Skip to main content

Full text of "North Atlantic biota and their history; a symposium held at the University of Iceland, Reykjavík, July 1962, under the auspices of the University of Iceland and the Museum of Natural History. Editors: Askell Löve and Doris Löve. Sponsored by the NATO Advanced Study Institutes Program"

See other formats

J JWjIj^ft 

Marine Biological Laboratory Library 

Woods Hole, Mass. 

Presented by 












A Symposium 

held at the University of Iceland, Reykjavik 

July 1962 

under the auspices of the University of Iceland 

and the Museum of Natural History 


AsKELL Love and Doris Love 


A Pergamon Press Book 


new YORK 

r- 1 n 


60 Fifth Avenue, 
New York II, N.Y. 

This book is distributed by 


pursuant to a special arrangement with 


Oxford, England 

Copyright © 1963 
Pergamon Press Limited 

Library of Congress Cord No. 62-22038 

Printed in Great Britain by Page Bros. (Norwich) Ltd. 

To the bicentennial anniversary of 

25 April 1762 to 23 April 1840 

the distinguished Icelandic naturalist 


pioneer glaciologist 


Foreword, by Askell and Doris Love ix 

Introduction, by Askell Love xi 

Some chapters of the Tertiary history of Iceland, by Trausti Einarsson 

(with 4 figures) 1 

The geological knowledge of the North Atlantic climates of the past, by 

Martin Schwarzbach (with 1 figure) 1 1 

The Atlantic floor, by Bruce C. Heezen and Marie Tharp (with 2 

figures) 21 

Recent studies on the geology of the Faeroes, by Joannes Rasmussen 

(with 6 figures) 29 

Phytogeographical connections of the North Atlantic, by Eric Hulten 

(with 24 figures) 45 

The problem of late land connections in the North Atlantic area, by 

Carl H. Lindroth (with 6 figures) 73 

Taxonomic differentiation as an indicator of the migratory history of 

the North Atlantic flora, with especial regard to the Scandes, by 

John Axel Nannfeldt (with 5 figures) 87 

Phytogeographical problems in Svalbard, by Olaf T. Ronning (with 2 

figures) 99 

Amphi-Atlantic zonation, Nemoral to Arctic, by Hugo Sjors (with 

3 figures) 109 

Distribution of the terricolous Oligochaetes on the two shores of the 

Atlantic, by Pietro Omodeo (with 15 figures) 127 

Historical and taxonomical aspects of the land Gastropoda in the 

North Atlantic region, by Henrik W. Walden (with 3 figures) 1 53 

Plant migrations across the North Atlantic Ocean and their impor- 
tance for the paleogeography of the region, by Eilif Dahl (with 2 

figures) 173 

Dispersal and survival of plants, by Doris Love 189 

On the history and age of some Arctic plant species, by Emil Hadac 

(with 2 figures) 207 

Problems of immigration and dispersal of the Scandinavian flora, by 

Knut F^egri 221 

Survival of lichens during the Glacial Age in the North Atlantic basin, 

by Zdenek Cernohorsky 233 



Recent discoveries in the south Norwegian flora and theirsignificance 

for the understanding of the history of the Scandinavian mountain 

flora during and after the Last Glaciation, by Rolf Nordhagen 

(with 12 figures) 241 

Survival of plants on nunataks in Morway during the Pleistocene 

Glaciation, by Olav Gjserevoll (with 22 figures) 261 

Phytogeography of Greenland in the Ught of recent investigations, by 

Tyge W. Bocher (with 5 figures) 285 

The elements and affinities of the Icelandic flora, by Eythor Einarsson 297 
Ice Age refugia in Iceland as indicated by the present distribution of 

plant species, by Steindor Steindorsson (with 19 figures) 303 

Some comments on the "ice-free refugia" of northern Scandinavia, by 

Gunnar Hoppe (with 5 figures) 321 

Field problems in determining the maximum extent of Pleistocene 
glaciation along the eastern Canadian seaboard — a geographer's 
point of view, by J. D. Ives (with 10 figures) 337 

Pollen-analytical studies on the vegetation and climate history of 
Iceland in Late and Post-glacial times, by Thorleifur Einarsson 
(with 5 figures) 355 

Palynology and Pleistocene ecology, by Gunnar Erdtman (6 plates) 367 

The Svinafell layers. Plant-bearing Interglacial sediments in Oraefi, 
southeast Iceland, by Sigurdur Thorarinsson (with 4 figures and 7 
plates) 377 

Conclusion, by Askell Love 391 

Appendix 399 

Author Index 403 

Subject Index 409 


pat er ok mannsins nattiira at forvitna ok sja pa hluti, er hanum eru sag5ir, ok 
vita, hvart sva er sem hanum er sagt e5a eigi. 


It is in man's nature to wish to see and experience the things that he has heard 
about and thus learn whether the facts are as told or not. 

King's Minor 
(Transl. by L. M. Larson: Scandinavian Monographs, vol. Ill, New York, 1917. 

These lines from the King's Mirror, the important Old Norse book which 
contains, among other things, valuable information on geography and 
natural history, express better than we can say what has been in our minds 
when we were editing this book. It is intended not only as a presentation of 
the current status of our knowledge on the distribution and history of 
plants and animals in the North Atlantic area, but also as a review of current 
trends in biogeographical and geological investigations concerning these 
problems. The papers included were delivered at a symposium held at the 
University of Iceland. Reykjavik, 12-25 July 1962. During the meeting 29 
lectures were given, followed by discussion periods, and one long and several 
smaller excursions were made to parts of Iceland of interest in this connection. 

Although the subject matter of the symposium. North Atlantic biota and 
their history, has been much discussed for almost a century, this is the first 
attempt to present in one volume a reasonably many-sided evaluation of the 
problems involved. It is evident that although many speciahsts from different 
branches of the life and earth sciences are represented in this book, the 
number of contributors had to be limited and, thus, also the points of view. 
There is, however, reason to believe that this compilation will be of value to 
those students who specialize in similar problems, and that it will help 
forward their research. Since the situation of Iceland and its unique Tertiary 
and Pleistocene deposits seem to make this country ideal for such studies, it 
was selected as the meeting-place for the symposium in the hope that this 
would advise the scientific world about the importance of much increased 
scientific studies of this and other North Atlantic "stepping stones". 

The symposium was organized by a small committee consisting of Askell 
Love, president; Armann Snaevarr and Sigurdur Thorarinsson, vice-presi- 
dents and representing, respectively, the University of Iceland and the 
Museum of Natural History in Reykjavik; Eythor Einarsson from the 
Museum of Natural History, secretary and organizer of the excursions ; and 
Sigurdur J. Briem, representing the Icelandic Ministry of Education. 



The committee acknowledges gratefully the encouragement and financial 
support received from the NATO Advanced Study Institutes Programme, 
which made possible both this symposium and the publication of this book. 
It is also indebted to the University of Iceland and the Museum of Natural 
History, under whose auspices the symposium was organized. Its gratitude is 
also extended to the Pergamon Press for valuable assistance in editing the 
papers, and to Miss Virginia Weadock, who had the task of correcting the 
language of the foreign manuscripts. Last but not least all the contributors 
are to be thanked for having given of their time and experience to make this 
venture possible. 

Montreal, August 1962 Askell Love 

Doris Love 



Institut Botanique de I'Universite de Montreal, Montreal, Canada 

The dispersal of plants and animals is one of the great problems in the field 
of evolution. It is also one of the most fascinating questions of the biological 
sciences, since it is concerned not only with understanding a distant past but 
also with knowledge of present conditions. In addition, it is not a problem to 
be solved by the biologist alone; his conclusions must be confirmed by aid of 
palynology telling us about biological and climatical changes in the recent 
past, and by aid of historical geology regarding the more distant past. The 
history and evolution of all living beings is closely related to their distribution, 
which in turn is intimately associated with the geological history of continents 
and oceans. 

It has long been known that a considerable number of species of plants and 
animals belonging to seemingly identical species inhabit both sides of the 
Atlantic Ocean. Already Humboldt has raised the interesting question 
whether any of these species are originally common to both continents or 
whether those species, externally so similar as to be known by the same 
name, are in fact identical to each other. 

The analogy of the animal kingdom seems to favor the negative of this 
question, since no quadruped or terrestrial bird, and even no reptile and not 
even an insect is said to be naturally common to the equinoctial regions of 
the Old and New Worlds. The same may be true also for higher plants in 
these regions. But as we go farther north and approach higher latitudes, the 
probabihty of finding animals and plants of identical species on both sides of 
the Ocean becomes increasingly greater. In northern Europe there are rare 
plants, not related to any others on that continent, but of identically the same 
species as are widespread in North America; similarly, there are plants in 
eastern North America, whose closest or even identical relatives all occur in 
Europe. The same is also the case for some lower animals which are unable to 
fly or swim across the Ocean and are as confined to a terrestrial habitat as 
ever any plants. 

The observation that identical animals and plants occur on both sides of the 
North Atlantic led, late in the last century, to the launching of a theory of 
so-called Pleistocene survival. However, this theory did not solve any ques- 
tions, it only moved the problem farther back in time. The outstanding 
unsolved problem in historical biogeography in the North Atlantic still is, 



whether certain flora and fauna elements on both sides of the Ocean reached 
their present areas by dispersal over the existing lands of the continents and 
subsequently became extinct in interior parts of these lands where they do 
not appear today, as maintained by some biogeographers, or, whether these 
continents in a not too remote past were in direct contact with each other; 
furthermore, in case such a contact existed, whether they might have been 
united by land-bridges that have later mysteriously disappeared or by a 
continuity that has been subsequently broken by some kind of a displace- 

In recent years so many new facts have been brought into hght regarding 
the old problem of the history of the North Atlantic biota that a fresh attack 
on their problems seems almost overdue. It also seems as if we were nearing 
the stage when the geological information permits us to construct a somewhat 
better timetable for possible dispersal periods. Even the biological facts and 
methods concerning studies of the evolution of the biota themselves have 
increased and improved so that we now are able to evaluate their true 
relationships better and attack many of the problems experimentally. 

Since the problems of the history of the North Atlantic biota cannot easily 
be understood by studying the continental conditions alone, it seemed 
advisible to search for a concrete foundation of the discussion and to restrict 
this broad subject in space and time. This is one of the reasons that we held 
this first symposium in Iceland, since this automatically concentrates the 
discussions to studies of the present conditions in light of our knowledge of 
the Pleistocene and the Tertiary. Not only is this island a "stepping stone" 
between the continents, but it may in itself harbor such data that are the key 
to the solution of many of the problems to be considered. 

This symposium was organized to clarify what are the facts, or to permit 
the presentation of available evidence on certain phases of an important 
scientific problem. It is meant as an opportunity to state in one issue what has 
been gathered and to gain new insights and outlooks from different fields of 
science. It is, however, not meant to prove a certain theory or to disprove 
another. A scientist's aim in a discussion with his colleagues is not to persuade 
but to clarify. It is the hope of the organizers and sponsor of this meeting that 
we will leave it better informed than we came, ready to reconstruct and 
construct anew on the foundations laid down here. 



University of Iceland, Reykjavik, Iceland 

The oldest rocks in Iceland are Tertiary plateau basalts. These basalts, which 
form the western, northern, and eastern parts of the country, are remnants of 
a 5-7 km thick pile or plateau of lava flows. The plateau must originally have 
had a much greater extension than the present island, as is evidenced by the 
truncation of the plateau at the coasts. The lowest accessible parts are of 
Early Tertiary, probably Eocene or possibly Uppermost Cretaceous age, 
while the highest parts are of Upper, possibly Uppermost Tertiary age. The 
lavas were formed entirely on land. Their base is unknown. 

The Icelandic basalts are usually grouped with similar rocks in north- 
western Britain, the Faeroe Islands, Greenland, Spitsbergen, and Franz 
Josef's Land under such names as the North Atlantic, Brito-Arctic, or 
Thulean plateau basalts. The lowest parts of the Icelandic plateau and the 
basalts in Britain, the Faeroe Islands, and Greenland seem to be of a similar 
age, whereas the basalts of Spitsbergen are somewhat older. In Iceland no 
marine fossils, nor fossil land fauna, are associated with the main part of the 
plateau and the age is based on lignites which are found at a number of 
horizons. This means that the age determinations are considerably uncertain. 

The assumption was made long ago that the plateau lavas were formed on 
an extensive land mass occupying the area of the present North Atlantic or 
at least large parts of it, and successive foundering of much of the land has 
been postulated. The question is naturally important in the present context 
and we shall therefore consider some aspects of it, although it must be said 
at the outset that we can say very Httle with any certainty. 

Postulation of a large-scale foundering of a continental area cannot be 
made without consideration of the isostatic equilibrium. This leads to specula- 
tion about processes in deep-crustal or sub-crustal regions. Foundering could 
have taken place if a relatively light deep layer grew thinner by spreading or if 
such a layer became denser by crystallization. Thus, foundering is not defin- 
itely excluded, but by postulating it we introduce inevitably deep- or sub- 
crustal processes of a wholly hypothetical character. 

Paleontologically, there is no strong evidence that Iceland was connected 



with Europe or America in the Lower Tertiary. On the contrary, the flora 
suggests lack of such a land connection. 

There is, in my opinion, one main evidence which suggests rather great 
Tertiary changes in the North Atlantic area. I am referring to the sea-bottom 
topography. This looks relatively fresh and is suggestive of considerable 
tectonic changes which might have taken place in the Upper Tertiary. There 
is one important tectonic event of which we know: the uphft of Iceland, 
Greenland, Scandinavia, and Britain from low lands to the present moun- 
tainous countries took place in the Upper Tertiary. At the same time it is 
possible that the ocean floor changed considerably. 

The geophysicists are today inquiring into the origin and history of 
ocean basins by way of seismic studies and through the examination of 
bottom sediments. It is advisable to await more such investigations in the 
North Atlantic and Arctic area before drawing further conclusions regarding 
the geographic development in this area in the Tertiary. But palynological 
studies also seem quite promising. 

As matters stand today, we must face the possibility that the North 
Atlantic Ocean existed in the Lowest Tertiary and that the various remnants 
of plateau basalts indicate as many separate and distinct volcanic regions. To 
see, for instance, how Iceland might have originated in a deep ocean, the 
following hypothesis may be considered (cf. Einarsson, 1960): 

Submarine volcanism produced a pile of relatively light pyroclastic material, 
and in spite of corresponding isostatic sinking of the base of the pile, a large 
island would eventually be formed. Assuming at the beginning a 3 km deep 
ocean, then a 6.2 km thick pile of density 2.2 would reach the surface of the 
sea. Adding a 6 km thick pile or plateau of subaerial basalts of mean density 
of 2.7 would, in equilibrium, give a land surface at 1100 m above sea level, if 
there was no compaction of the underlying pile. 

I shall now consider the Icelandic rocks more specifically. Intercalated 
between the lava banks are very often thin seams of terrestrial sediments. 
They consist mostly of windblown sand or dust, whereas conglomerates are 
very rare, except in the higher parts. The sediments of the lower parts hence 
suggest in the first place a dry lava desert, but in a few cases one also finds 
indications of the existence of rivers. The clearest case, I think, is the well- 
known locality Brjanslaekur. As shown in Fig. 1 we find evidence that three 
times a lava filled up the river bed until at last more intense volcanic produc- 
tion so completely altered the drainage that sedimentation at this place came 
to an end. 

In the Skardsstrond area we have sediments with a total thickness of about 
50 m. A few lavas flowed during the period of sedimentation and one of 
these is seen to have flowed over soft mud. The lignite seams found in this 
area are very variable from place to place. There is a seam of 80-90 cm 
thickness at one place (Tindar), whereas at another place (Nipur) there is a 


succession of many 5-10 cm seams separated by sandstone. I think that 
these sediments have been formed by a sluggish river in a wide flat depression. 
As long as there was only an occasional lava flow at extended intervals the 
course of the river changed little. But in the end more intensive volcanism set 





1 1 1 1 1 

5»in>-v ."^ 



Fig. 1. Section through the western part of the Brjanslaekur sediments (looking 
S.-SW.). On top of a lava (1) of normal magnetization, there is a mainly barren 
lower part (2) of sediments: dark, brown, and yellow clay. A new, normally mag- 
netized, lava (3) filled a groove (water course) which had been eroded into the 
lower sediments. This lava was previously considered a sill, but distinct magneti- 
zation of the clay at the lower contact, and absence of any magnetization at the 
upper contact as well as the difference of the under- and overlying sediments tend 
to confirm the present interpretation. The main fossiliferous sediments (4) follow. 
The lignite forms many thin layers, which are embedded in clay and consist to 
a considerable part of wood chips and numerous leaves, apparently carried by a 
sluggish river and buried in its muddy bottom. A new lava (5), of normal 
magnetization, again filled the water course and covered its surroundings. The 
lava has a blocky (kubbaberg) structure and is partly a typical "palagonite" 
breccia as might be found in the much younger "Moberg" formation. The river 
dug a new shallow bed at the same place as before and deposited a new layer (6) 
of coarser, fluviatile material. Again a lava (7), of reverse magnetization, filled 
the bed, but a clear trace of this is now lost in the section. A coarser sediment, (8), 
mainly non-fossiliferous, was deposited before a new and more intense volcanic 
period, represented by basalt lavas (10) put an end to sedimentation in this 
locality. At (9) is an intrusive basalt of reverse magnetization. 

in, there came a rapid succession of lava floods, with the result that sedimenta- 
tion came to an end here, i.e. the river was diverted to another area or 
perhaps it disappeared completely into the porous group of young lavas, the 
water flowing as ground water and not as a river. 

When speaking of a lava desert, I have in mind conditions somewhat 
similar to those of the present Odadahraun lava field. All the precipitation is 
lost into the porous lavas and flows as ground water on a deeper impervious 
floor until it emerges as large springs in the outskirts of the lava field. Within 


the dry lava field the lavas weather into sand or dust that is moved by the 
wind and settles in the depressions. Vegetation is naturally very sparse here, and 
fossilization would be most unhkely. Only around the springs at the rim of 
the lava field does the vegetation thrive and here it could be fossiUzed. 

The picture of the Lower Tertiary landscape which I have in mind is that 
of a very extensive flat lava desert. It is mostly dry and without vegetation, 
only in occasional depressions is there sufficient moisture to sustain an 
oasis. Rivers are formed during long periods of relative volcanic quiescence 
and deposit sand and mud in their lower parts. But when a new intensification 
of volcanism sets in these rivers may disappear. Thus the vegetation is subject 
to often repeated changes in the water supply. While an oasis or a river bank 
is destroyed by a lava flow in one place, a wet depression is formed far away 
at the front of a new lava flow. The species we would find here after a short 
time possibly give no true picture of the climate or the vegetation of the 
country as a whole but may rather reflect which species migrate most rapidly 
into a new, isolated, wet place. 

Practically every place, where we can study a thick section of the plateau 
basalts, we find that alternating with sections where a thin sediment separates 
every two lava banks, there are one or more such parts of the section, con- 
taining 10-20 lavas, with no sediments at all. These lava successions seem to 
represent great intensification of the volcanic activity and it seems most 
likely that during such periods vegetation must have been largely destroyed 
over an extensive area. 

How large such an area may have been we cannot easily answer at present, 
because the individual lava groups have not been traced in sufficient detail and 
over sufficiently large areas. 

In the lower parts of the plateau, traces of frost action are unknown, but 
in the topmost plateau group such signs are common. In addition there occur 
moraine-like conglomerates, and sometimes distinct moraines, resting on a 
striated floor. The conglomerates probably are mainly of fluvial nature, 
but they often seem to have been reworked by frost. In some cases they also 
show clear signs of erosion by sandstorms, among them the so-called "drei- 
kanters". Thus, in this topmost group we have clear signs of a severe climate. 

The age of this group is a very important question, but one not yet fully 
settled. It may be of Upper Pliocene age, and this is suggested by some 
approaches to the problem. On the other hand this plateau group is older than 
a complete peneplanation of the country, a differential uplift, a modelling of 
the landscape at a base level some 300 m above the present one, and finally a 
general uplift of 300 m in two or more steps, and a grading of the landscape 
to the lower base levels. One of the later episodes in this story was the forma- 
tion of a strandflat at a sea level some 100 m above the present one, and after 
this strandflat had been formed there flowed lavas that have reverse magnetic 
polarity which indicates Lower Pleistocene age. 


If we were free to base our estimate of age on this tectonic and erosional 
history, it seems most Hkely that a Lower PHocene or even Miocene age 
should be assumed for the topmost plateau group. But the relations to the 
Upper Pliocene sediments in Tjornes, although not at all very clear, seem to 
indicate a Late Pliocene age for the plateau group. This would demand on the 

Fig. 2. Localities for fossil plants (dots) and marine fauna (crosses) in Iceland 
from earlier than Late and Post-glacial times. A few uncertain localities of 
surtarbrandur on Snaefellsnes and the Northwest Peninsula have been ignored. 
Hatching indicates the areas in which Pflug's "first type" of Icelandic flora occurs 
(cf. the text, p. 5). L Brjanslaekur; 2, Skardsstrond ; 3, Litlisandur; 4, Hredavatn; 
5, Stafholt; 6, Sleggjulaekur; 7, Tjornes; 8, Vopnafjordur; 9, Jokuldalur; 10, 
Bessastadaa and Hengifoss; 11, Lungufell; 12, Holmatindur; 13, Gerpir; 14 


Other hand that peneplanation and valley erosion in Iceland had been extreme- 
ly rapid, and this is not at all easy to comprehend. Absolute dating of the 
topmost plateau group is very much to be desired. 

A fossil flora is found at a number of horizons in the plateau (Fig. 2). In 
the lowest ones, a temperate or warm-teinperate climate is indicated. But 
with increasing height the flora takes on a cooler character. Pflug (1959) has 
divided the flora into several types: 





••"■• ':•■.'•-. "v-- 


I . . . .-. ». . ;... 



.":."• •'..'. ; -'. ■'• 



■• ■'•■■■' '.■'••'. 


^^7T '-^"^^^~*'. ■?- 





■•■,»<■«>.• •A' .\i: 


.■■H'- TT' 

iTi i 1 1 i 



Ui I 

1 1 1 1 1 1 1 


T'"-' •■ ■•-■ 

300 m 

Fig. 3. The Litlisandur Glacial Horizon, 250-300 m above 
sea level, (a) Brown sandstone; (b) some lavas with strong 
reverse magnetization, basis of the regional magnetic group 
R^; (c) light and dark brown clay and fine sandstone, 20 m 
thick, the surface of which is distinctly glacially striated 
(direction N. 50" W.); (d) grey-brown conglomerate with 
striated basalt cobbles (moraine), J m thick; (e) a 5 m thick 
layer of reddish-grey clay with scattered striated basalt stones 
(moraine), the reddish color of which was acquired probably 
before the deposition of (f ). The surface of (e) is broken up 
(frost action), and debris from (e) is incorporated in (f); (f) 
conglomerate containing scattered basalt blocks, larger than 
are found in (e); (g) a 1 m thick, brown conglomerate with 
glaciallystriatedbasaltblocks(moraine);(h)a 1 mthickvarve- 
clay with grey and reddish layers; (i) brown sandstone, 2 m 
thick; (j) 1-2 m of fine basalt gravel with numerous rhyolite 
pebbles; (k) a 4 m thick brown sandstone layer; (1) a thin 
layer, 10-20 cm thick, of greenish-grey clay, possibly mud 
from the bottom of a lake; (m) 1 m light brown loess with 
plant remains (leaves), suggesting willows, birch and alder; 
pollen has not been found; (n) 4 m of brown sandstone; 
(o) a very thick layer of primary volcanic "palagonite" 
breccia of reverse magnetization. 

250 m 


(1) A Deep-Lower Tertiary flora of warm character. 

(2) A mixed flora with warm and cool elements. 

(3) A Tertiary flora of Hoifell and Tjornes type, ranging perhaps from 
Upper Miocene to Upper Pliocene. 

(4) An Upper Tertiary and Pleistocene type flora. 

(5) A Pleistocene type flora. 

This scheme represents in a very broad sense a chronological order. 
But it has been revealed that rather early there were considerable fluctuations 
between "warm" and "cool" floras, whether or not this is a reflection of 
climatic alternations or of the repeated and rapid changes in the other external 
conditions which 1 have mentioned earlier. 

In western Iceland the localities of Skardstrond and the Northwestern 
Peninsula belong to Pflug's first type. Hredavatn and Stafholt in Borgar- 
fjordur still belong to this group, but Sleggjulaekur, a little higher than 
Hredavatn, pertains to the second type. In the topmost plateau group we then 
have the horizon of Litlisandur (Hvalfjordur) with a leaf bed in loess above 
varve-clay and a moraine on a clearly glacier-striated floor (Fig. 3). This is 
the lowest known glacial horizon in western Iceland. Above it are certainly 
three reverse magnetic periods and three normal ones, and more probably the 
total number of magnetic periods above this horizon is closer to 10. Pollen has 
not been found in this horizon nor have the leaves been analyzed by experts. 
They seem to belong to willows, birch, and alder. 

Fig. 4. Drawing of the impression of a 
plant on the lower face of a basalt lava. 
The plant stem is about 25 cm long, and 
the width of it and the branches is 3-5 
mm From Grafardalur near Hvalfjor- 
dur. The lower surface of the lava must 
have consolidated before the plant was 

In eastern Iceland the lowest plant horizon is that of Gerpir. It is considered 
as Lowest Tertiary or even Uppermost Cretaceous by Pflug (1959). 

On top of this we have a 4500 m thick pile of lavas that have been mapped 
by Walker (1959). Near the top of the pile, in which Walker did not find any 
unconformity, are the lignite seams of Holmatindur and Tungufell. The 
former seems to belong to the Lower Tertiary on palynological grounds 
(Schwarzbach, private communication) and this is in keeping with the lack 
of an unconformity between it and Gerpir. But a widely divergent view has 
been expressed concerning Tungufell to which I shall return soon. 

Farther inland we have still higher members of the plateau basalts. These 


form a thick group that contains hgnites of a much cooler character than that 
of Gerpir; they belong to Pflug's third or fourth types. Rather high up in this 
group the locahty Bessastadaa has a flora comparable to that of the warmest 
zone of the Phocene Tjornes sediments (Pflug, 1959). In the close neighbor- 
hood, and almost certainly below Bessastadaa, the locahty Hengifoss shows a 
flora of a much cooler character. Pflug interprets it as a cold period at the 
beginning of or at the end of a glacial time. 

A number of localities at Jokuldalur and Vopnafjordur, which belong to 
this group of basalts, have a flora comparable to that of Bessastadaa, but 
it has also been found that the flora is horizontally quite variable (Jux, 1960). 
Pflug and Jux place this flora into the Upper Tertiary, or even the Upper 

Let us now return to Tungufell. According to the mapping by Walker, this 
locahty should be close to the Holmatindur lignite and we must put this far 
below the above-mentioned Upper Tertiary inland basalts. However, Meyer 
and Pirrit (1957) give an Upper Phocene or Lower Pleistocene age for 
Tungufell on a palynological basis. Also Jux (1960) concludes that the 
Tungufell flora shows a striking similarity to the flora of the inland basalts, 
which he is inclined to put into the Upper Pliocene. The conditions are thus 
quite perplexing. A possible solution of the difficulties is the assumption that 
the warm flora of the Lowest Tertiary in Iceland was very soon replaced by a 
much cooler flora that, with httle variation, persisted throughout all the rest 
of the Tertiary. 

In the Icelandic lignites Pflug has found some 50 poflen species, including 
nearly 20 new ones, i.e. not previously found elsewhere. He points out that 
whereas European and North American pollen types of this time are considered 
nearly identical, the Icelandic types show marked differences. They present 
closer affinity with the Paleocene of Spitsbergen and with the Lower Tertiary 
of Japan. Macroscopic remains have also rendered about 50 species if we 
include the old determination made by Heer (1868). 

The results of Pflug's work suggest that in the Lower Tertiary, and perhaps 
stiff later, Iceland was not connected with lands in the east or west but 
instead there were rather connections with more northerly lands. 

Finally, it may be recalled that, on the basis of the Eocene flora. Chancy 
(1940) concluded that the poles and the continents in the Arctic to Sub- Arctic 
areas were practically at their present relative position. The same data would 
seem, among others, to indicate that the influence of the Atlantic waters was 
felt in Spitsbergen as it is today. Provisional results of paleomagnetism in 
Iceland (Th. Sigurgeirsson, private communication) give for the Lower 
Tertiary pole a position of about 75°N., 70°W. (Smith Sound) which would 
imply a latitude of 70° for Iceland, instead of the present 65°. On the other 
hand, the paleomagnetically located pole for the Upper Tertiary in Iceland 
(Sigurgeirsson, 1 957) shows no sure difference from the present geographic pole. 



AsKELSSON, J. (1954). Some Tertiary plants from Iceland. Ndttiinifr. 24, 92-96. 

Chanf.y, R. W. (1940). Tertiary forests and continental history. Bull. Geol. Soc. Amer. 

51, 469-486. 
EiNARSSON, Tr. (1957). Der Palaomagnetismus der islandischen Basaltes und seine strati- 

graphische Bedeutung. Neues Jahrb. Geol. Paldontlog. Mh. 159-175. 
EiNARSSON, Tr. (1960). The plateau basalt in Iceland. Int. Geol. Congr. XXI. Guide to 

excursion No. A2 (Iceland): 5-20. 
Heer, O. (1868). Flora fossilis Arctica I. Ziirich. 
Jux, U. (1960). Zur Geologic des Vopnafjord-Gebietes in Nordost-Island. Geologie 9, 

Bh. 28, 1-57. 
Meyer, B. L. and Pirrit, J. (1957). On the pollen and diatom flora contained in the surtar- 

brandur of East Iceland. Proc. Roy. Soc. Edinb. 61, 262-275. 
Pflug, H. D. (1959). Sporenbilder aus Island und ihre stratigraphische Deutung. Neues 

Jahrb. Geol. Paldontolog. Abh. 107, 141-172. 
Schwarzbach, M. (1955). Allgemeiner tJberblick der Klimageschichte Islands. Neues 

Jahrb. Geol Paldontolog Mh. 97-130. 
Schwarzbach, M. (1959). Die Beziehungen zwischen Europa und Amerika als geolo- 

gisches Problem. Kolner Universitatsreden, No. 23 : 1 -39. 
Schwarzbach, M. and Pflug, H. D. (1957). Das Klima des jiingeren Tertiars in Island. 

Neues Jahrb. Geol. Paldontolog. Abh. 104: 279-296. 
Sigurgeirsson, Th. (1957). Direction of magnetization in Icelandic basalts. Phil. Mag. 

Suppl. 6 (22), 240-246. 
Walker, G. P. L. (1959). Geology of the Reydarfjordur Area, Eastern Iceland. Quart, J. 

Geol. 114, 367-393. 


Martin Schwarzbach 

Geologisches Institut, Universitat Koln, Cologne, Germany 

The climatic history of the northern Atlantic is of great importance for the 
understanding of the history of the biota in this region. But we can recon- 
struct it only in fragments. This is caused partly by the fact the the sea- 
covered areas of the Earth give only few outcrops and insights to the geologist ; 
the islands are an exception, but there are not many in the northern Atlantic. 
It is true that the situation has improved since it is now technically possible to 
bring up cores from the ocean floor. Thus we can gain at least some additional 
facts about the Quaternary and in part the Tertiary too, and this is, of course, 
the time which is especially interesting for us. But as concerns the Pre- 
Tertiary period, we can rely only on the adjacent continental areas, i.e. 
Europe and North America, and on Greenland and some other islands. 

Here we have to wholly disregard Iceland, at least at first, for there are no 
known sediments older than Tertiary. The oldest rocks of the island are 
basalts. It is not impossible that this volcanism began in the Upper Cretaceous 
as it is supposed to have done in Greenland, but that is without special 
importance for the climatic history, and only the plant-bearing Tertiary beds 
of Iceland reveal something about it. 

We can distinguish 4 or 5 great divisions in this climatic history. Their 
time-span varies greatly, and they are known to be very different in kind. 
In part they differ considerably as regards climate. Their respective boundaries 
are arbitrary, partly conditioned by the state of the investigation. These 
divisions are : 

1. Pre-Cambrian, 

2. Eocambrian. 

3. Paleo- and Mesozoic. 

4. Tertiary. 

5. Quaternary. 


The Pre-Cambrian, i.e. the period which is older than 600 million years. 
can be treated in a few sentences because we know nearly nothing of it. 
Neither the large Pre-Cambrian areas of Scandinavia nor those of Canada or 



Greenland with their generally highly metamorphic rocks give us sufficient 
paleoclimatic information. Moreover, we do not know the paleographic 
situation in the North Atlantic region at that time. It is true that the Lewisian 
gneisses of the Scottish Hebrides are regarded as equivalents of the Canadian- 
Greenlandian shield. Hans Stille (1958), for instance, draws an eastern 
projection of this shield which comprises the northern Atlantic including 
Iceland; he called it "Laurentia Minor" (in analogy to the projection of Asia 
Minor). But that is an hypothetical, though possible, idea. 

It is, therefore, impossible to say how far the North Atlantic region was 
temporarily influenced by a glacial climate like the one presumed for Canada 
in Huronian time — perhaps 1 billion years B.P. — taking the cobalt tillites into 


The first important fix-point at the turn of the Pre-Cambrian to Cambrian 
is in the so-called Eocambrian. This period is short, extending approximately 
some 100,000 or, at most, some million years; it has an age of perhaps 600 
million years. But in many places it is characterized very well by moraine-like 
sediments, the so-called "tillites". There is no doubt that many of these are 
not true morainal deposits but pseudo-tillites. The stratigraphical position is 
also uncertain in many cases, but there remains a lot of locahties where we 
must suppose glacial activities, especially in regions surrounding the North 
Atlantic. Investigation of those phenomena started first in the Norwegian 
mountains. Now we do also know such deposits from Sweden, Spitsbergen, 
eastern and northern Greenland, and farther — but more doubtfully- — from 
the British Isles, Normandy, and eastern Bohemia. At typical localities we 
find poHshed and striated boulders, striated pavement, and superposition of 
fossiliferous Cambrian. Not so clear are the occurrences in North America. 
But the above-mentioned tillites found between Greenland and Sweden make 
it nearly certain that there was an Arctic climate with big glaciers or inland 
ice in the North Atlantic area during the Eocambrian. 

It is remarkable that nearly all these tillites are situated in regions with a 
recent or Pleistocene glaciation. Therefore they should actually present no 
more problems than the Quaternary Ice Ages do. However, there are also 
occurrences in other continents, and these make the Eocambrian Ice Age a 


The following division of the Paleo- and Mesozoic comprises ca. 500 
million years — a very long period. But it is possible to treat it collectively for 
it presents rather uniform chmatic features. There is much climatic informa- 
tion available in Europe and North America, Greenland, and Spitsbergen. 


Not only floras and faunas, but also sediments give valuable information, 
especially the rather thick Hmestones and the evaporites. 

If we consider only Greenland and the other Arctic islands in this connec- 
tion, we must mention the limestones of the Cambrian and Ordovician, the 
Old Red beds in east Greenland, the coals of the Upper Devonian and Lower 
Carboniferous on Bear Island and Spitsbergen, gypsum beds in the Upper 
Carboniferous of Spitsbergen and Greenland, reefs in the Rhaeto-Liassic of 
Jameson Land, in the Cretaceous of Kome, Atane, and Patoot in western 
Greenland, Spitsbergen, and King Charles Land. 

All these occurrences prove the same thing : that for a very long time there 
was a climate completely different from today. There was no Arctic; not 
just a moderate, but a warm or very warm chmate existed. In another connec- 
tion 1 have explained comprehensively that the North Atlantic belonged to the 
tropical reef-belt during the Paleozoic; Franz Lotze (1957) has shown that 
also the Paleozoic belt of the northern evaporite zone, i.e. the hot desert belt 
of the Earth, was situated there. In the Mesozoic, the temperatures were 
already somewhat lower, e.g. in Greenland, and the northern boundary of 
the reef corals shifted farther to the south. But the Mesozoic floras prove that 
very favorable chmatic conditions were still present. 

The Rhaeto-Liassic floras of Jameson Land alone have revealed some 200 
diff'erent species. I mention further the famous leaves and fruit of the bread- 
fruit tree, Artocarpus, described in 1890 by Nathorst from the Cretaceous of 

I would hke to add that the modern direct measurements of temperature 
with the aid of O^^jO^^ isotopes fit in rather well with the geologically deter- 
mined climate. Belemnites from the Scottish Jurassic gave sea temperatures of 
17-23°C, from Alaska of 17°. That is somewhat more than 10° higher than 

It is not yet possible to say if there were large cUmatic fluctuations in the 
Paleo- and Mesozoic of the northern Atlantic. Their existence seems possible, 
especially at the turn of the Carboniferous-Permian with the big inland 
glaciations of the Southern Hemisphere. But the influence of the Gondwana 
Ice Ages must not have been large, in analogy with the small influence which 
the Quaternary glaciations did show in the tropics. 

In any case, we do not find positive indications for cool temperatures in the 
northern Atlantic and its surroundings. There are almost no tilhtes of 
Paleo- or Mesozoic age. What has been described as such is very uncertain, 
taking for instance the much-mentioned Squantum tillite near Boston; its 
glacial origin is as doubtful as its stratigraphic position. 

This is not the right occasion to discuss the cause of the warm climate in 
the North Atlantic region during Paleo- and Mesozoic time. But it must be 
emphasized that another position of the pole and the equator would be a very 
good explanation, at least in the Paleozoic. This fits in very well with modern 



paleomagnetic results as well as with paleoclimatic reconstructions in other 
continents. The climatic map of Devonian time which I published some time 
ago may illustrate this problem (see also Fig. 1). 

Recent climatic 






P a 

1 e o 

200 100 


million /s. BP 


Fig. 1. The climatic position of Iceland, according to geological results, during 
the last 600 million years. Only the recent position (shown in black) is definite, all 
others are more or less hypothetical, and more so the farther back we go into the 
past. Paleomagnetic data were not used in the construction of the figure. The 
shifting position may be due (especially in the Paleozoic time) in part to continental 
drift. Design by Dr. L. Ahorner. 


There is no sharp boundary between the Mesozoic and the Tertiary, a 
period which lasted from ca. 70 to 1 million years B.P. At least in the Lower 
Tertiary we find climatic conditions similar to those in the Cretaceous and 
the temperatures of high latitudes were much higher than today. There 
existed no polar ice caps, but instead rich tree vegetation even on the islands 
nearest the pole. 

Nevertheless we will consider the Tertiary separately; first, because the 
climate changed decisively in the younger Tertiary; second, because we now 
have much more and better climatic indicators, also from Iceland, as 1 
mentioned earlier. 

The Tertiary tree floras of Iceland were the very first to be known, for as 
early as in 1772 (190 years ago), Eggert Olafsson from Iceland carefully 
described the plant impressions of Brjanslaekur in northwestern Iceland. 
Afterwards many other polar floras were discovered. ''Polar flora" means all 


Arctic floras, especially of the Tertiary, but partly also of Pre-Tertiary age, 
which contrast so impressively with the poorer recent vegetation. The pioneer 
paleo-botanist of the polar floras was the Swiss Oswald Heer who pubUshed 
his famous Flora fossilis arctica in 1868 and later. 

We must mention here especially the following places where Tertiary 
Arctic floras have been found: Iceland. Greenland, Spitsbergen, King 
Charles Land and Grinnell Land. 

The southernmost occurrences are those in Iceland. Heer (1868) cited 41 
plant species from here; he and other authors mention Piinis, Picea, Abies, 
Tsuga, Sequoia, Cryptomeria, Liriodendroii, Lawns, Sassafras, Platanus, 
Planer a, Dombeyopsis. Acer, Rhus, Rlianmus, Vitis, Alnus, Be tula, Corylus, 
Fagus. Quercus, Juglans, Sali.x, Ulmus, Vaccinium, Viburnum. However, the 
determinations are based on leaf impressions and therefore they are in part 
very uncertain. Berry (1930), who revised critically all polar floras in 1930, 
let pass as plants that might justly be considered of a cool temperate climate 
only the following: Platanus, Liriodendron, Acer, Juglans, Gingko, Fra.xinus, 
and Hicoria. Heer {loc. cit.) inferred a chmate with an annual average 
temperature of at least 9 C. That may be rather true. 

Later on. Askelsson (1946) has described also some pollen, and at my 
suggestion Pflug (1956, 1959) studied in more detail the pollen floras of 
Iceland, especially from the lignites (Icelandic: surtarbrandur). Also Meyer 
and Pirrit (1957) and Jux (1960) made pollen-analytical investigations here. 
But the pollen has more importance as regards stratigraphy and not so much 
concerning paleoclimatology. 

Finally. E. Schonfeld (1956) studied fossil Icelandic woods. He found among 
others, Ilex and supposed that the Icelandic Ilex and also Picea had their 
nearest relations in North American species. 

In Greenland the lea\es of willow, poplar, birch, and hazel dominate 
according to the revisions of Berry (1930). But there also are represented 
Liquidanibar. Ulnnis. Platanus. Sassafras. Fraxinus. Cornus. Liriodendron. 
Acer, and I itis. 

In a rather new paper Schloemer-Jaeger (1958) cites from Spitsbergen 
above all Sequoia langsdorfic, Metasequoia occidentalis, and Cercidiphyllum 
arcticum. The average January temperature must have been higher than O'C. 
There are also pollen-analytical studies by Manum (1954). 

Grinnell Land is the locality nearest to the Pole, 82' N. Lat. According to 
Berry (1930) its Tertiary flora consists of Equisetum, Faxodiwn, Pinus, 
Abies, Populus. Be tula, and Corylus. 

This last locality, especially, shows that the Tertiary polar floras have 
nothing to do with tropical or even only subtropical vegetation; they are 
ordinary floras of moderate climate. They fit in with the picture of the Tertiary 
chmatic belts of the whole Earth, and these belts were generally shifted 
polewards (cf. Chaney. 1940; Schwarzbach, 1946). 


With higher temperatures in the polar region there is, of course, no change 
in the unusual distribution of day and night so characteristic for high latitudes ; 
that means we have the additional problem of the polar night, of short- and 
long-day plants. But, considering the successful cultivation of hundreds of 
plants by Icelandic greenhouse gardeners, among them many plants from 
subtropical regions, we must admit that this problem cannot be a difficult 

There exists the real difficulty that we do not know the exact age of the 
Tertiary polar floras. In part they seem to be of older Tertiary age. Pollen- 
analysis also points in that direction. But we can only express conjectures 
as long as we have no reference to securely dated beds of mammals or marine 
faunas or have no absolute age determinations. 

We must suppose — as everywhere on the Earth — that the younger Tertiary 
was cooler than the older, and that there was a gradual transition to the 
Quaternary Ice Age. The Pliocene marine beds of Tjornes in northern Iceland 
— the only Pre-Quaternary marine sediments of the island — indeed show such 
faunas. The faunas of the lower parts of the Tjornes beds required higher 
temperatures than the recent sea. (I have supposed a difference of 5°C 
according to the faunal list of Bardarson, 1925.) But the temperatures of the 
youngest Phocene can have been only slightly higher than now. 

Of special interest in this connection are the studies by Jon Jonsson (1954) 
in the region of Hornafjordur, southeastern Iceland. He found tilhtes there, 
overlain by hgnites. The pollen-analytic investigations by Pflug (1956) in the 
Geological Institute of Cologne showed that the hgnites may be from younger 
Tertiary, meaning that moraine-depositing glaciers already existed before 
this time (i.e. also in Early Tertiary time). We do not need to imagine a large 
Vatnajokull at this time, but at least we must take into account small Tertiary 
glaciers. The general gradual cUmatic deterioration of the Tertiary in Iceland 
led temporarily and locally to glacial chmatic conditions. We know nothing — 
at least not directly — of the other North Atlantic regions in this respect, but 
we can suppose similar conditions in Greenland and Spitsbergen; all the more 
so because Early Tertiary glaciers already existed in Alaska, according to 
studies by Miller (1953). 

For the problem of the development of biota in the North Atlantic it is of 
great interest to know a little not only about the Tertiary climate but also 
about paleogeographic conditions on the whole. Generally we can say that the 
climatic picture allows no room for continental drift, at least not on a big 
scale. That is in agreement with paleo-magnetic results. We must suppose that 
the cause of the warm Tertiary climate in Europe was partly a powerful 
Gulf Stream which could flow unchecked far to the east and northeast. 
Therefore, paleoclimatologists have no reason to construct a land-bridge 
between Europe and North America; on the contrary, the evidence speaks 
more against an emerged Faeroes-Iceland ridge. I mention, by the way, that 


the distribution of Tertiary mammal faunas in Europe-Asia-North America 
do not require such a land-bridge. 

it is to be hoped that the study of deep-sea cores will give more certain 
answers to these questions. 


The Quaternary Ice Age which began ca. 1 milHon years ago is of greatest 
importance for the recent flora and fauna of the North Atlantic regions. 
Nevertheless, we can treat it relatively briefly. One fact stands without doubt : 
that these areas were covered more or less completely by ice during Glacial 
times; looking out of the windows of the University of Reykjavik, we see 
ice-polished rocks all around us. 

But there remain some questions of special interest: 

1. What was the detailed course of the Ice Age, i.e. the succession of 
Glacials and Interglacials ? 

2. Where were the ice-free refugia? 

3. What about Quaternary land-bridges? 

The first question concerning the stratigraphy of the Quaternary Ice Age 
can be answered only on a large scale. Generafly ice-bordered and temporarily 
ice-covered areas are much less fit for a detailed stratigraphic division than 
more distant regions. This is due to the fact that ice-erosion often removes 
older deposits completely. But still it is certain also that in the polar regions 
Glacial periods alternated with Interglacials in which climate, and vegetation, 
were about the same as today or even a httle more favorable. There are 
several known Interglacials in Iceland, for instance from Snaefeflsnes and 
from the neighborhood of Reykjavik, and Thorarinsson will tell us something 
about new finds in southern Iceland. Likewise, it seems to be certain that 
these Interglacial floras and faunas belong to several, i.e. temporally different, 
Interglacial periods. 

