it is obvious that this percentage 
should decrease wath increasing height of the upper limit of submergence, 
anrl it is therefore useless to try to find a fixed figure of percentage 
which would suit all cases. 



26o IKIDIJOF NANSKN. M.-N. Kl, 

rro\i'lc'l lliat llic iiplicaval of the laii'l lias increased rej^ularly in- 
laii'l, ami pro\ iflcd tlial llic relation between the inclinations of the two 
planes of the shore lines has been similar to what is represented in Fig. 167, 
llu-n we should be able to conipule ilie probable height of the Tapes level 
at anv place where we know the upiur linnt of submergence. 

l)Ctween llelgoi and llavnnes (I'ig. 167) the upper limit of sub- 
mergence rose 43.9 metres (from 17 metres to 60.9 metres), while the 
Tapes-line rose 13.5 metres (from 9.3 metres to 22.8 metres). Hence, 
for every metre whicli the u[)per limit of submergence rises landwards 
the Tapes-line will rise 0.308 metre. If .r be the lieight above the sea 
of the ]>oint, where the planes of the u])per limit of submergence and of 
the Tapes-line intersect one another, and // be the height of the upper 
limit of submergence above sea-level, the height // of the Tapes-line will.be: 
// = (//— .V) -• 0.308 + X 

According to the above values of // anrl /; on Helgoi and at Havnnes 
.r is = 5.87 metres. Hence: 

// = {H — 5.87 )>< 0.308 + 5.87 

// = //X0.3 + 4 (i) 

By taking tlie mean of Hclland's measurements of both shore-lines 
at six places where the height of the upper shore-line was between 57 
and 62.7 metres and at seven places where it was between 28.7 and 38.4 
metres we find the formula to be: 

/i = //X0.315 + 3.4 I2) 

The values computed by this formula docs not differ much from 
those comj^uted bv the former formula. 

In order to test the correctness of our formula let us take some 
reliable observations of the upper marine limit and the Tapes-level in a 
region as far away from the Tromso district as the Christiania valley. 

At Skådalen Station, on the Holmenkollen tramway, near Christiania, 
P. A. Öyen has found the upper limit of submergence at a height of 220.8 
metres above sea-level. At Cllern, west of Christiania, he found the 
Tapes-level at a height of 69.5 metres. The distance between Skådalen 
and Ullern is about 4 kik^metres at right angles to the probable direction 
of the isobases in this region. If we assume the gradient of elevation 
to be about 0.7 per mille, the height of the upper limit of sul)mergence 
at Ullern should consequently be 218 metres. Hence the height of the 
Tapes-level at Ullern should be: 

h = 218x0.3 + 4 = 69.4 metres 
or 

/i = 2i8-^^<o.3i5 -|- 3.4 = 71.3 metres. 

This agrees well with ()ycn"s value, O9.5 metres, especially con- 
sidering that our formula is based only on some few observations by 
Heiland in the Tromso district. 



1 92 1. Xo. 1 1- 



THE STRANDFLAT AND ISOSTASV 



261 



Localitv 



X. Lat. 



Heiirht in Metres of: 



Upper 
Shore-line 



Lower Shore-line 
observed computed 



Hammerfest Island 
South of Hämmert". 

Vargsund south of 
Hammerfest 



Region outside 

Lyngen Fjord and 

Ulfs Fjord 



Near Tromso 

Lyngen Fjord 

Ulfs Bjord 
Malangen Fjord 
Inside Senjen 

South of Senjen 



J Molstrand 

J^Sjâholmen 

Kvalsund 

IKvænklubben 
Mainland opposite 
Storbekk Fjord . . 

Helgoi-Mikkelvik. . 

Kvitnes, Vannoi . . . 

Lanesoren, Vannoi . 

Karlsoi 

Reinsvoll, Reinoi . . 

Skjervoi 

Finkroken, Reinøi . . 

Glimma, Ringvarøi . 

jMovik 

(Kalsletta, Bals Fjord . 

jHavnnes 

(Rotsund 

Ulsnes 

j Greipstad 

(Ansnes 

Bukskind, Gi-sund 

Bjornera, Grj'toi 

Lundesnes » 

Kjeøi, Vågsljord 

Vinje 

AnderV'åg, A