There are also Interglacials in Greenland, from where Bryan (1954) 
described an occurrence of Picea mariana, and also in Alaska and Arctic 
Canada (Terasmae et al. ; cf. Craig and Fyles, 1961). 

Until now it has been completely impossible to equate all these Inter- 
glacials with the standard divisions of Europe or North America. 

But the studies of deep-sea sediments will perhaps help us to get better 
results. The cores from the floor of the oceans prove a repeated alternation of 
warm and cool periods, confirmed by foraminiferal faunas or directly isotopic 
temperatures. The core no. 280, for instance — from the northern Atlantic at 
35° N. Lat. — shows according to Emiliani (1958) a change between tempera- 
tures of 11° and 18°C. The 11° record corresponds to a Glacial, the 18° to an 
Interglacial time, with perhaps an age of 90,000 years. But most age calcula- 
tions are very uncertain, especially for the periods beyond C^'* determinations, 


i.e. older than 50,000 years. Therefore we will get valuable new results for 
Quaternary stratigraphy and chronology only with better age determinations 
and a more sure correlation with continental events. 

The second question concerns the ice-free refugia during Glacial times. They 
can originate where precipitation is too low to produce glaciers, though 
temperatures may be sufficiently low; for example, the recent Peary Land in 
north Greenland. Ice-free refugia can also occur where steep mountain 
ranges prevent snow accumulation. All regions with pronounced relief, for 
instance Iceland, had such places. Thorarinsson (1937) has given a map which 
indicates the rather large areas without inland ice or at best local glaciation. 
We can suppose that the glacial climate at these places was about the same as 
it is near the border of recent inland ice areas. 

Finally, the f bird question: Quaternary land-bridges have been demanded 
as an explanation in certain Ice Age hypotheses or to explain the peculiarities 
of the recent or Interglacial flora and fauna, especially in Iceland. But the 
geologist must be skeptical of such suppositions, for he cannot find any 
indications that at any time in the Quaternary parts of the Wyville-Thompson 
ridge emerged from the sea. Of course, this ridge was about 100 m shallower 
during Glacial time because of eustatic fluctuations. But for zoo- or phyto- 
paleography not the Glacials but, on the contrary, the Interglacials are 
important — and there we must expect even higher sea levels than today! 

Properly that paleoclimatologist now ought to treat the Post-glacial time 

accordingly, for this period has especially close relations to the recent 

distribution of flora and fauna. But there will be special papers regarding this 

problem in this symposium, and I can forego speaking on this last chapter of 

climatic history. 


AsKELSsoN, J. (1949). Er hin smasaeja flora surtarbrandslaganna vaenleg til konnunar? 

Sk rsla Menntaskolans i Reykjavik 1945 46, 1-9. 
AsKELSSON, J. (1956). Myndir ur jardfraedi Islands IV. Faeinar plontur iir surtarbrands- 

logunum. Ndtti'irufr. 26, 44-48. 
AsKELSsoN, J. (1960). Pliocene and Pleistocene fossiliferous deposits. Int. Geol. Congr. 

XXI. Guide to excursion No. A2 (Iceland): 28-32. 
Bardarson, G. G. (1925). A stratigraphical survey of the Pliocene deposits atTjornes in 

northern Iceland. Kgl. Danske Vidensk. Selsk. Biol. Mecki. 55 (5), 1-118. 
Berry, E. W. (1930). The past climate of the north polar region. Smithsonian Inst. Misc. 

Coll. 82 (6), 1-29. 
Bryan, M. S. (1954). Interglacial pollen spectra from Greenland. Danmaiks Geol. Vmleis. 

II. Raekke, No. 80. 65-72. 
Chaney, R. W. (1940). Tertiary forests and continental history. Geol. Soc. Anier. Bull. 51, 

Craig, B. C. and Fyles, J. G. (1961). Pleistocene geology of Arctic Canada. In G. O. Raasch 

(ed.). Geology of the Arctic, Toronto, 403-420. 
Emiliani, C. (1958). Paleotemperature analysis of core 280 and Pleistocene correlations. 

/. Geol. 66, 264-275. 
Heer, O. (1868). Flora fossilis arctica. I. Island. Zurich. 
JoNSSON, J. (1954). Outline of the geology of the Hornafjordur region. Geogr. Ann. 36, 

146 161. 


Jux, U. (1960). Zur Geologic des Vopnafjord-Gehietes in Nordost-Island. Geologie 9, Bh. 

28, 1-57. 
LoTZE F. (1957). Steinsalz iind Kalisalze. I. 2nd ed. Berlin 
Manum, S. (1954). Pollen og sporer i tertiaere kull fra Vestspitsbergen. Norsk Polarinsdtutt 

Ski: 79, 1-10. 
Meyer, B. L. and Pirrit, J. (1957). On the poiien and diatom flora contained in the surtar- 

brandur of East Iceland. Pioc. Roy. Soc. Edinb. B61, 262-275. 
Miller, D. J. (1953). Late Cenozoic marine glacial sediments and marine terraces of 

Middleton Island, Alaska. J. Geol. 61, 17^0. 
Nathorst, a. G. (1890). tJber die Reste eines Brotfruchtbaums, Artocarpus Dicksoni 

n.sp., aus cenomanen Kreideablagerungen Gronlands. Sv. Vet. Akad. Handl. N.F. 

27 (1), 1-10. 
Pflug, H. D. (1956). Sporen und Pollen von TroUatunga (Island) und ihre Stellung zu den 

pollenstratigraphischen Bildern Mitteleuropas. Neiies Jahrb. Geol. Paldontolog. Abh. 

102, 409-430. 
Pflug, H. D. (1959). Sporenbilder aus Island und ihre stratigraphische Bedeutung. Neues 

Jahrb. Geol. Paldontolog. Ahh. 107, 141-172. 
Schloemer-Jaeger, a. (1958). Alttertiare Pflanzen aus Flozen der Brogger-Halbinsel 

Spitzbergens. Palaeontogr. B 104, 39-103. 
ScHONFELD, E. (1956). Fossile Holzer von Island. Neues Jahrb. Geol. Paldontolog. Abh. 104, 

ScHWARZBACH, M. (1946). Klima und Klimagurtel im Alttertiar. Natiirwiss. 33, 355-361. 
ScHWARZBACH, M. (1949). Fossilc Korallenriffe und Wegeners Drifthypothese. Naturwiss. 

36, 229-233. 
SCHWARZBACH, M. (1955). Aligcmeiner Uberblick der Klimageschichte Islands. Neues 

Jahrb. Geol. Paldontolog. Mh. 97-130. 
ScHWARZBACH, M. (1961). Das KUmo der Vorzeit. 2 ed. Stuttgart. 
SCHWARZBACH, M. and Pflug, H. D. (1956). Das Klima des jungeren Tertiars in Island. 

Neues Jahrb. Geol. Paldontolog. Abh. 104, 279-298. 
Stille, H. (1958). Die assyntische Tektonik im geologischen Erdbild. Beih. Geol. Jahrb. 

22, 1-255. 
Terasmae, J. (I960). Contributions to Canadian palynology. No. 2. Geol. Survey Canada, 

Bull. 56, 1-41. 
Thorarinsson, S. (1937). The main geological and topographical features of Iceland. 

Geogr. Ann. 19, 161-175. 


Bruce C. Heezen and Marie Tharp 

Columbia University, Department of Geology and Lamont Geological Observatory, 

Palisades, New York 

Former continental connections across present seas have been frequently 
proposed by botanists, zoologists and paleontologists striving to understand 
the affinities and routes of dispersal of land biota. Land connections across 
present epicontinental seas have clearly occurred during the Pleistocene due 
to eustatic fluctuations of sea level. More difficult is the question of earlier 
connections across the deep seas (Heezen et al., 1959). 


Geophysicists and oceanographers have long been skeptical of sunken 
ancient continents or sinuous isthmian links across the deep sea (Bucher, 1952 ; 
Ewing, 1952). The structure of the ocean floor in depths greater than 4000 m 
is fundamentally different from that beneath the continents. The Mohorovicic 
discontinuity, the boundary between the crust and mantle, lies at some 30 to 
40 km below the continents. However, below the ocean's surface the Mohoro- 
vicic discontinuity lies at a depth of only 10 to 12 km. The rocks lying im- 
mediately above the Mohorovicic discontinuity are known entirely on the 
basis of their seismic-wave velocities. In a typical deep-sea crustal section i to 
1 km of sediment lies above a thin 1 to 2 km layer in which the seismic-wave 
velocity is between 4 and 5 km/sec (Hill, 1957). This second layer is sometimes 
ascribed to altered basalt and sometimes to lithified sediments, although it 
seems more likely that the material is igneous rather than sedimentary. 
Beneath this lies a layer 3 to 5 km thick which has a velocity of about 6.5 
km/sec. The mantle below the Mohorovicic discontinuity has a velocity of 
about 8.2 km/sec. Hundreds of measurements beneath the deep-sea floors 
have revealed this column to be essentially universal. However, in the 
central part of the Mid-Oceanic Ridge, in the Norwegian Sea, in the northern 
North Atlantic and in other areas (primarily where the depth is less than 4000 
m), the crustal structure is quite different, the seismic velocity of the deepest 
observed layer ranging between 6.8 and 7.5 km/sec. Recent seismological 
studies in Iceland have led Tryggvason (1962) to conclude that the 7.4 km/sec 
layer reaches to tens of kilometers depths below Iceland, an emerged portion 

* Lamont Geological Observatory Contribution No. 577. 



of the Mid-Oceanic Ridge. The same velocity material has been found 
beneath the crest of the Mid-Atlantic Ridge and over much of the floor of the 
Norwegian Sea and the Labrador Sea (Ewing and Ewing, 1959). 

If a continent subsided to oceanic depths (greater than 4000 m), one 
would expect the crustal structure to be markedly different in the area formerly 
occupied by the continent. In addition, in order to maintain isostatic balance, 
a yet unknown process, whereby a thick section of continental-type crustal 
rocks is transfornied into thin section oceanic crustal rocks, is required. Thus, 
in view of these difficulties, it seems necessary to limit any consideration of land 
connections across the deep sea to "island stepping stones" and to continental 

Flat-topped seamounts of the central Pacific were islands in the Cretaceous 
(Hamilton, 1956). Thus numerous "stepping stones" did exist across many 
parts of the Pacific in the Cretaceous. These flat-topped seamounts (guyots) 
are less plentiful in the Atlantic. Botanists in particular have serious reserva- 
tions about a "stepping stone" type of land bridge, for they maintain that 
small islands would not develop continental-type vegetation and therefore 
would not act as a bridge. 

The continental displacement hypothesis, once rejected for lack of a 
mechanism, has received new support in the last decade from the results of 
the studies of paleomagnetism. The gradual opening of the Atlantic by 2000 
miles in the Mesozoic and Early Tertiary is supported by a consistent dis- 
crepancy between paleomagnetic measurements made in North America and 
Europe (Runcorn, 1959). The fact that no sediments older than Cretaceous 
have been found on the Atlantic deep-sea floor is not in opposition to this 
view, although it might be argued that the few hundred outcrops of Tertiary 
and Cretaceous sediment constitute too small a sample upon which to base 
such an important generalization (Ericson et aL, 1961). Recently, detailed 
studies of the thickness of the sedimentary layer which lies above the oceanic 
crust have been made by the seismic-reflection technique (Ewing, 1961). In 
general it is found that there is an exceptionally homogeneous 1-2 km layer 
which overlies a layer of similar thickness which in turn rests on an intensely 
irregular sub-bottom topography. The material lying beneath the second 
sedimentary layer is presumably one which refraction measurements have 
indicated to have a crustal velocity of 4.5 to 5.5 km/sec. Both sedimentary 
layers thicken markedly upon approaching the continents, apparently in 
response to greater rates of sedimentation. 

When approaching the Mid-Oceanic Ridge, the thickness of sediment thins 
markedly. On the flanks of the Mid-Atlantic Ridge the irregular deeper layer 
apparently reaches the surface in many areas, sediment being restricted to the 
bottoms of occasional intermontane basins. Near the crest of the Mid- 
Oceanic Ridge sediment is virtually absent. The absence of sediment from the 
crest of the Mid-Oceanic Ridge has been taken as evidence of the recent 


origin of this topography (Heezen, 1960). The continental displacements which 
may have created the Atlantic in the Mesozoic and Tertiary may be the result 
of a process which continually added material along the crest of the Mid- 
Oceanic Ridge. The mechanism of this continual addition of new crust has 
been variously ascribed to mantle convection currents which rise beneath 
the ridge, diverge and carry the oceanic crust toward the continents (Dietz, 
1961). The sinking of these currents below the continents may, in turn, cause 
compression and uplift in the continents. However, others prefer to explain 
the displacement of continents through an overall expansion of the interior 
of the earth (Heezen, 1960; Wilson, 1960). Needless to say, for students 
of the Atlantic it makes little difference which of the two hypotheses is 
favored, for in regard to the Atlantic the effects of either mechanism would 
be identical. 

It might occur to some that the absence of sediments from the crest of the 
Mid-Atlantic Ridge could be explained in terms of a recent emergence. It 
might be argued that the sediments were eroded from the Mid- Atlantic 
Ridge by subaerial erosion and that the ridge has only recently been sub- 
merged beneath the ocean. However, if this were true one would expect to 
find exceptionally thick deposits of sediment on the margins of the Mid- 
Oceanic Ridge near the former shorelines. The recent data indicate that the 
entire width of the Mid-Atlantic Ridge from deep basin on one side to deep 
basin on the other is nearly devoid of sediment. This would require that the 
entire Mid-Atlantic Ridge some 1200 miles wide be raised 3 or 4 km above 
the adjacent basins in order to affect denudation and that the products of 
denudation lie in the basins. However, the pattern of distribution of sediment 
thickness in the basins does not support this view, for there is a gradual 
increase in thickness toward the continents and no evidence of thickening 
along the margins of the Mid-Oceanic Ridge. The discovery of a few fresh- 
water diatoms in a core from the crest of the equatorial Mid-Oceanic Ridge at 
one time led certain investigators to propose an emergence of a short duration 
(Kolbe, 1957). However, it need only be mentioned that, (1) the layer of 
freshwater diatoms is approximately a millimeter thick interbedded with 
typical deep-sea sediments, and (2) the winds blowing off Africa often carry 
such large quantities of diatom tests as to lay down layers of appreciable 
thickness on the decks of ships. 

We may conclude that land connections across the deep basins of the 
Atlantic have not existed in the form of sunken continents, isthmian links, or 
closely-spaced insular "stepping stones". But it is now probable that a dis- 
placement of Europe and North America has occurred and that at some time 
in the Paleozoic or Early Mesozoic parts of Europe and America lay adjacent 
to one another without an intervening ocean. It seems unlikely that once the 
two continents were displaced from one another that any land connections 
existed where there is now deep sea. 



lOOM ZOOM 200M-500M 500M-I000M >IOOOM 

Fig. 1. Bathymetric chart of the Faeroe-Iceland-Greenland Ridge. 



Fig. 2. Hypothetical bathymetric chart of the Faeroe-Iceland-Greenland Ridge if 
sea level were lowered 200 m. Current estimates of the maximum Pleistocene 
lowering of sea level do not exceed 160 m; thus, even if the maximum lowering of 
sea level during penultimate glaciation were 40 m greater than current estimates, 
the Faeroe-Iceland-Greenland Ridge would not provide a continuous land bridge 
from Europe to Iceland and Greenland. If such a connection is required by 
studies of the geology of the ridge or of the biota of the area, a substantial 
subsidence of this ridge must be assumed to have occurred in the Late Pleistocene. 



As shallow seas we will consider those areas considerably less than 4000 m 
in depth. For the Atlantic this actually consists of one connection across the 
Faeroe-Iceland-Greenland-Canada Ridge. Although it may well be that a 
connection existed between Spitzbergen and Greenland,* if a connection did 
exist across the Nansen Straits its disruption was probably due to the opposed 
continental displacements of Europe and North America. 

At the present time, the 500 m contour connects almost completely across 
the Faeroe-Iceland-Greenland Ridge from Europe to Greenland (Fig. 1). In 
two small gaps, one near the Faeroes and another in Denmark Strait, depths 
are slightly greater. Little can be said concerning the geological history and 
origin of this ridge for at present no cores nor dredge hauls of ancient rock 
have been reported from anywhere along the ridge. Since the top of this 
ridge lies close to sea level, minor vertical movements of the Earth's crust could 
either cause an emerged land link or a submerged sill. At the present time it 
would seem that the Tertiary history of a possible land bridge across the 
Faeroe-Iceland-Greenland Ridge could be best estimated from data obtained 
from the fossil and contemporary biota (Love and Love, 1956). 

On the other hand, we can perhaps be somewhat more definite about 
possible land connections in the Middle and Late Pleistocene. Due to the 
relatively short interval of time involved we can perhaps exclude really large 
tectonic changes in the absolute elevation of the ridge and restrict ourselves to 
considerations of the effects of glacial eustatic changes of sea level. In recent 
years studies of eustatic changes in sea level have been made in widely separ- 
ated parts of the world (Fairbridge, 1961). A prominent submerged shoreline 
found in depths of 160 m off of North America, South America and Africa 
is generally ascribed to the sea level associated with the penultimate glaciation. 
Studies of the probable ice volume based on the distribution of ice and 
studies of its probable thickness based on Post-glacial rebound of the land 
have allowed the calculation of the probable volume of water tied up in the 
glaciers of the penultimate glaciation (Donn et al, 1962). This volume as 
given by Donn, Farrand and Ewing is approximately 85-100 X 10** km^. 
This would account for lowering of sea level of approximately 140-160 m. 
Their estimates for the two uhimate glaciations are 70-84 and 75-88, with 
resulting sea level lowerings of 105-123 m and 114-134 m (Table 1). It is 
generally considered that the earlier glaciations were less extensive, that the 
maximum lowering of sea level should have occurred during the penultimate 
glaciation (Farrand, 1962). If the Iceland-Greenland Ridge has neither been 
elevated nor depressed nor seriously eroded during the past 200,000 years, 
then no land bridge could have existed across the Faeroe-Iceland-Greenland 

* Depths in this area exceed 3000 m, recent Soviet expeditions having disproved the 
existence of the so-called Nansen Sill (Hope, 1959). 


Table 1* 

Ultimate (Wisconsin) 





Ice Volumes 


Total glacier ice 

Volume of modern glacierst 

Volume of ancient glaciers minus 

modern volume 
Water equivalent (ice density 0.9) 

71-84 74-88 
28-35 28-35 

43-49 46-53 
38-45 41-48 



Sea-level lowering (meters) 

(Surface area of world oceans 
361 X 10«km-) 



* After Donn, W. L., Farrand, W. R. and Ewing, M. (1962). 

t Minimum volume after Crary (1960), maximum after Novikov (1960), estimates of 
Antarctic ice. 

Ridge (Fig. 2). Gaps of over 200 miles must have existed between Faeroes and 
Iceland and between Iceland and Greenland with prevaihng depths of 100 or 
200 m. Thus, if a nearly continuous land bridge between Europe and Iceland 
is required, it must be assumed that the Iceland-Faeroe Ridge has subsided 
slightly more than 200 m in the last 200,000 years since eustatic lowering 
alone is insufficient to account for the emergence of a land bridge during the 
penultimate glaciation. Such a subsidence cannot be considered geologically 
unreasonable, but the ultimate proof will lie in geological exploration of the 
ridge and in the results of the study of North Atlantic biota. 


BucHER, W. H. (1952). Continental drift versus land bridges. In E. Mayr (ed.): The 
problem of land connections across the South Atlantic, with special references to the 
Mesozoic. Bull. Amei: Mas. Nat. Hist. 99, 93-103. 

Crary, A. P. (1960). Status of United States scientific programs in the Antarctic. T.G.Y. 
Bull. 39, Amer. Geophys. Union Trans. 41, 521-532. 

DiETz, R. S. (1961). Continent and ocean-basin evolution by spreading of the sea floor. 
Nature 190, 854-857. 

Donn, W. L., Farrand, W. R., and Ewing, M. (1962). Pleistocene ice volumes and sea- 
level lowering. J. Geol. 70, 206-214. 

Ericson, D. B., Ewing, M., Wollin, G. and Heezen, B. C. (1961). Atlantic deep-sea 
sediment cores. Geol. Soc. Amer. Bull. 72, 193-286. 

Ewing, J. I. and Ewing, W. M. (1959). Seismic refraction measurements on the Atlantic 
Ocean basins, Mediterranean Sea, on the Mid-Atlantic Ridge and in the Norwegian 
Sea. Geol. Soc. Amer. Bull. 70, 291-318. 


EwiNo, M. (1952). The Atlantic Ocean Basin. In E. Mayr (ed.): The problem of land 

connections across the South Atlantic, with special references to the Mesozoic. 

Bull. Amer. Mus. Nat. Hist. 99, 87-91. 
EwiNG, M. (1961). Address to American Geographical Society, November 1961. 
Fairbridge. R. W. (1961). Eustatic changes in sea level. In L. H. Ahrens and others (ed.): 

Physics and Chemistry of the Earth. Vol. IV, 99-185. Pergamon Press, London. 
Farrand, W. (1962). Post-glacial uplift in North America. Amer. J. Sci. 260, 181-199. 
Hamilton, E. L. (1956). Sunken islands of the Mid-Pacific mountains. Geol. Soc. Amer. 

Memoir 64, 1-97. 
Heezen, B. C. (1960). The rift in the ocean floor. Scientific American 203, 98-110. 
Heezen, B. C, Tharp, M. and Ewing, M. (1959). The floors of the oceans. I. The North 

Atlantic. Geol. Amer. Spec. Paper 65, 1-122. 
Hill, M. N. (1957). Recent geophysical exploration of the ocean floor. In L. H. Ahrens 

and others (ed.): Physics and Chemistry of the Earth, 129-163. Pergamon Press, 

Hope, E. R. (1959). Geotectonics of the Arctic Ocean and the Great Arctic Magnetic 

Anomaly. J. Geophys. Research 64, 407-427. 
Kolbe, R. W. (1957). Fresh-water diatoms from Atlantic deep-sea sediments. Science 126. 

Love, A. and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic flora. Acta 

Horti Gotoburg. 20, 69-291. 
NoviKOV, V. (1960). The study of the Antarctic is continuing. Priroda 8. 43-52 (in Russian). 
Runcorn, S. K. (1959). Rock magnetism. Science 129, 1002-1012. 
Tryggvason, E. (1962). Crustal structure of the Iceland region from dispersion of surface 

waves. Seismo, Soc. Amer. Bull. 52 (2), 359-388. 
Wilson, J. T. (I960). Some consequences of expansion of the earth. Nature 185, 880-882. 



Joannes Rasmussen 
Museum of Natural History, Torshavn, Faeroes 

Apart from papers written during the seventeenth and eighteenth centuries on 
Faeroese minerals and coal measures, the geology of the Faeroe Islands is not 
mentioned in the literature until about 1800. 

Without going into tiring, historical details, I shall very briefly list the most 
important works on the geology of the Faeroes in the nineteenth century, 
namely the papers by Sir George Mackenzie and Th. Allan in 1814, by 
J. G. Forchhammer and W. C. Trevelyan in 1823-24, and by A. Helland and 
J. Geikie in 1880. 

Sir George Mackenzie, together with Thomas Allan, paid a visit to the 
Faeroe Islands in 1812. The purpose of their journey was to ascertain whether 
special geological features observed in Iceland, where Sir George Mackenzie 
had visited in 1810, were to be found also in the Faeroes. In his work, 
Mackenzie describes some geological observations and concludes that the 
islands are of submarine volcanic origin. Allan shares Mackenzie's view of the 
volcanic origin, but does not agree with his opinion regarding the submarine 
eruptions. Allan's observations of striae and his words concerning them are of 
interest: "The rock appears to have been worn down by the friction of heavy 

Eleven years later, in 1821, Forchhammer visited the Faeroe Islands 
together with Trevelyan, the British mineralogist and botanist. In Forch- 
hammer's paper (1824) we find the first complete account of the geology of the 
Faeroe Islands. Forchhammer divides the rocks into 4 groups: (1) Trap 
without glassy feldspar, (2) Coal measures, (3) Porphyritic rocks, and (4) 
Irregular trap (dykes and sills). He shows the regional extension of these 
rock types on an accompanying small map. 

During the next 58 years we find only a few fragmentary works on the 
geology of the Faeroe Islands, but in 1879 Helland and Geikie visited the 
Faeroe Islands. The particular purpose of the journey was to examine 
the glacial geology of the islands, which hitherto was as good as unknown. 
Basing their statements on the radiating direction of the striae and the total 
absence of foreign rocks, Helland (1880) and Geikie (1880) concluded that 
the Faeroes have had a glaciation of their own. In Helland's as well as in 



Geikie's works we find, besides the glacial geology, a comprehensive account 
of the general geology of the Faeroes. Helland retains Forchhammer's 
classification of the rocks, but he uses other terms, e.g. "Anamesite" instead of 
"Trap without glassy feldspar'', and "Dolerite" instead of "Porphyritic rocks". 

While the papers by Mackenzie (1814) and Allan (1814) cannot be consider- 
ed of essential importance to later geological works, this is, however, not the 
case with the publications by Forchhammer (1824), Helland (1880), and 
Geikie (1880). They have laid the basis for later studies even though points of 
view have changed a great deal, especially thanks to work by Peacock (1928), 
Walker and Davidson (1936) and subsequent investigators. 

The present short account of investigations on the geology of the Faeroes 
in more recent years, is chiefly based on results obtained during the systemati- 
cal geological mapping of the Faeroes under the auspices of the Geological 
Survey of Denmark. 

This project was begun in 1939, but was suspended during the war. In the 
beginning it was conducted by Professor Arne Noe-Nygaard, and later (from 
1951) it was carried on by this author at a permanent station in the Faeroes, 
and is now about to be concluded by both of us. 

It must be emphasized that the geological mapping of the Faeroes has 
not been concerned solely with rock types, but also with geological develop- 
ment, or rather, with development of the plateau basalt volcanism. Thus, 
the series delimited by the mapping represent stages in volcanic development. 

Attempts were made to make the map on the basis of different types of 
basalt, but in our experience, many difficulties arose and this system in several 
cases proved quite impractical. Among the problems encountered, the follow- 
ing may be mentioned: (1) the characteristic tapering out and overlapping of 
strata in the plateau basalts; (2) the very thin lava flows of varying types in 
certain series occurring in a large number; (3) the horizontal changes in the 
same lava flows, consequential to the sinking of heavier minerals. 

The Faeroes belong geologically to the large Brito-Arctic, North Atlantic, 
or Thulean igneous region and appear as a geographically bounded link 
between the Scoto-Irish region and Iceland. 

The geological structure of the Faeroes is very simple : a regular alternation 
of lava beds with pyroclastic materials, and very subordinate intra-basaltic 
sediments. The whole series, having a total thickness of about 3000 m, 
is traversed by intrusive bodies. No eruptive rocks are known from the 
Faeroes — neither effusive nor intrusive — of a composition other than basalt, 
and the substratum of these basalts is completely unknown. 

As far as the age of the Faeroes is concerned, it is difficult to determine, 
even roughly. Only two definable species of plant fossils have been found 
{Sequoia langsdorfii and Taxodium distichum). However, compared with 
conditions in the Scoto-Irish region and in Iceland, it is reasonable to estimate 
the age as Eocene-Oligocene. 



In the i'ollowing, 1 shall try to give a brief account of the geological develop- 
ment of the Faeroes, based upon a purely schematic cut through the whole 
sequence of rock layers and on surface map sections of some significant 


In all probability the volcanic activity started with large fissure eruptions, 
and this phase is represented by a basalt series nearly 1000 m thick and petro- 
graphically very uniform, referred to below as the Lower Basalt sequence. 

Fig. I. Schematic cut through the Faeroean sequence: Lower Basah sequence. 
Upper Basalt sequence (lower zone), and Upper Basalt sequence (upper zone). 
Black: Coal-bearing series; dotted: Tuff-Agglomerate zone; white: intrusions. 

The Lower Basalt sequence, the oldest stratum, is exposed on Suduroy, 
Mykines, and Vagar beneath the Coal-bearing series (Figs. 1 and 2). As the 
dip on Suduroy is 3°-6° NNE., NE., and ENE., on Mykines 8^-13° ESE. and 















•9 -^ 

r— 1 




o -a 












































'^ l-i 

2 ° 

3 60 

C 3 
I" 1—1 


SE. and on Vagar 3° ESE., this sequence goes below sea level in the northern 
part of Suduroy, and in the western part of Vagar. An excellent view of this 
sequence is to be had in the naked, steep, sometimes quite vertical, rocky 
wall to the west. The thickness of the individual basalt beds generally ranges 
from IC to 30 m, although thinner beds do occur; the greatest measured 
thickness of a basalt bed in this sequence is about 70 m. Petrographically the 
flows ar-? very uniform. They consist of homogeneous, hard, dark, fine 
grained, only exceptionally porphyritic basalts often with a well-developed 
columnar structure. The surface of the individual flows is slaggy, porous, red, 
due to either the oxidation or to the heating effect of the overlying flow. 

Intrabasahic sediments are of rather subordinate significance. Besides the 
prevailing tuff" layers, 1 to 4 m in thickness, intrabasaltic, fluvial, conglomer- 
ates and shales of almost the same thickness as the tuff layers occur, sometimes 
containing sporadic coal in very tliin layers or lenses. 

All the known part of the Lower Basalt sequence (about a 1000 m thick) is 
of subaerial, volcanic origin. However, since its substratum and consequently 
the absolute thickness of the whole sequence is unknown, as mentioned above, 
nothing can be stated regarding a possible earher submarine volcanic phase. 


Subsequent to the formation of the Lower Basalt sequence there occurred 
a quiet period in the volcanic activity, an interval of a rather long duration, 
represented by a Coal-bearing series (Fig. 1). This rests immediately on the 
surface of the Lower Basalt sequence and occurs thus in the northern part of 
Suduroy and in the westernmost part of Vagar. On Suduroy (Fig. 2) the 
Coal-bearing series has its highest position in the southwest at about 425 m, 
and on Vagar in the northwest at about 250 m above sea level. The surface of 
the Lower Basalt sequence has undergone a rather high degree of subaerial 
weathering, and the Coal-bearing series, therefore, rests on a strongly 
undulating surface. The coals in the Coal-bearing series are the only ones in the 
Faeroes of any economical interest. They are allochtonous, deposited in a 
basin or a lake, and cover an area of about 23 km^. In Fig. 3 is drawn a 
characteristic profile section through the Coal-bearing series, the thickness of 
which usually reaches 10-15 m. The sequence is as follows: 

1. Light grayish-yellow or gray bottom clay (Faeroic: Banki). 

2. Lower coal band (Faeroic: Stabbi). 

3. Dark shale (Faeroic: Rann). Some coals often occur in this shale, 
especially in the southern area. 

4. Upper coal band (Faeroic: Kolband). 

5. Roof clay (Faeroic: Tak). 

The roof clay is of rather variable nature, and sometimes fluvial conglomer- 
ates entirely of basaltic origin occur. They occur mainly towards the outer 



boundary of the coal area. The clays are all fire-resistant and without plasticity. 
The thickness of the individual coal bands in the profile section varies some- 

FiG. 3. Profile through the Coal-bearing series. 

what from place to place. In the northern area the lower coal band is thicker 
than the upper one, whereas the contrary is the case in the southern area. 


At the transition between the two areas several coal bands often occur. The 
total thickness of the two coal bands ranges generally from 50 to 150 cm with 
an average thickness of f m for the western part of the area. Towards the 
east and north the coals taper out. 

The coals must be classified somewhere between lignite and bituminous, 
and appear as two types: glossy (Vitrite) and dull (Durite). The glossy coals 
are lustrous, hard, with conchoidal fracture and very pure. Their caloric 
value lies between 6000 and 6300 kcal; the ash content being below 5 per cent. 
The dull coals are streaky, brittle, crumble easily, and are less pure. Their 
caloric value lies generally between 5000 and 5500 kcal and the ash content is 
often close to 20 per cent, sometimes even higher. 

Leaf fossils are very scarcely represented, possibly because the coals are 
allochtonous. As mentioned above, only two definable species have been 
found: Sequoia langsdorfii and Taxodium distichum. 


After the long rest period renewed volcanism set in, and the eruptive 
activity in the intial stage was highly explosive with a production of predomin- 
antly pyroclastic material. The deposits of pyroclastics overlying the Coal- 
bearing series and underlying the Upper Basalt sequence have been described 
as the Tuff-Agglomerate zone (Fig. 1). 

The Tuff-Agglomerate zone occurs on the east side of Suduroy, on Tind- 
holmur, and on the west side of Vagar. On Suduroy (Fig. 2) it appears as an 
elongated beh, 2 to 3 km in width and about 10 km long, from the north side of 
Trongisvagur. It can be studied in ravines and in coastal cross-sections, where 
it is evident that it overlies the Coal-bearing series and is overlain, some- 
times with interbeddings, by the Upper Basalt sequence. Since the deposit 
almost entirely consists of pyroclastic materials such as ash, lapilli, and volcanic 
bombs, the thickness varies considerably from place to place. Generally 
it is 20 to 30 m. However, thinner as well as much thicker layers occur. 

It is hkely that the Tuff Agglomerate zone on Suduroy and on Vagar, 
where the conditions are quite analogous, covers the eruption fissures which 
have been feeding the Lower Basalt sequence, as well as their closest surround- 
ings. Furthermore, since they appear along and follow the direction of the 
Suduroyarfjordur and the Mykines fjordur, it is tempting to suppose — as will 
be discussed later on — that the orientation of the inter-island straits is 
determined by the longitudinal bearing of these eruption fissures. 


The explosive eruption activity was succeeded immediately by a volcanism 
which resulted in the formation of the Upper Basalt sequence (of. Fig. 1). This 
is most likely what happened, since the Tuff-Agglomerate zone, at the bound- 
ary of the Upper Basalt sequence, alternates with thin beds belonging to this 


sequence. Likewise, it must be assumed that during the following stage, the 
volcanism was more localized along the old fissures, respresented by the numer- 
ous necks which are visible in cross- as well as longitudinal sections in coastal 
profiles along the straits running NW.-SE. between the northern islands. 

The basalts in the Upper Basalt sequence display far greater variation 
from bed to bed than do those of the Lower Basalt sequence. The following 
main types occur: (1) dense, dark basalts without phenocrysts; (2) plagioclase- 
porphyritic, grayish basalts with characteristic subtypes; (3) olivine basalts, 
varying from ordinary olivine basalts to oceanities. All transitions between 
these main types occur. 

While the thickness of the Lower Basalt sequence is about 1000 m, the 
Upper sequence is considered to be about 2000 m, so that the whole Faeroic 
sequence will be about 3000 m. However, in the geologic mapping work it has 
proved justified and adquate to divide the Upper Basalt sequence into a 
Lower and an Upper zone. In the Lower zone, comprising a httle more than 
two-thirds of the sequence, the individual lava flows are generally of slight 
thickness, often 1 to 2 m or less; however, beds of somewhat greater thickness 
occur. The very thin lava flows, about 20 to 30 cm thick, appear frequently in 
large numbers and may be explained as having run out through breaks in 
thicker lava layers. Tuff" layers are very uncommon in this zone, and the 
individual flows are usually separated by porous inter-zones and distinct 
ropy-lava surfaces. Therefore, it must be assumed that the whole zone was 
formed by a steady, continuous volcanic activity. In the Upper zone, com- 
prising barely one-third of the Upper Basalt sequence, the individual flows are 
usually about 1 m thick, and tuff" layers are common between the individual beds. 

The boundary between the two zones is usually distinct in the field and 
clearly expressed in the topography. The Lower zone is exposed on the 
northern part of Suduroy, on Vagar and on the northern part of Streymoy, 
whereas it ispartlycoveredbythe Upper zone on the southern part of Streymoy, 
Eysturoy, and the northern islands, except Fugloy and Svinoy where only the 
Upper zone prevails. 

After the basalt plateau was completed, unequal withdrawal of the sub- 
stratum resulted in readjustment movements which led to the formation of 
fracture fines (master joints). Further readjustment movements along the 
same lines of fractures results in "lamellae zones" which traversed the whole 
sequence. Subsequently we have basaltic intrusions in these "lameflae zones" 
and in other zones of weakness in the plateau. 


Intrusive formations occur as (1) irregular intrusive formations; (2)sills; and 
(3) dykes. 

The dykes are intruded in the above "lamellae zones" and traverse the 
entire plateau. The two other intrusion forms, irregular intrusive formations 


and sills, occur, as indicated by Fig. 1, in two different intrusion levels: a 
lower one, where the Coal-bearing series and the Tuff-Agglomerate zone have 
acted as a zone of weakness ; and an upper one, where the boundary between 
the lower and the upper zones in the Upper Basalt sequence apparently 
acted as a zone of weakness. 

Since the dykes have been observed as irregular intrusive formations as 
well as sills, we must assume that all the above forms of intrusions belong 
roughly to the same phase of eruption. 

Irregular intrusive formations. The irregularintrusive formations are intruded 
in the Coal-bearing series and especially in the overlying Tuff-Agglomerate 
zone. Since the latter is often strongly porous and not very resistant to 
magma pressures in different directions, these intrusions will have quite an 
irregular shape. Such irregular intrusive formations appear in the north- 
easterly part of Suduroy, onTindholmur, and in the western most part of Vagar. 

On Suduroy they are present at the coast on the northern side of Hvalbiar- 
fjordur and along the eastern coast of Suduroy to Hvannhagi, as shown by 
Fig. 2. On the northern side of Trongisvagur they can be studied in ravines in 
the vegetation-covered terrain. They are also visible on the south side of the 
fjord at the easterly boundary of the Coal-bearing series. In the north the 
intrusive formations reach a height of up to 35 m above sea level, and in the 
south they attain thicknesses of about 100 m at some places. By the intrusion 
the easterly coal-area has been strongly dislocated and destroyed with 
regard to coal mining. Occasionally long apophyses are visible in the Upper 
Basalt sequence. 

On Tindholmur and on the westernmost part of Vagar the irregular 
intrusive formations — like on Suduroy — are intruded in the Coal-bearing 
series and in the overlying Tuff-Agglomerate zone. 

Sills. Sills occur on Streymoy, Eysturoy, Svinoy, and Fugloy (Figs. 4-6). 

The large Streymore sill (Fig. 4) extends in an almost true NW.-SE. 
direction. It is about 9 km long and varies in width from 1 to 2 km in the 
north and from 2 to 3 km in the south. At one place the erosion has been so 
deep that the sill has been divided into a northerly and a southerly part. The 
lowest position of the sill is along its central part in the west, about 300 m 
above sea level, where it more or less follows the regular basalt bedding. To 
the north, south and east it is strongly transgressive. The thickness is greatest 
towards the west, up to about 60 m, whereas it is strongly decreasing towards 
the east, down to 10 m, in some parts even to 5 m. 

The Eysturoy sill (Fig. 5) extends about 6 km in a NW.-SE. direction and 
ranges in width between 2 and 3.5 km. It is most conspicuous towards the 
west where it forms prominent columnar structures. Like the Streymoy sill, the 
lower sill boundary on the west side has a smooth, winding course, whereas in 
the north, south, and east it is strongly transgressive. Again, like the Streymoy 
sill, the Eysturoy one has its lowest position and greatest thickness — up to 





500 1000 m 

Fig. 4. The Streymoy sill. Black: Sills. Double line: Dykes. Dotted line: Border 
between lower and upper part of Upper Basalt Sequence. Stippled areas: Settlements. 







500 1000 m, 

Fig. 6. The Fugloy and Svinoy sill. Black: Sills. Double line: Dykes. Stippled 

areas: Settlements. 


100 m — in the central area in the west. It decreases in thickness with in- 
creasing transgressivity in the east, where the peripheral part of the sill is 
locally almost perpendicular. It is most Ukely that the Fugloy sill and the 
Svinoy sill (Fig. 6) belong to the same intrusive formation, later on divided 
by the Fugloyarfjordur. The thickness of these sills is only 15 to 30 m. 

As is evident from what is said above we find a morphologically common 
feature in these sills, a characteristic semi-saucer-shaped form. The sills run 
rather concordantly in the central area, where they reach their greatest thick- 
ness, while in the peripheral area they become strongly transgressive, at the 
same time decreasing considerably in thickness. 

Dykes. Dykes occur everywhere, but are most abundant along a belt running 
NE.-SW. across the northern group of islands. They represent roughly the 
same main types as are found in the plateau basalt. The course of the dykes 
is usually a straight fined, but sometimes rather winding. It looks as if the 
feldspar dykes have a more winding course than the dense, dark basalt dykes. 
General thickness is 2 to 4 m; the greatest thickness observed is 20 m. Geikie 
(1880) gives the main direction as NNE.-SSW. and ENE.-WSW. Noe-Nygaard 
(1940) proves that the dominating direction of 76 dykes on Kallsoy and 
Kunoy is NNE.-ENE. to SSW.-WSW. On Suduroy the main orientation is 
approximately NW.-SE. 

After the intrusion of irregular intrusive formations, sills, and dykes, there 
occurred again a tectonic period. 

If we draw a line from Mykines to Fugloy, it will hit the deepest points 
in the straits and the greatest heights in the valleys running in the direction 
of the straits. North of this fine the dip is roughly towards the ENE., and 
south of this fine it is approximately ESE. The fine thus forms the present 
main water-shed of the islands. 

On western Mykines the dip is 13°, on eastern Mykines 8°, on western 
Vagar 3°, on StreymoyandEysturoy2°-3°, andon the northern islands the dip 
is even smaller. This indicates a movement along the fine with upheaval of the 
land to the west. By this movement fracture fines and "lamellae zones" have 
arisen in the plateau — running NE.-ENE. north of the fine and SE.-ESE. 
south of the fine — and where they cross dykes, horizontal dislocation of the 
dykes are visible in the direction of the "lameUae zones". Thus the movement, 
which has resulted in the above-mentioned water-shed in the plateau, is 
younger than the dyke intrusion and consequently is Late Tertiary. 

While the Tertiary geology of the Faeroes in recent years has been the 
object of a systematical geological mapping, the Quaternary geology has only 
been dealt with fragmentarily. 

As already shown by Helland (1880) and Geikie (1880) the Faeroes had a 
glaciation of their own. The aforementioned hinge-line (the water-shed of the 


Late Tertiary and the present), during the widest extension of the Pleistocene 
ice, formed an "ice-shed" for one ice flow running northwards and another 
southwards. However, at more advanced stages the ice direction was decided 
by the local topographic forms. 

Moraine deposits occur everywhere in the lowlands, but the cover is rather 
thin, only exceptionally reaching 4 to 5 m in thickness. Terminal moraines 
have not been observed, and the poor moraine cover can be explained, on the 
whole, by the fact that the greater part of the material was carried out to sea 
by the ice. As shown by previous investigators, no boulders of foreign origin 
occur in the Faeroic moraines. 

The problem of the thickness of the ice has frequently been touched upon 
in previous literature, and it has generally been assumed that the ice did not 
reach above 500 m since striae have not been observed above that height. 
Otherwise, however, observation of striae is difficult at that height on account 
of the extensive weathering, and locally the very dense, resistant rocks (as, for 
example, the sills) are distinctly glaciated up to far greater heights. 

As to the extension of the ice, naturally, nothing can be suggested, but 
remnants of cirques and valley slopes, particularly on the west and north 
coasts, show that the ice has extended a good deal farther than the present 
coast line, and that the postglacial coast erosion on the west and on the 
north is considerable. 

A glance at the map of the Faeroes shows an obvious NW.-SE. running 
dominance of the fjord direction. The same orientation prevails in the 
traversing valleys (e.g. Milium Fjarda, Sandsdalur). 

The origin of the fjord system, including the traversing valleys, has been 
discussed frequently. Mackenzie (1814) suggested that they resulted from 
removal by erosion of great dykes. Geikie (1880) showed that the ice followed 
the Pre-Glacial topography and explained the fjord system as a consequence of 
river action assisted by Glacial erosion. Grossman and Lomas (1895) pro- 
posed that the ice did little more than modify pre-existing valleys, Gregory 
(1913) assumed that the fjord system was preglacial in age, and that its 
formation thus could not be due to glacial erosion. Peacock (1928) suggested 
that the fjord system was derived from an earlier goe-system, and that it 
therefore had its ultimate origin in a system of master joints related to an 
episode of crustal movement which took place after the volcanic period but 
before the Ice Age, 

When mentioning the Tuff-Agglomerate zone it was pointed out that the 
agglomerates most likely covered the eruption fissures which were feeding the 
Lower Basalt sequence, and that those occur along the fjords following them. 
In the same manner the numerous vents, which probably were supplying the 
Lower Basalt zone in the Upper Basalt sequence, occur along the fjords. No 
vents have been observed outside the fjord system although the opportunities 
for observation are excellent. Tt might therefore be tempting to assume that the 


volcanic activity was localized throughout along the old fissures and that the 
ultimate origin of the fjord system was determined by these fissures. 

Lakes are numerous in the bottom of the traversing valleys as well as in 
the cirques. Most of them occupy true rock basins excavated under varying 
conditions. Formerly they were, no doubt, much more abundant, as many of 
them are now being silted up and replaced by alluvium. 

Alluvial deposits are rather insignificant in the Faeroes. Marine alluvium 
is quite lacking and freshwater alluvial layers show far less variety and far less 
thickness than elsewhere. 

Peat occurs everywhere in the lowlands along the slopes of the valleys as well 
as on their bottoms. The thickness of the peat is generally 1 or 1.5 m, rarely up 
to 3 m. The greatest thickness of biogene freshwater alluvium has been 
measured to 5.5 m. 

Our knowledge of the Faeroic bogs is limited to a joint study by Jessen 
and Rasmussen (1922) on a section of a bog in the Faeroes, and to work by 
Jessen (1923) alone on the stratigraphy of Faeroic bogs. 

The most important alluvial freshwater layers according to Jessen (1923) 
are: mud, rarely in thick layers; Equisetum-peaX, not common in thick layers; 
and low-bog peat, which appears in two forms, Sphagnum-peat and Erioplwrum 
angustifolium-peat. The latter constitutes the bulk of the peat in the Faeroes. 
The most important fossils in the Faeroic low-bog peat are Erioplwrum 
august {folium, Care.x stellulata, and Ranunculus flammula. Moreover, there is 
a Calluna-peai with twigs and roots of Calluna vulgaris and Juniperus com- 
munis. Pure Spliagnum-peat is very rare. Clay, sands, and gravel often form a 
considerable part of the bog sections. 

The Calluna-peat, according to Jessen and Rasmussen (1922), occurs as a 
horizon dividing the Faeroic bogs, and the authors are of the opinion that it 
is contemporaneous with Upper Forrestian and the Sub-Boreal period. 

Submarine bogs occur frequently along the eastern shores of the Faeroe 
Islands, indicating that a depression has taken place in Post-Glacial time 
According to Jessen and Rasmussen (1922), the land has been submerged at 
least 3.5 m after the formation of these bogs. 

Therefore, since distinctive features of a Post-Glacial upheaval of the land 
are visible along the western coasts (Mykines, Vagar, Hestoy) and recent 
movements can be indicated in the "lamellae zones", it is tempting to con- 
clude that the Late Tertiary movement is still active along the Mykines-Fugloy 
axis with an upheaval to the west and a submergence to the east. 


Allan, Th. (1814). An account of the mineralogy of the Faroe Islands. Trans. Roy. Soc. 

Edinb. 7, 229-67. 
FoRCHHAMMER. J. G. (1824). Om Faeroernes geognostiske Beskaffenhed. Kgl. Damke 

Vidensk. Selsk. Skr. 2, 159-206. 
Geikie, J. (1880). On the geology of the Faeroe Islands. Trans. Roy. Soc. Edinb. 30, 217-69. 

44 j6annes rasmussen 

Gregory, J. W. (1813). The Nature and Origin of Fiords. London. 

Grossman, K. and Lomas, J. (1895). On the glaciation of the Faeroe Islands. The Glaciol. 

Magazine, 1895, 1-15. 
Helland, Amund (1880). Om Faeroernes geologi. Geogr. Tidsskr. 4, 149-79. 
Jessen, K. and Rasmussen, R. (1922). Et Profil gennem en Torvemose paa Faeroeme. 

Damn. Geol. Undersog. 4 (1), 1-13. 
Jessen, K. (1925). De faeroske Mosers Stratigrafi. Det 17. Skand. Naturforskarmotet i 

Goteborg 1923, 185-190. 
Mackenzie, Sir George (1814). An account of some geological facts observed in the 

Faeroe Islands. Trans. Roy. Soc. Edinb. 7, 213-27. 
Noe-Nygaard, a. (1940). Om Gj6gv-Systemernes Alder paa Faeroeme. Medd. Dansk 

Geol. Foren. 9, 542-7 
Noe-Nygaard, A. (1946). Some petrogenetic aspects of the Northern basalt Plateau. 

Medd. Dansk Geol. Foren 11, 55-65. 
Noe-Nygaard, A. (1951). Den geologiske Kortlaegning af Faeroeme 1938-50. Medd. 

Dansk Geol. Foren. 12, 163. 
Noe-Nygaard, A. and Rasmussen, J. (1957). The making of the basalt plateau of the 

Faeroes. Congreso Geol. Intern. Mexico 1956, 299-407. 
Peacock, M. A. (1928). Recent lines of fracture in the Faeroes, in relation to the theories of 

fiord formation in the northern basalt plateau. Trans. Geol. Soc. Glasgow 18, 1-26. 
Rasmussen, J. (1951a). Nyere synspunkter vedrorende de faeroske kullags stratigrafi. Medd. 

Dansk Geol. Foren. 12, 164. 
Rasmussen, J. (1951b). Transgressive sillintrusioner i Faeroplateauet. Medd. Dansk Geol. 

Foren. 12, 164. 
Rasmussen, J. (1952). Bidrag til forstaaelse af den faeroske lagseries opbygning. Medd. 

Dansk Geol. Foren. 12, 275-83. 
Rasmussen, J. (1955). Nokur ord um gj^ir i Foroyum — upprana teirra og aldur (Notes on 

origin and age of goes). Frodskaparrit {Ann. Soc. Set. Faeroensis), 4, 108-24. 
Rasmussen, J. (1957a). Yvirlit yvir innskotin grotslog i Foroyum (General view of intrusive 

rocks of the Faeroe Islands). Frodskaparrit (Ann. Soc. Sci. Faeroensis), 6, 61-96. 
Rasmussen, J. (1957b). lies Faeroe. Lexiqiie stratigraphique international. Paris, 1, 13-17. 
Trevelyan, W. C. (1823). On the mineralogy of the Faeroe Islands. Trans. Roy. Soc. 

Edinb. 9, 461-4. 
Walker, F. and Davidson, C. F. (1936). A contribution to the geology of the Faeroes. 

Trans. Roy. Soc. Edinb. 58, 869-97. 


Eric Hulten 
Department of Botany, State Museum of Natural History, Stockholm 50, Sweden 

The present study is based on distributional maps of all plants known to 
occur in Iceland, Greenland, Spitsbergen, Jan Mayen, Bear Island, and the 
Faeroe Islands. The maps have been divided into 23 groups according to 
their general geographical area : Circumpolar, European, American, Amphi- 
Atlantic, and so on. 

The number of species belonging to each group and found in certain 
localities distributed all over the map of the northern Atlantic have been 
counted, and places with the same number of species in each category have 
been united with Unes. In this way an idea has been gained as to how many 
species of each group occur at different places. We find, for instance, how 
many Circumpolar plants occur on Iceland or on Spitsbergen, and how many 
species with their main area in continental Europe exist on these islands. The 
maps should give a good idea of the relationship of the floras of the above- 
mentioned Atlantic islands to each other as well as to adjacent continental 
Europe and America. 

To each map has been added a sketch of the world range of all species 
belonging to the group in question. As these ranges are very variable in 
different species, these sketches can only give a general idea of their distribu- 
tion. However, the sketches will help make the conditions understandable. 

Of fundamental importance for the phytogeography of the North Atlantic 
is the question as to whether the species are native or introduced. Com- 
munication between Iceland, Europe, Greenland, and America has been 
intense and has been going on for several centuries. Very many weeds, or 
plants otherwise not native, have been added to the natural floras. It is 
desirable to exclude them from this discussion as they do not concern the 
older history of the flora. About 210 such species, recognized by most 
botanists as not belonging to the native flora, occur within the area ; most of 
them are of European origin, and only about ten are American. 

In Iceland, Greenland, and Newfoundland, however, there are European 
species considered native by botanists dealing with the respective floras, but 
whose character suggests that they possibly have been originally introduced. 
They occur in the natural vegetation, but this is hardly a rehable indication 


46 ERIC hult6n 

that they belong to the native flora. If, for instance, a European species is 
introduced to Newfoundland and the climate there is suitable, it might well be 
able to compete with the native vegetation and become a member of the 
natural plant societies after some centuries. Personally, I believe that not a few 
of the European plants in Iceland, Greenland, and Newfoundland fit into this 
category, as for instance, some of the species occurring around Reykjavik 

As plants occurring in central Europe must have a different history from 
those not reaching south of Scandinavia, they have here been referred to 
other groups. The Circumpolar plants lacking in Greenland, for instance, 
have been divided into two sets, those occurring in central Europe (Fig. 1), 
and those not reaching that far south (Fig. 2). 

A review of the groups represented by the maps follows. 

In Fig. 1 are shown the ranges of those Circumpolar, or nearly Circumpolar, 
plants which do not occur in Greenland, nor in central Europe: 27 species. 
Only one of them. Ranunculus pallasii, exists in Spitsbergen but not in Iceland. 

Figure 2 includes Circumpolar plants not occurring in Greenland, but 
found in central Europe: 64 species. They are of a more southern affinity but 
16 of them have reached Iceland. 

Figures 3, 4, and 5 correspond to Figs. 1 and 2, but include species occurring 
also in Greenland. Figure 3, comprising those species not occurring in central 
Europe, has an Arctic or Arctic-montane character. Species found also in 
central Europe have been divided into two groups, namely one with an Arctic- 
montane character (Fig. 4) and another with a more boreal character (Fig. 5). 
No sharp line of demarcation can be drawn between Figs. 4 and 5, but as their 
areas are quite different it is impractical to include them in the same map. 

These three groups (Figs. 3, 4, 5) comprise altogether 197 species, all of a 
more or less Circumpolar character and all found in Greenland or on the 
Atlantic islands. They form the main element of the Atlantic flora. It should 
be noted that all of them have northern ranges around the Pole with 131 of 
them occurring on Iceland, 98 on Spitsbergen, 22 on Bear Island, 24 on Jan 
Mayen, and 58 on the Faeroe Islands. 

Figures 6 and 7 consist of nearly Circumpolar plants which show the same 
behavior in the Atlantic sector as the other Circumpolar plants but have a 
peculiar gap in their total range in northern Asia, as for instance, Silene 
acaulis, Loiseleuria procumbens, and Rhododendron lapponicum. Here they are 
treated separately only because it can be suspected that they have a different 
history from those with more or less continuous distribution all around the 
Pole. The total area of some of them approaches that of the Amphi-Atlantic 
plants. As seen from the maps illustrating their total ranges, they are Arctic- 
montane species, but sometimes with very split-up areas. They seem to have 
had an earlier continuous connection also in northern Asia, but at a later 
stage have lost part of their area owing to changes in the climate. From an 


Fig. 1. Circumpolar plants lacking in Greenland and central Europe. 



Fig. 2. Circumpolar plants lacking in Greenland but occurring in central Europe. 


Fig. 3. Clrcunipolar plants occurring in Greenland but lacking in central Europe. 




Fig. 4. Circumpolar plants occurring in Greenland and isolated in the mountains 

of central Europe. 


Fig. 5. Circumpolar plants occurring in Greenland and with lowland areas in 

central Europe. 



Fig. 6. More or less Circumpolar plants with large gaps in their area in northern 
Siberia, not occurring in central Europe. 


Fig. 7. More or less Circumpolar plants with large gaps in their area in northern 
Siberia, also occurring in central Europe. 


Atlantic point of view they might as well have been included in other Circum- 
polar groups. 

In Figs. 8, 9, and 10 are treated plants which have a more or less Circum- 
polar area with gaps frequently occurring in different places, but, contrary to 
the plants of the preceding groups, they appear as different races or as 
corresponding, slightly differing species on both sides of the Atlantic. There 
are altogether 107 of these taxa. 

Figure 8 includes 76 species not occurring on Iceland or in Greenland, 
while Fig. 9 includes cases where the European type reaches Iceland or 
Greenland: in all 21 species. Figure 10 represents American species reaching 
Greenland and Iceland: 10 species. It is remarkable that all plants with such 
different races on both sides of the Atlantic are comparatively southern 
plants, mostly of the Boreal type. Their Atlantic connections are very feeble. 

Figures 11 to 14 deal with the so-called Amphi- Atlantic plants. As they 
are of special interest in this connection, they have been subdivided into 
smaller groups. Thus, species not occurring in central Europe have been 
divided into the two groups: namely one having a northern distribution in 
Greenland, and another with a southern area there. Species found in central 
Europe have also been divided into two groups : one that reaches Iceland and 
Greenland, and one that does not. 

The sketches illustrating the corresponding total areas indicate that species 
occurring on Iceland and in Greenland are Arctic or Arctic-montane types, 
whereas those lacking there are of a Boreal, lowland type. All form a chain of 
species with the largest part of their ranges in the Atlantic sector, but with 
gaps in the Pacific sector. In my opinion they are remnants of earlier con- 
tinuously Circumpolar plants which have lost part of their area due to pressure 
of changing conditions, but are not remnants of plants which once inhabited a 
land-bridge in the Atlantic. They comprise 71 species of which 25 occur on 
Spitsbergen, 7 on Bear Island, 6 on Jan Mayen, 41 on Iceland, and 26 on the 
Faeroe Islands. Their Atlantic connections are thus not very strong. This is, 
however, perhaps of less importance, as it must be assumed that their areas 
became much reduced when the continental climate that must once have 
existed in what is now the North Atlantic changed to a coastal one when the 
land-bridge disappeared. Their total areas do not indicate that they are 
truly continental plants; on the contrary, their ranges even now are reduced in 
the interior of the continents. 

Figure 15 shows the distribution of American or American-eastern Asiatic 
plants which do not reach Greenland or the Atlantic islands, and Fig. 16 
indicates the corresponding European plants which do not extend out into the 
Atlantic. Both contain a great number of species. Their approximate numbers 
in different places have been indicated on the map in order to stress the great 
difference in the flora of the Atlantic shores in Europe and in America. 

Figures 17 and 18 show the areas of American plants (with variable ranges 


Fig. S. Plants with difterent races or corresponding species on both sides of the 
Atlantic, not occurring in Greenland or Iceland. 



Fici. 9. Plants with different races or corresponding species on both sidei of tlie 
Atlantic. The European type reaching Iceland-Greenland. 


Fig. 10. Plants with different races or corresponding species on both sides of the 
Atlantic. The American type reaching Greenland-Iceland. 



Fig. 11. Amphi-Atlantic plants lacking in central Europe with a northern area in 



Fig. 12. Amphi-Atlantic plants lacking in central Europe with a southern area in 




Fig. 13. Amphi-Atlantic plants lacking in Greenland but occurring in central 



Fig. 14. Amphi-Atlantic plants occurring in Greenland and also in central 




Fig. 15. American plants lacking in Greenland. 


Fig. 16. European plants lacking in Iceland, Greenland and E. America. 



Fig. 17. American plants reaching northern Greenland. 


Fig. 18. American plants reaching southern Greenland-Iceland-Europe. 


on that continent) which reach respectively northern and southern Greenland, 
and comprise altogether 74 species. Only 6 of them have extended as far as 
Iceland, and 4, Spiranthes Romanzojfiano. Potaniogeton epi/iydrus, Eriocaulon 
septentrionale, and Sisyrinchium sp., to the British Isles. A strong connection 
between America and Greenland is thus demonstrated. 

Figures 19 and 20 represent the corresponding European or Eurasiatic 
elements penetrating out into the Atlantic. The plants in Fig. 19 reach Iceland 
or Greenland but not America, and those in Fig. 20 extend also to America. 

It is especially in these groups that the question about native or introduced 
species is particularly important. In my opinion, many plants occurring in 
Iceland have been introduced to that island, because the composition of the 
group seems to indicate that it consists to a large extent of plants favored by 
human activity. On the other hand, a few certainly do not belong to an 
introduced flora, for instance, Saxifraga hypnoides, Vaccinium Myrtillus, 
Salix lanota. Orchis macu/ata, Paris quadrifolia, Saxifraga Cotyledon, Hydro- 
cotyle vulgaris, Gentiana aiirea, Gentiana detonsa, Veronica fruticans. Geranium 
silvaticum, and a few others. 

Figure 20 includes those European species which occur in eastern America 
and are considered by Fernald in his Manual to be native there. It seems to me 
that most, if not all, of them probably have been introduced into America; 
those which have the best chance to be of native origin are Pedicularis 
silvatica, Juncus capilatus, J. bulbosus, J. subnoduJosus, and Ranunculus 

In Figs. 21 and 22, respectively, are considered Atlantic Circumpolar and 
coastbound plants. It seems preferable to treat coastbound species separately 
because their distribution does not always conform to that of other plants. 
Several of these species have large inland areas on salt soi! in the center of the 
continents. A few have a fairly unbroken connection from the Atlantic to the 
Pacific along the shores of the Arctic Ocean, whereas others show large gaps 
there. This indicates that during a warmer period the latter may have had such 
an unbroken connection which was later broken up by the deterioration of the 

Figures 23 and 24 represent endemics in Greenland or on Iceland-Faeroes. 
They are all very weak neo-endemics, mostly of critical genera, and cannot 
support a land-bridge theory. 

A few plants have unique areas and could not be placed in any of the 
above-mentioned groups; thus, PotentUla stipularis and Draba sihirica have 
their main ranges in northern Asia with only single outpost localities in 
Greenland, whereas Cakile edentula occurs on the American coast with different 
types in different places. It has reached Iceland, and furthermore it has also 
extended to the Azores, apparently by means of sea currents. 

Summarizing the above, the following can be stated concerning the 
phytogeographical connections of the North Atlantic: Very many species 


Fig. 19. European plants reaching Iceland-Greenland. 



Fig. 20. European plants reaching easternmost America. 


Fig. 21. Atlantic coastbound plants. 



Fig. 22. Circumpolar coastbound plants. 


19 tax£ 

Fig. 23. Endemics in Greenland. 

that do not occur in Europe attain the shores of the Atlantic in America, and 
similarly, many European plants reaching the Atlantic shores do not occur in 
America. A high number of Circumpolar plants with a Boreal area of a usually 
somewhat continental type have different races in eastern America and in 
Europe. Many other species are represented by corresponding, but different, 
counterparts on both sides of the Atlantic. The differences are thus large and 

6 taxa 

Fig. 24. Endemics in Iccland-Faeroe Islands. 


certainly old. That diverse types have developed to such a large extent indicates 
that a very long time has been at their disposal. 

The plants which show the most obvious connections over the Atlantic are 
those with a more or less Circumpolar area, at present having a northern 
boundary line fairly far to the north. The distance between the American and 
European continents is smallest in the north where the connection is strongest. 
Plants like Poteiitil/a stipularis and Draba sibirica show that occasional 
introductions are possible over long distances there. It seems probable that 
the Circumpolar flora may have been spread by the wind blowing over the 
frozen Polar Sea or by floating on ice. We know that nowadays on the large 
floating Ice Islands in the Arctic Sea, Uving, well developed, flowering plants 
occur, drifting around the Polar Basin. 

As the flora around the Polar Basin apparently is an old one only a few 
introductions are necessary in every century to account for a fairly even 

The strongest botanical argument for a land-bridge over the North Atlantic 
is the Amphi- Atlantic plants although only little of the supposed connection 
is to be seen today, even in this group. Many species are missing in Greenland, 
on Iceland, and ,6n the other Atlantic islands. The present ranges, however, 
could be interpreted as centering around the Atlantic, thus indicating that 
they once spread from a center there. But they may also be explained as 
reductions of former Circumpolar ranges caused by changing climatical 

T^.endSmics occurring in the Atlantic sector are very weak indicators of a 
vious connection. 

The phytogeographical conditions around the North Atlantic thus, give 
poor support for a land-bridge that could have existed in Quaternary or Late 
Tertiary times. 


Carl H, Lindroth 

Zoological Institute, University of Lund, Lund, Sweden 

The idea of an earlier land connection between Europe and North America 
first occurred to biogeographers who wished to explain the striking similarity 
of the flora and fauna of the two continents. It found its most ardent supporter 
in R. F. Scharff (1907, 1909, 1911) and according to Th. Arldt (1917, p. 83) the 
Tertiary North Atlantic land-bridge constituted "eine der am sichersten 
feststehenden palaeogeographischen Tatsachen". The latest to express his 
acceptance of this hypothesis was the Norwegian botanist E. Dahl (1958). 
An Early Tertiary trans-Atlantic land connection has also been assumed on 
paleontological evidence (cf. Simpson, 1940, p. 149; 1947, pp. 658, 666). 

An analysis of the animal species common to Europe and North America 
(Lindroth, 1957) has shown, however, that the majority of these belong to 
either of two about equally large groups: one, whose present distribution is 
a result of unintentional transport by man, mainly in the direction from 
Europe to North America; another, having a more or less complete circum- 
polar area, where the migration between the two continents went through a 
"back door", via Siberia and the Beringian land-bridge. This bridge is now 
considered by geologists as well as biogeographers almost unanimously to 
have functioned as a continuous land connection during each of the Pleisto- 
cene glaciations. 

Exceptions from the rule that the faunal and floral exchange between 
Europe and North America took place by the way of Asia may be expected 
in the group of Amphi- Atlantic species (cf. Hulten, 1958) — provided the 
plants and animals are indigenous in both continents. This is an important 
restriction since several examples of Amphi- Atlantic distribution, such as that 
of the common garden snail, Cepea hortensis L., used as arguments in favor 
of a Eur-American connection, are now supposed to be the results of 
introductions into North America. 

However, originally Amphi-Atlantic species do no doubt exist among 
animals as well as plants though they are not numerous. It is our task to 
explain their history and to decide whether an earlier continuous land connec- 
tion between the two continents can be regarded as responsible for their 
present distribution. If so, the islands of the North Atlantic, as remnants of a 




supposed "bridge", would be the most suitable starting-point and we should 
begin the discussion with an analysis of their faunas and floras. 

The Faeroes and Iceland are inhabited by almost purely European biota. 
The few undisputable cases of Nearctic elements on the islands are listed in 
Table 1. 

Table 1. Nearctic Elements in the Flora and Fauna of Iceland and 

THE Faeroes 
(lacking on the European mainland) 




Vascular plants* 

Leiicoichis straminea (Fern.) Love 




Habenaria hyperborea L. 




Epilobiiiin latijolium L. 





Heteronieyenia ryderi Potts 





Diaptomus minutus Liljeb. 





Colymbetes dolabratus Payk., sbsp. 





Crino sommeri Laf. 




Cryinodes exulis Laf. 




Rhyacia qiiadmngula Zett. 





Siimiliiini vittatum Zett. 




Aves (breeding) 

Gavia inimer Briinn. 




Biicephala islandica Gm. 




Histiionicus histrionicm L. 




* The number of Nearctic vascular plants may actually be higher in Iceland 
(Love and Love, 1956, p. 171). At least Sali.x glauca callicarpaea (Trautv.) 
and Galium bievipes Fern. & Wieg. (Biandegei auct. p.p.) should perhaps be 
considered; the former is accepted by Hulten (1958, p. 186). On the other 
hand, 1 have included Leucoichis straminea above, though it is not kept 
separate from albida L. by Hulten Hoc. cit. p. 116), Bocher, Holmen and 
Jakobsen (1957), and others. Also, Carex Lyngbyei Hornem., occurring on 
Iceland and the Faeroes, seems to be Nearctic (Hutten, loc. cit., p. 292). 

On the Faeroes, they consist of four species: one Orchid with extremely 
minute seeds; two Noctuid moths with excellent flying ability; finally, the 
freshwater sponge Heteromeyenia ryderi, not found on Iceland but known 
from Ireland and the Isle of Mull in the Inner Hebrides. It has been suggested 
(Arndt, 1928, p. 159; Lindroth, 1957, p. 245) that the occurrence on the 
Faeroes, at least, is the result of bird transport of the resistant "gemmulae"' 
of this sponge. 

Iceland possesses a somewhat larger American element (Table 1): three 
vascular plants (at least); one freshwater crustacean of the genus Diaptomus; 
the eight remaining species being flying animals, insects and birds. The 



Diaptomus species are able to produce thick-walled winter eggs, very resistant 
to desiccation (Wesenberg-Lund, 1937, p. 523, etc.). Considering the geo- 
graphical position of Iceland, much closer to Greenland {ca. 300 km) than to 
the European mainland (Norway ca. 950 km) and the British Isles {ca. 800 km), 
the extremely poor representation of a Nearctic element is indeed surprising. 
Furthermore, it seems to consist exclusively of species with above average 
ability of dispersal. There is little doubt that the Nearctic animals and plants 
of Iceland were able to invade the island by active flight or by passive, "non- 
human" transport during present-day conditions. 

Table 2. Representative Groups of the Terrestrial Greenland Fauna 
Indigenous species and subspecies. Endemic subspecies of birds are distributed among the 
three geographical groups according to their taxonomic relation to other subspecies (from 

Lindroth, 1957). 






Aves (Salomonsen and Gitz, 1950) 
Macro-Lepidoptera (Henriksen, 1939) 
Coleoptera (Lindroth, 1957) 
Collembola (Hammer, 1953) 
Araneae (Braendegaard, 1946; 
Holm, pers. comm.) 


37 = 45% 
7 = 29% 
27 = 59% 
21 = 42% 

12= 10% 
3 = 12% 
11 = 46% 
13 = 28% 
11 = 22% 

19 = 28% 

9 = 35% 

6 = 25% 


18 = 36% 



106 = 50% 50 = 23% 

58 = 27% 

The fauna of Greenland, if distributed among geographical groups (Table 
2), is a veritable "mixture". Roughly counted, it consists of one-half Circum- 
polar taxa, and about one-quarter each of Nearctic and Palearctic forms, 

However, it is easily observed that different animal groups behave differently 
in this respect. The highest percentages of Palearctic forms are found among 
Coleoptera and Collembola; among Nearctic forms the highest frequency 
occurs in Araneae, Lepidoptera, and Aves. This is, of course, a biological and 
not a taxonomical feature. It is an expression for differences in dispersal 
ability : The Nearctic element of the Greenlandic fauna consists of a high 
proportion of easily dispersed animals, able to traverse at least moderate 
distances over the sea by active flight or by passive, aerial transport. The 
Palearctic element includes many soil-bound species not easily dispersed by 
these or other methods. 

A closer examination of the entire Coleopterous fauna of the North Atlantic 
islands provides a clearer picture (Fig. 1). The purely Palearctic character of 
the fauna of the Faeroes and Iceland is strongly manifest and, what is more 
itnportant, the species concerned are to a great extent flightless and thus not 



easily spread. In principle, this applies also to the Coleoptera of Greenland. 
There, the Nearctic element contains a single flightless beetle, the StaphyUnid 
Micralymma brevilingue Schio., fit for hydrochorous dispersal in salt water. 

Fig. ]. Species and subspecies of indigenous Coleoptera of the North Atlantic 
islands and their occurrence in selected regions of the Holarctic. 

White = flying; black = flightless. (From Lindroth, 1957.) 

Essentially ditferent is the fauna of Baffin Island, separated from 
Greenland by a strait only 350 km at its narrowest (that is, approximately 
the same distance as from Iceland, ca. 300 km). The Coleoptera of Baffin 
Island are truly Nearctic having one (Circumpolar) species in common with 
Iceland, none with the Faeroes, and, respectively, two and three with the 
British Isles and Scandinavia. All of them are flying forms. Actually, the 



fauna of Baffin Island shows much more relationship to east Siberia than to 

As far as the Coleoptera of Iceland, Greenland, and Baffin Island are 
concerned, the situation may be illustrated, somewhat roughly, in the 
form of a map (see Fig. 2). It demonstrates the blocking effect west of Green- 
land by Davis Strait, which may be regarded as the most effective faunal 
barrier of the entire Circumpolar area (Lindroth, 1957, p. 264). 

Fig. 2. The Palaearctic and Nearctic elements of indigenous Coleoptera in Green- 
land and adjacent areas. Black sectors = flightless forms. Size of circles in 
proportion to number of taxa (from 1 to 11). (Data from Lindroth, 1957.) 

The distribution in the same area of Nearctic and Palearctic elements of 
vascular plants (Fig. 3) gives less evidence because most plants are apparently 
better suited than soil-animals to passive long-distance dispersal (Lindroth, 
1960). But the principle remains the same: plants with low abihty of dispersal 
are to a considerable degree blocked by Davis Strait; the Palearctic element of 
Greenland includes proportionally more forms not adapted to long-distance 
dispersal than does the Nearctic element of that island. 

This is, indeed, surprising, considering the geographical position of Green- 
land. It is tempting to resort to the strong pressure of conventional thinking 
and suppose that if Greenland had belonged to North America politically and 
not only geographically, everybody would have noticed the strange composi- 
tion of its fauna and flora. 

What ought to be expected is, indeed, that the long distance from Europe — 
against prevailing winds! — could have been successfully conquered only by 
organisms with extraordinary means of dispersal, whereas an immigration 



from Baffin Island and the North American mainland should have been much 
easier to accomplish. Thus, the actual situation is contrary to all calculations. 
Either of two explanations seems possible: 

1. The influence of man has been underestimated. The Palearctic element of 
plants and animals in Greenland is largely a result of introduction by the old 
Norsemen (and the same would apply to Iceland). 

2. A considerable part, notably the Palearctic element, of the flora and 
fauna of Greenland (and Iceland) has immigrated during a period of geo- 
graphical conditions fundamentally different from those of the present time. 
This is the so-called "Land-bridge Theory". 

Fig. 3. The Palearctic and Nearctic elements of indigenous vascular plants in 

Greenland and adjacent areas. Black sectors = species without adaptation to 

long-distance dispersal. Size of circles in proportion to number of taxa (from 3 to 

55). (Data from Lindroth, 1960.) 

Commenting on point (1), it should be remembered that the Palearctic 
plant species of Greenland have been thoroughly scrutinized by M. P. Porsild 
(1932), considering the possibilities of introduction by man from Europe. 
Similar calculations on the insect fauna have been made by myself for Iceland 
(Lindroth, 1931, p. 516) and for Greenland (Lindroth, 1957, p. 268). The con- 
clusions drawn are that both islands possess an old and indigenous Palearctic 
element. Among other arguments, it is impossible to accept that a haphazard 
dispersal by anthropochorous transport could result in the rather homo- 
geneous fauna of Greenland, Iceland, and the Faeroes, that have so many 
flightless species in common. 

We have reached our main problem here and I would like to declare my 
position at once: The fauna and flora of the North Atlantic islands, notably of 



Iceland and Greenland, claim decidedly a continuous land connection with 

This is, it should be observed, not the continuous Eur-Anierican land- 
bridge postulated by Scharff( 1907, 1909, 1911), and others. As here proposed, 
it includes Greenland but does not extend beyond Davis Strait. 

The bottom configuration of the North Atlantic Ocean seems rather 
favorable for the appearance of a land-bridge (Fig. 4). A positive displacement 


Fig. 4. The bottom configuration of the North Atlantic. (From Lindroth, 1957.) 

of the shore-line amounting to less than 600 m would connect the Faeroes, 
Iceland, and Greenland with Europe along the present Wyville-Thompson 
ridge ("the Greenland-Scotland ridge"). Now, 600 m would seem like a modest 
depth in comparison with ordinary deep-sea values exceeding 2000 m even 
within short distance and on both sides of the Wyville-Thompson ridge. 
But it is nevertheless considerably more than the changes in sea level regarded 
as acceptable in late geological times, at least during the Pleistocene. 

The eustatic depression of the sea level, due to storage of the precipitation 
in form of ice on the continents, has been calculated as hardly exceeding 
100 m during each of the last two (and possibly earlier) Pleistocene glacia- 
tions. There is no reason to believe that similar events have occurred in Terti- 
ary time. Pliocene, or earlier, glaciations (if a reality; cf. Schwarzbach and 
Pflug, 1957, p. 295) were under all circumstances less extensive. 

No isostatic movements of the Earth-crust, caused by ice depression and 
simultaneous upheaval of marginal areas, can have exerted any noticeable 
effect on the now submerged parts of the Wyville-Thompson ridge. 

There is also a special biological reason for denying the importance of 



iso- and/or eustatic movements effecting positively the rise of a North Atlantic 
land connection: The Palearctic element of the Greenland -Iceland biota, the 
history of which we are here trying to trace, is not Arctic, not even Sub- 
Arctic; the insects, at least, are members of a Boreo-Temperate fauna. 
Therefore, they must have immigrated, not during a Glacial, but an Inter- 
glacial period, possibly in Preglacial time. 

Since iso- and eustatic changes of the sea level fail to explain the assumed 
land connection, the remaining possibility is that it was created by tectonic 
movements within the area, or — more adequately expressed — that continuous 
land once existed but was broken down by tectonic processes. 

This is perhaps more than a mere hypothesis. From northern Ireland and 
western Scotland over the Faeroes, Iceland, and Greenland west to Disko 
Island in Davis Strait runs a chain of volcanic rock-occurrences, mainly 
basalt, formed in Tertiary time (Fig. 5). Whether these are remnants of a 

Fig. 5. Tertiary volcanic rock (mainly basalt) of approximately similar age in the 
North Atlantic area. (Compiled from different sources.) 

continuous land area occupying large parts of the present North Atlantic 
region, is still a matter of dispute (Einarsson, 1961), but some geologists 
(e.g. Schwarzbach, 1959, pp. 32, 36) think it is probably so. 

The crucial question from the point of view of a biogeographer is how 
long the assumed land connection existed. To geologists, this question is of 
subordinate interest and only few of thein have tried to provide an answer. 
Schwarzbach and Pflug (1957, p. 296) do not think that connection between 
Iceland and the British Isles was probably after Eocene, and later, Schwarz- 


bach (1959) seems inclined to remove a possible Eur-American "bridge" to 
Pre-Tertiary time. 

Unfortunately, we do not know even approximately how old the animal 
species are that now inhabit the Earth. The rate of evolution has been of quite 
different magnitude in different groups as well as within different genera of the 
same group. In spite of this, some idea of the minimum of time required for 
the formation of a "new species" would be extremely valuable for the 
dating of zoogeographical events. 

As far as insects are concerned it is safe to say that no evidence for species 
formation in Pleistocene time has been brought forth. The fossils from Inter- 
glacial or Interstadial deposits so far investigated (from Scandinavia, Great 
Britain, and central Europe) seem in no case to be different from now living 
representatives of the species. 

On the other hand, the insect fauna of Late Eocene, as amply illustrated by 
thousands of very well-preserved fossils in the Baltic Amber, was quite 
distinct on the species level. Most genera were identical with those of the 
present time but very few are regarded as belonging to the same species 
(e.g. among the Coleoptera, only one or two). 

Thus, species formation among insects in general seems to have taken place 
after the Baltic Amber time, but before the Pleistocene, i.e. during a period of 
between 40 and 1 million years ago. 

If the postulated North Atlantic bridge with its biota lasted only during 
Early Tertiary and was broken down before the Oligocene period, and if 
part of the present fauna and flora had survived from that remote time, then 
there is no doubt at all that they would have contained a considerable number 
of taxonomically well distinguished endemic species. For comparison we 
need only refer to the numerous endemic insect species which survived, often 
restricted to single mountain peaks, throughout the Pleistocene within the 
massifs de refuge in the European Alps. 

Several biogeographers (e.g. Love and Love, 1956, p. 235, etc.; Larsson, 
1959, p. 36, etc.; Einarsson, 1961, p. 46) have assumed that the land-bridge 
persisted into Late Tertiary time with subsequent survival of the biota through 
all Pleistocene glaciations. As far as I can see, however, old Tertiary faunal and 
floral elements would have been left on Iceland and Greenland also in case 
the bridge lasted into Pliocene time, with a peculiar flora in Iceland, rather 
different from that of contemporary Europe (Jonsson, 1954; Schwarzbach 
and Pflug, 1957, p. 289, etc.). 

It is also noteworthy that the shallow and narrow sound between Greenland 
and Ellesmere Land in the High North was not used as a by-pass for the 
Icelandic-Greenlandic Palearctic elements. This is easily understood if it is 
assumed that the species in question immigrated from Europe as far as 
Greenland during the Pleistocene, when the climate was never much warmer 
than now. But it is much harder to realize that this northern connection would 


not have been biologically effective, if the trans-Atlantic land connection is 
removed entirely to Tertiary, Early Tertiary, or even to Cretaceous time, i.e. 
to periods with a warm-temperate climate also on the North Atlantic islands 
(Schwarzbach, 1955, p. 101, etc.). 

The most satisfactory explanation of present faunal and floral conditions in 
Iceland (and Greenland) would be to assume that the Tertiary biota were 
destroyed and followed by a new invasion from Europe. It is difficult to 
imagine such a total destruction by any other agency than a glaciation. 

According to this hypothesis, the land connection Greenland-Iceland- 
Scotland persisted (or was re-established) in at least one of the Interglacial 
periods, offering an immigration route for Boreo-Temperate biota from 
Europe. It is known from plant fossils that the climate of Iceland was slightly 
warmer than now during at least one Interglacial period (with Pinus and 
Picea in deposits). It was with some hesitation referred to the "Giinz-Mindel" 
by Askelsson (1960). Alnus appeared in the "Mindel-Riss" Interglacial 
(Thorarinsson, 1958). During the last Interglacial, the "Riss-Wurm", at least 
Betula trees grew on the island (Thorkelsson, 1935). 

I am perfectly well aware that the hypothesis here emphasized (and 
maintained also earlier: Lindroth, 1931, p. 551; 1957, p. 253, etc.) is against 
the opinion of almost all geologists. It would be interesting to live long enough 
to see how close to the truth it may be ! 

Since every idea of a North Atlantic land-bridge in Post-glacial time, 
previously assumed by some older authors (e.g. Simmons, 1905), is a geological 
impossiblity, the hypothesis stressed above implies that the main part of the 
indigenous fauna and flora of the North Atlantic islands survived at least 
one glaciation in situ. In this respect, conditions in Iceland are more easily 
understood than in Greenland and have been treated by many biogeographers. 

As usually, only the effect of the Last Glaciation (WUrm) can be recon- 
structed with some claim of reliability. 1 would like to add one more reason to 
those proposed by several specialists in favor of a "WLirm-hibernation" in 
Iceland: A deposit is known at Elhdavogur (near Reykjavik) from the 
"Riss-Wiirm" Interglacial (Thorkelsson, 1935). It contains fragments of 
seven beetle species, six of which are members of the present indigenous fauna 
of the island; the remaining one is a species of flying water-beetles, genus 
Hydroporus, unidentifiable as to species, but apparently not identical with the 
single present representative of this genus in Iceland, H. nigrita F. Of the six 
species in common with the present fauna, three are flightless, i.e. Nehria 
gyllenhali Schnh., Pterostichus diligens Sturm, and Tachimts corticinus Gr. Is it 
really within the limits of probability to assume that the Riss-Wiirm Inter- 
glacial fauna of Iceland was exterminated to be substituted later, in Post- 
glacial time, by oversea immigration by the very same, in part flightless, 
species? To accept this would, for me, be a clear underestimation of the 
random character of long-distance dispersal. 



The locating of the Icelandic Wiirm-refugia has been a matter of some con- 
troversy. Botanists (e.g. Steindorsson, 1954) have paid most attention to the 
isolated occurrences of certain Arctic-Subarctic plants in the northern 
parts of the island and, consequently, placed most of the refugia there. This 
seems to be supported also by geological data (cf. map by Einarsson, 1961, 
p. 44). The terrestrial fauna, however, shows a clear concentration of flightless 

Fig. 6. Percentage of indigenous flightless Coleoptera in different parts of Iceland. 

Dimorphic species counted, if occurring in the short-winged form. 
Three species, found isolated in the north, are omitted since regarded as transported 
by man: Trechus obtiisus Er., Othius melanocephalus Gr., Apion cruentatuin Walt. 

or otherwise not easily dispersed forms in the south and the southeast. For 
that reason, I have earlier (Lindroth, 1931, p. 481, etc.) tried to locate the 
two main faunal Wiirm-refugia to the surroundings of Eyjafjallajokull and 
Hornafjordur, respectively. 

The distribution of flightless Coleoptera in dilferent districts of Iceland is 
illustrated here (Fig. 6). The two main features are: (1) the high proportion of 
the said element in the south and, as could be expected, notably in the 
southeast; and (2) the very low figure for the northeast. This is remarkable, 
since botanists have claimed several refugia in this area (Einarsson, 1961). It 
seems to be another proof of the fact that the "botanical refugia" in the north 


were of quite subordinate importance as far as the land fauna is concerned. 
Judging from its present distribution on the island, a single beetle species 
(Notiop/uliis aquaticus L. ; a wing-dimorphic Carabid and the only, entirely or 
partly, short-winged beetle of the Icelandic fauna that in Scandinavia reaches 
the regio alpina superior) may be suggested as a hibernator in the north. 

At least during the Last Glaciation, there were apparently two different 
kinds of refugia in Iceland: (1) northern refugia, both on the coast and in 
isolated nunatak areas with Arctic and Subarctic conditions; (2) coastal 
refugia in the south and southeast inhabited mainly by Boreo-Temperate 

It is quite possible that conditions have been similar during the Riss 
Glaciation, but this cannot be inferred from the present species distribution. 
Thus, from a purely biogeographical standpoint, the most satisfactory 
explanation of the character of the biota on Iceland as well as on the other 
North Atlantic islands would be: (1) a land connection that persisted into 
Early Pleistocene; (2) survival of the main part of the now existing floras and 
faunas on the isolated islands through the Riss and Wiirm Glacials. 

Finally, I would like to summarize what has been said above: 

1. Conditions of faunal and floral distribution in the North Atlantic area 
are extraordinary, with European biota predominating in Iceland and reaching 
Greenland, but completely blocked by Davis Strait. 

2. An extraordinary explanation is therefore called for: a former land 
connection from the European mainland to Greenland. 

3. The fauna of the Atlantic islands gives the impression of youth since it 
does not include any endemic species. It can hardly be regarded as directly 
descended from Tertiary time, but it is thought to have immigrated in one of the 
Interglacials and thereafter to have survived at least one Glaciation, probably 
more, in situ on coastal refugia in Iceland, Greenland, and probably also in 
the Faeroe Islands. 


Arldt, T. (1917). Handbuch der Palaeographie. I, 1-3. Leipzig. 

Arndt, W. (1928). Der Siisswasserschwamm Heteroineyenia ryderi Potts. &c. Zool. 

Anzeiger 77, \56-l66. 
AsKELSSON, J. (1960). Pliocene and Pleistocene fossiliferous deposits. Int. Geol. Congr. 

Exc. Guide A2 (Iceland), 28-32. 
BocHER, T. W.. HoLMEN, K. and Jakobsen, K. (1957). Gronlands flora. Copenhagen. 
Dahl, E. (1958). Amfiatlantiske planter. Problems of amphiatlantic plant distribution. 

Blyttia 16, 93-121. 
EiNARSSON, Th. (1961). Pollenanalytische Untersuchungen zur spat- und postglazialen 

Klimageschichte Islands. Sonderveroff. Geol. Inst. Univ. Koln 6, 1-52. 
HuLTEN, E. (1958). The amphiatlantic plants and their phytogeographical connections. 

Kgi Svemka Vetensk. Akad. Handl. IV, 7(1), 1-340. 
JoNSSON. J. (1954). Outlines of the geology of the Hornafjordur region. Geogr. Ann. 36, 

Larsson, S. G. (1959). Coleoptera 2. General remarks. Zool. of feel. HI. 46b. 1-85. 


LiNDROTH, C. H. (1931). Die Insektenfauna Islands und ihre Probleme. Zool. Bidr. jr. 

Uppsala 13. 105-589. 
LiNDROTH, C. H. (1957). The Faiiual Connections between Europe and North America. 

Stockholm and New York. 
LiNDROTH. C. H. (1960). Is Davis Strait — between Greenland and Baffin Island — a floristic 

barrier? Bot. Notiser 113. 129-140. 
Love. A. and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic flora. Acta 

Norfi Gotob. 20. 65-291. 
PoRSiLD. M. p. (1932). Alien plants and apophytes of Greenland. Medd. om Gronl. 92 

(1), 1-85. 
SCHARFF, R. F. (1907). European Animals: Their Geological History and Geographical 

Distribution. London. 
ScHARFF, R. F. (1909). On the evidence of a former land-bridge between northern Europe 

and North America. Proc. Irish Acad. 28B, 1-28. 
ScHARFF, R. F. (191 1). Distribution and Origin of Life in America. London. 
ScHWARZBACH, M. (1955). AUgemciner Uberblick der Klimageschichte Islands. Neues 

Jahrb. Geol. Paldontol. Mh. 97-130. 
ScHWARZBACH. M. (1956). Das Vulkangebiet von Hredavatn (West-Island). Neues Jahrb. 

Geol. Paldontol. 104, 1-29. 
SCHWARZBACH, M. (1959). Die Beziehungen zwischen Europa und Amerika als geologisches 

Problem. Kolner Univ.-Reden 23, 1-39. 
SCHWARZBACH, M. and Pflug, H. D. (1957). Das Klima des jiingeren Tertiars in Island. 

Neues Jahrb. Geol. Paldontol. 104, 279-298. 
Simmons, H. G. (1905). Har en landbrygga ofver Nordtatleanten funnits i postglacial tid? 

Ymer 25, 150-155. 
Simpson, G. G. (1940). Mammals and land bridges. /. Wash. Acad. Sci. 30, 137-163. 
Simpson, G. G. (1947). Holarctic mammalian faunas and continental relationships during 

the Cenozoic. Bull. Geol. Soc. Amer. 58, 613-687. 
Steindorsson, S. (1954). Um aldur og innflutning islenzku florunnar. Arsrit Riektunarfel. 

Nordurl. 51. 3-23. 53-72, 101-115. 
Thorarinsson, S. (1958). The Oroefajokull eruption of 1362. Acta Nat. Isl. 2 (2), 1-99. 
Thorkelsson, Th. (1935). A fossiliferous interglacial layer at Ellidaarvogur, Reykjavik. 

Soc. Scient. Isl., Greinar 1 (1), 1-14. 
Wesenberg-Lund, C. ( 1937). Ferskvandsfaimaen, biologisk belyst. 2. Udgave. Copenhagen. 





J. A. Nannfeldt 
Institute of Systematic Botany, University of Uppsala, Uppsala, Sweden 

In this everchanging world the distribution of a taxon is determined not only 
by the sum of all its ecological demands and reproductive properties but also 
by historical factors. The environment is changing and so is also the genetical 
constitution of every taxon, but it is only rarely that the changes in distribu- 
tion can be proved by direct observations. 

To keep to the area and era closest to us, the North Atlantic area and the 
Quaternary era, we know a good deal about, for example, our trees and their 
immigration after the last glaciation from pollen-analysis and other fossil 
remains. We know also a good deal about the first plants to invade south 
Scandinavia after the retreat of the last ice-sheet. Already Nathorst and his 
contemporaries found Dryas octopetala, Salix herbacea and other high- 
mountain species, now growing in the Scandes. Iversen and Erdtman have 
recently, by pollen-analytical methods, found a rich, very early steppe flora. 
At least one of its species. Ephedra distachya, has long ago disappeared again 
from Scandinavia. For at least one of the other steppe species, Centauiea 
Cyaiius, it seems very improbable, to say the least, that the present-day 
population of Scandinavia has any genealogical connection with that early 
Post-glacial population. Also, as for Dryas and the other mountain plants now 
growing in the Scandes, there are no proofs or even indications that there are 
direct genealogical connections between those early populations of south 
Scandinavia and the present-day populations of the Scandes. In the opinion of 
numerous students, amongst them myself, there are none. 

About most of our species we know, in fact, nothing of that kind. 

One of the first distributional problems in the Scandes to attract the 
interest of scientists was the presence of what has become known as a "West 
Arctic Element" in the flora. These plants cannot reasonably have reached 
their present areas from the south or the east. Neither can their occurrence in 
the Scandes be explained by chance dispersal from a remote west. The 
possibility and importance of such long-distance dispersal should of course 



not to be denied. Is there any better explanation for Dryopteris fragrans in 
Utsjoki (northernmost Finland) or for Oxytropis deflexa in Kautokeino 
(northernmost Norway) or for the short visit of Arctostaphylos alpina to a 
spot in west Jutland? 

The "West Arctic Element", however, cannot be explained in that way. 
The number of species is too large, and they do not occur haphazardly but 
are restricted to special areas which are inhabited also by other species 
possessing restricted and disjunct part-areas. A theory explaining the occur- 
rence of the "West Arctic" species is not acceptable, if it does not at the same 
time explain also the occurrence of these other species. In the mind of 
numerous phytogeographers, including myself, the only possible explanation 
is the existence during the Last Glaciation of ice-free refugia west, and perhaps 
even north, of the ice-sheet, not too far from the areas where these plants 
grow nowadays. How they survived the previous glaciations is impossible to 
know, but I feel sure that they, or at least most of them, did survive in north- 
west Europe. I feel sure that the Scandinavian mountain plants have a long 
history with us and that few, if any, species have reached us from the south as 
late as during or after the Wurm Glaciation. Possible exceptions are such 
species as Campanula harhafa, Gentiana purpurea and Ranunculus platani- 
folius. These are far from High-Alpine and are restricted to the south part of 
the Scandes, except for a most isolated locality in Finnmark for the Ranunculus 
(chance dispersal ? ?). Several species have certainly perished from the 
severe vicissitudes during the glaciations, and a new glaciation would certainly 
impoverish our flora still more. This is the only explanation of the absence 
from the Scandes of such species as Alopecurus alpinus, Cluunaenerium 
latifolium and Lomatogonium rotatum. 

Taxonomists and phytogeographers often plead for a broad species 
concept, lest the general survey and the understanding of the natural con- 
nections should be lost. It is true that a species concept such as used by 
numerous Russian authors who give specific rank to almost every geographic- 
ally isolated population, leads to a loss of the understanding of the phylo- 
genetical connections. The treatment of all agamospecies as normal "full" 
species may have the same effect, especially if the agamospecies of, for 
example, the Ranunculus auricomus-complex were treated on the same level 
as the species within the larger, sexual part of the genus. But the scope of the 
species is in itself of little importance, if only the lower recognizable units are 
not neglected, for these smaller units give often important, in many cases 
perhaps the most important, clues to the migratory history of a species. I shall 
here give some few examples from the North Atlantic area, especially relating 
to the flora of the Scandes. My first examples aim at showing the closer 
floristic connections of the Scandes with Iceland and Scotland than with 
the Alps and other central and south European mountains. 


The Poa laxa group or Poo sect. Oreinos forms an excellent example. This 
group has its centre in central Europe and has certainly originated there. 
It is totally absent from the Arctic. To be sure, Poa laxa has repeatedly been 
reported from Greenland and Arctic America, but in so far as I have seen the 
voucher specimens they have been misnamed. Hulten (1942) claims that closely 
related plants occur in Alaska and Kamtchatka but as far as I understand 
they have nothing to do with sect. Oreinos. Now, some 25 years ago I (Nann- 
feldt, 1935) was able to show that the populations of Scandinavia. Iceland and 
Scotland are very homogeneous both intra and inter se and that they show 
sharp though small differences from the populations of the Alps and the other 
southern mountains. Some years later, Nygren (1950, 1955) showed that the 
northern taxon {Poaflexiiosa or P. laxa subsp. Jiexuosa) is hexaploid (2/z = 42) 
in contrast to the true southern P. laxa, which is tetraploid (2n = 28) and 
perhaps also diploid. A very isolated population occurs in eastern North 
America. I found it to represent a distinct taxon {Poa fernaldiana or P. laxa 
subsp. fernaldiana) and Nygren (1955) to be hexaploid. All these facts seem to 
indicate that the northern populations have been isolated from the southern 
for a very long time, and this conclusion is strengthened by Poa jemtlanclica 
(Nannfeldt, 1937; Nygren, 1950). This viviparous taxon has certainly arisen 
as a hybrid between P. flexuosa and P. alpina and is so uniform in all mor- 
phological characteristics that it must be of monophyletic origin. It propagates 
exclusively by vivipary. It occurs in Scandinavia and Scotland but is unknown 
from Iceland. Its Scandinavian (Figs. 1 and 2) and Scottish part-areas are 
smaller than those of P. flexuosa but fall completely within them. Its present 
distribution affords thus an additional proof of a connection between 
Scotland and the Scandes and is strongly indicative both of its own high 
age and of the high age of P. flexuosa in the North Atlantic area. 

Trisetwn spicatuni is a grass with an unusually large, almost world-wide 
Arctic-montane distribution. This species is certainly very old. Hulten 
(1959a) has recently treated its racial differentiation, distinguishing no less 
than 14 subspecies. For my purpose it is sufficient to mention that the type 
subspecies has an Arctic -Subarctic distribution reaching farther south in 
the central Asiatic mountains and in the western American mountains, whereas 
the Pyrenees, the Alps and the Caucasus are inhabited by a separate taxon, 
subsp. ovatipaniculatum. It may further be mentioned that Iceland and south 
Greenland are reached by a northeast American taxon, subsp. pilosiglume. 
The cytological features of this complex are still very imperfectly known. 
The subsp. ovatipaniculatum is unknown cytologically. The type subspecies is 
tetraploid, and a hexaploid is known from Greenland. 

Several additional examples could be given even if those just discussed are 
the most instructive. It should not be concealed that there are species, in 
which, at least so far, no significant morphological differences have been 
found between the populations of Scandinavia and the Alps. Such is the case 





The Scandinavian distribution of 
Poa f/exuosa Sm. 

The circles indicate doubtful though not unlikely 
localities outside its certain area. 




•^••*4 iff, 







Fig I. The Scandinavian distribution of Poa f/exuosa. (Mainly after Nannfeldt, 




with Diyas octopetala, also recently studied by Hulten (1959b). But nobody 
can predict the results of detailed and careful cyto-taxonomic studies. 

Fig. 2. The Scandinavian distribution of Poa jemtlandica (Almqu.) Richt. (After 

Nannfeldt, 1937.) 

1 turn now to the problem whether the taxonomic differentiation between 
populations in different parts of the Scandes can give clues to an under- 
standing of their migratory history. The now classical example is the mountain 
poppies, studied by Nordhagen (e.g. 1931) from the late twenties onwards and 
then given a detailed cyto-taxonomical treatment by Gunvor Knaben (1958, 
1959 a and b). Love (1955, 1962) has studied these poppies in other part-areas. 



It must first be pointed out that all North Atlantic mountain poppies are widely 
different from those of the Alps, the latter all being diploids (2/7 = 14), 
whereas those in the north are octoploids, decaploids and dodecaploids. 

Fig. 3. The total distribution of the mountain poppies in Scandinavia (After 

Knaben, 1959a). 

• Fapaver radical iii>i sensu Nordhagen and Knaben. 
■ P. clahtianuiu. o P. lapponiciim ssp. laestac/iaiiiiiii. 
A P. lapponictini ssp. scandinaviciini. 

The populations growing in Scandinavia (Fig. 3) can be grouped into three 
polymorphous species, one is octoploid, viz. P. lapponicum (incl. P. laestadia- 
num). Love (1952) claims that the name P. radicatum should be transferred 
to this species. The two other Scandinavian species are decaploids. One is 



p. dahlianum, and the third is the species that Nordhagen and Knaben 
consider to be the true P. radkatum but for which Love (1955) has coined the 
new name P. nordhagenianum. If Love's views on the typification of P. 
ladicatum are correct and if P. nordhagenianum is taken in the broad sense 
that Knaben and Love now take it, its correct name seems to be P. relictum. 

The last-named species has a total distribution comparable to that of 
Poa flexuosa, except that it does not grow in Scotland, instead the Faeroe 
Islands is an additional area. Its distribution in the Scandes differs from that 
of the Poa, as the poppy is bicentric and the Poa southern. Moreover, the 
poppy is very polymorphous and split up in a number of races with limited 
distribution. With the more detailed and refined studies their number has 
increased to ten, six in the southern part-area (Fig. 4) and four in the northern. 

Fig. 4. Distribution areas of the six south Norwegian subspecies of Papaver 

radkatum (After Knaben 1959). J. relictum. 2. intermedium. 3. ovatilobum. 

4. GJaere colli 5. groevudalense 6. oeksendalense 

But it is not sufficient with this splitting-up. The poppies are very rare plants. 
They inhabit very restricted localities, and the individuals in each locality 
are very few. These local populations are very homogeneous in themselves 
but differ markedly from the populations of even the most adjacent localities, 
although the differences inanifest themselves very little in the physiognomy of 
the plants but almost exclusively in their chromosome structure. The differences 
between the races mentioned above and equipped with taxonomic names 
are of a higher class of magnitude, they manifest themselves clearly in the 
physiognomy and also the differences in chromosome structure are much 


larger. At least five of the six southern races show a distinct affinity to each 
other and contrast markedly to the four northern which show a similar 
affinity inter se. Dr. Knaben's conclusion is that each of the ten taxonomically 
recognized races survived the Last Glaciation in a refuge of its own and that 
the lesser differences between the local populations have arisen on the spots 
where they now grow. This theory is very suggestive but sounds perhaps too 
good to be true in every detail. It is impossible to calculate even roughly how 
long these two stages of differentiation have taken, but it is now known 
that in very small populations differences in chromosome structure may arise 
in a surprisingly short time. Nevertheless, her establishing of the two 
stages of differentiation is a most important discovery. The only possible 
explanation to this remarkable fact is that it is connected with the migratory 
history of the poppies. The additional fact that the northern races group 
themselves into one group and the southern races (or at least five of the six) 
into another, is most tempting to try to explain by historical reasons as well. 
Also the two other species of Scandinavian mountain poppies behave in a 
similar way, although they are not bicentric but northern. Our knowledge of 
the three species outside Scandinavia is not as detailed but their behavior 
there seems to be modified, i.a. by the circumstance that the populations are 
often much larger and not so isolated from each other. 

Another group showing differentiation within the Scandes is the poly- 
morphous Poa arctica-comphx, studied by me some 20 years ago (Nannfeldt, 
1940) from a taxonomic point of view and then submitted to a cytological 
study by Nygren (1950). This complex is totally absent from the Alps and the 
other southern mountains of Europe, and in the Scandes it is typically bicentric 
(Fig. 5). In the southern part there occur three very distinct and uniform 
taxa, viz. subsp. depauperata, subsp. elongata and subsp. stricta, whereas in 
the northern part the polymorphy is more continuous, only a minority of the 
specimens being referable to distinct lower taxa. The explanation of this 
difference is given by Nygren's cytological studies. One of the southern taxa, 
subsp. stricta, is viviparous and has a very small area. It has the lowest 
chromosome number (2/? = 39) known in the whole complex and forms its 
embryo-sacs sexually. Due to its vivipary no seeds are ever formed and it 
propagates exclusively by bulbils. Morphology, chromosome number and 
lack of aposporous embryo-sacs suggest a very isolated and probably very 
old type. The two other taxa are non-viviparous and propagate thus exclusively 
by seeds. Their areas are small but much larger than that of subsp. stricta, and 
one of them, subsp. depauperata, has been found on Iceland by Love (1947), 
which affords still another example of a close connection between the Scandes 
and Iceland. Nygren has found that in both the embryo-sac mother-cells always 
degenerate very early and are substituted by aposporous embryo-sacs in 
which the egg-cell divides so early that chance fertilization becomes impossible. 
All embryos are thus formed asexually. 



Fig. 5. The 
distribution of 
Poa arclica and 
its subordinate 
taxa. (After 
N an n f eld t, 





^^ ' 



The Scandinavian distribution of 
Poa arctica R. Br. 

• locality for ' stricta (Lindeb) Nannf 
V • caespitans (Simm) Nannf 

A __ ^, • tromsensis Nannf 

total area of ' depauperata (Fr) Nannf 

.-- •elongata (Bl) Nannf 

O locality for other races. 
— approximate area of tfie northern uni- 
centric' (and northern part-area of the 
bicentric") mountain species 


In the northern population the reproductive features are different. Besides 
the aposporous embryo-sacs there may now and then — though rarely — be 
formed sexual embryo-sacs, and even the egg-cells of aposporous embryo-sacs 
may occasionally become fertilized. 

One of the recognizable northern taxa, subsp. caespitcms, is most outstanding 
morphologically, i.a. by the always empty anthers. It has very few localities in 
Scandinavia, and grows there together with other very rare plants, but it has a 
wide West Arctic distribution and seems to be common in part of its area. 
Both normal and aposporous embryo-sacs are formed, and the latter are 
able to form both embryo and endosperm without fertilization. Otherwise 
the apomictic types, are as a rule, pseudogamous. The subsp. caespitans is 
thus able to breed true and — in spite of its empty anthers — to propagate 
even in areas where there grows no other type that can supply serviceable 
pollen. This seems to be the situation in, for example, EUesmereland. In, 
for example, Scandinavia the situation is different, for there both normal and 
aposporous embryo-sacs may become fertilized. Nygren has in one of the 
Swedish localities found specimens similar to, but not identical with, caespitans 
and having a somatic chromosome number of 86-88. Such specimens have 
probably arisen from caespitans by fertilization of aposporous embryo-sacs by 
alien pollen. 

It seems clear that the populations of the south and the north of the Scandes 
are both old and have developed independently for a long time, and that the 
wide West Arctic distribution of caespitans proves a high age of that taxon 
just as the occurrence of depauperata in Iceland proves a high age of that 

These examples are selected from plants intensely studied, in the cases of 
Papaver and Poa both taxonomically and cytologically, in the case of Papaver 
also genetically. 

There are a number of species awaiting similar studies, and I am sure that 
most of our mountain plants will repay generously the labour devoted to 
them. It is especially important that the work is not too much concentrated 
upon the rarest species with widely isolated, very small populations, for in 
those we run the risk that the special features of "small populations" may 
overshadow more general trends. Such described taxa as Artemisia norvegica 
var. scotica, SteUaria crassipes var. dovrensis and Oxytropis deflexa subsp. 
norvegica exemplify certainly the evolution within such small populations. 
In several cases, the distinguishing marks of such isolated populations have 
been found to break down, when the variability within the main area has 
been studied more in detail. I shall take an example from another part of 
Scandinavia. When Orchis Spitzelii was found on the Swedish island of 
Gotland this population was described as a var. gotlandica, but later Bengt 
Pettersson (1958, pp. 77-82) could show that closely corresponding individuals 
occur also in the south of Europe. Also in other species such described local 


races may meet the same fate, but their describing has not been useless, as it 
has stimulated further research. 

When all or at least most of our mountain species have been studied 
carefully and in detail we shall certainly be able to draw more certain con- 
clusions about the history of our flora. 


(Only a few recent papers and papers specially referred to are cited here. The problems 
relating to the mountain flora of Scandinavia have been discussed so often that the pertinent 
literature is easily found.) 
HuLTEN, E. (1942). Flora of Alaska and Yukon. II. Acta Univ. Luiul. N.F. II, 37, 1, 

HuLTEN, E. (1958). The Amphi-Atlantic plants and their phytogeographical connections. 

Kimgl. Sv. Vet.-Akacl. Hand!. 4, 7, 1. 
HuLTEN, E. (1959a). The Tiisetum spicatiim complex. Trisetum spicatiim (L.) Richt., an 

Arctic-montane species with world-wide range. Sv. Bot. Tidskr. 53, 203-228. 
HuLTEN, E. (1959b). Studies in the genus Divas. Sv. Bot. Tidskr. 53, 507-542. 
Knaben, G. (1958). Papaver-studicr, med et forsvar for P. radicatiim Rottb. som en 

islandsk-skandinavisk art. Blyttia 16, 61-80. 
Knaben, G. (1959a) On the evolution of the radicattmi-group of the Scapiflora Papavers as 

studied in 70 and 56 chromosome species. Part A. Cytotaxonomical aspects. Op. Bot. 

{Lund) 2, 3. 
Knaben, G. (1959b) On the evolution of the mdicat um-group of the Scapiflora Papavers 

as studied in 70 and 56 chromosome species. Part B. Experimental studies. Op. Bot. 

{Lund) 3, 3 
Love, A. (1947). Heimskautasveifgras {Poa arctica) R. Br.) fundid a Hornstrondum. 

Ndtfurufr. 11, 17-21. 
Love, A. (1955). Cytotaxonomical notes on the Icelandic /'«/:'flrc/'. Nytt Magas. f. Bot. 4, 

Love, A. (1962). Typification of Papaver radicatiim — a nomenclatural detective story. Bot. 

Notiser 115, 113-136. 
Love, A. and D. (1961). Chromosome numbers of Central and Northwest European plant 

species. Op. Bot. {Lund) 5. 
Nannfeldt, J. A. (1935). Taxonomical and plant geographical studies in the Poa laxa 

group. Synib. Bot. Ups. [1], 5. 
Nannfeldt, J. A. (1937). On Poa jemtlandica (Almqu.) Richt., its distribution and possible 

origin. A criticism of the theory of hybridization as the cause of vivipary. Bot. Not. 

1937, pp. 1-27. 
Nannfeldt, J. A. (1940). On the polymorphy of Poa arctica R.Br., with special reference 

to its Scandinavian forms. Symb. Bot. Ups. 4, 4. 
NoRDHAGEN, R. (1931). Studien iiber die skandinavischen Rassen der Papaver radicatiim 

Rottb., sowie einige mit derselben verwechselte neue Arten. Vorlaufige Mitteilung. 

Berg. Mas. Art., Natiirv. r. 2. 
Nygren, a. (1950). Cytological and embryological studies in ArcUc Poae. Symb. Bot. Ups. 

10, 4. 
Nygren, A. (1955). Chromosome studies in the Poa laxa group. Ann. Roy. Agric. ColL 

Sweden 22, 359-369. 
Pettersson, Bengt (1958). Dynamik och konstans i Gotiands flora och vegetation. Acta 

Phytogeogr. Siiec. 40. 


The Royal Norwegian Society of Sciences, Botanical Department, Trondheim, Norway 

The Norwegian name Svalbard is a collective appelation given to all the 
islands situated in the Arctic Ocean between 10° and 35° E. Long, and 
between 74° and 81° N. Lat. It includes Bear Island in the south, the Spits- 
bergen Archipelago, and several other smaller groups of islands around 
Spitsbergen, i.e. King Charles Land, Hope Island, Northeast Land, etc., with 
a total area of about 62,000 km'^. 

Svalbard belongs to the High Arctic region, but has a climate influenced 
by the Gulf Stream. One of the branches flows along the western coast of 
Spitsbergen and keeps the water open in summer farther north than anywhere 
else on the Globe. Cold, Arctic streams run especially on the east, south and 
north sides and on the west side between the Gulf Stream and the land. 

The mean temperature from 1912 to 1930 (Gronfjorden, 78° 30' N.) was 

— 7.6°C for the whole year, the average for July being +5.4°C and for March 

— 19.0°C. The precipitation is very low, with an average of 287 mm per 
annum, but both temperature and precipitation vary greatly from one part of 
the area to another. 

Looking at the Svalbard flora as a whole, we find between 155 and 160 
indigenous vascular plants, according to the concept of species now most 
commonly accepted. This means that in spite of the latitude and the severe 
climatic conditions, the flora is rather rich. In the south it is related to that of 
northern Scandinavia or the Scandinavian mountain areas, in the west to the 
flora of east Greenland. To the east we find many plants common to Svalbard 
and the Novaya Zemlya islands, but farther east the relationship is not so 

About 120 of the ca. 160 species reported from Svalbard are found also on 
the neighbouring European mainland, i.e. northern Scandinavia. The rest, 
about 35 species, have a pronounced High Arctic distribution and do not occur 
in northern Scandinavia. Among them are: Alopecurus alpinus, Dupontia 
fisheri, Festuca baffinensis, F. brachyphylla, Pleuropogon sabinei, Poa abbreviata, 
PuccineUia angustata, Carex ursina, Cerastiwn regelii, Mimiartia rossii, 
Draba sitbcapitata, Saxifraga flagellar is. Taraxacum arcticum, and others. 

A considerable number of the remaining species belong to the group of 
plants with a centric distribution in Scandinavia, which means that they are 
mainly within one limited area in southern or northern Scandinavia. In the 



Svalbard flora about 12 species belong to the group called bicentric in the 
sense of Scandinavian botanists, i.e. they occur in both of the above-mentioned 
areas. Here we mention only: Luziila arctica, Carex parallela, Sagina caes- 
pitosa, Draba nivalis, Saxifraga hieraciifolia. Campanula uniflora, and others. 
Plants distributed only within the northern one of the two Scandinavian areas 
are represented in Svalbard by a larger number, i.e. about 20 species; Cassiope 
tetragona, Hierochloe alpina, Papaver dahlianwn, Erigeron imalaschkense and 
others are examples belonging to this group. 

The plants mentioned above have gaps in their distribution areas either 
from southern to northern Scandinavia and then to Svalbard or from 
northern Scandinavia to Svalbard. We have, however, among the Arctic- 
Alpine plants one small, but distinct, group comprising only three species, 
Draba gredinii, Kobresia simpliciuscula, and Phippsia concinna, which have an 
even larger gap in their distribution, stretching from southern Scandinavia to 
Svalbard. In a way, these latter might also be called bicentric, but with a gap 
between their areas 600-800 km larger than that of the other group of 
bicentric plants. The distance between their centers is about 1600-1800 km. 
One of the three species, viz. Phippsia concinna, is also known from Bear 
Island. We may therefore say that we know three groups of Arctic-Alpine 
plants which are (a) common to southern Scandinavia and Svalbard, (b) 
common to northern Scandinavia and Svalbard, and (c) common to both 
southern and northern Scandinavia and Svalbard, but with a gap between 
each center. For a fourth group we may place the ubiquitous Scandinavian 
species, common also in Svalbard. 

On the whole, we may say that the Scandinavian centric species constitute 
the phytogeographically most important group in Svalbard as well, and it 
should be noted that it comprises a great part of the flora there. If these 
plants survived the last glaciations in Scandinavia, we are led to the assump- 
tion that there must have been chances for a third center farther north in ihe 
Arctic, with conditions suitable for a possible survival of plants. This leads us 
also to conclude that a possible connection between the flora of Svalbard and 
Scandinavia must have existed before the Last Glaciation or even in Early 

Within the flora of Svalbard proper, plants can be grouped according to 
their distribution. The southernmost part of Svalbard, Bear Island, has a few 
plants known from northern Scandinavia, but not occurring farther north in 
Svalbard. They are: Cerastium cerastoides, Hippuris vulgaris, Luzula arcuata, 
and Stellaria calycantha. 

In the southernmost parts of the island of West Spitsbergen in the Spits- 
bergen Archipelago are found two species which are isolated in this area. 
They are : Salix herbacea and Ranunculus glacialis, and are known only from a 
few isolated localities. One of them, viz. Salix herbacea, occurs also on Bear 
Island. According to my opinion, this southernmost area of the Spitsbergen 



Archipelago is concomitant with the occurrence of the plants mentioned 
from Bear Island. Southwest Spitsbergen and Bear Island are now separated 
only be a rather shallow sea with depths between 10 and 100 m. The plants 
may be regarded either as southern remnants of a previously continuous 
Arctic vegetation, or as northern outposts of a flora connected with the 
Scandinavian mainland. 

About half of the Svalbard plants are more or less ubiquitous species and 
are distributed over the greater part of the Archipelago. But there are about 
35 species with a more limited distribution, confined mostly to the inner 
fjord districts, especially between Van Mijenfjord and Isfjord, including the 
innermost branches of Wijdefjord on the island of West Spitsbergen (Fig. 1). 
In this area are distributed the greater part of the most interesting plants 












Fig. 1. Areas in Spitsbergen with an especially rich occurrence of plants. Dot 
indicates locality with hot springs. Bear Island is not indicated on the map. 

within the flora of Svalbard, e.g. Arctagrostis latifolia, Hierochloe alpina, 
Eriophorum triste, Kobresia simpliciuscula, Juncus arcticus, J. castaneus, 
Luzula wahlenbergii, Potentilla crantzii, Empetrum hermaphroditum, Cassiope 
hypnoides, Polemoniiim boreale. Campanula imiflora, etc. 

The concentration of plants in these areas is due not only to better 
ecological conditions. Suitable substrata and climate no doubt exist also in 



Other regions of Svalbard, but there is on the whole a striking difference 
between the floras of these fjord districts and other parts of Svalbard, e.g. 
the fjords farther south, Hornsound, etc., and the north and east coasts where 
the floras are much poorer in species. Hadac (1944) found a total of 114 
species in the Sassen area of the inner fjord districts. In my opinion the 
reason for this peculiar distribution must be sought within the historical 
factors, as these areas must have provided possibilities for a survival of plants 
during the glaciation. If a migration from the south had taken place recently, 
we should expect the plants instead to be distributed especially in the southern 
parts of the Archipelago or scattered all over the area where suitable sub- 
strata occur. 


A remarkable, isolated occurrence is characteristic of the three species, 
Carex capillan's, Euphrasia arctica and Sibbaldia prociimbens. During an 
expedition in 1960. I visited some small hot springs situated near the head of 
the Bockfjorden on the north coast of Spitsbergen. Here, the only locality for 
these three species was close to the springs and nowhere outside the limited 
area influenced by the heat from the soil. 

The occurrence of these three species, of which Euphrasia arctica is one of 
only two annuals in the flora of Svalbard (the other being Koenigia islandica), 
represents a phytogeographical problem of its own. The nearest locaHties of 
the three species are in northern Norway, and a dispersal to that region by 
long-distance transport of seeds cannot be accepted. 

In my opinion, the only way to explain the occurrence of these three species 
is to regard them as remnants of a previously larger vegetation distributed 
all over Spitsbergen. But as the climate deteriorated, these species survived 
within the area influenced by the heat from the hot springs. This explanation 
is supported by the fact that one of the species, Euphrasia arctica, is an annual, 
and that the hot springs are situated not far from the area previously outlined 
as a center of rare plants in Spitsbergen, i.e. the inner fjord districts north and 
south of the large Isfjord. If we presume that this area in a previous epoch with 
a lesser degree of glaciation had a much richer flora than today, and that the 
three plant species mentioned also occupied that area, it will be easier to 
understand how they still can occur in such an isolated locahty. The deteriora- 
tion of the climatic conditions caused other less hardy plants to die out, but a 
few species succeeded in surviving in this limited area. 

From the same point of view it will be possible to consider the existence on 
Bear Island of the few plants mentioned as occurring there, but nowhere 
else in Svalbard, viz. Cerastiwn cerastoides, Hippuris vulgaris and Stellaria 
calycantha. These plants also must be regarded as remnants of a previous 
flora extending continuously northwards to the Spitsbergen Archipelago. 
At least tv» o of them, viz. Hippuris vulgaris and Stellaria calycantha, are not 


truly Arctic species, and are confined to more temperate or Low Alpine 
regions, e.g. in Scandinavia. This greater demand for a mild climate could be 
the reason why they occur today in Svalbard only on Bear Island. 


It is generally assumed that the frequency of endemic plants within a 
flora is an indicator of its age and of a long-time isolation. If the flora of 
Svalbard has been isolated for a considerable length of time, the occurrence 
of endemic species within the Svalbard area must be expected. During the 
last years, in the course of my investigations concerning the flora of Svalbard, 
I have discovered, as far as I can see, that such an endemic element exists, 
comprising plants ranking both as species and as taxa of lower rank such as 
varieties. It is in this respect first necessary to look upon the Svalbard flora 
separately. Here we have two species endemic to that area, viz. Puccinellia 
sialbardensis Ronning and Ranunculus spitsbergensis Hadac. Besides there are 
two endemic varieties, viz. Puccinellia angustata var. decwnhens and Colpodium 
vahlianuni var. pallida. 

However, the two groups of islands, Svalbard and Novaya Zemlya, are 
closely related both geographically and floristically. Many plants are common 
to the two areas, but Novaya Zemlya, especially in the southern part, has a 
greater number of eastern plants than Svalbard. Their phytogeographical 
connection is further emphasized by the fact that within Novaya Zemlya and 
Svalbard taken as one unit, there are some endemic plants: Pedicularis 
dasyantha, Colpodium vacillans and Puccinellia phryganodes of the Spitsbergen 
type (in the sense of Sorensen. 1953, and Ronning, 1962) are examples of 
such species, endemic to this larger area. 

It is thus possible to separate two areas, the larger comprising both Svalbard 
and Novaya Zemlya, and the smaller one only the Spitsbergen Archipelago. 
It is not possible to say exactly how the isolation has taken place, but the 
endemic plants mentioned give a true indication that these areas once did 
exist. It is also very likely that even more endemic plants, ranking either as 
species, subspecies, or varieties, will appear as the investigation of the 
Svalbard flora continues. This is even more probable since such taxonomically 
difficult genera as Draba, Potent ilia, and Poa have not yet been thoroughly 
worked out, and some of their varieties may turn out to be endemic to the 
area in question. 


The greater part of the plants common to Svalbard and northern Scan- 
dinavia have a wide distribution and must be regarded as ubiquitous in 
Svalbard or at least in the Spitsbergen area. Examples of such plants are 
Equisetum vaviegatwn. Deschampsia alpina. Salix polaris. Oxyria digyna. 


Ranunculus pygmaeus, Saxifmga cernua, S. oppositifolia, S. groenlandica, and 
many others. 

Among the ubiquitous species in Svalbard are also many that in Scandinavia 
belong to the centric species mentioned above. 

Another group of plants of a striking importance in Scandinavia is dis- 
tributed mostly along the coast, or not far from the coast, and is common also 
to the adjacent Arctic islands and northern Scandinavia. The most important 
members are Arctagrostis latifolia, Arenaria pseudofrigida, Chrysospleniwn 
tetrandrum. Braya purpurascens, Papaver dahJiamun, and Stellaria humifusa. 
Though some of them can be found at a distance from the coast or fjord areas 
in northern Scandinavia, they must all be regarded as lowland plants, and as 
such they represent taxa with a strikingly different ecology. They are also 
found farther east in northern Europe reaching as far as the Kola peninsula 
and Vaygach, but not much farther. In my opinion, also these plants point 
sometimes to a close connection between the floras of northern Scandinavia 
and the neighbouring Arctic, and show that this connection includes not only 
Alpine plants, but also lowland or sea shore plants with an ecology different 
from that of the Alpine ones. 

To summarize the phytogeographical features of the Svalbard region we 
may state that: 

1. The number of indigenous species of Svalbard is between 155 and 160, 
according to the species concept commonly accepted. 

2. Among the species, about 35 have a distinct High Arctic distribution and 
do not occur on the European mainland, i.e. northern Scandinavia. 

3. Many of the species common in Svalbard are plants having a centric 
distribution in Scandinavia. They occur either in both southern and 
northern Scandinavia and Svalbard, or only in northern Scandinavia 
and Svalbard. or, only in southern Scandinavia and Svalbard. 

4. Groups of species with isolated occurrences in Svalbard are found: 
(a) in Bear Island, (b) in southernmost Spitsbergen, (c) especially in 
the inner fjord districts (a larger group with a lower degree of isolation); 
(d) around the hot springs at the head of Bockfjorden. The existence 
of this last isolated group is contingent upon the heat of the soil. 

5. Distinct groups of endemic species are found, one endemic to the Spits- 
bergen Archipelago only, and one endemic to both Spitsbergen and 
Novaya Zemlya. 

6. A group of plants, not strictly Alpine in northern Scandinavia, also 
shows a close connection between the floras of the two areas, Svalbard 
and Scandinavia. 

From the facts presented above it is evident that within the flora of Svalbard 
distinct groups of plants exist, and that they show a close connection to the 
flora of the European mainland, i.e. northern Scandinavia. 


The problem of how and when this connection has taken place is still a 
matter of dispute and only a little can be contributed to help solve it. It seems 
to me that time is the only factor that can give a satisfactory explanation of 
these problems. 

As early as 1933, Lynge, on the basis of the distribution of some Arctic 
lichens, concluded that ice-free refugia had existed on the north coast of 
Spitsbergen (Lynge, 1939). He accordingly supposed that the hchens were 
relic plants of a very high age. Another interesting fact concerning these 
north coast lichens is that when occurring in Scandinavian mountains, they 
are not at all High Alpine, but more or less continental Subalpine. Lynge 
{loc. cit.) found no other explanation than "that the area, or a part of it, 
should have been ice-free refugia during the last glaciation, perhaps all through 
the time subsequent to the Tertiary age, and that these lichens should be 
relics which persisted, at least, from the last Interglacial down to the present 

In 1869 Fries had already launched the hypothesis that some of the Svalbard 
plants were of a relic nature, and several later authors, among them Nord- 
hagen (1935), have discussed the possible migration tracks for the Scandinavian 
West Arctic plants. In my opinion, the close relationship shown between the 
floras of the two areas makes it most likely that in an earlier geological epoch 
there existed a connection between northern Scandinavia and the islands of 
the European Arctic. Later geological conditions have split up this continuous 
area into several isolated ones. The question as to when this happened remains 
still unsolved, but is of special importance from a phytogeographical point of 

Nansen (1920) suggested that the Barents Sea area of the Late Tertiary 
period was situated 400-500 meters higher than today. Orvin (1940, p. 54) 
says: "From the presence of large submarine valleys in the Barents Sea we 
may conclude that this area in comparatively recent times has been at a level 
about 500 meters above the present. This happened probably in the latter 
part of the Tertiary." 

Horn and Orvin (1928, p. 44) also agree with Nansen that the elevation 
mentioned could have existed in the Tertiary period. However, one important 
feature concerning the Bear Island needs emphasizing: during the Pleistocene 
glaciations this island was more or less covered with an ice sheet and was 
partly submerged. But although the island was covered with ice during the 
deepest submergence, the mountains in its southern part protruded above the 
glacier (Horn and Orvin, loc. cit., p. 53 and Fig. 44). 

Today the Barents Sea (Fig. 2) is a shallow body of water with large areas 
less than 200 m deep, especially to the east. Perhaps a continuous area of land 
once stretched from northernmost western Europe over the Barents Sea east 
to Novaya Zemlya and west to the other Arctic islands around the Spitsbergen 
Archipelago. This land area must have offered possibilities for plant migration 



from south to north and vice versa. It is also possible that such a land mass 
was invaded by plants from both east and west. Most probably this occurred 
in the Late Tertiary. 

Fig. 2. Barents Sea with depth contour lines. 

Later, geological events such as submergence of the land and glaciation 
during the Pleistocene led to the splitting-up of the flora. Part of the flora from 
then on has survived the glaciations in refugia most probably where, or close 


by where, they occur today. This leads us, for example, to the conclusion that 
there have been conditions for plant life both in refugia along the coast and 
on nunataks in the inner fjord districts. The plants common to the total area 
are probably of a very great age but the endemic species mentioned above 
must be the result of an evolution after the Tertiary. 


Fries, T. M. (1869). Tillagg till Spetsbergens fanerogam-flora. Kgl. Sv. Vetensk. Akaci 

Forhand26, 121-144. 
Hadac, E. (1944). Die Gefasspflanzen des "Sassengebietes", Vest-Spitsbergen. Norges 

Sialbaid og hhavsumlersokelser, Skrifter No. 87, 1-71. 
Hadac, E. (1946). The plant-communities of Sassen Quarter, Vest-Spitsbergen. Stiiclia 

Botanica Chechoslovaka, 7, 127-164. 
Horn, G. and Orvin, A. K. (1928). Geology of Bear Island. Skv. oiu Svalbarcl og Ishavet 

15, 1-152. 
Lynge, B. (1939). On the survival of plants in the Arctic. Norsk Geografisk Tidsskrift 7, 

Nansen, F. (1920). En ferd lil Spitsbergen. Kristiania. 
Nathorst, a. G. (1883). Nya bidrag till kannedomen om Spetsbergens karlvaxter, och 

dessvaxtgeografiskaforhallanden.A"^/. SiTM^A'fl Vetensk. — Akad. Handl. NF 20(6), 1-88. 
Nordhagen, R. (1935). Om Arenaria humifusa Wg. og dens betydning for utforskningen 

av Skandinaviens eldste floraelement. Bergens Museums Arbok J 935. Naturv. Rekke 

No. 1, 1-185. 
Orvin, A. K. (1940). Outline of the geological history of Spitsbergen. Skr. om Svalbard og 

Ishavet 78 1-. 
RONNING, O. I. (1959). The vascular flora of Bear Island. Acta Borealia A. Scientia No 15, 

RoNNiNG, O. I. (1960). The vegetation and flora north of the Arctic Circle. Norway North 

of65\ Oslo, 1960. pp. 50-72. 
RoNNiNG, O. I. (1961). Some new contributions to the flora of Svalbard. Norsk Polarinstifutt 

Skrifter 124, 1-20. 
RoNNiNG. O. I. (1962). The Spitzbergen species of Colpodium Trin., Pleuropogon R. Br. and 

Puccinellia Pari. Det. Kgl. Norske Vidensk. Selsk. Skrifter 1961, No. 4, 1-50. 
Sorensen, Th. (1953). A revision of the Greenland species of Puccinellia Pari. Medd. om 

Groenl. 136 (3), 1-179. 



Hugo Sjors 

Institute of Plant Ecology, University of Uppsala, Uppsala, Sweden 

Whereas in Canada a book called North of 55° (Wilson, 1954) tells the story 
of a still almost uninhabited country, in Europe about 70 million people 
live and work above this parallel. The difference in the northward extension of 
natural ranges and vegetational zones is not quite so evident as the difference 
in distribution of human population. Still, a considerable disparity also exists 
between the two continents with respect to climatic conditions and natural 
resources from a biogeographer's point of view. 

On the other hand, the two continents show very similar situations regard- 
ing regional zonation. The European zonation is a mirror image of the 
American one, much displaced toward the north, but Europe's Oceanic 
West has no counterpart in North America (the "Maritime Provinces" of 
Canada are far less oceanic). 

Broadly speaking, the southern Great Lakes area, extending northward 
into the Niagara Peninsula of Ontario, has its European equivalent in south- 
central Europe, as far north as southern Germany. The Canadian provinces 
of Quebec and Ontario correspond to east-central Europe and the Baltic- 
Scandinavian area as far north as Finland and northern European Russia. 
The "Prairie Provinces" and northwestern Canada are better compared to 
central and eastern European Russia and western Siberia. 

It seems of basic importance for any kind of detailed comparison that those 
points, lines, and zones which correspond most closely to each other bio- 
geographically should be determined on each side of the Atlantic Ocean. If 
this were possible, much more of the experience from one side could be readily 
used on the other, not only in pure biogeographical and ecological science but 
also in practical management of arable land, pasture and forest. 

There is a serious obstacle against the direct use of climatic figures in this 
respect. Northeastern America, except for a narrow zone confined mostly to 
the "Maritime Provinces", is rather continental as to temperature but has 
everywhere an adequate rainfall. In Eurasia we must go far east to obtain the 
same degree of continental temperature, but then summer rainfall becomes 




Individual elements of climate thus fail to give the same pairs of corres- 
ponding points. But vegetation, being an integration of cHmatic as well as 
other factors, may show the total effect of climatic impact on life better than 
single variables do. 

The vegetational unit most suitable for such a comparison is the regional 
(vegetation) complex (term by Du Rietz, in mimeographed lectures) which 
occupies a vegetation region, in the sense used by many phytogeographers, 
notably those of Scandinavia (cf. Du Rietz, 1930, pp. 497-502). If we include 
animal life and environment, and thus refer to the whole ecosystem — a 
heterogeneous ecosystem covering vast expanses of land — we may speak of a 
hiotic region (Sjors, 1955, p. 163). Biotic regions situated in different longitudinal 
sectors of the globe may occupy corresponding positions as to zonation in 
latitudinal direction. They are regarded as parts of the same biotic zone 
(cf. Rousseau, 1952, p. 437). In mountainous countries, biotic regions homolo- 
gous in their altitudinal position constitute a biotic belt (cf. Du Rietz, 1930, 
p. 499). There are cases when mountain biotic belts correspond to, or even 
gradually merge into, biotic zones of the lowland farther north (or south, in 
the Southern Hemisphere). 

In forested biotic regions, forests are usually more important than other 
kinds of vegetation. Leading tree species become significant for the delimita- 
tion of these regions. Usually the dominance, although only partial, of a tree 
species is more important for this purpose than a scattered occurrence. 
However, sub-regions within the Boreal coniferous zone (see below) must 
often be determined from the areas of subordinate species, because the 
dominant conifers are the same. Also, criteria taken from the physiognomy 
and total composition of vegetation may be adduced for similar purposes 
(Kujala, 1936: Hare, 1950, 1954, 1959; Kalela, 1958, 1961 : Ahti, 1961). 


Following Regel (1950, 1952) the author prefers to designate as Nemoral 
the three great parts of the Northern Hemisphere where temperate deciduous 
trees are prevalent. According to Regel (1952, p. 38) the term "Nemoral" was 
introduced by Russian authors. Regel includes the Boreo-nemoral zone in the 
Nemoral zone as a special sub-zone. The frequently used term "deciduous 
forest zone" is open to some criticism. Thus, considerable coniferous forests 
(pines, etc.) occur within this zone, particularly in North America and Japan. 
There exist also deciduous forests of quite different types, e.g. those in the 
Sub-tropics and the seasonally moist Tropics. Nor should the deciduous 
birchwoods of Sub-Alpine Fennoscandia and Kamchatka be included. 
Strictly speaking, also Larix is deciduous. The word Nemoral is free from such 
objection, and emphasizes the adaptation of the vegetation and flora to mild- 


temperate, fairly moist conditions, and the formation of a humus layer of the 
mull type and a brown forest soil profile, as is typical of most stands of 
genera such as Fraxiims, Acer, Ulmus, Tilia, Fagiis, and certain sections of 
Que re us. 

The Nemoral zone is discontinuous (since Quaternary time) but its three 
parts, in eastern North America, Europe and the Far East, are remarkably 
similar in spite of considerable climatic difference. Each of the three parts 
shows a south-north subdivision into two or more biotic regions, but these 
conditions, although indicated on the maps. Figs. 2 and 3. are not dealt with 
further in this paper. 

With reduced rainfall and increased continentality, the Nemoral regions 
change into Steppe or Prairie regions. The usually quite extensive ecotones 
include the warmer parts of the Woodland Steppe regions (European Russia, 
and the Middle West of the U.S.A.). 


Between the three parts of the Nemoral zone and the true Boreal zone, 
transitional biotic regions occur, here called Boreo-nemoral. They occupy a 
considerable width in Europe and in the Far East (Manchuria and Amur 
districts) but the American Boreo-nemoral region is quite narrow. As this 
part of the zonation is frequently misunderstood, it is desirable to state more 
precisely what areas belong here and thus correspond on both sides of the 

Because ofearlier abundance ofwhite pine (PrnM^^/ro^M^) and the occurrence 
of other conifers (P. resinosa, Tsuga eanadensis. Thuja occidentalis), the 
Great Lakes-St. Lawrence area has sometimes been regarded as representing 
this zone in America, but this is only partly correct. The above-mentioned 
conifers are essentially non-Boreal, and (except Thuja) do not reach much 
farther into the Boreal zone than do several of the Nemoral hardwoods 
(e.g. Fraxinus nigra, Ulmus americana, Populus grandidentata, Betula lutea). 

The Boreo-nemoral zone is met with a short distance north of the St. 
Lawrence and Ottawa Rivers, e.g. at St. Donat and Maniwake. In Quebec 
the change is abrupt due to the rise in level from the Paleozoic Ottawa-St. 
Lawrence valley to the uplands of the Pre-Cambrian Shield. The difference in 
bedrock and Quaternary deposits greatly amplifies the combined effects of 
latitude and altitude on vegetation. The shift of dominance from exacting, 
mainly deciduous trees to hardy, less demanding Boreal conifers is thus 
comparatively abrupt and was evidently still more impressive before so much 
of the latter were cut. 

Farther west the Boreo-nemoral zone is broader, extending also south of 
the Ottawa River (Petawawa, Algonquin Park), and crossing Lake Superior 
to reappear in southeastern Manitoba and northeastern Minnesota. 


The Boreo-nemoral region in America is a tract of land where Boreal 
conifers {Picea glauca, P. mariana, Abies halsamea) are well distributed and 
frequently tend to be predominant, even if now largely replaced by secondary 
hardwood forests. The latter are extensively formed by trembling aspen 
{Popuhis tremuloides) which, like the white birch {Betula papyrifera), is a 
"neutral" tree of little zonal importance. Also some southern hardwoods have 
expanded considerably into the widespread secondary forests, notably sugar 
maple {Acer saccharum). Other southern tree species common in most of the 
Boreo-nemoral region are Quercus borealis, Tilia americana, Populus grandi- 
dentata, Betula lutea, Fraxinus nigra, Ulmus americana, and the conifers 
Pinus strobus, P. resinosa. and Tsuga canadensis. Some of these even reach 
quite a distance into the Boreal region farther north (see below). 

Most of the "Acadian Forest Region" (Rowe, 1959) in the Maritime 
Provinces of Canada can be classed as part of the Boreo-nemoral zone. It is 
rich in hardwoods in the lower hills, whereas conifer forests are often pre- 
dominant on the coast, in the valleys, and in the interior upland. Red spruce 
{Picea rubens), a non-Boreal species, is often dominant. 

The spruce-fir forests of the northeastern U.S.A. are found at increasingly 
higher elevation towards the south (Shantz and Zon, 1924) and are here 
mapped as Montane (Fig. 2). They merge into the Boreo-nemoral zone 
towards the north and are not truly Boreal. Picea rubens is one of the chief 

In Europe, conditions are simpler. The European vicariants of Pinus 
strobus (i.e. P. peuce), and P. resinosa (i.e. P. nigra) are confined to the south 
for historical (not climatic) reasons and do not interfere. P. silvestris occurs, 
chiefly on sandy soils, even in some parts in the Nemoral zone, but in the rest 
of the Nemoral area, the forests were deciduous almost everywhere before 
conifers were planted. Passing into the Boreo-nemoral zone, there is a very 
sudden shift to predominant Boreal conifers: Norway spruce {Picea abies) 
and Scots pine {Pinus silvestris). Hornbeam {Carpinus befulus), beech {Fagus 
silvatica) and durmast oak {Quercus petraea) occur only in small western 
parts of the Boreo-nemoral. Most of the other southern hardwoods continue 
their distribution farther north (or rather northeast). The Boreo-nemoral 
region in Europe, contains much Quercus robur, Fraxinus excelsior, Ulmus 
glabra, Tilia cor data, Acer platanoides (not A. pseudoplatanus), Corylus 
avellana, etc. Although generally subordinate, these southern hardwoods may 
be of great local importance, particularly on calcareous soils (e.g. on the 
island of Oland, Sweden, which is crossed by the borderline between the 
Nemoral and Boreo-nemoral regions). 

The Boreo-nemoral region in Europe is narrow in the west and the east 
but quite broad in the central part. It reaches from southernmost Norway over 
most of the southern third of Sweden (the southwestern and southern 
coastal parts belong in the Nemoral zone). The southwestern corner of Finland 


belongs in the Boreo-nemoral (Jalas, 1957), which further comprises the former 
Baltic States, northeastern Poland, and large parts of western and central 
European Russia east to the Urals. The northern limit of Quercus robur is 
generally considered as its boundary towards the Boreal zone proper, but in 
Russia the latter boundary is drawn some distance farther south (Lavrenko 
and Sochava, 1954). 

From Manitoba westward, and east of the Urals, the Nemoral zone is 
absent, being replaced by prairies (steppes). The Boreo-nemoral zone in 
turn is replaced by a transition zone between the Boreal coniferous forest 
zones and the steppes, a zone where both most Boreal conifers and Nemoral 
hardwoods are absent or scarce. Betula and Popiilus remain, and aspen 
groves are particularly characteristic {P. tremuloides and P. tremula, respective- 
ly). In the Canadian Prairie provinces this zone is represented by the Aspen- 
Oak Section (with Quercus tnocrocarpa, U/mus americana, Acer negundo, 
etc.) and the Aspen Grove Section of Rowe (1959). The west Siberian counter- 
part is divided into a northern region of "parvifoliate forest" (Lavrenko and 
Sochava, 1954) and a Woodland Steppe region between this and the treeless 


Encircling the globe south of the Arctic there is a Boreal zone (Hare, 1954) 
where forests are usually formed by a very limited number of species belonging 
to a few coniferous and hardwood genera : Picea. Larix, Pinus, Abies, Betula, 
Populus, Ahuis. In addition, species of Salix, Sorbus {aucuparia type), and 
Prunus {padus type) reach tree size. Only very few of the species of these 
large genera are actually Boreal, as emphasized by Hare {loc. cit.). Some of 
these, e.g. the Populus species, extend much farther south; others are pre- 
vailingly Boreal, but all of them extend south at least into the Boreo-nemoral 
zone. This is true also of many Boreal species of smaller size. 

As expressly stated by Hare {loc. cit.) a large part of the conifer forests of 
the world should not be included in the Boreal (or Boreo-nemoral). This is 
true of the Pacific forests, and of the pinewoods occurring in the Nemoral and 
still warmer zones. More related to Boreal forests are some Montane or Sub- 
Alpine forests at the middle latitudes (e.g. in the Rockies, the Alps, Altai, etc.), 
and the oceanic pinewoods of Scotland and westernmost Norway, but they 
are kept separate in this paper. 

The Boreal forests, after having reached maturity, are usually made up of 
conifers, but some areas belonging here are permanently covered by birch- 
woods, alderwoods, or groves of poplar. The Boreal zone is also very rich in 
bogs and fens. It comprises by far the largest part of the peatland areas of the 
world (Fig. 1, from Sjors, 1961b). 

The Boreal zone may be subdivided into three or more east-west sectors 



with different floras. Some species of small size, but not trees, occur in all 
sectors. Alnus rugosa var. americana and Sorhus decora of northern Canada, 
for example, are more similar to the Eurasian A. incana and Sorhus aiicitparla 
than admitted by leading present-day dendrologists, and probably would have 
been regarded as conspecific if they had been small plants. However important 
from the point of view of historical plant geography, this longitudinal sub- 
division is less important than the latitudinal zonation ecologically and also for 
economic forestry. Such a zonal subdivision of the Boreal zone has been 
independently proposed in most Boreal countries. 

Peatland dominant 
■■Peatlands large and abundant 
: The same, but distribution more loca' 

Fig. 1. Extension of the Boreal zone (between the full lines). South of, or below, the 
tree-line, woodland-tundra or Sub-Alpine areas extend to the northern broken 
line. South of the Boreal zone, Boreo-nemoral and Aspen-Birch Woodland 
regions reach to the southern broken line. The abundance of peatland is indicated 
on the figure. Most lowland in Iceland and the Shetlands is south of the tree-line. 

Boreal of North America 

In Canada propositions regarding a subdivision of the Boreal have been 
advanced by Halliday (1937), Hustich (1949, 1951), Rousseau (1952). Hare 
(1950. 1954, 1959). Ritchie (1956. 1959, 1960), Rowe (1959). and others. The 


approach and the terms of these authors differ widely but nevertheless there 
is a fairly close similarity in the resuhs arrived at, as far as strictly zonal 
divisions are considered. Space does not allow a discussion, but references 
may be given to Hare (1959, p. 34). Three sub-zones are more or less generally 
accepted, but the first of these is further sub-divided in the present paper. 

(a) The southern Boreal sub-zone. Here several of the species mentioned for 
the Nemoral and Boreo-nemoral zones occur, usually in low frequency: 
Pinus strobus, P. resinosa, Acer saccharum, Betula lutea, Fraxinus nigra, 
Uhnus americana, etc. The dominant conifers are Picea mariana, P. glauca, 
and Abies balsamea, with Pinus Banksiana on rocks and sandy soils and on 
disturbed sites; birch and particularly aspen are prominent in secondary 
growth. The sub-zone is typically developed east of Lake Superior. North and 
northwest of the latter it is virtually absent, only Pinus strobus, P. resinosa, Pop- 
uhis grandidentata, Uhnus americana, and Fraxinus nigra sparsely representing 
the southern tree species here. In Manitoba the sub-zone reoccurs, although 
the southern species are largely different: Acer negundo, Quercus macrocarpa, 
Uhnus americana, and Fraxinus pennsylvanica var. subintegerrima. This 
sub-zone is again lacking in Alberta and most of Saskatchewan. 

(b) The Main Boreal sub-zone. As the preceding sub-zone, the Main 
Boreal is characterized by closed-canopy coniferous forests except on burned 
sites and bogs and fens. The Boreal conifer species are the same. Abies 
balsamea is an important forest tree mainly in the east. There is often a com- 
plete dominance of the two spruces, with Picea mariana on poor soils and 
P. glauca on the better soils which are more widespread on the sedimentary 
rocks of the western interior. In some areas Pinus banksiana prevails, and in 
the westernmost parts Pinus contorta var. latifolia is abundant. The southern 
species are lacking or represented by very few outposts, mainly of Fraxinus 
nigra and Uhnus americana; only Thuja occidentalis and Acer spicatum are 
still fairly common in some parts. Conditions for forest growth are quite 
satisfactory within this sub-zone which extends across Canada from New- 
foundland to the upper Mackenzie and the Liard valleys, and possibly even to 
some of the deep valleys in the mountainous Yukon Territory. 

(c) The Sub-Arctic sub-zone. This is characterized by open, parklike stands 
usually formed by low black spruce {Picea mariana) with a Cladonia under- 
growth. These stands are regarded as woodland, but not as true (commercially 
valuable) forests, although pulpwood size is usually reached. On more 
favourable sites white spruce {P. glauca) is found, and the undergrowth is 
often richer in species. Boggy sites are extremely common, both wooded 
("black spruce muskegs") and treeless. Larix laricina is frequent on peat, 
especially in wooded fens. Along rivers grows balsam poplar {Populus 
balsamifera). All four tree species occur throughout the Boreal (and Boreo- 
nemoral) zones. The other Boreal trees are confined to certain parts of the 
Sub-Arctic, not reaching its northern outskirts everywhere. The sub-zone 





cr 10' 20* 30- 40" 50" 

Fig. 3. Biotic zonation of north, central and east Europe. 

1. Arctic zone. 

2. Alpine belts. 

3. Sub-Alpine and Montane belts of 
non-Boreal mountains. 

4-8. Boreal zone: 

4. Sub-Alpine Birch Woodland 
belt of Fennoscandia. 

5. Woodland-tundra sub-zone. 

6. Sub-Arctic and Boreo-montane 

7. Main Boreal zone. 

8. Southern Boreal sub-zone. 
9. Boreo-nemoral zone. 

10. Birch-Aspen Woodland region of 

west Siberia. 
11-12. Nemoral zone: 

1 1 . Northern sub-zone. 

12. Central sub-zone. In North 
America: Qiierciis-Caiya and 
Castanea - Qiierciis - Liiiodend- 
ron\ in Europe: Qiiercus 
pubescens, Juglans, cultivation 
of grapes, etc, (sub-Mediter- 

13-14. Woodland Steppe zone: 

13. Northern sub-zone. In North 
America: Pine, aspen, birch 
and a few southern species, 
e.g. Acer negimdo, Qiiercus 
maciocarpa, etc.; in Europe: 
Qiiercus robur; in Europe and 
west Siberia: Tilia cordata. 
The North American region 
equates both regions (10) and 
(13) of Siberia. 

14. Southern sub-zone. In North 
America : Oak-hickory groves ; 
Europe: oak and pine with 
sub-Mediterranean species. 

15. Steppe (Prairie) zone. 

16. Steppe Desert zone. 

17. Desert Scrub zone. 

18. North Atlantic Pine-Birch Wood- 
land and Heath region (with 
Qiiercus, Taxus, Hedera, Ilex in 
Scotland and west Norway). 


extends from the east coastal parts and the uplands of Newfoundland and 
Labrador to Alaska. 

Rowe (1959) and Hare (1959) disagree in the mapping of easternmost 
Labrador. The "coastal tundra" is probably caused here by high winds and 
lack of soil, and has only partly been mapped as Woodland-Tundra sub-zone 
on Fig. 2. The woodland farther inland is largely regarded as Sub-Arctic, 
even if fairly dense, because of low stature. The Hudson Bay Lowland is 
divided between the Main Boreal and Sub- Arctic (Sjors, 1959, 1961a and in 
prep.), and is not as entirely Sub-Arctic as one may believe (Hustich, 1957; 
Rowe, 1959). 

(d) The Hemi-Antic (Rousseau, 1952) or Woodland-tundra suh-zone. Here 
tree growth is reduced to scattered clumps and low, shrubby stands in 
sheltered localities, e.g. along riversides. Nearly or totally treeless barrens 
occur between these outposts. Near the tree-line, the tundra areas become pre- 
dominant and only a small percentage is actually covered by the woodland. 
Any of the four species mentioned for the Sub-Arctic may form the actual 
tree-line (or rather the outposts of potentially tree-forming species), but the 
white spruce (Picea glauca) is the most common, particularly close to the sea 

The typical woodland-tundra is a feature of flat country, either continental 
or adjacent to Arctic coasts. However, Rowe (1959) describes an "Alpine 
Forest-Tundra Section" on the slopes west of the lower Mackenzie Valley, 
and similar sites occur near the tree-line in interior Yukon Territory (Porsild, 
1951), both areas being extremely continental as to climate. In mountainous 
or hilly districts the altitudinal tree-line tends to be sharper than the latitudinal 
tree-line in flat country, but some of the tree species develop a shrubby 
appearance; what actually corresponds to the woodland tundra is often a 
belt of stunted woods and, on exposed localities, even shrubs. Porsild (1951, 
pp. 27-28) describes the dense shrub of alpine fir {Abies lasiocarpa) near the 
tree-line in southeast Yukon. The Picea mariana and Abies balsaniea wood- 
lands of coastal eastern Labrador and some parts of Newfoundland are 
described as often very dense but dwarfed, and interspersed with barren 
heaths, due to a wet. cool and windy climate (Hustich, 1939, 1949, p. 12; 
Rowe, 1959; Ahti, 1959, p. 2). 

Russian Boreal, or Taiga 

As east Siberia is very aberrant, only west Siberia and European Russia are 
considered here. Russian authors (cf. Lavrenko and Sochava, 1954; Sotchava, 
1954; Tikhomirov, 1960) subdivide the Boreal zone, which they call the Taiga, 
into the southern, central, and northern Taiga, in addition to which comes the 
Lyeso-tundra (Sylvo-tundra, forest- or woodland-tundra). This subdivision 
corresponds very well to the Canadian one. but the southern Taiga is a more 
extensive sub-zone than the Canadian counterpart (a) mentioned above, 


whereas the central Taiga, in Siberia, may be less w ide than the Main Boreal 
(b) of the Canadian West. 

The conifer trees of the Taiga (except east Siberia) are only five. Piceo is 
mainly represented by P. ohovata, which is better regarded as a sub-species of 
P. abies. The two "species" of Larix (L. Sukatchewii in Europe. L. sihirica in 
west and central Siberia) are probably also conspecific races. Pinus sihirica is 
very close to the central European P. cembra. The other two trees are Pinus 
silvestris and Abies sibirica. Picea and Laiix reach the woodland-tundra. 
In northwest Russia (Kola Peninsula and Karelia) Abies is absent and Larix 
and Pinus sibirica are extremely rare. For geological and phytogeographical 
reasons this area is better included in Fennoscandia (as is always done by 
western botanists). Pinus silvestris is prevalent in most of the Russian Fenno- 
scandia, except on eastern Kola where Picea reaches farther coastward. but 
both are ultimately superseded on Kola by the mountain birch [Betula 
pubescens ssp. tortuosa). Elsewhere in Russia, birch only exceptionally forms 
the northern tree-line. 

The northern limit of the woodland-tundra in Fig. 3 was taken from a 
sketch-map by Andreyev (1956; reproduced in Tikhomirov. 1961). It is 
somewhat more northern than on the vegetation map by Lavrenko and 
Sochava (1954). 

Boreal Fennoscandia 

Zonal problems in northern Fennoscandia are simplified by the low number 
of coniferous trees, only Pinus silvestris and Picea abies occurring there. 
Otherwise, the situation is quite complicated. 

Two vegetation regions are generally accepted for the Fennoscandian part 
of the Boreal zone, viz. the "Northern coniferous forest region (without oak)" 
and the "Sub-alpine birch woodland region". Only Hustich (1960) gives a 
slightly different treatment. The former region has been subdivided both in 
Finland and Sweden (Kalela. 1958, 1961: Du Rietz, 1950, 1952; Sjors, 1950. 
1956). but unfortunately the proposed sub-regions agree neither in extent nor 
in number; probably none of them corresponds exactly to any of the above- 
mentioned Canadian and Russian sub-zones. 

The Boreal sub-zonation in Fennoscandia is largely due to other factors 
than latitude. In Norway and Sweden altitude is apparently more effective 
than latitude, and the influence of the altitude is also evident on recent Finnish 
maps (Kalela. 1958, 1961). Moreover, the absence of spruce in several 
western and northern parts of the Scandinavian Peninsula has often been 
considered of regional iniportance. 

Both in Norway and in Sweden as well as in Finland, a transitional 
Southern Boreal sub-zone has some scattered occurrences of Nemoral species, 
particularly Tilia cordata, Acer platanoides, and the scrub Corylus avellana. 
The upper and northern limit of this sub-zone is very ill-defined in Norway 


and Sweden, due to the broken topography. The wide occurrence of Tilia in 
Finland makes it broader there. 

The Main Boreal sub-zone corresponds, in Sweden, both to the upper part 
of the area where Myrica gale still occurs, and to the central Norrland or 
middle sub-region which lacks Myrica and a number of other southern 
species. The nearest Finnish equivalent is the sub-region of Ostrobottnia 
(Pohjanmaa), according to Kalela (1958, 1961). It borders the central Taiga 
towards the east, but seems to be somewhat more northern in latitudinal 

In northern Sweden, open pine-lichen woodlands of Sub-Arctic type are 
closely confined to coarse, sandy and gravelly soils, and thus do not form a 
regionally characteristic feature. According to Ahti (1961), they occur also on 
other soils in northernmost Finland, but lichen woodlands are not so wide- 
spread as in more continental parts of the Sub-Arctic areas. However, even 
forests of a more mesic appearance tend to be widely spaced in northern and 
high upland parts of Boreal Fennoscandia. The spacing of the forest ought to 
be used as a phytogeographical regional criterion. Its importance in Swedish 
forestry has only recently been duly emphasized (Ebeling, 1962). Stature and 
rate of growth have long been considered important for site classification 
(edaphical as well as climatic). 

Thus the uppermost and northernmost coniferous woodlands in Fenno- 
scandia are of Sub-Arctic type with regard to the wide spacing, slow growth 
and low stature of the conifers. In Sweden, the "Pre-Alpine conifer forest 
sub-region" (sensu Du Rietz, 1950, p. 7; Swedish = fjdllbarrskogsregionen, 
Du Rietz, 1942b, 1952 or "upper sub-region of the northern coniferous 
forest region", Sjors, 1950, p. 177; 1956, p. 202) is largely identical with the 
area above the economic limit for artificial regeneration (Swedish = skogs- 
odlingsgrdnsen) in the State forests of Sweden, as drawn by Hojer (1954). 
It seems likely that some more areas in the high uplands (outside the 
mountainous district proper) should be included in the area equivalent to the 
Sub-Arctic, with respect to the poor growth of the forest. Du Rietz (1952, p. 7) 
gives the lower limit at about 350 m in the north and 600 m in Harjedalen. 
In Finland, according to Ahti (1961), both the coniferous woodland area of 
Lappland and that of "Perapohjola" (farther south; terms by Kalela, 1958) 
as well as some of the birchwoods (see below) belong in the Sub-Arctic (or 
northern Taiga) sub-zone. 

Practically everywhere in Fennoscandia (also the Russian parts, i.e. iCola 
Peninsula), birchwoods {Betula pubescens ssp. tortuosa) extend beyond the 
limit of coniferous trees. The vertical extension of this zone, which is com- 
monly regarded as Sub-Alpine, is very variable, from almost none to several 
hundred meters, a fact showing that it is hardly a zone of equal significance in 
all parts. If a Woodland-tundra sub-zone exists in Fennoscandia, it is located 
in the upper part of the birch belt (and near the Polar coast) where birches 


are often interspersed by barren areas, and are in any case low and nearly 

This uppermost Sub-Alpine sub-zone (or rather "Hemi-Alpine" ecotone 
because the open patches are outposts of Alpine plant communities) is usually 
of narrow vertical extension in the Scandes (Scandinavian mountains) proper 
but is more extensive on the elevated plateaux of northeasternmost Fenno- 
scandia. This type of vegetation even goes down to sea-level on the Polar 
coasts of Norway and the Kola Peninsula. 

The lower parts of the birchwoods are continuous with the birches usually 
of small tree size, reaching as much as 10 m or more in sheltered valleys and 
slopes with rich, moist soils. These lower parts are better regarded as 
equivalent to the Sub-Arctic. 

In a recent work Hustich (1960) regards most of northernmost Fenno- 
scandia, with woodlands formed by birch and some pine, as a Sub-Arctic 
region which he considers different from the Sub-Alpine region and equivalent 
to the woodland-tundra farther east. This seems to be an underestimation. In 
some of his earlier papers (e.g. 1949, p. 52) on Canadian phytogeography, 
Hustich used the rather ambivalent term "Sub-Arctic" expressly in the 
meaning "Woodland-tundra". However, in 1957, he included some more 
"southerly" and even Main Boreal vegetation under the heading of Sub- 

The Sub-Alpine -Sub- Arctic birchwoods are a feature of cool, oceanic 
areas. Outliers of them are found on southwest Greenland and on Iceland. 
The impact of severe over-grazing has probably caused them to disappear from 
the Shetland Islands (Spence, 1960), and much of the Scottish Highlands. 
However, there is no definite evidence that a birch belt has existed above 
the pinewoods of Scotland, except in the extreme north and west (McVean 
and Ratcliffe, 1962). Birchwoods are also locally present, usually near 
river estuaries and deltas, within the Siberian woodland-tundra, where the 
latter comes close to the Arctic Sea (Lavrenko and Sochava, 1954). Betiila 
Ermanii birchwoods also grow in Kamchatka and some of the Kuriles 
together with Sub-Alpine alder thickets. 


The zonal subdivision of the Arctic is far from clear. It is evident that 
altitude and maritimity influence Arctic vegetation as much as latitude, 
and thus regional subdivision cannot result in regularly consecutive zones on 
a small-scale map except on flat continents or very large islands. Polunin 
(1951) has proposed a subdivision into three zones (Low Arctic, Middle 
Arctic, and High Arctic), but it is not known exactly how these zones are 
distributed. A more detailed subdivision has been carried out and mapped for 


the Russian Arctic (Lavrenko and Sochava, 1954; Sotchava, 1954; Tik- 
homirov, 1960). 

For the Alpine barrens, a subdivision into three regions or belts is now 
unanimously accepted in Scandinavia (e.g. Du Rietz, 1928, 1930, 1942a and b, 
1950; Nordhagen. 1936, 1943). Each of the Low Alpine, Middle Alpine, and 
High Alpine belts has a phytogeographical significance about equal to one of 
the zones of the forested country. Similar belts are expected to occur in other 
northern mountains, and also in high mountains at somewhat lower latitude, 
e.g. the Alps (see a comparison by Du Rietz. 1930). A detailed comparison of 
the Arctic and Alpine zonations must be left to the botanists of the future. 


Ecological conclusion 

On comparing the two maps. Figs. 2 and 3, the general similarity is striking. 
A closer analysis reveals two interesting features. 

1. All zones bend southward when near cold seas. The eflfects of Hudson 
Bay (incl. James Bay), the Atlantic off Labrador and Newfoundland, and the 
White Sea are evident. The Woodland-tundra in these sectors (and even the 
Sub-Arctic near James Bay) is extraordinarily narrow due to this depression. 
The Baltic, although cool in spring, has an opposite effect, presumably 
because of comparatively long-lasting autumnal warmth. 

2. When not influenced by the cold sea effect, the width of the zones and 
sub-zones is dependent on slope. If the general slope is southern, they tend to 
be narrow (Boreo-nemoral of Quebec, sub-zones of the Boreal in Sweden). 
Great width is often related to a northward slope (Main Boreal of Canada 
west of Quebec, Sub-Arctic of Labrador Peninsula, Russian Main Boreal and 
Sub-Arctic). This, of course, is an effect of altitude, and shows that such an 
effect is present not only in mountainous areas. Even moderate elevation has 
considerable impact on vegetation, an impact that has not always been duly 

It would require too long a report to discuss the climatic factors by which 
the zonations are conditioned. It should only be mentioned that at least for 
the Boreal sub-zonation the length of the growth period appears to be highly 
important. It is defined as duration of the period when mean temperature is 
above +5.5" or 6". A lower temperature limit makes cool oceanic areas seem 
more favourable than they are. Summer heat seems to be more important 
near and above the tree-line. Winter cold is hardly of any general importance, 
and there is little relation to the distribution of permafrost (Brown, 1960). 

Historical conclusion 

The present zonation of the northern vegetation is young. It was practically 
non-existent as late as 9000 years ago; only 4000 and even 3000 years ago it 


was considerably different from modern conditions. Even if some of the 
species probably are very old, their re-combination into plant communities 
thus is comparatively recent. The elements of the Boreal or Taiga zone, for 
instance, must have survived the latest Glacial Epoch in several widely separ- 
ated refugia where ecological conditions and types of vegetation were 
necessarily different in various respects from those prevailing in the present 
Taiga. It is astonishing that the products of the re-combination of these ele- 
ments is so uniform, and that such a great zonal parallelism exists in areas as 
much as 8000 miles apart in longitudinal direction. 

The history of the vegetation is the evident starting-point of ecology. On 
the other hand, the ecological requirements of both single taxa and whole 
plant communities, as reflected in their present zonal distribution, put down 
limits which historical deduction should not overstep without very strong 
reasons. Therefore, this somewhat lengthy exposure of largely well-known 
facts may be of some value as a background for the historical discussions of 
these problems. 


Ahti, T. (1959). Studies on the caribou lichen stands of Newfoundland. Ann. Bot. Sac. 

Vanamo 30 (4), 1-44. 
Ahti, T. (1961). The open Boreal woodland sub-zone and its relation to reindeer husbandry. 

Arch. Soc. Vanamo 16(suppl), 91-93. 
Andreyev, V. N. (1956). Zaselenie tundry lyesom v sovremennuyu epokhyu (Penetration of 
forest into the tundra in the present epoch). In RastiteVnost Kraynevo Severa 

S.S.S.R. i yeyo osvoynie {The vegetation of the Far North of the U.S.S.R. and its utilisa- 
Brown, R. J. E. (1960). The distribution of permafrost and its relation to air 

temperature in Canada and the U.S.S.R. Arctic 13, 163-177. 
Du RiETz, G. E. (1928). Fjallens vaxtregioner. Naturens liv i ord och bildU, 800-811. 
Du RiETZ, G. E. (1930). Classification and nomenclature of vegetation. Sv. Bot. Tidskr. 

24, 489-503. 
Du RiETZ, G. E. (1942a). Rishedsforband i Tornetraskomradets lagfjallbalte. Sv. Bot. 

Tidskr. 36, 124-146. 
Du RiETZ, G. E. (1942b). De svenska fjiiliens vaxtvarld. Ymer 1942, 169-190. 
Du RiETZ, G. E. (1950). Phytogeographical excursion to the surroundings of Lake Torne- 

trask in Torne Lappmark (Northern Sweden). Seventh Int. Bot. Congr. Stockholm 

1950, Exc. Guid. C III c, 1-19. 
Du RiETZ, G. E. (1952). Vegetations- och odlingsregioner som uttryck for klimat och 

jordman. In Trddgardskonst I. Vci.xt material och anldggningsteknik. Stockholm. 
Du RiETZ, G. E. (mimeographed); Vdxtgeografiens griinder. Uppsala, finished 1961. 
Ebeling, F. (1962). Bestandsbehandling i norrliindskt storskogsbruk. In Skogen och 

Skogsbruket. Stockholm. 
Halliday, W. E. D. (1937). A forest classification for Canada. Canada Dept. Res. and 

Developm., Forest Res. Div. Bull. 89, 1-50. 
Hare, F. K. (1950). Climate and zonal divisions of the Boreal forest formations in eastern 

Canada. Geogr. Rev. 40, 615-635. 
Hare, F. K. (1954). The Boreal conifer zone. Geogr. Studies 1, 4-18. 
Hare, F. K. (1959). A photo-reconnaissance survey of Labrador-Ungava. Canada Dept. 

Mines and Techn. Surv., Geogr. Branch, Mem. 6, 1-64. 
HoJER, E. W. (1954). Skogsodlingsgransen pa kronoparkerna i Norrland. Norrt. Skogs- 

vardsforb. Tidskr. 1954, 233-253. 


HusTiCH, I. (1939). Notes on the coniferous forest and tree limit on the east coast of 

Newfoundland-Labrador. Acta Geogr. 7 (1), 1-77. 
HusTiCH, I. (1949). On the forest geography of the Labrador peninsula. A preliminary 

synthesis. Acta Geogr. 10 (2), 1-63. 
HusTiCH, I. (1951). The lichen woodlands in Labrador and their importance as winter 

pastures for domesticated reindeer. Acta Geogr. 12 (1), 1-48. 
HusTiCH, L (1952). Barrtradsarternas polara grans pa norra halvklotet. Comin. Inst. 

Forest. Fenn. 40, 29, 1-20. 
HuSTiCH, L (1957). On the phytogeography of the Subarctic Hudson Bay Lowland. Acta 

Geogr. 16(1), 1^8. 
HusTiCH, L (1960). Plant geographical regions. In A Geography of Norden. Oslo. 
Jalas, J. (1957). Die geobotanische Nordostgrenze der sog. Eichenzone Siidwestfinnlands. 

Ann. Bot. Soc. Vanamo 29 (5), 1-32. 
Kalela, a. (1958). tJber die Waldvegetationszonen Finnlands. Bot. Notiser 111, 353-368. 
Kalela, a. (1961). Waldvegetationszonen Finnlands und ihre klimatischen Paralleltypen. 

Arch. Soc. Vanamo 16 (suppl.), 65-83. 
Kujala, V. (1936). Tutkimuksia Keski- ja Pohjois-Suomen valisesta kasvillisuusrajasta. 

Conim. Inst. Forest. Fenn. 13. 
Lavrenko, E. M. and Sochava, V. B. (ed.) (1954). Geobotanicheskaya karta SSSR, scale 

1 :4,000,000. Akad. Nauk Moskva-Leningrad. 
McVean, D. N. and Ratcliffe, D. A. (1962). Plant communities of the Scottish Highlands. 

Monogr. Nat. Conservancy 1. 
NoRDHAGEN, R. (1936). Vcrsuch einer neuen Einteilung der subalpinen-alpinen Vegetation 

Norwegens. Bergens Miis. Arb. 1936, Naturv. Rekke 7, 1-88. 
NoRDHAGEN, R. (1943). Sikilsdalcn og Norges fjellbeiter. Bergens Mus. Skr. 22, 1-607. 
PoLUNiN, N. (1951). The Real Arctic: suggestions for its delimitation, subdivision and 

characterization. J. Ecol. 39, 308-315. 
PoRSiLD, A. E. (1951). Botany of southeastern Yukon adjacent to the Canol Road. Nat. 

Mus. Canada Bull. Ill, 1-400. 
Regel, C. (1950). Dynamik von Klima und Pflanzendecke in Nordeuropa. Ber. Geobot. 

Forsch.-Inst. Riibel 1949, 11-23. 
Regel, C. (1952). Botanische Betrachtungen auf einer Reise in Schweden. Ber. Geobot. 

Forsch.-Inst. Riibel 1951, 35-55. 
Ritchie, J. C. (1956). The vegetation of northern Manitoba. I. Studies in the southern 

spruce forest zone. Canad. J. Bot. 34, 523-561. 
Ritchie, J. C. (1959). The vegetation of northern Manitoba. III. Studies in the Subarctic. 

Arct. Inst. N. Am. Techn. Paper 3, 1-56. 
Ritchie, J. C. (1960). The vegetation of northern Manitoba. V. Establishing the major 

zonation. Arctic 13, 210-229. 
Rousseau, J. (1952). Les zones biologiques de la peninsule Quebec-Labrador et Themi- 

arctique. Canad. J. Bot. 30, 436^74. 
RowE, J. S. (1959). Forest regions of Canada. Canada Dept. North. A_tf. and Nat. Resourc, 

Forest. Branch Bull. 123, 1-71, map: Forest classification of Canada 1957. 
Shantz, H. L. and Zon, R. (1924). The natural vegetation of the United States. Atlas of 

Am. Agric. I. E. 
Sjors, H. (1950). Regional studies in north Swedish mire vegetation. Bot. Notiser 1950, 

Sjors, H. (1955). Remarks on ecosystems. Svensk Bot. Tidskr. 49, 155-169. 
Sjors, H. (1956). Nordisk vdxtgeografi. Stockholm. 

Sjors, H. (1959). Bogs and fens in the Hudson Bay lowlands. Arctic 12, 2-19. 
Sjors, H. (1961a). Forest and peatland at Hawley Lake, northern Ontario. Nat. Mus. 

Canada Bull. Ill, 1-31. 
Sjors, H. (1961b). Surface patterns in Boreal peatland. Endeavoin- 20, 217-224. 
Sjors, H. (in prep.): Bogs and fens on Attawapiskat River, northern Ontario. Nat. Mus. 

Canada Bull. 
Sotchava, V. B. (1954). Les principes et les problemes de la cartographic geobotanique. 

Essais de Botanique I, 273-288. 


Spence, D. H. N. (1960). Studies on the vegetation of Shetland. III. Scrub in Shetland and 

in South Uist, Outer Hebrides. J. Ecol. 48, 73-95. 
TiKHOMiRov, B. A. (1960). Plantgeographical investigations of the tundra vegetation in the 

Soviet Union. Catiad. J. Bot. 38, 815-832. 
TiKHOMiROV, B. A. (1961). The changes in biogeographical boundaries in the north of 

U.S.S.R. as related with chmatic fluctuations and activity of man. Bot. Tidsskr. 56, 

Wilson, C. (ed.) (1954). North of 55\ Toronto. 




P. Omodeo 

Institute of Biology and Genera! Zoology, University of Siena, Italy 

In order to describe the distribution of the terricolous OHgochaetes that 
populate the two coasts of the Atlantic. I think the best system is to list the 
forms that populate the various faunistic districts of America and to describe, 
case by case, their relationships with those found on the other side of the 

To carry out this review I shall follow Wallace's zoogeographic classifica- 
tion, which lends itself splendidly to our purpose. 

I shall begin with the Nearctic region (map. Fig. 1). which can be most 
clearly and briefly dealt with because of its extraordinary poverty. In this 
region there are 27 species of the Holarctic family Lumbricidae; one species 
for each of the genera Criodrilus and Sparganophi/iis. of uncertain systematic 
position but undoubtedly related to the Lumbricidae; about 20 species 
of Diplocardia (fam. Acanthodrilidae. subfam. Acanthodrilinae); four 
species of Ilyogenia (fam. Acanthodrilidae. subfam. Ocnerodrilinae); 
five species of Plutellus (Megascolecidae). Only the genus Diplocardia is 
endemic in the Nearctic region. 

These animals are not uniformly distributed: in the Canadian subregion 
we find only a few earthworms belonging to European species, whereas in the 
more southern Alleghany subregion. which is richest in forms, we find, in 
addition to the European species, which are always predominant, many 
endemic species: eight of the family Lumbricidae, twenty of the genus Dip- 
locardia, and one of the genus Sparganophilus. 

Very few of these forms (Sparganophilus eiseni and four species of Dip- 
locardia reach the nearby subregion of the Rocky Mountains, where they 
occupy several stations on the mainland of Mexico and in Lower California; 
in Lower California we also find four endemic species of the genus Ilyogenia, 
particular to the Neotropical region and equatorial Africa. Apart from these 
infiltrations, which concern only the southern districts, the better part of the 
fauna of the subregion of the Rocky Mountains is made up of European 

The Oligochaete fauna of the Californian subregion is also made up of 





European earthworms, but — a very curious fact — there we find five endemic 
species of the genus P/utellus, of which genus the other endemic species are 
found principally in AustraUa, Tasmania, and New Caledonia. The northern- 
most species, which is also the only one endemic in Canada, is found on 
Queen Charlotte Island. It is important to note that the genus Plutellus is 
one of the very few that include forms that tolerate brackish and salt water. 

The earthworm fauna of the Neotropical region follows the classical 
subdivisions quite well, with the difference, however, that the border line 
between the Brazilian fauna and that of Chile-Patagonia runs farther south 
than shown on maps drawn by most biogeographers (map. Fig. 2). It is 
worth noting that along the same line the distribution of certain groups of 
freshwater teleostei, amphibians, and terricolous molluscs comes to a halt. 

The peculiarity of the Neotropical Ohgochaete fauna lies in this fact: 
the differences from subregion to subregion are very clear cut and do not 
correspond to any existing geographical barriers. Nevertheless, whoever is 
familiar with the biogeography of the OUgochaetes will not be surprised, 
because a similar state of affairs occurs in many parts of the world. 

The Brazilian subregion is the one that has the most typical fauna, repre- 
sented by an endemic family, Glossoscolecidae, and by members of two sub- 
families of the Acanthodrilidae. 

The family Glossoscolecidae is made up of about 160 species belonging to 
22 genera. The two most important centers of endemism of this family are 
found in the high basin of the Amazon and in the low basin of the Parana/ 
Uruguay. To the north the Glossoscolecidae extend into Mexican and 
Antillean subregions, but it is important to note that there exist no endemic 
genera, and the only two endemic species are found in Costa Rica and on the 
Island of Barbados (the latter usually not included in the Antillean region). 
Some Glossoscolecidae extend into the northwestern corner of the Chilean 
subregion, formed by the Peruvian and Equadorian Andes: there the endemic 
species are numerous. No species of Glossoscolecidae has ever been found 
south of the fine that joins Bahia Blanca with Antofagasta. 

The Ocnerodrilinae are represented in the Brazilian subregion by four 
genera of which five are endemic {Haplodrilus, Kern'oua, Queclniona, Paulistus, 
Liodrilus) and one is found also in Lower CaUfornia. However, the taxonomy 
of this genus (Eukerria) needs some amendment (cf. Gates, 1957), which may 
lead to the two Californian species being separated. We find other genera of 
this subfamily in central America. 

Another genus endemic in the Brazilian subregion is the aquatic Dri/ocrius, 
of uncertain systematic position, but undoubtedly closely related to the 
African genus Alma: the northernmost species, found in Costa Rica, {D. a/fari 
Cognetti) has a morphology such that, if it had been found in Africa, it 
would certainly have been ascribed to the genus Alma. 

The last Ohgochaete group of the Brazilian subregion belongs to the 



uj IS < 

i i <^^' 


subfamily Benhaminae and consists of two species of the genus Neogaster and 
three of the genus Wegeneriella. These five species are found along the Atlantic 
coast from Darien to the mouth of the Amazon River. The other four 
species of these two genera are distributed along the Atlantic coast of Africa 
from Cameroon up to Guinea. The American species of Neogaster are 
extremely similar to the African ones; there are however, differences of 
subgeneric rank between the species of Wegeneriella of the two opposite 
shores. But one can say about these animals too, that the taxonomic distinc- 
tion would not have developed if the species had been gathered on the same 
continent: other genera of the subfamily are more heterogeneous than 
Wegeneriella, and no one has bothered to subdivide them. 

Facts regarding the Oligochaetes of the Brazilian subregion can be sum- 
marized in these words: the fauna has a very high degree of endemism, has 
nothing in common with the Chilean-Patagonian subregion, and very little 
relationship with the fauna of the Mexican and Antillean subregions; the 
latter is probably due to relatively recent migrations. A certain relationship 
between the earthworm fauna of the Brazilian subregion and the Palaeo- 
tropical fauna is shown by the two genera Wegeneriella and Neogaster, 
common to both, and the close resemblance between the American genus 
Drilocrius and the African genus Alma. 

In the Chilean-Patagonian subregion, excluding the corner north of 
Antofagasta and the region around the River Plata estuary, the Glossoscole- 
cidae, Ocnerodrilinae, Benhaminae, and the genus Criodrilus disappear; that 
is, all the taxonomic groups represented in the Brazilian subregion disappear, 
whereas a new family appears, the Acanthodrilidae. 

The Chilean-Patagonian Acanthodrilidae include more than fifty species 
belonging to five genera. The genus Yagansia is endemic, and its fifteen species 
are distributed from the Tierra del Fuego up to Titicaca; two other genera 
(Chilota and Parachilota) occur from Chile to South Africa over the islands 
between the two continents: in the southern part of South America, on the 
Falklands and in South Georgia, twenty-two species of Chilota and one 
of Parachilota are endemic. The genus Microscolex has a similar but much 
more extensive distribution, reaching the islands of Crozet and Kerguelen 
south of Africa, and the islands of Macquarie, Auckland and Campbell south 
of New Zealand; ten species of Microscolex are endemic to the southern 
point of America. On the adjacent islands, no less than three are endemic on 
the small island of Possession in the Crozet group (maps Figs. 3-5). 

The last genus, primitive as its name Eodrilus indicates, has a still greater 
distribution and reaches Madagascar, New Zealand and Australia. Four 
species of Eodrilus are endemic to the Chilean-Patagonian subregion. 

Thus, the tip of South America has four genera out of five in common with 
South Africa; on the other hand, it does not have any genus, not even a 
subfamily in common with the Brazilian subregion. T can add that the 

132 P. OMODEO 

subregion of Cape of Good Hope shares with the Chilean-Patagonian sub- 
region four out of seven genera, but has no genus and only one of its two sub- 
families in common with the rest of the African continent. Because of this, an 
oligochaetologist could mistake a collection from Patagonia for one from the 
Cape of Good Hope, but he could never confuse a collection of Chilean 
earthworms with one from Brazil. 

To complete the picture, I want to add some information regarding the 
biogeography of an austral family of limicolous Oligochaeta, the Phreodrili- 
dae, sub-divided into four genera, Hesperodrilus, Phreodrilus, Gondwanae- 
drilus, and Phreodriloides. The first genus has endemic species spread through 
Chile, the Tierra del Fuego, the islands of South Georgia, Crozet, Kerguelen, 
and Campbell, New South Wales, and Ceylon. The genus Phreodrilus has 
endemic species in New Zealand, South Africa, the Tierra del Fuego, and the 
Falklands. Gondwanaedrilus and Phreodrilus are endemic, respectively, in 
South Africa and New South Wales (map. Fig. 6). It is notable that two 
families of fish, Haphochitonidae and Galaxidae, and one family of crayfish, 
Parastacidae, have distributions almost identical with those of the Phreo- 
drilidae and the Acanthodrilinae (cf. Joleaud, 1939). 

To summarize: the Austral fauna of Oligochaetes extending from Chile and 
Patagonia to South Africa across the Falkland islands and New Georgia, 
and from South Africa to New Zealand across the Crozet, Kerguelen, 
Macquarie and Campbell islands is one of the most uniform and typical 
despite its great geographical splitting. 

The situation in the Mexican and Antillean subregions is much more 
singular than in the rest of America and also a bit more difficult to unravel 
because of the disorder that remains in some of the taxonomic categories, 
whereas, it must be stated, the systematics of the South American and 
South African Oligochaetes revised in the two large monographs by Michael- 
sen (1917) and Pickford (1937) are in perfect order. 

The Oligochaetes of these central American subregions (which will be 
dealt with together because they are essentially alike) belong to one family 
only, the Acanthodrilidae. The recent infiltrations from the South, which 
reach up to Costa Rica and the island of Barbados, and the admixtures from 
the north, which have come as far as Guatemala, are of course not counted. 
They make up a negligible part of the fauna and, except in a couple of 
cases, do not involve endemic species. 

The four subfamilies of the Acanthodrilidae are represented in central 
America as follows: there are about a dozen endemic species of Acantho- 
drilinae, most of which are usually attributed to the genus Eodrihis, but 
which, in my opinion, would be better placed in a separate genus. There are 
twenty-two endemic species of Benhaminae belonging to the genus Dicho- 
gaster found also in equatorial Africa (map. Fig. 7). The Ocnerodrilinae 
are represented by the genus Nematogenia, which lives exclusively in central 






05 n 
















tf o 

;"6 -o 


<j -Si 

(U <u 

so s: 




45 ^ 

.s ^ 


138 P. OMODEO 

America and western Africa (map, Fig. 5) and by the genera Ilyogenia (Fig. 
10). Ocnerodrilus, and Gordiodrilus, which Hve throughout tropical Africa 
also. Lastly there are six to eight species of Octochaetinae. Some of these 
have been attributed to the Indo-Malagasy genus Howascolex and one to 
the Indian genus Ramiella. It seems to me that they should be placed to- 
gether in a genus of their own; at any rate their relationship with the Malagasy 
and Indian forms is certain. 

To conclude, in central America and the Antilles we find fauna similar to 
that of Africa with Indo-Malagasy elements as well as elements related to 
that of the Chilean -Patagonian subregion wedged in between a commonplace 
Holarctic fauna and a Neotropical fauna with a high degree of endemism. 

I should like to emphasize the relationship with the African fauna which 
concerns five genera out of eight. We find that these five genera are abundant 
precisely on the western coast of Africa from Guinea to Angola. Apart from 
the representatives of the family Eudrilidae, confined to the equatorial strip of 
the Paleotropical region, there were represented in some West African 
collections that I have had an opportunity to study, eleven genera of which 
five were common to Africa and central America and two to the northeastern 
coasts of South America and Africa. 

To show the particular situation of the central American Oligochaetes 
more clearly, I shall momentarily modify the method of exposition that I have 
been following and instead of listing the faunas of each region and their 
relationship, I shall here describe the distribution of the family Acantho- 
drilidae (cf. Table 1). This family, as mentioned above, is divided into the 
subfamily Acanthodrilinae (the most primitive), Ocnerodrilinae (generally 
specialized for limicolous life), Octochaetinae, and Benhaminae. The 
Acanthodrilinae are distributed in the southern countries, from Chile to New 
Caledonia, and have in addition several genera in central and northern 
America: the Diplocardiacea section is endemic to this latter area. The 
Ocnerodrilinae occupy all of central America, the Brazilian subregion, 
Africa, Madagascar, and the Oriental region. Contrarily, the Octochaetinae 
are peculiar to New Zealand, the Oriental region, Madagascar, and are 
missing on the African continent but reappear in central America. The 
Benhaminae have the same distribution as the Ocnerodrilinae but they are 
very scarce in South America (map, Fig. 7). 

Only in central America are found all four subfamilies of Acantho- 
drilidae: following an old rule of biogeography, one should conclude that 
this district was the center of origin and diffusion of the entire group. This is 
possible but there remain some doubts because of the great gap between the 
central American and austral Acanthodrilinae and between the central 
American and Indo-Malagasian Octochetinae. However, it must be kept in 
mind that out of all the families of earthworms the one of which we are 
speaking here has the widest distribution and includes forms with the most 



















; -, ■- 



+ + + 










o -f- o 





+ o o + 






+ + + + 



O + o 



O ; 



O 0^ + 

0000 0+00 



















1 . 1 

1 -^ 1 


E 1 -r + + + 

ca 1 






+ + + + 






1 1 


. ! 1 






n- + + + 

«MO ■« 



-t- 1- -1- 



c« =« a •/= 


. « 



d cs 


o c 


+ ■ :^ 







o OO 









o o o 







D =! 









140 p. OMODEO 

simple morphology; therefore it is probably very ancient. That makes 
the point more clear because it is known that other more or less ancient 
systematic groups (reptiles, fresh water fish and amphibians) have a similar 
discontinuous distribution in central America and Madagascar since they 
have found conditions suitable for their survival only in these rather isolated 

At this point we can begin trying to reconstruct the genesis of the profound 
and undeniable relationship responsible for the fauna of Guinea being more 
similar to the American fauna than to that of East Africa, and for the 
Patagonian fauna being almost identical to that of the Cape area but entirely 
different from that of Paraguay. 

The explanations proposed for problems of this sort can be classified in 
two categories, which we will call static and dynamic. The static explanations 
presume that in the past the continents had more or less the same configura- 
tion and relative position as they do now; the dynamic ones presume that 
there have been substantial modifications in the configurations of the conti- 
nents and especially of their relative positions. 

If one admits the stability of the continents, the Amphi-Atlantic relationship 
of the terricolous Oligochaetes can be explained either by the hypothesis of 
passive transport across the ocean, or by Matthew's (1915) theory according 
to which certain systematic groups reached their present centers by radiating 
from Arctic Circumpolar regions; the subsequent extinction of the northern 
taxa would have lead to the present discontinuity. 

On other occasions (Omodeo, 1955, 1957) I have maintained that similar 
interpretations cannot be applied to the earthworms. I am still of the same 
opinion, but now I am disposed to grant a minimal margin of probability to 
these interpretations. 

If Simpson (1952) is right in believing that highly improbable events become 
almost certain given a very long period of time, it is also true that if the 
probability of an event is nil, eternity will not be enough to make it come true. 
It is a fact that earthworms, with the exceptions of very few species, have no 
possibihty of crossing the ocean haphazardly since they do not tolerate 
immersion in salt water. If they get bathed in it, they die immediately and the 
same happens to their eggs. 

Their transport across the ocean on rafts or logs is therefore unthinkable, 
and even if, for the sake of argument, we would accept it as possible, we 
would still have to explain how they could have crossed the beaches or the 
wave-beaten rocks on which they were stranded. 

Earthworms and their eggs dwell underground and they do not have any 
chance of getting stuck to the feet of some birds, and what is more, if they are 
exposed to the air, they dehydrate and die in a few minutes, or even faster if 
the air is moving. Thus their transport from one continent to another by 


aid of birds appears decidedly impossible, and even more so since no case 
of bird migration across the middle of the Atlantic is known. 

It is well known that the Oligochaetes are exclusively terricolous or fresh- 
water taxa and everything that is known about their phylogenesis leads us to 
assume that they have evolved exclusively in such habitats. Only a few forms 
of hmicolous Oligochaetes and two genera of Megascolecidae have acquired 
secondarily a certain tolerance of brackish and even salt water. The two 
genera of Megascolecidae which tolerate sea water are Pontodrilus (the name 
is significant) and Plutellus, often considered as one. Only for these forms is 
the passage from one continent to another thinkable, but mainly by means of 
slow displacement along the shores : the presence of five endemic species of 
Plutellus in the Cahfornian subregion can be explained by postulating the 
arrival of an ancestor from the eastern coasts of Asia. This is the only case 
where I consider it permissible to suppose than an earthworm was able to 
move, if not across the sea, then at least along its shores, to reach America. 

I want to add one last objection before moving on to other points : if we 
admit that the relationship between the earthworms of the two Atlantic 
coasts is due to passive transport of some ancient forms across this ocean, 
how do we justify the great differences existing between the earthworm 
faunas of Madagascar and the African mainland, which are separated by a 
body of water so much narrower than the Atlantic? 

The hypothesis that the principal zoological groups of the two coasts of 
the Atlantic came from the north — a hypothesis which has found its most 
authoritative defender in Matthew (1915) and has recently been supported 
by Darhngton (1957) — does not offer as serious theoretical difficulties as the 
preceding idea. Nevertheless, it raises myriads of doubts and problems calUng 
for very complicated, additional hypotheses. 

To explain the presence of the Acanthodrihnae in South Africa and in the 
Chilean-Patagonian region, intentionally ignoring their presence in New 
Zealand and other lesser southern islands and admitting that they started 
from the northern continents, we must explain also why they have disappeared 
from immense continental areas where once they must have been common. 
If we assume that the environmental conditions became hostile to them in 
these zones and that the appearance of new zoological groups has completely 
replaced them, we must ask ourselves why it is that two species of Acantho- 
drihnae, Microscolex phosphoreus and M. dubius, recently have been able to 
invade Europe, North America, and many tropical countries where they now 
flourish and compete successfully with indigenous forms. 

Certain facts regarding the biology of earthworms, which it would be well 
to make clear immediately, make the application of Matthew's (1915) theory 
to our case difficult. Earthworm species have often a latitude of adaptation 
unparalleled by any other invertebrate: Dendrobaena rubida lives on the 
island of Disko off Greenland, on the Himalayas and throughout the tropical 

142 p. OMODEO 

Malayan peninsula; the African species Eudrihis eugeniae, dispersed volun- 
tarily and involuntarily by man, colonizes much of the United States, where- 
as its original environment is the equatorial forest. Such an extraordinary 
adaptability makes it hard to imagine a climatic modification capable of totally 
destroying the populations of a genus or a family of earthworms living on a 
continent: it is almost certain, as we shall see, that the Lumbricidae of Green- 
land and Iceland have survived in situ all of the Last Glacial and maybe the 
entire Quaternary. 

The competition among these saprophagous species is very slight: in an 
area of a few square meters, in Africa as well as in Europe, it is possible to 
collect individuals belonging to 15-20 different species and also to two or 
more families. In any Tuscan garden it is normal to find living together 
species of Lumbricidae, Microchaetidae and Acanthodrihdae (the latter 
introduced). This being the situation, the substitution of an entire family of 
earthworms by another over a whole continent does not appear probable. In 
this connection, I think it is worth while to recall the case of the family 
Phreorictidae (or Haplotaxidae) — certainly very ancient and primitive and 
probably the ancestor of all living families of earthworms — containing 
endemic species belonging to two or three genera in all corners of the world : 
from Europe to New Zealand, from Sumatra to Japan, from the Cape to 
Guinea. In colder climates these species take shelter in phreatic waters, in 
hotter climates they occupy surface as well as subterranean waters; in any 
case they always have been found wherever an accurate investigation of the 
fauna has been carried out. 

Summing up. we can say that although Matthew's (1915) theory appears 
plausible for an explanation of the discontinuity of the neotropical Acantho- 
drilinae or the absence of the Octochaetinae in equatorial Africa, it is com- 
pletely inadequate to justify the principal pattern of earthworm zoogeography. 

It remains to consider the theories that suggest major modifications of 
the continental areas and of their reciprocal relationship through the geo- 
logical eras: the theory of the intercontinental land bridges, and that of 
continental drift. 

Only geophysicists and geologists can decide which of the two theories is 
correct or at least preferable; however, while waiting for them to come to an 
agreement on this point, I shall state my point of view. 

Personally I prefer a certain eclecticism and would tend to explain the 
genesis of the earthworm fauna of the central and southern countries facing 
each other on both sides of the Atlantic Ocean on the basis of Wegener's 
theory and the genesis of the fauna of the North Atlantic countries on the 
basis of the land-bridge theory. 

A glance at the present distribution of earthworms superimposed on 
Wegener's paleogeographic maps (maps, Figs. 8-10) is more satisfactory 
than a long discussion. It demonstrates how simply and completely can be 




Fig. 8. Distribution of the endemic species of the Lumbricidae. of the Acantho- 

drilinae genera Microscolex and Chilota, and of the Benhaminae genera Neogaster 

and Dichogaster, on Wegener's Mesozoic map. 

Fig. 9. Distribution of the Acanthodrilinae genera Eodrihis and Paracltilota, 

of the Benhaminae genus Wegeneriella, and of the OcnerodriUnae Nematogenia, 

on Wegener's Mesozoic map. 





'y'^'» HTM! ISLfUJU 

Fig. 10. Distribution of the Ocnerodrilinae genus Ilyogenia and of two genera of 
Phreodrilidae (freshwater Oligochaeta), on Wegener's Mesozoic map. 

solved what Michaelsen (1911) called the great puzzle of earthworm bio- 
geography. Naturally, the explanation based on Wegener's theory does 
raise important collateral problems, and this has often been reiterated. The 
main objection is that, if the continental drift theory is correct, there should be 
a greater uniformity in the distribution of the zoological groups, especially in 
the Southern Hemisphere. 

This objection can be answered by stating that the genera of earthworms are 
extremely ancient and have evolved very slowly, whereas other systematic 
groups have become so greatly differentiated that the ancient resemblance of 
the faunas has become almost obsolete. 

The great tardiness in the evolution of terricolous Ohgochaetes is docu- 
mented by innumerable circumstances. One of these, I think, must be apparent 
to whoever has followed me up to now : the distribution of single genera of 
earthworms corresponds generally to the distribution of superior taxonomic 
categories, families and superfamilies, of vertebrates or arthropods. Other 
indirect data on the antiquity of the Ohgochaetes can be gathered from the 
study of various fossil types of soil. It is a fact that no fossil remains of 
earthworms are known but fossil remains of soil types produced by the work 
of earthworms are well known : gyttja and mull. We know also that these 
soil types evolved during the Mesozoic in correspondence with the evolution 
of the modern Spermatophyta. 



Critical analyses of these phenomena have been carried out satisfactorily by 
Wilcke (1955) but it must be remembered that some of the Oligochaetes 
dwell in peat or moor soil, which was certainly the first environment that they 
occupied; even now the Phreorictidae — presumed ancestral group of all the 
living famihes of terricolous Oligochaetes — live in this environment; other, 
more modern groups, e.g. the Ocnerodrilinae, the genera Alma and Drilocrius, 
etc., have invaded this important ecological niche secondarily, but without 
doubt this occurred in a very ancient time while other parallel groups were 
speciahzing themselves in their present habitats. 


40 J 





i i i 


P i^'i 

i i i 



i mm 
i I 

■ j 


Fig. 1 1 . Chronology of Mammals, of Terrestrial Oligochaeta, and of three types 

of soil in millions of years. The data for Mammals after Rensch (1954), the data 

of soil types after Wilcke (1955). 

The diagram in Fig. 1 1 establishes the parallel between the chronology of 
the Mammal evolution as schematized by Rensch (1954), and the chronology 
of the Oligochaete evolution as it can be reconstructed from their geographical 








+ + + + MS 1 1 1 1 1 1 1 1 1 1 1 


+++++I+ I++II++ MM 


+ + + + -]- 4-+ + + + + +I+ +I+ + 


H- + + + + + + + ^- + + + + + + + + + 


1 + + + 1-1+ Mh+ M- M M M 1 


+ + + + M-+ + + + 1 4- 1 M M M 



M Mh M 1 1 + M 4- 1 + 1 M 1 1 

1 + 1 + M 1 ++ M + 1 M M M 



++++I 1 1 +++++I+I+I 1+ 


+ + + + + + + + + + + + + + + H- + + + 


+ + + + + M- 4-+ M + 1 + 1 M M- 



F + + + + 1 + _ + + \ 1 + + 4 1 1 i Mh 


AUolobophora rosea 
A. caliginosa 
Eiseniella tetraedra 
Dendrobaena rubida 
Octolasiiim lacteum 
O. cyaneiim 
Eisenia foetida 


Lumbricus terrestris 
L. rubelhis 
L. castaneus 
L. festivus 

Dendrobaena octaedra 
D. hortensis 

AUolobophora chlorotica 
A. limicola 
A. longa 

Bimastus muldali 
Dendrobaena mammalis 



distribution, and the ancient history of soil types as summarized by Wilcke 
(1955, 1960). 

As I have already mentioned, the validity of the Wegenerian hypothesis can 
be confirmed only by geophysicists but it seems reasonable for me to under- 
line the extreme simplicity and completeness with which it resolves certain 
zoogeographical problems, something which other Oligochaetologists, 
Michaelsen (1922, 1928) and Cernosvitov (1936), have already done with 
greater authority. 


Fig. 12. Distribution oi AUolobophora longa and A. liinicola. 

But putting aside these purely conjectural solutions, let us return to the 
first zoogeographical problem: the sameness of north European and North 
American earthworm faunas. Gates (1929, 1959) thought he had solved the 
questions by stating that all European species in North America were intro- 
duced by European settlers during the last two centuries; Lindroth (1957) 
accepts this point of view without reservations. 

I am very skeptical of this solution and feel obliged to ask : is it possible 
that a whole continent was almost entirely devoid of earthworms up till 200 
years ago? Or, is it possible that the European Lumbricidae annihilated the 
autochton taxa almost everywhere in North America ? 

To the first question the answer is: no. Any pedologist would refuse to 



accept that the soil of the North American grass and forest land was produced 
differently from that of other continents. Nor would it be possible to justify 
the presence in North America of certain higher animals, such as moles, whose 
diet is made up almost exclusively of earthworms, if none were there. 

To the second question the answer is another: no. More than 200 years 
ago European earthworms were introduced into central and South America, 
where they have competed successfully with the indigenous fauna, however, 
without ever replacing them, even in limited areas. 

Fig. 13. Distribution of Liimbriciis castaneiis and L. rubellus 

W/A £«mt«.. ^.lan 

On the other hand, the European species present in North America do not 
have the random distribution that would be expected if they had been intro- 
duced exclusively by the European settlers: they all occur in the northern part 
of Europe and many do not go beyond the northeastern part of North 
America (maps, Figs. 12-15). Of course, most of the traffic went on precisely 
between the northern countries of Europe and New England, but this is not 
enough to explain the massive presence of these animals in northern localities 
where farming is very recent. It must be remembered that traffic has been 
going on between Italy and Sardinia for two millenia and has not yet resulted in 
the introduction into Sardinia of species that are common on the Italian main- 
land (and found also in Canada). Table 2 eloquently sums up the situation. 



Usually, the species of Lumbricidae have a very wide distribution and 
therefore the Palearctic earthworm fauna is very uniform. For example, the 
earthworms of Afghanistan (cf. Omodeo, 1959) belong to species, all but one of 
which are found also in Italy; most of the Manchurian earthworms belong to 
species that we find in Caucasia as well as in the Urals; etc. Therefore, all 
premises exist for considering the North American populations of European 
earthworm taxa an integral part of the Holarctic, ancient fauna. 

Fig. 14. Distribution of Luinbriciis festivus and L. terrestris. 

In addition to indirect proofs and merely inductive arguments, there exist 
also very convincing direct proofs: Allolobopliora caliginosa, a very common 
earthworm throughout the Palearctic, is represented in the United States by 
two types, molita and anwldi (Gates, 1952), which have already evolved to a 
subspecific or maybe even specific rank. We also have the case of the North 
American genus Sparganophilus, represented in England and France by an 
endemic species. Finally we have, scattered all over the arc from the Faeroes 
over Iceland to Greenland, a large number of stations of Euro-American 
species. The biometric and caryological studies of the populations of Iceland 
and Greenland permit us to state that they are not as similar as would be 
expected had they developed recently from introduced material. Instead, 



they differ significantly both in morphology and in degree of ploidy (cf. 
Omodeo, 1957). 

In my opinion, the sum of these arguments points to only one explanation 
for the type of distribution we are dealing with, the existence of a land-bridge 

Fig. 15. Distribution of Dertdrobaena octaedra and Spaiganophilus. 

across the North Atlantic along which the earthworm fauna of Europe moved 
to North America, and a few rare American species went in the opposite 
direction, in a relatively ancient age, but not so ancient as to permit specific 
differentiation of most of the Lumbricidae. 


Cernosvitov, I. (1936). Notes sur la distribution mondiale de quelques Oligochetes. 

Mem. Soc. Zool. Tchechosl. 3. 16-19. 
Darlington, P. J. (1957). Zoogeogiaphv. New York. 

Gates, G. E. (1929). Earthworms of North America. /. Wash. Acad. Sci. 19, 339-347. 
Gates, G. E. (1952). New species of earthworms from Arnold Arboretum, Boston. Breviora 

Mas. Conip. Zool. Cainhr. 9, 1-3. 
Gates, G. E. (1957). Contribution to a revision of the earthworm family Ocnerodrilinae. 

The genus Nematogenia. Bid/. Miis. Comp. Zool. Caiiihr. 117, 427-445. 
Joleaud, L. (1939). Atlas de paleobiogeographie. Paris. 
Lindroth, C. H. (1957). The Faimal Connections between Europe and North Anierica. 

Matthew, W. D. (1915). Climate and evolution. Ann. N. Y. Acad. Sci. 24, 171 318. 


MiCHAELSEN, W. (1911). Die OiigochiUen des inneren Ostafrika unci ihre geograpliischen 

Beziehung. IViss. Erg. Deutsch Zential Afrika Exp. 3, 1-60. 
MiCHAELSEN, W. (1917). Die Lumbriciden mit besonderer Beriicksichtigung der bisher als 

Familie Glossoscolecidae zusanimengefassten Unterfamilien. Zool. Jahrh. (Syst.) 41, 

MiCHAELSEN, W. (1922). Die Verbreitung der Oligochaeten im Lichte der Wegener"schen 

Theorie der Kontinentai-Verschiebung, etc. Verh. Naturwiss. Ver. Hamburg, 29, 1-37. 
MiCHAELSEN, W. (1928). Oligochaeta. Kiikenthal: Handh. d. Zoologie. Berlin u. Leipzig. 
MiCHAELSEN, W. (1933). Die Oligochaetenfauna Surinames mit Erorterung der verwandt- 

schaftlichen Beziehungen der Octochatinen. Tijdsclir. Ned. Dierk. Vereen. 3, 112-131. 
Omode(i, p. (1955). Nuove specie dei generi a distribuzione anfiatlantica Wegeneriella e 

Neogaster. Ann Jst. Mus. Zool. Univ. Napoli, 7 (3), 1-29. 
Omodeo, p. (1957). Lumbricidae and Lumbriculidae of Greenland. Medd. oni Groenl. 

124 (6), 1-27. 
Omodeo, P. (1959). Oligocheti delPAfghanistan. Boll. di. Zool. 26, 1-20. 
PiCKFORD, G. E. (1937). A monograph of the Acanihodrilinae earthworms of South Africa. 

Rensch, B. (1954). Neiiere Prohleme der Ahstamnuingslehre. Die transspezifisehe Evolution. 

Simpson, G. G. (1952). Probabilities of dispersal in geologic time. Bull. Amer. Mus. Nat 

Hist. 99, 163-176. 
WiLCKE, D. E. (1955), Bemerkungen zum Problem des erdzeitlichen Alters der Regen- 

wiirmer (Oligochaeta opisthopora). Zool. An:. 154, 149-156. 
WiLCKE, D. E. (1960). Fossile Lebensspuren von Regenwijrmern. Decheniana 112, 255-269. 




Henrik W. Walden 
Museum of Natural History, Gothenburg, Sweden 

A ZOOGEOGRAPHICAL divisioii with respect to land Gastropoda rather markedly 
differs from the conventional one, which is based mainly on the distribution 
of vertebrates (cf. Darlington, 1957). However, more thorough studies of 
the differences show that an essential causal consistency exists. Somewhat 
schematically, the terrestrial Gastropoda can be characterized as a conserva- 
tive group, whose actual distribution and taxonomy to a rather remarkable 
degree reflect the past. On the contrary, they less obviously react to short- 
term changes. The rich paleontological evidence offers substantial basis for 
speculations concerning the faunal development. 

For an understanding of the recent distribution of the gastropod fauna on 
both sides of the North Atlantic it is necessary to follow its development 
since the late Mesozoic. At that time a terrestrial gastropod fauna, with 
obvious points in common with the present one, appears for the first time. 

At this era the gastropod faunas in Europe and North America are already 
thoroughly distinct from each other, and they have largely remained so till 
the present time. However, this does not imply a parallel development — on 
the contrary, very marked differences in the trends of development exist. These 
differences are of importance to the recent distribution of the land Gastro- 

Owing to the presence of modern, comprehensive literature, it is easier to 
demonstrate how the gastropod fauna has been formed in North America 
than in Europe. The European literature in the field no doubt is much richer 
than the American, but for obvious reasons it is highly diverse from the 
point of language as well as quality. These circumstances should be kept in 
mind in the following discussion. 

Taxonomy, distributional and paleontological data mainly follow the 
fundamental works of Wenz (1961-2), and Wenz and Zilch (1959-60), in 
addition to which Pilsbry (1939-48), and Henderson (1935), constitute the 
main sources concerning North American Gastropoda. Certain modifica- 
tion, especially concerning fossil records, has been undertaken in accordance 



with Hibbard and Taylor (1960); Licharev and Rammelmeyer (1952); and 
Taylor (1954). 

It may also be pointed out that, diverging from Wenz and Zilch, the 
subgenus Calidivitrina has been placed under Vitrina (cf. information given 
by Hubendick, 1953). Consequently Semilimax has been regarded as an 
endemic European genus. Lehniannia has been classified as a subgenus of 
Umax (cf. Walden, 1961). Finally, the records assigned to the highly problem- 
atic genus Brachyspira have been disregarded. 

As it seems still to be current in the literature, attention Is drawn here to an 
old opinion thai the early European gastropod fauna has much In common 
with the recent American (cf. Ehrmann, 1914. and authors mentioned by 
him). It was based on superficial shell resemblances, but its ground has 
nowadays been radically demolished by the gradually deepened taxonomical 
knowledge. It is not excluded that future taxonomical revision will corre- 
spondingly modify the conception of some further taxa, whose common 
presence on both sides of the Atlantic today seems rather enigmatic. 



in all, more than 750 terrestrial gastropod species are known from North 
America. Of these, about 50 have been introduced by man in historical time. 
The occurrence of this element has been so exhaustively elucidated by 
different authors (see Lindroth, 1957; Pilsbry, 1939-48; and Quick, 1952) 
that it can be disregarded in the present account. The remaining more than 
700 species are dispersed through 107 genera. 

The North Anierican land Gastropoda are dominated by two pronouncedly 
endemic groups, which together comprise about 80 per cent of the indigenous 
species. In relation to the European Gastropoda the endemism occurs largely 
on the family level. — The first group is eastern, with the Appalachians as its 
center, and comprises about 250 species. Among these the members of the 
autochtonous family Polygyridae dominate. Only a minor fraction of the 
endemic species, about 50 in all, have Old World relatives. In most instances, 
however, those belong to different genera or subgenera. 

The other. West American group, comprises about 300 endemic species, 
localized on the Pacific coast and in the Rocky Mountains. This group is 
dominated by two families of very old, autochtonous development, viz. the 
Helminthoglyptidae and the Camaenidae. The latter is represented by other 
genera in Australia and East Asia. Holarctic genera are less well represented 
in western North America than in eastern. On the contrary, the influx of 
central or South American genera is much richer in the southwest as a conse- 
quence of the land connection there. 

Figure I illustrates the development of the terrestrial gastropod fauna in 














UI )3UI)X3 








BOuaiuV N 
UI iua33^ 








rt y -^ 1^ ^ = 


o^ 1 io 

eiaua9 (O in I a> I CO ot 



10 % 

Genera with 
affinities to 
Europe and/ 
or East Asia. 










UC3U3UJV in rv c^ — ^ tti 
nuap puB s o* I "" 







s 1 « 

9 " 

M. i 

O 01 


i r 

1 1 

■1 UJ 







CO i en 
CM ! — 

2 CN g 





1 1 





CM — 


^ o> 




















5 1 





'^ \ 
































P -c < 'C '5 ~ 
.. ^ r -4^ -c 

5 12 

c3 uT i: 

O o re rt — ^ 

^ ^ ^ '^ fc CA 

rt O r- -^ CJ 

60 ^ O C CJ "p::; 

^ .^ ;* u _: - i; 

•- — re 

c c 

ij i: g w ^ 

2 "^ — 

> c 


15' X 

2 ° 

■y^ 60-- t: 

c3 c .^ ^ ^ re 

^■E re^J^ 

y; u > > -^ T3 
= > ^ ^ u 

g fe^ii ^^ 

re± w e ^ a. 
— 5^ re c: '' 
re — o _ 

2 o 

<=: zZ^ 
■-■- 't-- =■ 


C ,, tJ t^ 

<u aj {U 

_^ E 3 E 

"O 4J 

o ? ^ 
u c '^ 

c/: ; — 

d 60X) 

'^ '? o 

r; o 

CA C 1- CJ 

(U 1) ^ 

o t-> •- 

■o o [;? 

o ,o 


.. _ : E ■£ K 

E ^|>^.E^ 

.i: -S ^ *- y <u 

., C J'- 

N ^ -a ^ 

>^£ n 

i= c. p 

^^l E "^ 

^ '^ E c 

!" o« 

t; D >- lA 

^7 x-2 

U ^ 60 

^ ou ^ 

■^ ^ s 

^ S o 

..-- C x, 

cc re -^ 

re • >- 

_ tA re 

cj C~ 

< o E 

60 " 

■- O J= 
C C- ■" 

cs 60 ^ - c: «J 
.:^ c re 60 *fi ^ 

w ^ ^ u re ^ 

<u y c 60 u 

^ re o c 

^ x: 



OJ '?^ "P 

u re rt =: 

^ 60« ii 

tiS J= .i: u £; X) t^ 


North America. From its early beginning it seems to be entirely distinct 
from the Palearctic one. At the end of the Mesozoic, tropical genera dominate, 
which seem to be purely American. Already some early representatives of the 
two major endemic groups appear. These increase rapidly, whereas the 
Tropical genera decrease correspondingly. Soon, in the Eocene, the propor- 
tions between endemic North American and essentially Tropical genera are 
near those of the recent period. At the same time the first representatives of 
the Holarctic genera appear in the strata. 

In addition to these trends in the development it is striking that the number 
of now extinct genera has always formed a minor fraction. The recent fauna, 
essentially, is composed of direct descendants of the early Tertiary fauna, 
which have been able to adapt themselves from the original, almost tropical, 
conditions to the present temperate ones. 

Henderson (1931) suggests as the basic reason for the American endemic 
pattern the probably profound separation of land masses in the Cretaceous. 
The largely remaining differences between the eastern and western gastropod 
groups have been maintained by unfavorable conditions in a broad area, 
east of the Rocky Mountains, which seem to have existed rather continuously 
since the Eocene. The principal exchange of land Gastropoda has taken place 
from the east to the west. The forcing of the Rocky Mountains by some 
eastern genera had largely come to an end in the Miocene, resulting in a 
today partially highly endemic offshoot from the eastern group. The effect of 
the interchange between the two major endemic groups never reached beyond 
the North American continent. 

Table 1 gives an account of the extracontinental relations on the generic 
level for the North American land Gastropoda. As the biogeographical 
boundary of North America the border of Mexico has been used, in accor- 
dance with Pilsbry (1939-48). Only six of the essentially endemic North 
American genera exceed this border. 

The dispersal of the tropical American element falls beyond the present 
theme, and will not be discussed further. It comprises about 50 species, most 
of them in the extreme south of North America. The Tertiary Tropical 
gastropod fauna was composed largely of species closely related to the recent 
ones, but the latter are late invaders, probably mainly Post-glacial. The signs of 
endemism are weak, or non-existent. 

For the remaining non-endemic element three possible ways of dispersal 
must be considered : 

(1) The North Pacific route. 

(2) The North Atlantic route. 

(3) Across the subtropical or tropical parts of the Atlantic. 

The Holarctic genera comprise somewhat over 100 species in North 
America. These represent all stages from morphological identity in the New 

Table 1 


^"""^^--.-^^^ Number of genera 
Distribution type ^^^--^..^ 

Extinct in 
N. America 

Recent in 

N. America 



Archeo- or holotropical* 
Central or South American 
Mainly North American and 

East Asiatict 









Endemic North American^ 




* In the Archeo- and Holotropical group, marine littoral genera play a major role. 
This refers to four extinct and six recent genera. They are useful as climate-dependent 
indicators in faunal development, but do not tell much about the ways of dispersal of the 
land fauna proper. Apparently, their often very wide distribution is a consequence of the 
fact that they are far better adapted to the possibilities of passive dispersal offered, e.g 
by ocean currents, than the other, truly terrestrial genera. In fact, some littoral species are 
rather amphibious than terrestrial. Two of the strictly terrestrial genera are regarded here 
as Archeotropical, viz. Pseiidocohimna and Protornatellina, Paleocene and Cretaceous 
respectively. To the Holotropical group are assigned Pupoides, Piipisoma, Cecilioides, 
Lamellaxis and Opeas. Of these, Pupoides appears in the Eocene, the others in N. America, 
are known from the Holocene only. 

t In the North American-East Asiatic group the record of Rhiostoma, from Oregon 
Miocene, must be regarded as somewhat problematic. Calinella, with a Pacific center, has 
been included in this group. The remaining genera are Hendersonia, Strobilops and Gastro- 
copta. The last two have a wider distribution and a complicated history, briefly discussed in 
the text. 

+ Some of the endemic North American genera show significant extra-continental 
relations. Philomyciis, Anadenuhis and Prophysaon have their closest relatives in East or 
Central Asia. Concerning the relations to certain European genera of Gastrodonta and 
Euglandma (the latter assigned to the Central and South American group), see text. 

Cf. also the caption to Fig. 1. 

and the Old World to distinct species and subgenera. Apparently their 
dispersal must have proceeded during a very long time, a statement which is 
verified by paleontological evidence. A remarkable feature in the distribution 
of the Holarctic genera is their dominance in the gastropod fauna north of 
47° N. Lat., thus largely within the area of the Pleistocene glaciations. 

However, not only in the Holarctic element, but also in the Tropical and 
East Asiatic, affinities to Europe are evident. These are partly of a highly 
intricate nature. But before the routes of dispersal can be discussed, the 
European gastropod fauna and its history must be regarded. 


The history of the European land gastropod fauna shows a pattern which 
is very different from that of the North American. In detail it is much more 


complicated in Europe; this is reflected in the markedly richer taxonomical 
differentiation. In recent times, 229 genera are indigenous to Europe. For the 
number of species, however, no actual, reliable figures exist, but they may be 
estimated as more than double those of North America. On an average, 
however, the geographical splitting of taxa takes place on a lower taxonomical 
level than in North America. The major causal background — the highly 
complicated history of the European mountain upfoldings and seashore 
dislocations — has led to isolation processes which were richly varied, but 
only moderately persistent. However, the details are of infra-European 
interest only, and consequently they can be largely ignored here. 

The principal traits in the development of the European land Gastropoda 
is demonstrated by Fig. 2. 

As already pointed out in the preceding paragraph, the land gastropod 
fauna which appears in Europe in the Late Mesozoic had no points in 
common with the contemporaneous North American fauna. But, in addition, 
it has very little in common with the recent European. Except for some 
unimportant, very specialized remnants, the Mesozoic genera today are 
quite absent from Europe. In 88.5 per cent (24 of 27 genera), the connections 
were Tropical, either with the recent South Asiatic and African gastropod 
fauna, or, for the greater part, representing an Archeotropical element which 
is difficult to evaluate taxonomically. 

Some authors have made the fascinating, though very hypothetical sugges- 
tion that this Archeotropical element partly represents an old "Gondwana" 

However, from the Paleocene onwards the recent European genera appear. 
At first, such genera, which today are Holarctic, form a relatively large 
fraction. Later, more and more endemic European genera are added. The 
Archeotropical element rapidly disappears, whereas still in the Eocene that 
which has connections to South Asia and Africa constitutes nearly 40 per 
cent of the land gastropod fauna. Then it continuously decreases, a process 
which no doubt reflects the climatic development during the Tertiary. In this 
respect the trend entirely parallels that of the marine Mollusca (Davies, 1934), 
and of the flora (Reid, 1935). 

Up to and including the Miocene nearly half the endemic European 
element is constituted by subsequently extinct genera. After this period, 
however, their number rapidly decreases. Also this process probably should 
be considered against the background of the climatic development. In the 
Phocene, the European land gastropod fauna has essentially its recent charac- 
ter; 87 per cent of the genera of this period still live in Europe. From the 
Pleistocene only recent genera are known, if the very doubtful record of 
Archaeoxesta in Germany is disregarded. This record may be due to redeposi- 
tion from Tertiary strata. 

To conclude, the European land gastropod fauna has kept less of its earlier 









(D CM 







CM 00 





ur 13UUI3 














ui )uS3aij 










o ja 


« 0) 
■X c 






CM 00 



I s. 





8 1 



o c 

-5 " c 


P — 


1 , 






CM — 

1 1 

leiiJ u< 













c 5 

c £ 
3 " 






















3 P"" S 



ailBisv s 

3 < 


pue uS3ri)y 





H Ui 




















1 H^ 









S 1! 
£ « 









< s. 








g 1 




8 " 














^-^^^^^^^Bn ^ 





















































o <I> 

ii a: 








— I 

1 -J 



3 O 

>- o 

•r; oj 
.3 -^ 

*^ C/5 

UJ .ti 

:=: O ^ 

r^ — 



=« c ►^ 

S >= "" 

"" c/i >> 
60 =^ 15 
iZ TJ '^ 

o -o o 

:; !3 ^ 

U 60 >i 

— • CJ 03 

oj (u x: 
— X •-: 

U j= t^ 



character than has the North American. But in spite of this it is remarkably 
conservative in comparison with most other animal groups, known by 
fossil evidence, as well as with the vascular plants. Szafer (1954) stated that 
the Euro-Asiatic element in the Pliocene constituted only between 6 and 20 
per cent of the flora in southern Poland. For the rest, now disappeared 
American and East Asiatic species dominated. 

In Table 2 the extracontinental relations of the European land Gastropoda 
are elucidated. It may be emphasized that Transcaucasia and Macaronesia 
(the Azores, Madeira, and the Canary Islands) have been counted with 
Europe. In this connection it is worth mentioning that the Macaronesian 

Table 2 

""^"-^^..^^^ Number of genera 
Distribution type^--^^^^ 

Extinct in 

Recent in 



Archeo- or holotropical* 
African and South Asiatic 
South and East Asiaticf 
Mainly North American and 
East AsiaticJ 













Endemic European-Palearctic§ 




* Marine littoral genera, with their special dispersal ecology, play a certain role also in 
the European gastropod fauna. Among the Archeo- and Holotropical genera four extinct 
and four recent ones represent this group. The two recent, strictly terrestrial genera, here 
classified as Holotropical, Cecilioides and Coilostele, are discussed in the text. 

t The South and East Asiatic group includes nine marine littoral genera, all extinct 
in Europe. 

1 Concerning the heterogenous group of North American-East Asiatic genera (Gastio- 
roplo, Strobilops and Catinella), cf. second note to Table 1, and the text. 

§ Among the recent European-Palearctic genera, 124 are exclusively European, the re- 
maining ones occurring also in NW. Africa and/or the Orient, in a few instances also with 
adjoining areas in tropical Africa. Six genera have a more or less wide distribution also in 
temperate Asia. The two genera reaching North America, Umax and Cepaea, are discussed 
in the text. 

Cf. also the caption to Fig. 1. 

gastropod fauna, which is the taxonomically most independent and most 
isolated partial fauna of the region, is decidedly an offshoot of the early 
Tertiary Mediterranean and west European fauna. The affinities to other 
regions are very insignificant. 

In a case where an entirely extinct genus shows clear taxonomical affinities 
to, for example, recent South Asiatic genera, it has been classified as South 
Asiatic, even if it has only been found in Fin-opc. 


Table 2, as well as Fig. 2, particularly well demonstrates the importance of 
the interchange in a southeastern direction. A long series of Tertiary European 
genera, or relatively closely related forms, today are met in South and even 
East Asia. To these is joined a more limited element, which occurs both in 
South Asia and in Africa, mainly in the eastern parts. From the point of 
view of principles it is of secondary importance whether the genera are 
European emigrants, or if they have spread from, for example, an Asiatic 
center of origin to Europe, and later become extinct there. 

An exclusive African affinity, wliich does not seem possible to connect 
to the South Asiatic route of dispersal, is shown by very few genera only. 
Except for three Paleocene genera which, indeed, are rather difficult to 
delimit from the previously regarded Archeotropical element, only the recent 
genus Lauria belongs to this group. However, as this genus is comprised of 
small forms well adapted for passive dispersal, for instance, by birds, its 
distribution is not very significant. In fact, also some Paleo- or Holarctic 
genera, with similar qualifications for passive dispersal, have adjoining 
areas in Africa, mainly on high mountains. For one genus, Truncafellina, 
the African distribution is so extensive that an African origin is not 

These conditions in terrestrial Gastropoda are quite compatible with 
modern conceptions regarding the pattern of evolution of tropical African 
biota (cf. Moreau, 1952), dominated by the long and efficient isolation. 

However, in recent time the European land gastropod fauna is as completely 
isolated from the South and East Asiatic one as it has been from the African 
one throughout the Tertiary. The part of the Mediterranean fauna, which has 
its center in northwestern Africa, has scarcely more points in common with 
the tropical African one; its affinities are almost exclusively European. 

For the Holarctic genera, which will be mentioned later, the dispersal 
over the Euro-Asiatic territory has been of fundamental importance. These too 
can be seen as an example of interchange between Europe and Asia, though in 
a more northerly latitude, and less influenced by climatical and geographical 
barriers during the Tertiary. 

Disregarding the possibility that an interchange of Holarctic taxa has also 
taken place in the North Atlantic area (though, for several reasons, to a 
limited degree only), there are no traces whatever of any contributions to the 
European land gastropod fauna from North America. Likewise, in the 
opposite direction, there are no incontrovertible examples of dispersal, 
except the doubtful, and in every case late, instances of Cepaea and Umax 
which will be attended to further on. 

However, a careful examination makes evident that for some taxa, which 
are common to the warmer parts of Europe and America, the possibility of a 
trans-Atlantic dispersal must be considered. But the nature of this dispersal 
seems enigmatic in the light of the facts that are known at present. 



For a correct understanding of the distributional history of the land 
Gastropoda their means of dispersal must be regarded. Small forms, especially 
when exposed, e.g. by climbing up into the vegetation, are subject in a con- 
siderable degree to passive dispersal, by birds and other higher animals, or 
even as aerial plankton. In all regions it is mainly the minute forms that 
exhibit wide distributions. This is pertinent on generic as well as specific levels. 

Increasing size limits the chances of passive dispersal, and the animals 
mainly become dependent on their capacity to actively migrate. It might be 
superfluous to state that this is rather restricted for land snails. However, 
the ability to self-fertilize must essentially favor dispersal. Large species may 
easily be transported passively when juvenile, but if capable of self-fertiliza- 
tion a single specimen can be the origin of a new population. It is known that 
different species behave differently in this connection, but unfortunately the 
knowledge is too Mmited yet to allow a broad apphcation to zoogeographical 

Hydrochorous dispersal, e.g. by drifts, is of great importance in the case of 
fresh water. But for transport over the oceans it can almost be discounted, 
partly because of the marked sensibility to NaCl-exposure of most land 
Gastropoda, partly due to their specialized ecological requirements. The 
great majority of species definitely avoid sea-shore habitats. Also when the 
transport (for instance under the bark of a floating log, cf. Kew 1893), may 
have been successful, the chance of reaching a proper habitat is very low. 

The marine littoral species constitute exceptions. As a consequence of the 
possibilities of dispersal which thus are opened for them, they often are 
very widely distributed. Being irrelevant to the present subject they will be 
disregarded here. 

All stages between excellent fitness for passive dispersal and an almost 
absolute dependence upon their own active migration are met among the land 

The Bering Strait route without a doubt has played an important role in the 
dispersal of land gastropods. Certainly it has been of essential importance for 
the development of the Holarctic element. The Holarctic genera, 16 in all, 
are presented in Table 3, with data for their earliest occurrence in European 
and North American strata respectively. 

For at least 12 genera earlier, often considerably earlier, records are 
present from Europe. It seems justified to assume a preponderantly Pale- 
arctic origin for the recent Holarctic element. Also, the indication that already 
in the Paleocene those genera reached their relative maximum (15 per cent) 
in the European gastropod fauna, whereas their North American relative 
maximum (30 per cent) was reached as late as in the Pleistocene, points in the 
same direction. 



Most of the Holarctic genera are characterized by their considerable age; 
thus in Europe they belong to the oldest persisting element in the gastropod 
fauna. With some generaHzation, the hypothetical nature of which may be 
admitted, the recent Holarctic element can be regarded as a relic Mesozoic to 
early Tertiary gastropod fauna of northern latitudes. It is probably not 
accidental that this gastropod element is more or less associated with the 
plants which characterized the corresponding Arcto-Tertiary flora. 

The species which belong to the Holarctic gastropod genera are largely well 
adapted for passive dispersal. Besides favoring extensive distribution, this 

Table 3 

Earliest European 




North American 























Paleocene ? 



Eocene ? 






Paleocene ? 

Eocene, possibly 
















* The absence of fossil records of Zoogenetes must be seen against the background that 
its shells, owing to shortage of lime, are not suited for fossilification. 

must have counteracted genetic isolation and thereby taxonomical differentia- 
tion. Fischer (1960) has argued that the repeated interruption of biotic 
evolution by climatic disasters in the Circumpolar region tends to keep the 
biota on an immature, relic level. In fact, this fits very well to the ancient 
nature of the Holarctic Gastropoda, at the same time as it must be intimately 
linked with their fitness for passive dispersal, which apparently must be of 
selective value in regions of violently fluctuating physical conditions. 

The recent distribution of Holarctic gastropod taxa must have reached its 
pattern to such a great extent during the Quaternary that it hardly gives any 
indications for, or against, a Tertiary dispersal across the North Atlantic. 
Reasonably, however, the presence of passive dispersal, a characteristic of the 
group, may have played a role also in the Tertiary. 


But it is quite evident that no interchange of the endemic European and 
North American elements in the North Atlantic region took place during 
the Tertiary. The profound character of the endemism is elucidated by the 
fact that it is almost total, also on the family level. Besides the Holarctic 
group (and the doubtful cases of Limax and Cepaea) none of the families, 
which dominate on the respective sides of the North Atlantic are represented 
on both. As already mentioned this separation has existed throughout the Ter- 
tiary; indeed, the initial differences in the late Mesozoic seem to have been 
even more profound. The few common features which exist, besides the 
Holarctic group, refer to a more heat-requiring element, for which dispersal 
via more lengthy routes must be considered. 

On the Pacific side, on the contrary, common features are more manifest. 
For a number of important families there are clear affinities to Australia and 
tropical East Asia; these, however, fall beyond the present scope. Pertinent in 
connection with a probable North Pacific dispersal is a group of five genera 
(cf. the Notes to Table 1), one of which is now extinct. To be correct, the 
relationship is not very close. The forms on the two sides of the Pacific are 
placed in different subgenera, or even genera. This indicates that the time for 
the interchange must be remote, probably in the Early Tertiary. The genera in 
question belong to the warm-temperate to Subtropical element, and the 
later apparent checking of the interchange via the Bering Strait bridge 
certainly must be seen in connection with the Tertiary climatic development. 
In the Middle and Late Tertiary the climate in the Bering Strait area apparently 
was too cold to allow passage for other than cold-adapted, or generally hardy 
animals. This has been very clearly demonstrated for mammals by Simpson 

In relation to the total endemic land Gastropoda, in North America as 
well as in East Asia, this "Amphi-Pacifi" (in fact the recent occurrence of 
some of the genera is remote from the Pacific coast) element constitutes an 
inconsiderable fraction only. 

More doubtful is the evidence given by certain other genera which occur in 
North America and East Asia, though not exclusively. Strobilops, which 
appears in the North American Pliocene, already existed in Europe in the 
Eocene, though it became extinct there in the Pliocene. Pilsbry (1948, p. 853) 
suggests an Asiatic center of origin for this genus, from which it has radiated 
on the one hand to Europe, on the other to East Asia and via the Bering 
Strait bridge to North America. However, a careful consideration shows that 
two of the American subgenera occurred also in the European Tertiary, 
whereas the genus in East Asia is represented by subgenera which do not live 
in America. Instead, also one of the East Asiatic subgenera is known from the 
European Tertiary. As a consequence, the reasons for a North Pacific dispersal 
of Strobilops scarcely are convincing, whereas it seems difficult to reject the 
possibility of a trans- Atlantic dispersal within the warm latitudes, the more so 


since the subgenera in question occur also in South and Central America. 
The alternative of a North Atlantic dispersal is certainly out of the question 
for the essentially warm-temperate to tropical genus Strobilops. 

A similar distributional pattern and history is also shown by the richly 
differentiated, preponderantly tropical genus Gastrocopta as well as by 
Pupoides whose origin, however, may be American. There are no indications 
that the latter genus has ever lived in Europe, 

Some further genera, which may be characterized as chiefly Pantropical, 
have a distribution which is difficuh to understand without assuming a 
trans-Atlantic dispersal. As an example, the subterranean-dwelling Cecilioides 
may be mentioned. It is represented by different subgenera in Europe and 
America, the European one having advanced as far north as the southern Baltic. 

However, it must be admitted that all these genera, by their very extensive 
distribution and their good qualifications for passive dispersal, show rather 
vague indications of faunal history. Stronger evidence for trans-Atlantic 
dispersal is given by the American genera Gastrodonta and EitgJandina. 
The former genus shows affinities to Jamdus of Madeira and the Canary 
Islands, which was widely distributed in Europe during the Tertiary. Euglan- 
dina, along with several tropical American genera, is related to Poiretia in the 
Mediterranean territory. These genera suggest a trans-Atlantic dispersal, but 
it must have been remote, perhaps in the Mesozoic. 

To conclude, there are no facts which unambiguously indicate that the 
southern element in the land gastropod fauna, which Europe and North 
America have in common, has followed the North Pacific route. The alterna- 
tive, a trans-Atlantic dispersal within the warm latitudes is, on the contrary, 
difficult to reject. But the nature of this dispersal yet seems enigmatic. 

The evidence is too scattered to support assumptions of any kind of land 
bridge. Rather it points to factors facilitating the passive dispersal, e.g., by the 
presence of interjacent islands. The idea that the Macaronesian archipelago 
once offered a connecting link can be rejected. When regarding their gastropod 
fauna it appears to be, as already emphasized, a decided derivative of the 
Early Tertiary Mediterranean and west European fauna. Probably this has a 
bearing on the occurrence of Jamdus too. 

The arguments for a connection between tropical Africa and South 
America, which have been brought forward, can be ignored here. They have 
no bearing on the actual problem, because the European land gastropod 
fauna always has been so profoundly isolated from the tropical African one. 
Besides, South America has relatively few groups in common with Africa, its 
main affinities being Australian and East Asiatic. 

However, these problems will not be discussed further here, as they refer to 
a faunal element of only peripheral importance in connection with the North 
Atlantic biota. It has seemed justified to draw attention to them because 
otherwise there would be the danger of overstressing the importance of the 



North Pacific route. The indications given by the land Gastropoda may be 
kept in mind when discussing other animal groups for which one must depend 
largely on the recent distribution, in the absence of fossil evidence. 

Finally, two further genera, Catinella and Coilostele, are common to 
Europe and America. Both have a phylogenetically ancient character. 
Catinella has a Pacific center and is represented by different subgenera in 
North America and northwestern Europe respectively. Coilostele occurs 
within a series of seemingly isolated areas from southernmost Spain to 
Timor; in addition there is, according to Pilsbry (1948, p. 1051), a quite 
isolated occurrence in Mexico. 

The distribution of these genera exhibits a pattern so decidedly relic that it 
seems hardly possible to discuss them in connection with any specific alterna- 
tive of dispersal. 


From Figs. 1 and 2 it is obvious that the Pleistocene caused no major 
changes in the composition of the gastropod fauna, neither in North America, 
nor in Europe. There occurred no mass extinction of species or radical 
dislocations of the gastropod faunas during the Glacials. Evidently the local 
Tertiary faunas largely survived within, or in relative proximity of the recent 
areas. The idea of radical changes in the distribution of the biotas during the 
Pleistocene, advocated especially by Deevey (1949), has a very moderate 
bearing for the terrestrial Gastropoda. However, in the North Atlantic 
region, in a narrow sense, their distributional pattern has been formed in the 
Quaternary, perhaps throughout in the Post-glacial time. 

Figure 3 illustrates the distribution of the gastropod species in the area. 
Excluded from the material are all settlements due to anthropochorous 
dispersal. Sometimes the origin is difficult to determine, and in such instances 
the occurrence within the territory in question has been indicated by hatching. 
The diagram is based on information given by Brooks and Brooks (1940), 
Ellis (1951), Forcart (1955), Lohmander (1938), Mandahl-Barth (1938), 
Oekland (1925), and Pilsbry (1939-48), in addition to which unpublished 
material in Scandinavian and American museums has been considered. 

The material has been divided into four categories, viz. endemic European, 
endemic American, and Holarctic species, plus representatives of the Holarctic 
form complexes, which seem to be specifically distinct on the two sides of the 
Atlantic. It may be admitted that our present knowledge does not always 
allow us to draw a sharp border between the last two categories. Furthermore, 
the question as to whether the genetic connection is continuous within the 
Euro-Siberian territory is left open. Endemic representatives for Holarctic 
genera are assigned to the European and the American groups respectively. 



I. Holarctic species. 

Vallonla oostsU (KUIler) . 
Tsrtlgo pvgii«e« (Erap.) ... 
Vsrtlgo alpestris Alder ... 
Vsllonla pulobella (UUller) 
Zoogvnstes harpa (Say) .... 
Zonltoldes nltldui (llUllerj 
Coohiloopa lubrlca (UUller) 
Coluse^la edentula (Drap.) 

Pupillc ::iu800njlD (L.) 

DarooaniB lar7« (HUller) .. 
Vertigo BOdesta (Say) 9. 1. 
Buconulua fulTui (MUller) . 


z ■ 


c 1 «" 


« s .2 
.. o 
o u ■? 
■otn S 

























Vullonia eoi 
Vertigo pye 
Vertigo alp 

ta (imller) 
a (Drap.) 

„ ,,_-rl« (Alder) 

ValloiUa pulchella (UUller) 
Zcogeneteo harpa (Say) 
Zonltoldee nltldus (KUIler) 
Cochllcopa lubrlca JKUllor) 
edent-ila (Drap.) 

Pup Ilia 




(Say) a. 

g (mller) 

II. Specifically distinct representatives for holarctic form complexes. 

Columella altlcola (Ingereoll) 
Diacui cronUiltel (Bimo.) 
HesoTltrea eleotrlsa (Sould) 
Pu&ctuBD mijTutlaelJiruB (Lea) 
Tltrlna lluplda Oould 5. 1. 


Endemic species. 

olunella columella (UartenB 
iBCua ruderatufl (Hartmann) 
Ltrea hassDonls (StrOm) 
m pygmaeua (Drap, ) 
la pellucHa (WlQler) 

IV. European: 

Carychius exile Lea 

Vertigo milluni (Gould) 

Vertigo nylanderi Sterki 

Vertigo bolleeiana (Horse) 

Gaetrocopta contracta (Say) 

Gaetrooopta tappaniana (Adajoa) 

Pupoidea allilabrlB (Adana) 

Strobilope labyrlnthioa (Say) 

Cisoua croijLh. catoklllenaia (Pilabry) 

PMlomycui oarollnlanua (Booo) 

Palllfera dorealia (BInney) 

Meaomphii inomatufl [Say) 

Ueeomphia cupreua (Bafineequa) 

Paravitrea multidentfa ta (Binney) 

SeaoTltrea binneyana (Voree) 

MyphyallBla rhoadsl (Pilabry) 

31yphyallnla indentata (Say) 

Baplotrema concarum (Say) 

Stenotrena fraternun (Say) 

Keeodon eayanua (Pilsbry) , 

Triodopoia tridentata (Say) 

Triodopaia denotata (F^r.) 

Trlodopele albolabrls (Say) 

TrlodopBiB dentifera (Btnney) 

Vertigo ovata Say 

Vertigo Tentri^oaa (Korae) 

Vertigo tridentata Wolf 

Gaetrocopta anBifera (Sayl 

Gaetrocopta pentodon (say) 

Succlnea bayardl Vanatta 

Anguleplra altemata (Say) 

Strlatura ferrea Borae 

Carychlun ezlguum (Say) 

Vertigo elatlor Sterki 

Vertigo perryi Sterki 

Vertigo gouldl (Binney) a. 1 

Vallonla aibula Sterki 

Planogyra aateriacua (Korea) 

Succlnea ovalio Say 

Succlnea a»ara Say 

Oxylona verrllll (Bland) 

Ozyloms decampl Tryon 

Hellcodlecua paralleluB (Say) 

Uawalia minuscula (Plnney) 

Strlatura exlgua IStljcpaon) 

Strlatura miliujo (Korae) 

Zonltoldes arboreua (Say) 

Cepaeo l.irtenais IKUller) 
LUoai marglnatua (KUIler) 
Oiyloma pfeifferi EoaaB. s. 1. 
Ciychiluji alllariua (Killer) 
Deroceraa agreete (L. ) 
Arlon Intermediua Rormand 
Arlon autfuacua (Lrap.) 
Vltrea cryetalllna (BUller) 
Vltrea contracta (Weeterlund) 
Baiea perreraa (L. ) 
Arlanta arbuetorulD (L. ) 
Arlon ater (L.) 
Acicula fuoca (Kontagu) 
Carychlum nlnlmum KUIler 
CarychiulD trldentatuo (Elsso) 
Atelia goodalll (P<r.) 
Cochllcopa lubrloella (Porro) 
Pyrafflldula rupestrlB (Drap.) 
Truncatelllna cyllndrlca \?tT.) 
Vertigo puallla KUIler 
Vertigo antivertlgo (Drap,) 
Vertigo eubotriata (Jeffraya) 
Vertigo lllljeborgl (Weaterlund) 
Vertigo ronnebyenaie (Weeterlund) 
Vertigo anguatlor Jeffreys 
Uiurla cyllndracea (Da CoBta) 
LeloBtyla an^lica (Wood) 
Vallonla eicentrlca Sterki 
Acanthinula aculeata (KUIler) 
Spermodea lamellata (Jeffreys) 
Ena obocura (KUIler) 
Catlnella arenaria ( Bouch. -Chant .) 
Succlnea oblonga Drap, 
Succlnea putria (1.) 
Diecua rotujldatu* (MUller) 
Arlon circumacrlptua Johnaton a 
Arlon hortenaie ?(t. 
Tlteovitrea petronella (Charp. ) 
Aegopinella jura (Alder) 
ketlnella nltidula (Drap.) 
CiychlluB cellariuB (KUIler) 
CxycMlufl helveticus (Blun) 
Zonltoideo eicaTatua (Alder) 
Llmax maxiiiruB L. 
Llnax clnereonlger Wolf 
Llmax tenellua Nllaaon 
Luconulue f'^Tua alderl (Gray) 
'.-ochlodlna laminate (Montaguj 

Iphlgena ilicatula (Drap.) 
ClBjsniu Mdentata (Str^im) 
Clbueilla dutla (Drap.) 
ClauBllla ;unlla Pfclffer 
c:l^uBiUa cruclata (studer) 
lirujybaena fruticuD (KUIler) 
Helicellu cujerata iKontagu) 
Cocl.llcella acuta (KUIler) 
tloniicha granulata (Alder) 
Trichiu Btrlolata (Pfeltfer) 
Trlchla hla[ila (L. ) 
tuotilhjlla Btrlgella (Drap.) 
Hellclgorj laploida (L. ) 
Cepaea nenoralls (L.) 

Fici. 3. Terrestrial gastropoiJ species in the North Atlantic region. 


Detailed comment on the diagram might be unnecessary. The extensive 
negative faunal character of Greenland is conspicuous. Obviously this is the 
consequence of combined cHmatic and dispersal obstacles. In principle the 
categories of land Gastropoda show a similar reaction, but most strikingly 
in the climatically more pretentious endemic groups. 

On the American side the Davis Strait forms a nearly absolute barrier. 
Numerically the decrease in species from Scotland and Norway to Iceland is 
as striking, but relatively it is less well marked. Then, among the Icelandic 
species only a minor fraction has spread to Greenland. Among these, two 
{Cepaea hortensis and Oxychilus alliarius) have been regarded by modern 
authors as probably introduced to Greenland. 

All definitely indigenous species in Iceland and Greenland must be regarded 
as well adapted for passive dispersal. For two species in Iceland, Arion ater 
and Arianta arbustorum, a spontaneous passive dispersal seems improbable. 
However, Lohmander (1938), the author who has most thoroughly penetrated 
the problem of Iceland's gastropod fauna, is of the opinion that for these very 
species anthropochorous dispersal in historical time is a probable alternative. 
A random, passive dispersal is also indicated by the absence of certain species, 
which, as Lohmander assumes, are well able to endure Icelandic conditions. 
For the idea of a Pleistocene landbridge to Iceland, and subsequent refugia, 
the land Gastropoda offer no positive evidence. The nearest known peri- 
glacial deposit of non-marine MoUusca is that of Lea Valley, England 
(Kennard and Woodward, 1912), the age of which has recently been dated to 
28,000 ± 1500 years, or early main Wurm (Godwin and WiUis, 1960). It 
represents a poor fauna, but some of the species are definitely not Arctic 
(Piipilla muscorum, various Succinea species, also some of the freshwater 
forms). If existing, an Icelandic refugial fauna most probably would have 
been still more depauperated. The presence of some remarkably demanding 
forms (the Vitrea species, and Arion intermedius) suggests that the Post- 
glacial Hypsithermal was of outstanding importance for the establishment 
of the gastropod fauna in Iceland. 

Against this general background concerning the distribution of land 
Gastropoda in the North Atlantic region, the presence of the European 
Cepaea hortensis and Limax marginatus in northeastern North America is 
very puzzling. American authors especially have been inclined to regard them 
as indigenous. L. marginatus seems to be limited to a very narrow strip in 
Newfoundland, whereas C hortensis has a rather extensive and relatively 
continuous distribution from Newfoundland to New England. The conclusive 
arguments have been concentrated in the discussion on the latter species. 

The opinion that C. hortensis has an indigenous distribution in America 
in based upon observations by different workers who have studied its occur- 
rence in natural habitats and pointed out that this species, contrary to 
definitely anthropochorous species, seems to have a continuous, "mature" 


distribution; to some extent this conclusion is also based on subfossil records 
(cf. Pilsbry, 1939, p. 8, and references given by him). Lindroth (1957, p. 234, 
etc.), however, finds the evidence weak and incomplete and suggests an 
entirely anthropochorous origin. A supposed Pleistocene fossil thus may 
equally well be of 19th century origin. 

There is no doubt that part of the population is of a recent anthropochorous 
nature. On the other hand, radiocarbon dating shows that the species is 
definitely of pre-Columbian age in Nova Scotia (Clarke and Erskine, 1961 : 
700 ± 225 years; later dating 600 ± 45 years B. P., according to a personal 
communication by Clarke). This may be thought to speak in favor of the old 
idea of dispersal by aid of the Vikings. That idea, however, is nothing more 
than an entirely unproven hypothesis and, furthermore, there exists no 
positive evidence that the Vikings ever visited Nova Scotia. The actual 
shells of C. hortensis were found in the camps of the Micmac Indians. When 
these matters are considered the statement of their pre-Columbian age 
evidently is in favor of an old, indigenous occurrence. 

With regard to the possibility of passive dispersal, an indigenous American 
occurrence is not definitely excluded. It is not difficult to raise various hypo- 
theses, regarding the recent area as relic, due to dispersal via Greenland- 
Labrador during a Post-glacial period, or even an Interglacial one. But owing 
to lack of really conclusive evidence it is very unproductive. The only way to 
solve the problem of C hortensis — which would be of certain interest owing 
to its crucial significance — is new, careful field research, by which special 
regard must be given to the existing rich experience of the ecological occurrence 
of C. hortensis in Europe. 

For land gastropods a solution of problems of this kind evidently is 
attainable. Owing to the slow rate of estabhshment, also when highly favored 
by passive dispersal, introduced gastropods very long maintain an immature 
pattern of distribution. Practically always, it seems possible to give a definite 
answer to the question whether or not a species is introduced into an area, if 
only its distribution on an ecological basis has been accurately clarified. 

A field survey would result in a more correct picture )of several further 
species ( Vallonia pulchella and costata, Vertigo pygmaea and alpestris, also the 
above-mentioned Limax marginatus) whose faunistic state in northeastern 
America is yet obscure. Of course, the distribution and history of the endemic 
American gastropod fauna would also be elucidated. 

Surprisingly, the interchange of endemic gastropod species in the North 
Pacific territories seems to be very limited. The American Zonitoides arbor ens 
and Hawaiia minuscula, according to Licharev and Rammelmeyer (1952), 
have an inconsiderable spontaneous distribution in the Far East. The prepon- 
derantly Siberian Succinea strigata reaches Arctic North America.* It is not 

* Pilsbry (1948, p. 811) cites S. strigata from Igaliko Fjord, Greenland. However, the 

actual sample, by revision, has proved to belong to Succinea pfeifferi subsp. groenlamlica. 


excluded that forthcoming faunistic and taxonomical research will unveil 
some further examples within the Holarctic genera, but it will scarcely change 
much in the picture. The favorable situation for dispersal, due to the narrow 
separation of the land masses and the late direct connection, apparently has 
been of limited importance for the Gastropoda when the ecological conditions, 
mainly climatic, are preventive. Obviously this has a bearing on the situation 
in the North Atlantic region too. 

It must be admitted that the discussion which can be carried out from 
present knowledge concerning the land Gastropoda within the Arctic and 
northerly Boreal realm is of rather general character. Few positive conclusions 
are possible. The key to several problems, which also concern the North 
Atlantic region, may lie within the immense territories in northern Asia and 
North America from which extremely little is known about the land Gastro- 
poda (as well as most other animal groups). Data are needed equally concern- 
ing distributional patterns, taxonomy on ± species level, and fossil occurrence. 
The fact that the Pleistocene refugia, most firmly supported by geological 
evidence, existed in the Bering Strait region may be especially considered in 
this connection. Also, for an understanding of prospective North Atlantic 
refugia, data from those of the Pacific may be beneficial. 

It is a matter of course that the detailed field research on the first hand 
must be concentrated on geographical sectors and habitats which are thought 
to be especially profitable, as well as on species of outstanding significance. 
But such research can attain its full value only against the background of a 
fairly homogeneous knowledge about the whole territory, also including the 
extensive "less interesting" areas. Strictly speaking, no area in which biota 
exist is uninteresting to the biologist. It depends upon the problem to which 
the approach refers. 

With the resources of transport available today there are scarcely any 
technical obstacles to bringing forth an acceptably homogeneous fund of 
data from the Arctic and northernmost Boreal regions. Thereby it would be 
possible to adjust a good deal of the obvious imbalance in recent discussions 
of zoogeography and faunal history. 


Baker, F. C. (1920). The life of the Pleistocene or Glacial period. Bull. Univ. III. 17, 41, 

Brooks, S. T. and Brooks, B. W. (1940). Geographical distribution of the recent Mollusca 

of Newfoundland. Ann. Carneg. Mils. 28, 53, 53-75. 
Clarke, A. H., and Erskine, J. S. (1961). Pre-Columbian Littovim littoiea in Nova Scotia. 

Science 134, 393-394. 
Darlington, P. J., Jr. (1957). Zoogeography : the geographical distribution of animals. 

New York. 
Davies, a. M. (1934). Tertiary Faunas. Vol. II. Guildford. 
Deevey, E. S., Jr. (1949). Biogeography of the Pleistocene. I. Bull. Geol. Soc. Amer. 60, 9, 



Ehrmann, P. (1914). Giimdziige einer Entwickhmsgeschichte der Tierwelt Deiilschlands. 

Lit.-Gesellsch. Neue Bahnen. Leipzig. 
Ellis, A. E. (1951). Census of the distribution of British non-marine Mollusca./. Conch. 

23, 6-7, 171-244. 
Fischer, A. G. (1960). Latitudinal variation in organic diversity. Evolution 14, 1, 64-81. 
FoRCART, L. (1955). Die nordischen Arten der Gattung Vitrina. Arch. Molhiskenk. 84, 

4-6, 155-166. 
Godwin, H., and Willis, E. H. (1960). Cambridge University natural radiocarbon measure- 
ments II. Anier. J. Sc, Radiocarbon Siippl. 2, 62-72. 
Henderson, J. (1931). Moiluscan provinces in the western United States. Univ. Colo. Stud. 

18, 4, 177-186. 
Henderson, J. (1935). Fossil non-marine Mollusca of North America. Spec. Pap. Geol. Soc. 

Anier. 3. Baltimore. 
Hibbard, C. W., and Taylor, D. W. (1960). Two late Pleistocene faunas from South- 
western Kansas. Contrib. Mus. Geol. Univ. Mich. 16, 1, 1-223. 
HOPPE, G. (1959). Nagra kritiska kommentarer till diskussionen om isfria refugier. Stat. 

Naturvet. Forskningsrads Arsbok. 12, 123-134. 
Hubendick, B. (1953). The relationships of the East African Vitrinae with notes on the 

taxonomy of Vitrina. Ark. Zool. Ser. 2, 6, 51, 83-96. 
Kennard, a. S., and Woodward, B. B. (1912). Appendix III.— The Mollusca. In: Warren, 

S. H.: On a Late Glacial stage in the valley of the River Lea, subsequent to the epoch 

of River-Drift Man. Quart. J. Geol. Soc. London. 68, 2, 234-240. 
Kew, H. W. (1893). The dispersal of shells. Internat. Scient. Ser. 75. London. 

Vidensk. Akad. 1. Mat.-Naturv. Kl. 8. 
LiCHAREV, I. M., and Rammelmeyer, E. S. (1952). Nazemnyje Molljuski Fanny SSSR. Acad. 

Scient. SSSR. 43. Leningrad. 
LiNDROTH, C. H. (1957). The Fainial Connections between Europe and North America. 

Lohmander, H. (1938). Landmollusken aus Island gesammelt von Dr. Carl H. Lindroth 

(1929). Goteborg Vetensk-Samh. Handl. Ser. B. 6, 2, 3-52. 
Mandahl-Barth, G. (1938). Land and freshwater Mollusca. The Zoology of Iceland A, 651, 

Moreau, R. E. (1952). Africa since the Mesozoic: with particular reference to certain 

biological problems. Proc. Zool. Soc. 121, 869-913. 
Oekland, F. (1925). Die Verbreitung der Landgastropoden Norwegens. 5A/-. Norske 

Vidensk- Akad. I. Mat. Naturv Kl. 8. 

Pilsbry, H. a. (1939-48). Land Mollusca of North America (north of Mexico). Acad. Nat. 

Sci Monogr. 3, I-Il. Philadelphia. 
Quick, H. E. (1952). Emigrant British snails. Proc. Malac. Soc. 29, 5, 181-189. 
Reid, E. M. (1935). British Floras Antecedent to the Great Ice Age. In: Discussion on the 

Origin and relationship of the British Flora. Proc. Roy. Soc. London. Ser. B. 118, 

808, 197-202. 
Simpson, G. G. (1940). Holarctic Mammalian faunas and continental relationships during 

the Cenozoic. Bull. Geol. Soc. Amer. 58, 7, 613-688. 
Sparks, B. W. (1957). The Tale Gravel near Thriplow Cambridgeshire. Geol. Mag. 94, 3, 

Szafer, W. (1954). Pliocene flora from the vicinity of Czorstyn (West Carpathians) and its 

relationship to the Pleistocene. Prace Inst. Geol. 11, 1-238. 
Taylor, D. W. (1954). Nonmarine Mollusks from the Upper Miocene Barstow Forma- 
tion, California. Prof Pap. U.S. Geol. Siov. No. 254-C. Washington. 
Walden, H. W. (1961). On the variation, nomenclature, distribution and taxonomical 

position of Lima.x (Lehinannia) valentianus Ferussac (Gastropoda, Pulmonata). Ark. 

Zool. Ser. 2, 15, 3, 71-95. 
Wenz, W. (1961-2). Gastropoda. Teil 1, Allgemeiner Teil/Prosobranchia. Handbuch der 

Paldozoologie 6, 1. 2nd edition. Berlin. 
Wenz, W., and Zilch, A. (1959-60) Gastropoda. Teil 2, Euthyneura. Ibid. 6, 2. 




ElLIF Dahl 

Department of Botany, Agricultural College, Vollebekk, Norway 

It is a striking fact that the Arctic-Alpine floras on both sides of the North 
Atlantic Ocean are very similar while this similarity does not extend to more 
Temperate floras. A Scandinavian botanist feels quite at home in south 
Greenland, whereas a French botanist working in areas of eastern America 
with a cHmate corresponding to his homeland probably would recognize 
most of the native genera but a few of the indigenous species. 

Many Arctic-Sub- Arctic species grow on both shores of the North Atlantic 
Ocean, some being confined to that area while others have large gaps in their 
distribution in Siberia or in western America. Such plants are said to have 
an Amphi-Atlantic distribution pattern which has recently been beautifully 
mapped by Hulten (1958). 

From Scandinavia to Scotland, the Faeroes, Iceland, east Greenland, 
west Greenland, and into Canada, there seems to be a gradual transition 
from a predominantly European flora in the east to a predominantly American 
flora in the west. There are also very close phytogeographic connections 
between Spitsbergen and northeast Greenland. This presents a problem since 
it suggests that a direct migration has taken place. But the area is split apart by 
long stretches of water and it is commonly supposed that migration of plants 
and animals does not easily take place across large sea areas. This problem 
has been discussed by numerous workers ; from earlier times might be mentioned 
Hooker, Darwin, Warming, Nathorst, Blytt, and Ostenfeld and more recently 
Fernald, Lynge, Nordhagen, Hulten, Nannfeldt, Bocher, and Love and Love. 
Lindroth (1957) has recently treated the zoological and also some botanical 
aspects of the problem in a very inspiring manner. 



Many plants and animals have been carried across oceans by human 
traflic. Distribution patterns resulting from this transport should be excluded 
from consideration in discussing the old phytogeographic connections between 



Europe and America. This brings up the question of how to discriminate 
between native and introduced plants. 

This is not a serious question as far as recent dispersal is concerned where 
often, if not the actual introduction, the subsequent phase of dispersal in the 
new country has been observed. It is much more difficult if the introduction 
took place long before any systematic botanical investigations were carried 

In general more species seem to have been spread from the east towards the 
west than vice versa and the problem is not very difficult as far as American 
species introduced to Europe is concerned. 

The question of the impact on the flora of the Norse colonists in Green- 
land and Iceland has been discussed by many authors. Ostenfeld (1926) 
believed that as much as 14 per cent of the Greenland flora had been introduced 
this way whereas Porsild (1932) thinks this figure too high, probably not more 
than 5 per cent. Modern intensive studies in the infra-specific races throw 
hght on these questions since among the anthropochorous species (species 
following human occupation) some races may be anthropochorous and others 
not. Also palynological research is of help ; e.g. Angelica archangelica which 
has been suspected of being introduced to Greenland by Norse colonists 
has been identified in pollen deposits considerably older than the Norse 
colonization (Iversen, 1953). But still, several doubtful cases are known. 
I have here accepted the evaluation given by Bocher, Holmen and Jakobsen 
(1957) regarding Greenland, and Love and Love (1956) regarding Iceland. 

In Newfoundland and adjacent areas a peculiar isolated European element 
of plants occurs which Fernald (1929) considered a relict, indigenous element. 
However, Lindroth (1957) has pointed out that many of the species were 
probably introduced during the period of fishing as early as in the sixteenth 
century by British, French, and Iberian fishing boats carrying ballast west- 
wards and throwing it ashore and bringing fish back. For this reason, all 
plants apparently native only in the Newfoundland-Nova Scotia area in 
America, but common somewhere along the coast from south England to 
Portugal and taxonomically indistinguishable from the European populations, 
have been considered here as introduced by man during the early fishing 
period. This makes a list of about 60 taxa (species and taxonomic entities of a 
lower rank than species) to be excluded from the number of possible Amphi- 
Atlantic plants. 




A western Amphi-Atlantic element in the flora of Europe can be recognized, 
consisting of taxa found in eastern America and not occurring east of the 
River Lena, the Carpathians and the Balkans in the Old World. Altogether 83 




taxa belong here and Fig. 1 gives the number of taxa in this element in the 
different parts of Europe. The highest number is found in north and central 
Scandinavia, a fair number also in Scotland, whereas the number decreases 
eastwards and southwards, in the south they are mainly confined to the 

Within the western Amphi-Atlantic element sub-elements with more res- 
tricted distribution can be discerned. Four taxa grow only in Spitsbergen east 
of the Atlantic Ocean. Another sub-element grows in Scandinavia, a few 
species of it also in Novaya Zemlya and Scotland, but it does not reach Ural 
or the Alps. This element has been called "West Arctic" by Blytt (1876) and 
34 taxa belong to it. Finally there are five species restricted to the British 
Isles east of the Atlantic Ocean. 

Corresponding to the western Amphi-Atlantic element in Europe, there is 
an eastern Amphi-Atlantic element in America. Its species are native in 
Europe and also in eastern America, but not farther west than the Great 
Plains and a fine east of the Rocky Mountains from the Great Plains to the 
mouth of the Mackenzie River. Figure 2 gives the number of members 
belonging to this element in America. It will be seen that the highest number 
is found in south Greenland with a fair number also in Newfoundland, and 
that the number drops westwards and southwards. The number is also rela- 
tively low farther north, but this is not pronounced if the element is taken as a 
percentage of the native flora. 

Within the eastern Amphi-Atlantic element in America sub-elements with 
more restricted distribution can be recognized. Eight taxa are confined to 
northeast Greenland on the west side of the Atlantic Ocean and an additional 
30 taxa do not reach farther west than Greenland. Twelve taxa have a very 
restricted distribution along the Atlantic seaboard. The three sub-elements 
mentioned form an American counterpart of the West Arctic element in 
Europe (see Lindroth, 1957, p. 237). Around the Gulf of St. Lawrence and in 
Newfoundland a European element of five taxa form a counterpart to the 
American element in the British Isles. 

From the data given it will be evident that the Amphi-Atlantic element 
essentially is Arctic-Sub-Arctic in character and very few plants from the 
temperate regions exhibit an Amphi-Atlantic distribution pattern. 


Different hypotheses may be invoked to explain the present-day distribu- 
tion patterns of the Amphi-Atlantic plants. One main hypothesis suggests that 
the Amphi-Atlantic species once migrated across the North Atlantic Ocean, 
either by long-distance dispersal or across a former land connection between 
America and Europe. The other main hypothesis suggests that they once 
migrated across the Bering Strait between Asia and America and subsequently 



died out in areas other than those which they now occupy. The long-distance 
dispersal hypothesis will be considered first. 

Plants can be spread over long distances in different ways. Anemochorous 
species are carried by wind and have very light and small seeds or spores 
(less than 0.2 mm) or have wings or hairs attached to seeds or fruits which 

Fig. 2. Number of taxa from the eastern Amphi-Atlantic element in the floras of 

diff'erent parts of America. Further explanation in the text. (Base map Denoyer- 

Geppert, by permission). 

reduce the settling velocity in air. Zoochorous species aretransported by animals 
and have fruits or seeds with either edible, fleshy parts or hooks which attach 
them to animals passing by. Or, they may be spread by sea currents; it is then 
important that the seeds or fruits float well in sea water. Unfortunately, not 
enough data are available for classifying all the plants concerned according to 
buoyancy of seeds and fruits. But it seems evident that halophilous plants 
growing on the sea shores in haline environment would have a particular 
advantage of dispersal in this way since the fruits or seeds would tend to 
land in an environment favorable for further growth and reproduction. 



Lastly. I have considered limnic species growing in fresh water as adapted to 
long-distance dispersal since there is some evidence that such plants are 
easily spread by water birds. 

If the Amphi-Atlantic plants attained their present area by long-distance 
dispersal one would expect plants with adaptations to long-distance dispersal to 
be more numerous within the Amphi-Atlantic elements as compared to other 
elements which presumably immigrated across more or less continuous 
land connections. Table 1 gives a breakdown of the Scandinavian Arctic- 
Alpine flora according to adaptations to distance long-dispersal and phyto- 
geographic elements. A two-way grouping is obtained, a so-called contingency 

Table 1 

Comparison of the Arctic-Alpine Elements in the Flora of Fenno-Scandia as to 
Phytogeographic Elements and Adaptations to Long-distance Dispersal 

Taraxacum and Hieraciiim excluded from the enumerations. 
[Upper figure == observed, lower figure (in brackets) = statistically expected number of 

species per element.] 

No adap- 
to long- 






West Arctic 








Other western 








Other elements 







i 78.4% 









table, and by standard methods it can be tested whether there is any correla- 
tion or contingency between the subdivision in phytogeographic elements and 
dispersal groups. The figures expected under assumption of no correlation or 
contingency are given in brackets. 

In general the numbers observed agree quite well with those expected, 
assuming that no contingency and no statistically significant discrepancies 
occur. There is a slight under-representation of types adapted to long-distance 
dispersal in the western Amphi-Atlantic element and this observation can be 
used as an argument against the hypothesis that they immigrated by long- 
distance dispersal. 



Table 2 gives a similar break-down of the flora of Iceland. In the Amphi- 
Atlantic group are included species belonging either to the western Amphi- 
Atlantic element in Europe or the eastern Amphi- Atlantic element in America. 
Some European species reach their westernmost and some American species 
their easternmost Umit in Iceland. 

Table 2 

Comparison of the Different Elements in the Flora of Iceland (based on Love, 1945; 
Love and Love, 1956) as to Adaptations to Long-distance Dispersal. 

Taraxacum and Hieraciiim excluded from the enumerations. 

[Upper figure = observed, lower figure (in brackets) = statistically expected number of 

species per element.] 

No adap- 
to long- 
















European species 









American species 








Other species 
















The European character of the Icelandic flora is clearly evident ; there are 
about 12 times as many European as American species. The same facts 
apply to the Icelandic fauna (Lindroth, 1957). There is some over-representa- 
tion of hmnic species in the other flora elements, and a tendency to over- 
representation of types adapted to long-distance dispersal among the American 
species, but the numbers are too low to yield any statistical significance. 
However, considering also taxa of lower rank than species, 16 American taxa 
are found in Iceland of which only 4 are not adapted to long-distance dis- 
persal. This is certainly significant. Similarly, Lindroth (1957) found that the 
American animals in Iceland are adapted to long-distance dispersal. In the 
Amphi-Atlantic and European elements in the Icelandic flora there is an 
under-representation of types adapted to long-distance dispersal. Thus, there 
is no evidence that the Amphi-Atlantic and European species reached 



Table 3 

Comparison of the Different Elements in the Floras of Greenland (based on Bocher, 
HoLMEN and Jakobsen, 1957); Newfoundland and Labrador (Mainly based on 
Rouleau, 1956) and Gaspe (Mainly based on Scoggan, 1950) as to Adaptations to 

Long-distance Dispersal. 

Polymorphic apomictic groups {Taraxacum and Hieiaciuin in Greenland, Oenothera, 

Crataegus and Rubus subg. Eubatus in other areas) excluded from the enumerations. 
[Upper figure = observed, lower figure (in brackets) = statistically expected number of 

species per element.] 

No adap- 
to long- 











Other elements 



























Other elements 


























Other elements 













95.1 % 











Other elements 

























Iceland by long-distance dispersal, whereas there is evidence that this was the 
case by the immigration of the American element. 

Table 3 gives the breakdown of the floras of Greenland (Bocher, Holmen 
and Jakobsen, 1957), Labrador and Newfoundland (Rouleau, 1956) as 
well as Gaspe Peninsula (Scoggan, 1950) in a similar manner. The eastern 
Amphi-Atlantic elements in all these areas are consistently under-represented 
in types adapted to long-distance dispersal; in several instances the differences 
between observed and expected numbers are statistically significant. The 
percentage of Amphi-Atlantic species in the floras decreases from north 
towards south. The percentage of anemochorous and zoochorous species 
increases towards the south while the percentage of species not adapted to 
long-distance dispersal decreases. This trend also is significant. In part, this may 
be the reason why the Amphi-Atlantic element, which is essentially an Arctic- 
Sub-Arctic element, is under-represented in types adapted to long-distance 
dispersal. But there is no evidence to suggest that the eastern Amphi-Atlantic 
species reached their present stations by long-distance dispersal. 

There appears to exist a discrepancy between the zoogeographical and 
phytogeographical observations regarding Davis Strait as a biogeographic 
barrier. According to Lindroth (1957) the Greenland fauna consists of about 
one half Holarctic species and about one quarter each of Nearctic and 
Palearctic species. The Palearctic influence is particularly evident in the 
soil-bound fauna which is not easily spread across the sea, while most of the 
Nearctic fauna can be more easily dispersed. The fauna of BaflSn Island, 
however, has an almost purely Nearctic character. Lindroth (1960) has also 
performed an analysis of the botanical aspects of the problem. 

I have examined the data presented by Lindroth from a statistical point of 
view and found their significance doubtful. However, Lindroth treated the 
flora of Greenland as a whole. Smith Sound between Greenland and Ellesmere 
Island is hardly an important barrier for Arctic plants. If Davis Strait has been 
a barrier this should be most clearly evident among more Low Arctic plants. 
It should also be remembered that during the Post-glacial climatic optimum 
(the Hypsithermal) many plants could grow in the Smith Sound area, but 
they are now found only farther south. For this reason, I have made an 
analysis of the Low Arctic element, here defined as species with a northern 
limit in Greenland at or south of the Nugssuak Peninsula and on Baffin 
Island at or south of the Pangnirtung Peninsula. The result of this analysis is 
presented in Table 4. 

From the table it will be seen that no enrichment of types adapted to 
long-distance dispersal is evident among the eastern Amphi-Atlantic taxa 
penetrating into America. However, among the western plants penetrating 
into Greenland and Iceland there is an over-representation of types adapted to 
long-distance dispersal. Using the entire flora of Greenland as a basis for 
comparison the difference becomes clearly significant. Taking the entire flora 



Table 4 

Comparison of Low-Arctic Elements in the Flora of Greenland and Eastern North 
America as to Adaptations to Long-distance Dispersal. 

Further explanation in text. 

No adap- 
to long- 












Eastern Amphi- 
Atlantic taxa: 

Reaching east 



Reaching west 







With limited 





area in eastern 

With wider area 
but not west of 
Hudson Bay 




Reaching west of 
Hudson Bay 


















Western taxa reaching 
Greenland and Iceland 














of Labrador as a basis for comparison results in some over-representation of 
types adapted to long-distance dispersal, but the difference is not statistically 
significant. This is, however, probably due to the higher representation of 
types adapted to long-distance dispersal within the more temperate elements 
in the flora of Labrador. 

From the data presented the following conclusions can be drawn: 

1. The Arctic-Sub-Arctic flora in general and the Amphi-Atlantic elements 
in particular are under-represented in types adapted to long-distance 


2. There is evidence of over-representation of types adapted to long- 
distance dispersal within western Low Arctic elements penetrating to 
Greenland and Iceland. 

From this it can be concluded either that the adaptations recognized as 
favoring long-distance dispersal are of no biological significance, or that the 
problem of Amphi-Atlantic plant distribution is not a matter of long- 
distance dispersal. 

The first conclusion is difficult to accept. As to the anemochorous species 
the reader is referred to the work by Wilhelm Schmidt and others (cf. Geiger, 
1961, p. 50 etc.) regarding probable travelling distances of seeds and spores as 
a function of turbulence and settling velocity in air. These considerations are 
borne out by finds of numerous far-travelled spores and pollen. Also, the 
re-colonization of the flora of the island of Krakatoa shows that species 
apparently adapted to long-distance dispersal had a considerable advantage 
in colonizing the island after its flora was destroyed by a volcanic eruption in 
1883 (cf. Dahl, 1959). Thus, it is concluded that, whatever is the explanation of 
the Amphi-Atlantic distribution pattern, it is not a matter of long-distance 



Only two hypotheses now remain to be considered in order to explain the 
distribution of the Amphi-Atlantic plants. One is the hypothesis of a former 
migration across the Bering Sea, and the other concerns migration across a 
land connection between Europe and America. 

The first hypothesis involves very extensive plant dispersals with the 
result that populations on both sides of the Atlantic Ocean must have been 
isolated genetically for a long time. Especially in polymorphic groups of 
species, and in taxa of lower rank than species, one would expect to find a 
diff'erentiation of the populations on both sides of the North Atlantic Ocean. 
On the contrary, if plants had dispersed directly across the North Atlantic 
Ocean one would expect to find closely related taxa on both sides of the 
Ocean differing from the populations farther east in Asia, in the Pacific area, 
and in western America. 

The Eastern Amphi-Atlantic element in North America can be sub-divided 
into the following groups, mainly based upon information from Hulten (1958): 

1. Amphi-Atlantic taxa of lower rank than species with 
vicariants in the Pacific area (Alaska and easternmost 

Asia) 13 taxa 

2. Amphi-Atlantic taxa of lower rank than species with 
vicariant taxa elsewhere in Europe, Asia, or America but 

absent in the Pacific area 8 taxa 


3. Amphi-Atlantic species with vicariant taxa of lower rank 

in eastern America and western Europe 10 species 

4. Species of polymorphic groups with vicariants in the 

Pacific area 22 species 

5. Species of polymorphic groups with vicariant species else- 
where in Europe, Asia and America but not in the 

Pacific area 25 species 

6. Other Amphi-Atlantic species 33 species 

Of the groups listed above, only Group 3 can be said to support the 
hypothesis that the plants dispersed across the Bering Sea, whereas Groups 
1 and 2 and, to some extent, also Groups 4 and 5 contradict the hypo- 
thesis. It will be seen that the data clearly suggest a direct migration between 
Europe and America, 

Within Group 3, supporting the hypothesis of dispersal across the Bering 
Sea, are seven temperate plants and only three Arctic-Sub-Arctic ones. 
As a comparison, in Group 1, supporting the hypothesis of a direct Atlantic 
dispersal, all except one species (the sea-shore plant Ligusticum scoticum) 
have an Arctic-Sub-Arctic type of distribution. Thus, the hypothesis of 
migration across the Bering Sea seems more likely concerning temperate 
than Arctic-Sub-Arctic plants. 

The outline given above can be exemplified by numerous individual 
examples. Here, I shall consider only one example, where more detailed and 
advanced information is available. 

In the Papaver radicatum complex studied by Knaben (1959a, b) one 
species, P. lapponicum, is Amphi-Atlantic, as far as is known. In Europe it 
grows in northern Scandinavia as far east as Kola, whereas in North America 
it has a wide area including most of Greenland, northern Labrador and the 
Canadian Arctic Archipelago as far west as Banks Island. Knaben {he. cit.) 
recognizes one subspecies in northern Norway and three different subspecies 
west of the Atlantic Ocean. The morphological differences between the 
American subspecies seem to be about as great as the differences between the 
Scandinavian and the American subspecies. 

More quantitative information on the degree of genetic differentiation is 
available through crossing and subsequent cytological analysis. When 
mutations take place the pairing of corresponding chromosomes in meiosis 
of hybrids is often impaired and the number of bivalents, thus formed, in the 
meiosis of such hybrids seems to be a fair measure of the degree of genetic 
differentiation. Knaben has carried out a number of intra-racial and inter- 
racial crosses, and Table 5 is based on information contained in her work. 
The maximum number of bivalents is 28. The number of bivalents in intra- 
racial crosses of ssp. occidentale is fairly low, suggesting that it is genetically 
heterogeneous and may one day be split up into more subspecies. But the 

Table 5 


Number of Bivalents in Intraracial and Interracial Crosses in Papaver Lapponicum 

(based on Knaben, 1959). 

The mean number of bivalents with standard error of the mean given together with number 
of PMC's counted and number of crosses. 





ssp. scandinavicum 
(North Norway) 

25.2 ± 0.35 
17, 3x 

16.5 = 0.74 
21, 5x 

14.5 ± 1.55 

7, 2x 

ssp. occidentale 
Arctic Canada) 

19.1 ± 0.79 
12, 3x 

15.8 ± 1.07 
13, 4x 

ssp. porsildii 
Arctic Canada) 

26.0 ± 0.82 
6, 2x 

22.3 ± 0.53 
15, 5x 

ssp. labradoricum 
(South Greenland- 

important point is that the genetic differences between ssp. scandinavicum 
and the American ssp. occidentale and labradoricum is about the same as the 
genetic difference between ssp. occidentale and ssp. labradoricum themselves. 
\^ Papaver lapponicum once dispersed in both directions from the Pacific area, 
the isolation between ssp. occidentale and labradoricum must be of a younger 
date than the isolation between these two subspecies and ssp. scandinavicum. 
This is not borne out by the observations made. 

As pointed out above, there is a fairly gradual transition in the flora from 
Scandinavia across Scotland to Iceland and Greenland, and this applies also 
to the fauna. This general feature can hardly be explained by the hypothesis 
that the plants came from the Pacific area. The strength of the hypothesis is 
that it avoids creating a new land-bridge, but in order to explain, for instance, 
the European element in the flora of Iceland, it seems necessary to postulate 
some land connection across the longest stretch of water now separating 
America and Europe. 


By process of elimination, the conclusion is reached that the only hypothesis 
capable of explaining the Amphi-Atlantic biota is that of a former land con- 
nection across the North Atlantic Ocean. Since there is evidence that long- 
distance dispersal has been of importance in the immigration of the western 


elements into Iceland and Greenland, it is concluded that the land connection 
was broken first between Iceland and America, and the strong European 
influence in the biota of Iceland corroborates this. Consequently, the western 
Low Arctic elements in Greenland and Iceland must be yoanger than the 
Amphi-Atlantic and eastern elements and this is borne out also by the endemics 
in Iceland, most of which have many close relatives on the European side of 
the Ocean but only a few on the American side. It is not denied that some 
species originally might have dispersed across the Bering Sea or have attained 
their distribution by long-distance dispersal, but for the majority of the biota 
this explanation does not seem feasible. 

It has been emphasized that the Amphi-Atlantic plants are Arctic -Sub- 
Arctic, whereas very few Temperate plants have this distribution pattern. 
This implies that the climate on our hypothetical land connection must 
have been favorable to Arctic-Sub-Arctic plants but not to Temperate 
plants. The same fact is brought out by an analysis of altitudinal limits in the 
Scandinavian flora; very few species unable to grow in the birch belt occur 
west of the Atlantic Ocean. In this element there is a high over-representation 
of types adapted to long-distance dispersal (Dahl, 1959). It is concluded that 
the climate on the land-bridge was about the same as is found in the birch 
belt of Scandinavia or of Iceland today. 

This permits us to propose an age for the land connection. In Mid-Tertiary 
times the climate was warmer than at present and only in later times, during 
Pliocene and Pleistocene, do we find climatic conditions comparable to the 
present-day climate. It is therefore concluded that the connection existed at 
least s late as the Pliocene. 


The biogeographic data presented suggest the existence of a former land 
connection between Europe and America as late as in the Pliocene. However, 
one cannot conclude from the distribution that the land-bridge existed and 
then explain the distribution pattern by the means of the land-bridge. This 
becomes circular reasoning; thus independent support in the form of geological 
observations is necessary in order to establish the existence of a land connec- 

It is tempting to invoke the Wegener hypothesis of continental drift to 
explain the features, and the data presented above may be taken as support 
for the hypothesis of continental drift. However, this is a very controversial 
matter and at present it seems unwise to build too much on the Wegener 

There are, however, other indications suggesting the existence of a land 
connection between Europe and America during Tertiary times. In the Eocene 


large eruptions took place in an area extending from Scotland to east Green- 
land and a huge basalt plateau was formed. Fossiliferous deposits between the 
basalt layers permit its dating. Since no deposits of Mid-Tertiary age are 
known from this region, it is believed that the plateau existed as a land area. 
The basalt plateau was later broken up by renewed volcanic activity in 
Pliocene or Pleistocene times and only some remnants are preserved. 

Barth (1941, p. 8) has given this account of the geologic history of Iceland; 

Across Iceland, from the south to the north, there is a broad belt with mountains and 
plateaus with sharp peaks and jagged crests. The rest of the country to the east and 
to the west consists of flat-topped mountains made up of layers of solidified lava which 
came up from the interior of the Earth abouc 50 million years ago, long before man 
appeared on Earth. The material making up these mountains was molten lava, stream 
after stream gushed out, gave ofl" the heat to the Cosmos, and became solid rock; 
layer upon layer were formed and huge flat plateau mountains built up. In this way the 
Old Iceland was born, a huge stone plateau which filled almost the whole of the North 
Atlantic Ocean; Old Iceland was much larger than the Iceland we know today; it 
stretched northwards and eastwards with a land connection to both Jan Mayen and the 

After this, all became quiet for a while; the eruptions stopped and the lava retreated 
slowly into the interior cf the Earth. There it remained for millions of years, until it 
once again came back up to the surface. As the spring flood breaks the ice so that the 
floes are broken and raised on end, thus were the plateau mountains of Old Iceland 
broken up by new volcanic activity and the molten lava pressing upwards; from 
innumerable cracks and crannies the lava again flowed out over the surface. The 
foundations of the stone plateau gave way, and, like another Atlantis, Old Iceland sank 
into the sea. What remains is chiefly the Faeroes and the large island now known as 
Iceland. This happened less than a million years ago, and volcanic activity has con- 
tinued with undiminished force ever since. (Translated from the Norwegian.) 

As may be seen, Earth's outline, based on geologic evidence, fits well the 
results obtained by biogeographic reasoning. The idea that a land connection 
existed between Europe and America as late as the Pliocene or the Pleistocene 
must be said to rest upon strong foundations. 

Many questions remain unanswered, however. The most important con- 
cerns the time when the connection was broken. The close relationships in 
biota between Iceland, Scotland and Scandinavia suggest that this might have 
happened in relatively recent times. The connection probably did not exist 
during the Post-glacial climatic optimum (the Hypsithermal). Marine deposits 
along the coast of Norway from this period contain molluscs of a southern 
distribution type. If a connection between Iceland and Scotland had existed, 
the Atlantic warm water could not enter the Polar Basin, and the effect of this 
should be noticeable in the marine fauna. The connection between eastern 
America and western Europe was probably not operative during the Glacial 
Ages since the ice masses on Iceland and Greenland then ought to have formed 
obstacles to plant and animal inigration. It seems unlikely that the connection 
existed after the Last Glacial Age, but precisely when it was broken remains 
an open question. The answer is of considerable evolutionary interest, since 
it could afford us with a means to gauge rates of evolution. Also, the question 


of where the Amphi-Atlantic plants and animals survived the glacial ages 
after the connection had been broken is interesting. 

The most reasonable method to obtain the information desired seems to be 
by means of deep sea borings since a connection between Scotland and 
Iceland and also between Iceland and Greenland must have affected pro- 
foundly the oceanographic situation in the seas adjacent to the bridges, and 
this should be noticeable in the character of the sediments. 

Some relevant information might also be obtained by closer study of 
Icelandic Interglacial peat deposits which have been preserved from subse- 
quent destruction by lava flows. By proper dating and analysis of such 
deposits more might be learned about the time when the different components 
of the fauna and flora immigrated to Iceland. 


Barth, T. F. W. (1941). Island. Oslo. 

Blytt, a. (1876). Essay on the Immigration of the Norwegian Flora during Alternating Dry 

and Rainy Periods. Christiania. 
BocHER, T. W., HoLMEN, K. and Jakobsen, K. (1957). Gronlands Flora. Kobenhavn. 
Dahl, E. (1959). Amfiatlantiske Planter. Problems of Amphi-Atlantic plant distribution. 

BIyttia 16, 93-121. 
Fernald, M. L. (1929). Some relationships of the floras of the Northern Hemisphere. 

Proc. Int. Congr. Plant Sci. 2, 1487-1507. 
Geiger, R. (1961). Das Klima der bodennahen Lnftschicht. Die Wissenschaft 78, 646 p. 
HuLTEN, E. (1958). The Amphi-Atlantic plants and their phytogeographical connection. 

Kgl. Svenska Vetensk. Akad. Handl. Ser. 4, 7(1), 1-340. 
Iversen, J. (1953). Origin of the flora of Greenland in the light of pollen analysis. Oikos 4, 

Knaben, G. (1959a). On the evolution of the radical um-group of the Scapiflora Papavers 

as studies in 70 and 56 chromosome species. Part A. Opera Botanica 2(3), 1-76. 
Knaben, G. (1959b). On the evolution of the radicatiim-group of the Scapiflora Papavers 

as studied in 70 and 56 chromosome species. Part B. Opera Botanica 3(3), 1-96. 
LiNDROTH, C. H. (1957). The Faunal Connections between Europe and North America. 

Stockholm and New York. 
LiNDROTH, C. H. (1960). Is Davis Strait — between Greenland and Baffin Land — a floristic 

barrier? Bot. Notiser 113, 130-140. 
Love, A. (1945). Islenzkar iurtir. Copenhagen. 
Love, A. and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic flora. Acta 

Horti Gotoburg. 20, 69-291. 
Ostenfeld, C. H. (1926). The flora of Greenland and its origin. Biol. Medd. Dansk. Vidensk. 

Selsk. 6(3), 1-71. 
PoRSiLD, M. P. (1932). Alien plants and apophytes of Greenland. Medd. om Gronl. 92(1), 

Rouleau, E. (1956). A check-list of the vascular plants of the province of Newfoundland 

(including the French islands of St. Pierre and Miquelon). Contr. de I'lnst. Bot. de 

rUniv. de Montreal 69, 41-105. 
ScoGGAN, H. J. (1950). The Flora of Bic and the Gaspe Peninsula, Quebec. Nat. Mus. 

Canada Bull. 115, 1-399. 


Doris Love 

Institut Botanique de rUniversite de Montreal, Montreal, Canada 

Speed of movement has become an integral part of human hfe. At present 
we tend to think of distances as a matter of hours rather than miles. Man is 
now able to swing around the globe in less than 2 hr, but he really becomes 
aware of distances only if he has to walk on foot. Still, both man and other 
animals have the ability of a more or less speedy transfer from one point to 
another, an ability which can assume life-saving proportions in cases of food- 
shortage, inclement weather, or other adversities. 

It is different with plants. Once they have taken root, it is touch and go. 
They do not have the choice of deciding where to go next. It is purely a matter 
of chance as to where the wind will waft the seeds, the water carry them, or 
how far that animal, to whose feathers or pelt the seed sticks, will travel. 
Chance alone decides even how far a seed will be conveyed between being 
consumed and deposited. Furthermore, plant dispersal is a feature of genera- 
tions of plants, and only a new generation can bring the species a step farther 
on its way. 

The area with which we are concerned here is mainly covered with water. 
It has been subjected to heavy glaciation over and over again, in parts it is 
still in the grip of the Last Ice Age. Yet, we find on both sides of the Atlantic, 
as well as on the islands in its northern parts, several plant species which are 
practically undifferentiated. In some way, or at some time, these species must 
have been able to disperse over all this area, even if it consists of landmasses 
so far apart today that it may seem impossible for plants to bridge the 
distances between them. 

Only a few plant species are actually adapted for long-distance dispersal, 
having seeds or other reproductive parts which can be carried far away from 
the immobile mother plant. Many more species are adapted to dispersal over 
relatively short ranges, from a few meters up to several kilometers, but in our 
particular area the shortest span between two landmasses is the 1 1 km gap 
between the islands in the Spitsbergen chain. The other distances are con- 
siderably longer, most of them over 300 km and all the way up to the 1761 km 
between Jan Mayen and Iceland (cf. Table 1). 

Some plants have reproductive parts which seem truly adapted for transport 
over long distances by wind or water, or by sticking to feathers or pelts. 




Quite a number of plants and seeds, if eaten, can be accidentally dispersed 
over considerable distances. 

Considering first the role of animals as a transportation medium, we find 
that at the present time man is by far the most active plant disperser, even in 
the northern latitudes with which v/e are concerned. But his role is of a very 
late date, seen against the geological time scale, and for our particular purpose 
if is actually insignificant. I will therefore disregard man-made dispersal. 

Table 1 

Exact Distances in Miles and Kilometers between the Nearest Points of the Land- 
masses IN THE North Atlantic Area. Courtesy Dr. J. D. Ives, Geographical Branch, and 
the Computer Center of the Department of Mines and Natural Resources, Ottawa, Canada. 





Scotland (Hebrides, Luchruban) 

Iceland (Hornsvik Light) 



Scotland (Mainland, Cape Wrath) 

Faeroes (Akraberg) 



Scotland (Mainland, Duncansby 


Orkney (S. Ronaldsay) 



Orkney (Westray island) 

Faeroes (Akraberg) 



Orkney (N. Ronaldsay Island) 

Shetland (Scatness) 



Shetland (Stanshi Head) 

Faeroes (Akraberg) 



Shetland (Unst Island) 

Norway (Storoy Island) 



Faeroes (Mykines Island) 

Iceland (Kambar Light) 



Faeroes (Fugloy Island) 

Norway (Bremanger Island) 



Iceland (Rekjavik Light) 

Greenland (Kap Grivel) 



Iceland (Fontur Light) 

Jan Mayen ^Kikut) 



Iceland (Einstakafjall Arm) 

Norway (Stadlandet) 



Greenland (Cape Ammen) 

Ellesmere Island (Wrangel Bay) 



Greenland (Inugsugtussoq) 

Baffin Land (Cape Dyer) 



Greenland (Nordostrundningen) 

Spitsbergen (Amsterdam Island) 



Greenland (Rathbone Island) 

Jan Mayen (Hoybergodden) 



Jan Mayen (Norkapp) 

Spitsbergen (Sorkapp) 



Jan Mayen (Austkapp) 

Bear Island (Cape Duner) 



Jan Mayen (So. austkapp) 

Norway (Heggelvoer Island) 



Spitsbergen (Mainland) 

Spitsbergen (W. Storoya) 



Spitsbergen (Storoya E.) 

Spitsbergen (Kvitoya W.) 



Spitsbergen (Kvitoya E.) 

Frans Joseph's Land (Mary Hams- 

worth Cape) 



Spitsbergen (Sorkapp) 

Bear Island 



Bear Island 

Norway (Ingo Island) 



Among other mammals, polar bears and foxes are the only ones that are 
occasionally carried by ice between the North Atlantic islands (and even from 
Greenland to Iceland). But during the time when these animals venture so far 
from land (and the possibility of eating vegetable matter) that they are caught 
in the drift ice, they subsist exclusively on animal food. We can therefore 
safely ignore them as possible plant dispersers. 

With birds it is another question. They travel freely between landmasses in 


our area, and there is no doubt that they are active as transportation media 
for both ingested and externally carried plant parts. Over short distances their 
role is unquestionably of very great significance (Samuelsson. 1934; Ridley, 
1930). But how important is it, when it comes to long distances? It is necessary 
to consider not only which birds can act as carriers in our area, but also what 
they eat and where and how they travel. Due regard must be given to such 
phenomena as behavior, food intake, rate of metabolism, flight speed, 
migration dates, etc. 

Most of the migratory birds in our area are shore- and waterbirds. Many 
subsist on a diet of plankton, fish, or other animal matter, augmented only 
occasionally by a nibble of berries, grain (e.g. from a shipwreck), and garbage 
(Witherby et al.. 1939). But there are also some, which are mainly vegetarian, 
like swans, geese, and ducks. Smaller, vegetarian landbirds, as well as gulls 
and terns, which occasionally eat vegetable matter, can be disregarded because 
their metabolism is too fast in relation to their flight speed to permit them to 
carry anything internally between our landmasses (Ridley, 1930). 

Experiments with captive ducks, geese, and swans have shown that the pas- 
sage of food in these large birds from the time of ingestion to evacuation takes 
from 3 or 4 hr up to, in extreme cases, 7 hr (Ridley, 1930). Their flight speed 
during migration is not too well known. Over land it seems, according to Hoch- 
baum (1955). to be about 80 km hr at an altitude of 300-1000 m. Over open sea 
speed as low as 50 km hr and altitudes of only 50 70 m have been measured 
by radar (Buss, 1946; Yocom, 1947) for mallards (Anas platyrhyncha L.) 
and pintails {Anas acuta L.). If, therefore, one of the large birds consumed 
a seed just before take off, it is possible that, flying at maximum speed, it 
could carry the seed as far as 600 km in 7 hr before depositing it. But it is 
likely that the rate of metabolism is considerably increased during a strenuous 
flight, and if the flight speed over oceans really is slower than over land 
(Buss, 1946; Yocom, 1947), the distance covered will be considerably shorter, 
down to as little as 150 kni. 

Furthermore, it has been observed that ducks, swans, and geese do not 
eat immediately prior to taking off on a migratory flight (Ridley. 1930; 
Hochbaum, 1955). There are even reports that they start on an empty 
stomach (Andersson in Ridley. 1930). 

The question must also be asked whether seeds will pass unharmed through 
the viscera of these large birds. The action of their crop is very energetic, and 
most soft seeds will inevitably be crushed although hard-shelled seeds may 
escape damage. This has been observed in domestic geese (Ridley, 1930). 
Too little is actually known about the diet of wild swans, geese, and ducks 
(cf. Table 2; also Ridley, 1930; Schaaning, 1933 a, b; Witherby era/., 1939; 
Durango, 1953; Lowenskjold, 1954; Hochbaum. 1955; Sladen, 1960) to 
evaluate the chances for a substantial part of consumed seeds to escape 
unharmed through the digestive channel of these birds. 




o M - a 

5-* CU . ^ ^ 

5 fcr. a' « 


> l« 

3 ^ 

ri— IN 

3 if 
;| .S ^ 

So ~ 
a 2 














»- "8 sr 

.'rt ^ 5 

S ~ S? ? =« 


l^ ^ S 5 '^ 
a 5u - •=: 5r 

y o -t: =>• 

s -3 

^ 3 c <u 

Co cr Q o 

S <*H ^ — 

!^ O c» <U 

-s; •- O 



s: ^3 r*r '^ 
IS ^. 

is .5 o< s: 

=>• s iy v 

2 § I s 
ttj ^ (J 

2 fe) 

5 -S 


03 S ^ 

3 a 

vh Q ^ 

S 2 N 

o ~ a 

<>5 CS 

W o 
"2 W 

o « 


•i -ss 

!-<' 3 U O 









t/5 U5 

S o 



(U 1/5 

« (S 2 ^ 

I § 

.« S 

to i; VI 

to ^ a — 

3 J" 'S: *- 

"a -. Q S 

5 O 









pq Uh 





a3 C C a> 
o cs 3 ,™ 

^ m 

C T3 •— "O rt ._ 

« ^iS g £ § 

i-H (^ 

^ o 


.. 3 

S 3 < 

I S" 
0) (in 


-o S 

"c 3' 

(U OJ 
(U Sfl 

« 5 

35 3 

^ 3 

— O 

<U (U 
c/5 -^ 



3 .^ 

o . 

(U 3 
O <-> m 

<D > 
(U ■> 

(i> 3 




S 3 O. 
3 "u t= 

03 <u 3 rt 

y > -s 3 -o 


hJ 2 13 

E ^ M >" .S ^M 

2 -3 w ,c^ 

, O ,03 _ 

•o OS 5? > 

03 ^' ^, 3 

O Xi H^ C/3 

(U S K 
o 5? 3> 

I— I W5 O 

--- 3 lU ''-' 

" (U 60 ~ 

^ 3 





3 ^- ^ 

03 -a h •- 

S 3 

S o 



to O, 




?s '^ Si 


-- O 
O Q 
o =-• 
60 -^ 








(O ^ 

t>« ^ t^ 03 
U DQ '^ O 

= 8:5 c'S 3 










■^ O 3 

i X) 

oq m 

















It must also be remembered that most migratory flights take place early in 
the spring, before there is a fresh crop of seeds and berries, and that the birds 
at this time of the year usually eat the soft, fresh vegetative parts of the plants, 
like young shoots of grasses, waterplants, etc. Occasionally, however, old 
seeds from the preceding fall must be consumed, but they do not make up a 
large part of the diet. In the autumn, when the migration goes from north to 
south, it starts in many places so early in August that during some years seeds 
and fruits may not yet be fully ripe. In other years a substantial part must have 
ripened and then no doubt fruits and seeds make up a considerable bulk of 
the bird diet. 

A number of Arctic seeds, particularly those that ripen early, are able to 
germinate at once. Others which ripen late need in most cases a rest period 
and exposure to cold before they can germinate (Bliss, 1958). We do not know 
what effect the passage through a bird can have on these seeds, but it seems 
likely that those seeds which need a rest period would have a better chance of 

It is worth mentioning in this connection that 45 km to the north of Iceland 
there is a small island. Grimsey. It supports large colonies of birds (gulls, 
terns, ducks, ravens, etc.) and many of them fly back and forth to the main- 
land. In spite of seemingly perfect ecological conditions this island still has no 
Empetnim, though the mainland abounds in this genus and birds are known 
to greedily devour the fruits in the fall. 

It must also be remembered that birds theoretically are able to carry seeds 
and other plant parts capable of vegetative reproduction externally, attached, 
for example, to feathers, feet, or beaks. At Delta, Manitoba, Canada, it has 
been observed that birds had Lemna in their feathers when shot down 
(Hochbaum, pers. comm.). Especially when the birds are suddenly scared 
away from their feeding grounds, the accidental transport of plants or parts of 
plants from one pond to another is likely to take place (Samuelsson, 1934; 
Hochbaum, pers. comm.). Farmyard birds also have been observed with 
lumps of clay, containing live seeds, sticking to their bodies (Ridley, 1930), 
but it is a fact that wild birds (as all other wild animals) keep themselves much 
better groomed than captive ones. Wild birds usually preen themselves 
meticulously before taking off" on a flight, and it seems especially so before a 
migratory flight (Hochbaum, pers. comm.). 

It is therefore highly unlikely that any larger bits of plants are carried by 
migrating birds. Even if this were the case, these plant parts would have to 
endure several hours of drying winds, during spring migration often combined 
with temperatures below freezing. It is hard to guess what effect the wind- 
chill factor can have on thus exposed plant parts, but most likely it would kill 
them within few hours. Seeds, on the other hand, would probably survive, 
and such genera as Potamogeton, Zannichellia, Sparganium, etc., could 
theoretically have been spread in this manner (Samuelsson. 1934). However, 


among the North Atlantic islands only the Faeroes and Iceland harbor any 
of these genera. 

A group of plants in Great Britain were believed to have been brought 
attached to the bodies of the Pink-footed geese from North America via 
Greenland to Ireland (Heslop-Harrison, 1953). Later, some of these taxa 
have been shown to consist of different species on both sides of the Ocean 
(Love and Love, 1958): Sisyrinchium august ifolium. In = 96 chromosomes, in 
North America, 5". montanum, 2n = 32, in Greenland, Sisyrinchium hiberni- 
cum, 2/7 = 64, in Ireland; Eriocaulon septangulare, 2n = 64, in Europe, 
E. Parkeri, In = 32, and E. pellucidunu 2// 48, in North America (cf. 
also Love and Love, 1961, and unpubl.), etc. The remaining species still 
need further investigation in order to clarify their relationship across the 

It seems thus, when closely considered, that there is a relatively small 
chance for any large number of plants to have been dispersed by birds over 
distances as great as the present ones between the North Atlantic islands. The 
role of the birds as dispensers over shorter distances is no doubt much more 
significant (Samuelsson, 1934) and they must have had a much more important 
function if, at some time, these distances were shorter than now (Love and 
Love, 1956; Dahl, 1958; Hadac, 1960). 

Based on a native flora of ca. 565 species (including ferns and fern-allies, 
Juniperus, and Angiosperms except Taraxacum and Hieracium) in the Faeroes 
(Rasmussen, 1952), Iceland (Love. 1945; Love and Love, 1956), Jan Mayen 
(J. Lid, Oslo, pers. comm.). Bear Island (Ronning, 1959), Spitsbergen (Scho- 
lander, 1934; Dahl, 1937; Hadac, 1944; Dahl and Hadac, 1946; Hagen, 1952) 
and Fianz Joseph's Land (Hanssen and Lid, 1932), not more than at most 
10 per cent can be referred to a category of plants which in recent (^ Post- 
glacial) time possibly have been dispersed, internally or externally, by birds 
(cf. also Table 2). 

Many plants are said to be dispersed by water, streams, currents, even 
frozen in ice, etc. In most cases such dispersal takes place in fresh water, 
which, however, in our area is of considerably limited importance. Except in 
Iceland, there are virtually no rivers or lake systems which can possibly act as 
transportation media. 

Long-distance dispersal by water in our area is therefore limited to the 
carrying capacity of sea-currents, and thus at present only to the warm Gulf 
Stream, and perhaps (though highly unlikely) to the cold currents around 
Greenland and in the northernmost parts of the North Atlantic and the 
Polar Sea. 

Few land plants have seeds which can stand immersion in salt water for 
even a short time without losing both their buoyancy and germination 
ability (cf. Salisbury, 1942), but there are seeds which can float for a long 
time. CakUe edentula has such seeds, and this species has evidently been 


dispersed by the Gulf Stream from the coast of North America, as late as in 
Post-glacial and Present time, as far north as Spitsbergen (Love and Love, 
1947; Hadac, 1960). 

Other shore plants which can fit into a similar distribution pattern, but 
probably one of an older date, belong to genera such as Cochlearia. Honckenya, 
Mertensia and Glaux (Ridley, 1930), of which the first three reach the Arctic 
part of our area, but Glaux so far is found only in Iceland, and not yet in the 

The large size and weight of Honckenya and Cakile seeds (cf. Table 3) 
indicate that they are well adapted to floating. Those of Glaux are smaller, 
individually, but usually united in a cluster of five and most likely therefore 
quite buoyant. Mertensia seeds are lighter in weight, but of a relatively large 
size and may possibly float well. Cochlearia groenlandica is found as a shore 
plant throughout our area but it consists of a complex, circumpolar group of 
subspecies whose areas are not yet well known, and it is somewhat uncertain 
whether it should actually be included in the group of plants dispersed by 

Salt water plants are of course naturally adapted to dispersal in sea-water, 
and the genus Zostera produces seeds (with a corky appendage) and vegetative 
parts which seem to have the necessary buoyancy for long-distance dispersal 
by sea-currents. In Iceland (Love and Love, 1956) and perhaps also in 
Greenland (Bocher, Holmen and Jakobsen, 1957; Jorgensen, Sorensen and 
Westergaard, 1958) the genus is represented by Z. stenophylla, an American 
species. Whether the Faeroes plants are true Z. marina, as indicated by the 
Flora of the Faeroes (Rasmussen, 1952), or are identical with the American 
one, is not yet known to me. None of these species, however, has reached the 
Arctic islands of the North Atlantic. 

Ruppia spiralis and Hippuris lanceolata (= H. tetraphylla p.p.) probably 
belong to this group also, the former so far known only from the Faeroes, 
Iceland and Greenland in our area. None of the two species is represented on 
the Arctic islands, but apparently good H. vulgaris has been collected on 
Bear Island (Ronning, 1959). H. vulgaris, being a freshwater plant does not 
belong in this group, but H. lanceolata is found in brackish and salt water 
also along the Arctic coasts of America and Eurasia. No Hippuris of any 
kind is found at or around the Faeroes, Jan Mayen, Spitsbergen and Franz 
Joseph's land. 

The number of plant species dispersed over long distances inside our area by 
sea-currents, thus, are quite few, at most 1.5 per cent of the total flora (cf. 
Table 3). It has been impossible, furthermore, for this author to find any 
evidence whatsoever for transport over the open seas of living plant material 
frozen in ice. It seems highly unlikely that, even if seeds frozen into the ice of 
a calving glacier could float around in the North Atlantic or the Polar Sea. 
they would ever reach fhore again (Osborne. 1855: cf also Hulten. 1962). 





H . 

z z 

< o 

H fc 








H O 


z ^ 

- ^ 


\- > 

z H 

W H 

es Z 

^ LU 

3 tu 

O ai 





1 1 1 1 1 1 1 X 1 1 

1 a 


1 1 1 X 1 X 1 X 1 X 




1 1 1 1 1 1 1 X 1 1 

1 1 1 X 1 1 1 X 1 X 





X 1 X X 1 1 X X 1 X 

^ e 

■a c 
(u .a 



-, ^ o - 
1 1 1 1 1 6 1 1 1 1 

■^ " <N Tt 








•^ * 

o -^ 

=^ X 

* + -- * •^- 
q ^ ;;] Jn'^ 

rn Tf "^ u-i — ' 

vD O OO • r<-i ^. p 
—I (N CO "^ (N — ' -^ 
X 1 X X -^ XX X 
00 p ^_ ^ '^. ^. ^. 

O rj r4 w ^' O r4 




■S § 2 § 
"S ^ "5 -2 "•^ 

1 &l§5^-5 1 S; 2 


t i i ^ i :^ :$ :|| 1 
§■111-2 -^-g ^:S-| 

O , O to 

•S 2 O' £ 


Table 4 
Natural Dispersal Limits of Fruits, Pollen, and Spores (after Geiger, 1950) 



Settling rate in 
cm/sec at 6 m.p.h. wind 

Average dispersal 
limits in km 

Fraxinus excelsior fruits 



Abies pectinata fruits 



Picea excel sa fruits 



Be tula verrucosa fruits 



Taraxacum officinale fruits 



Pinus silvestris pollen 



Lycopodium spores 



Polytrichiim spores 


19 000.0 

Lycoperdon spores 


460 000.0 

Then, seeds blown over sea-ice by the wind would seem to have a better 
chance, but this will be considered below 

The role of the wind as a long-distance dispenser of plant material is 
perhaps the most important one (Ridley, 1930; Sahsbury, 1942, etc.). It is 
well known that very many plants are equipped for wind-dispersal, and a 
lot of seeds and fruits have extra appendages making this sort of dispersal 
especially feasible. 

There is, however, a substantial difference between wind-dispersal over 
moderate distances and dispersal over such long ones as we deal with here. 
Fruits and seeds equipped with plumes, wings, and similar arrangements may 
not fly long distances at all, because their settling rate, which is dependent on 
the weight and dimensions of the seeds, is too fast (Ridley, 1930; Geiger, 
1950). The only particles which actually can serve as a sort of "air-plankton" 
and be carried very far from the mother plants are spores, pollen, and similar 
microscopical bodies (cf. Tables 4 and 5). But even these do not go very far if the 
wind dispersing them is faint or occurs in stratified layers as is not seldom the 

Table 5 
Experimental dispersal limits of Papaver seeds (after Salisbury 1942). The seeds 



Distance in cm 

Number of viable seeds which would be furnished by one plant 

P. dubium 

P. argemone 

P. hybridum 

over 375 














































® 1 X 1 1 1 







£ s 








^ 1 X X 1 1 













•^ 1 X 1 1 1 











1 1 1 1 1 








r<-> 1 1 1 1 1 










P5 X X X X 1 











fsj X X X X X 














— Tl- <N , , 



o o o 1 1 








































* * -H 

"^ ^. d 
— o o 



















c3 X 



<-> X -1— X 


D. w-i 






o. -^"^^ "^ 











o "^ °°. d "^ -: 













o o o o o 


2 ■ X '^ X X 



















.ii (^ m rN m — ' 




^ — ' 


' — ■ 

' — ' 

1— ' 


■ — ' 


■ — ' 

£ d d d d d 
























i5 a -5 -S ^ ^ 
u -S .5 c S 5 

Q 0^. C^. :/^ >; t^ 





o. ~ ~ ~ ? 15 

s" 5:2 ^ ^ i; SJ 

5 "^ 




I I X 1 I I I I X X I I I I I I I I I I 

xllxxl llxxxlllllxlll 

lllxll llxxxllllllll 

xllxxl I I X I I I I II I I X I 

I I I 

r- f^i — 

o o o o o 

o o o o o o o 

o o o o o o 

(N p 
O <N 

d d 

o — o — — 

(3 (^1 ^c _: — re ^ ^. 

r^ (N 1/-1 

d d (5 

■ c ;■: ''. d d 

\0 — ■ — r^ ■^ m rn 

do d d d d 






O O 




o . 



f ) 






>. S -2 

^ ~ -C C; C 

•-- ? s :c 

•? 55 ? -^SS.-^ 

^ ■§ g 1 "§' 

S"S"~S'5 5:^"SSS-^R 

-5 S S C>:;g 5 

.2 =^> ? 5 c 

5, fe^siSjbfS: 5^5^^ 5 2 > 

= .i: .i .i .i: ts s ^ s -Si .5 •= 

"^ -s s s -S ^ 

S^^^-^"5~5:s^ ^ ? £ 

••c to .5 ~ ^ ? 

■S^H!<>;i:i:>?^. -2:=- 


I cc 



case on a quiet, warm day (Geiger, 1950). On the other hand, if there is a 
high degree of turbulence in the air near the ground, even light seeds can 
eventually be hfted to sufficient height for a long-distance dispersal to take 
place (Ridley, 1930; Dahl, 1958). 

It seems that such seeds must have approximately the same properties as 
loess or very fine sand, i.e. no diameter in any direction over 0.2 mm (Dahl, 
1958), or a density of not more than 2.6 gr/cm^ with a largest diameter below 
0.6 mm (J. Elson, Montreal, pers. comm.). Very few plant species have seeds 
wliich come inside these narrow limits, and in our area only species of 
Orchidaceae and certain Juncus, Sagina, Drosera, Pyrola, Phyllodoce, Loise- 
leuria, and Harrimanella seem to qualify outright (cf. Table 6). 

Air turbulence is often created by thunderstorms in lower latitudes, but 
the violent hurricanes in the North Atlantic area certainly create enough 
turbulence in the ground layers of the atmosphere to move dust-seeds out of 
their capsules and up to a sufficient altitude for long-distance dispersal 
(cf. also Sverdrup, 1957). 

The quantity of spores (of ferns, clubmosses, horsetails, mosses, and 
lichens, etc.) no doubt exceeds in relative amounts the mass of airborn 
seeds, including dust-seeds (Ridley, 1930). Some must be regarded as ever- 
present "air-plankton" (e.g. Lycoperdon, cf. Table 4, and Geiger, 1950), but 
ecological and climatological factors certainly hmit the areas where both 
spores and dust-seeds can develop into mature plants (SaHsbury, 1942). 
Though some species of, for example, Orchidaceae, Lycopodium, Equisetum, 
and various ferns reach very high latitudes, the bulk of them belong to more 
southern areas. 

It cannot be denied, however, that there are a number of species which 
could have [spread over our area very easily if the distances between the 
landmasses at some time in the past were shorter than now (Love and Love, 
1956; Dahl, 1958; Hadac, 1960). Under such circumstances it would be much 
easier to account for the present distribution of species whose seeds are 
carried moderate distances from the mother plant by the wind, and it would 
explain, for instance, why we have a dominance of windspread species in the 
western element of this flora (Dahl, 1958). The winds in the North Atlantic 
are predominantly westerly, although occasionally easterly winds of carrying 
capacity blow for short intervals (Orvig, Montreal, pers. comm.). 

But wind-dispersal does not consist only of material blown through the 
air, it also includes heavier particles blown over the surface of the ground. It 
is self-evident that open water limits this form of dispersal in most of our 
area, but in the far north the continuous ice-cover over the Polar Sea might 
provide means for ground dispersal of seeds over relatively long distances. 
Thus, some Russian scientists (Tikhomirov, 1951; Aleksandrova, 1960) 
believe that all the present flora of Novaya Zemlya has blown in from the 
mainland over the Kara Sea and Strait in Post-glacial time. Whether this 


form of distribution is able to convey seeds from, for example, Greenland to 
Spitsbergen, and from there on to Franz Joseph's Land or vice versa cannot 
be definitely established, but the possibility should not be totally excluded. 

It must be remembered, however, that the sea-ice provides a far from smooth 
surface over which seeds can slide as easily as over lake-ice. Sea-ice is very 
rugged, ;^full_ of relief, ridges, etc. (Sverdrup, 1957), that will easily trap the 
seeds. It is, furthermore, a rough surface even on a minute scale and would 
act as[an abrasive on the material being pushed over it. Thus, probably only 
seeds with heavy coats will be able to withstand this treatment for a sustained 
period of time. 

Not being able to estimate the number of plants dispersed by blowing over 
the polar ice, we can calculate that at most 10 per cent of the species in our 
area are able to spread over long distances by air. 

Considering the three media for long-distance dispersal of plants and seeds, 
it is evident that only a fraction of the present day distribution of plants in the 
North Atlantic area can be explained as a result of Post-glacial wind-, water-, 
or animal-dispersal over the present distances. Those plants, which cannot have 
made use of any of the above-mentioned transportation media, or which have 
not been brought around by man during very recent times, must therefore 
have come to the Atlantic islands over land-connections, or, over a system of 
landmasses at considerably shorter distances from each other than at present. 
This author does not doubt that such conditions did exist in some form or 
another at some time or another, but probably so early that it has been neces- 
sary for the main part of the present, native flora to survive all or at least part 
of the Pleistocene Ice Age in the area. 

That this survival has resulted in the loss of a great deal of a previous 
flora is beyond question; it is more surprising that so much has managed to 

In our area it can be said that the Ice Age still reigns in certain parts, as in 
Greenland, Spitsbergen, and Franz Joseph's Land. Some authors doubt that 
the climate in the Arctic was as severe as at present during the time when there 
were continental ice sheets in Eurasia and North America, and it has been 
designated as an area for possible plant refugia during the Ice Age by, for 
example, Fernald (1925), Hulten (1937), Marie-Victorin (1938), Ewing and 
Donn (1956), to mention only a few. Voices to the contrary have of course 
also been heard (Flint, 1947; Savile, 1961). 

When an area is studied the average temperature is often used as an 
indicator to the rigor of its climate, but the annual average is a poor figure so 
long as the amplitude of the temperature variations is not given. If the plants 
have any frost tolerance at all, it does not seem to matter so much how deep 
the temperature dips. Three species of cacti easily survive winter temperatures 
down to —45°C in Manitoba, Canada, Summer temperature, duration of 
vegetative season, and precipitation are factors that largely determine 


distribution areas (Jeffre, 1960). Even the average air temperature of the 
summer months may not give a full answer to the tolerance of a given species. 
Under conditions which may seem forbidding to a human, a microclimate 
may exist that is very tolerable for a plant (Monteith, 1960). Floras, from a 
dozen to well over a hundred species, exist today in areas where the tempera- 
tures at meteorological stations indicate averages only slightly above the 
freezing point throughout the growing season (cf. Fristrup, 1952, for Peary 
Land, Greenland: June 2.6°, July 6.3°, Aug. 3.6^C; Aleksandrova, 1961, for 
Great Lyakhovsky Island, N. of Siberia: June 0.2°, July 3.5°, Aug. 2.5°C; 
Savile, 1961, for Isachsen, Queen Elizabeth Islands, Canada: June 0.38°, 
July 3.5°, Aug. 1.25°C). 

Aleksandrova's (1961) detailed analysis of the phenology of Great Lyak- 
hovsky Island in the Novosibirsk Archipelago in the summer of 1956 demon- 
strates the close relationship between temperature at ground level and plant 
development, and shows that even in this rigorous climate a rhythmic and 
seasonal development of the flora takes place. During the short vegetative 
season, thus, both Ranunculus sulplmreus and R. Sabinei had two different 
flowering periods, a spring anthesis 23 June to 25 July and a fall anthesis 
8 Aug. to 20 Aug., both apparently temperature-regulated. She noticed a 
difference also in the rhythmic development and length of the vegetation 
periods between various plant communities. The ones on the quickly thawing 
and warming polygon-tundra were richest in species, first to start growth, and 
first to reach a fall aspect, whereas communities in depressions with a long- 
lasting snow-cover, in spite of starting vegetative development under the 
snow, were slower in maturing and had shorter, more compressed seasons, as 
well as a lesser number of species. 

Tikhomirov, Shamurin and Shtepa (1960) have recently used micro- 
thermo-couples to measure the temperature in various parts (buds, leaves, 
stems, roots, etc.) of Arctic plants (e.g. Sieversia glacialis in E. Siberia), and 
found their temperature to be up to several degrees (Centigrade) higher than 
ihe surrounding air temperature. The differences were most marked on clear, 
quiet days, but noticeable also on overcast and windy days. It is interesting 
that the roots, too, had higher temperatures than the soil around them. 

At high altitudes in the Himalayas (Swan, 1961) it has been demonstrated 
that a favorable microclimate and access to some water will permit wind- 
dispersed plant species to take hold, even where the success seems bleak 
indeed, judged by human standards. 

A high degree of adaptation against climatic rigors: polster-growth, much 
hairiness, dark buds, characters which we regard as "protective measures", 
are of course often found among plants thriving in our area. But there are 
also several plants which lack these qualities and still survive well (cf. Bocher, 
1938; Sorensen, 1933). 

It is, however, easily forgotten that these same plant species, subjected lo 


such cold treatment, have also had to go through a period of considerably 
higher temperatures than at present, the Hypsithermal. It may be said that 
the plants have had their tolerances tried to the utmost. 

That many species, which we regard as confined to cold climates, do 
have an almost incredible tolerance, has been demonstrated to me by some 
wild plants at present cultivated in our greenhouse at the Montreal Botanical 
Garden. These plants were brought down from the top of Mt. Washington, 
New Hampshire, U.S.A., from an altitude of 1918 m and a cHmate with a 
yearly average temperature of — 2.8°C (July 9.5°C, extremes Jan. — 43.9°C, 
July 21.7°C). They have been able to survive three years of maltreatment in 
small pots, scorched by summer temperatures of over 50 °C, erratically over- 
and under-watered, occasionally frozen, subjected to extraordinary variations 
in air-moisture, and still they go on existing, growing, and even flowering and 
fruiting. Among them are such species as Silene acaulis, Saxifraga hyperborea. 
Campanula dubia, Bistorta vivipara, Juncus trifidus, Hierochloe orthantha, all 
normally considered as requiring Arctic, or at least High Alpine, conditions 
for their existence. 

If one dares to judge from the few examples given above, the tolerance of 
many of the species which grow in the North Atlantic area is very wide 
indeed. Most likely it is the result of strong selection over a very long period of 

The fact that we have a high percentage of polyploids in this flora may 
also be an indication of its genetical amplitude for varying conditions at the 
same time as it may possibly indicate its old age (Love and Love, 1943, 1949; 
Love, 1959). 

In the opinion of this author it seems therefore that, dispersal and survival 
conditions considered particularly, there are very strong indications that the 
present native flora in the North Atlantic area is old, well established, and 
in the majority of cases a relict from a time preceding the Pleistocene Ice 
Age or at least its latest phases. 


Aleksandrova, v. D. (1960). Some regularities in the distribution of the vegetation in the 

Arctic tundra. Arctic 13, 147-163. 
Aleksandrova, V. D. (1961). Seasonal dynamics of plant associations in the Arctic (in 

Russian). Problem i Severa 4, 59-24. Acad. Sci. U.S.S.R. 
Beijerink, W. (1947). Zadenatlas der Nederlandschen Flora. Veenman & zonen, Waagenin- 

Bliss, L. C. (1958). Seed germination in Arctic and Alpine species. Arctic 11, 180-189. 
BocHER, T. W. (1938). Biological distributional types in the flora of Greenland. Medd. om 

Groenl. 106, 1-339. 
BocHER, T. W., HOLMEN, K. and Jakobsen, K. (1957). Gronlands flora. P. Haase & Sons 

Forlag, Copenhagen. 
Buss, T.O. (1946). Bird detection by radar. The Auk 63. 315-318. 


Dahl, E. (1937). On the vascular plants of Eastern Svalbard. Norges Svalbard og Ishavs- 

undersukelser. Skr. Svalbard og Ishavet 75, 1-50. 
Dahl, E. (1958). Amfiatlantiske planter. Problems of Amphiatlantic plant distribution. 

Blyttia 16, 93-121. 
Dahl, E. and Hadac, E. (1946). Et bidrag til Spitsbergens flora. Norges Svalbard og 

Ishavsimdersokelser . Medd. 63, 1-15. 
DuRANGO, S. (1953). Faglarna ifdrg. 2nd ed. Almquist & Wiksell, Stockholm. 
EwiNG, M. and Donn, W. L. (1956). A theory of Ice Ages. Science 123, 1061-1066. 
Fernald, M. L. (1925). Persistence of plants in unglaciated areas of boreal America. 

Mem. Amer. Acad. Arts & Sci. 15, 241-342. 
Flint, R. F. (1947). Glacial geology and the Pleistocene epoch. John Wiley, New York. 
Fristrup, B. (1952). Physical geography of Peary Land. Medd. om Groenl. Ill, 1-143. 
Geiger, R. (1950). The Climate near the Ground. Harvard Univ. Press. 
Hadac, E. (1944). Die Gefasspflanzen des "Sassengebietes" Vestspitzbergen. Norges 

Svalbard og Ishavsimdersokelser. Skr. Svalbard og Ishavet 87, 1-71. 
Hadac, E. (1960). The history of the flora of Spitzbergen and Bear Island and the age of 

some Arctic plant species. Preslia 32, 225-253. 
Hagen, a. (1952). Plants collected in Vestspitzbergen in the summer of 1933. Norsk 

Polarinstitiitt Medd. 70, 1-13. 
Hanssen, O., and Lid. J. (1932). Flowering plants of Franz Josefs Land. Norges Svalbard og 

Ishavsimdersokelser. Skr. Svalbard og Ishavet 39, 1-42. 
Heslop-Harrison, J. (1953). The North American and Lusitanian elements in the flora of 

the British Isles. In J. E. Lousley (ed.): The Changing Flora in Britain, 105-123. 
HocHBAUM, H. a. (1955). Travels and Traditions of Waterfowl. Univ. of Minnesota Press. 
HuLTEN, E. (1937). Outline of the History of Arctic and Boreal Biota during the Quaternary 

Period. Stockholm. 
HuLTEN, E. (1962). Plants of the floating Ice-island "Arlis 11" Svensk Bot. Tidolir 56 

Jeffre, E. p. (1960). A climatic pattern between latitudes 40" and 70" and its probable 

influence on biological distribution. Proc. Linn. Soc. London 111, 89-121. 
JoRGENSEN, C. A., SoRENSEN, T., and Westergaard, M. (1958). The flowering plants of 

Greenland. A taxonomical and cytological survey. Dansk Vidensk. Selsk. Biol. Skr. 9, 

Love, A. (1945). Islenzkar jurtir. Einar Munksgaard, Copenhagen, 
Love, A. (1959). Origin of the Arctic flora. McGill. Univ. Mus, Publ. 1, 82-95. 
Love, A., and Love, D. (1943). The significance of differences in distribution of diploids and 

polyploids. Hereditas 29, 145-163. 
Love, A. and Love, D. (1947). Studies on the origin of the Icelandic flora I. Cyto-ecological 

investigations on Cakile. Icel. Univ. Inst. Appl. Sci. Dept. of Agric. Rep. B,2, 1-29. 
Love, A. and Love, D. (1949). Geobotanical significance of polyploidy. 1. Polyploidy and 

latitude. Portug. Acta Biol. {A). R. B. Goldschmidt Vol, 273-352. 
Love, A. and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic Flora. Acta 

Horti Gotoburg. 20, 65-291. 
Love, A. and Love, D. (1958). The American element in the flora of the British Isles. 

Bot. Notiser 1958, 376-388. 
Love, A. and Love, D. (1961). Chromosome numbers of central and northwest European 

plant species. Opera Botanica 5, I- VIII, 1-581. 
LowENSKJOLD, H. L. (1954). Studies on the avifauna of Spitzbergen. Norsk Polarinstitiitt 

Skrifter 103, 1-131. 
Marie-Victorin, Frere (1938). Phytogeographical problems of eastern Canada. Amer. 

Midi. Nat. 19, 489-558. 
MoNTEiTH, J. L. (1960). Micrometeorology in relation to plant and animal life. Proc. Linn. 

Soc. London 171, 71-82. 
Osborne, S. (1855). Mirgration of plants on ice Journ. of H. M. Sledge. 
PoRSiLD, A. E. (1920). Sur le poids et les dimensions des graines arctiques. Rev. Gen. 

Botanique 32, 97-120. 
Rasmussen, R. (1952). Fdroya flora. 2nd ed. Torshavn, the Faeroes. 


Ridley, H. N. (1930). The Dispersal of Plants throughout the World. L. Reeve & Co. Ltd., 

Ashford, England. 
RoNNiNG, O. I. (1959). The vascular flora of Bear Island. Acta Borealia A. Scientia 15, 1-53. 
Salisbury, E. J. (1942). The Reproductive Capacity of Plants. G. Bell, London. 
Samuelsson, G. (1934). Die Verbreitung der hoheren Wasserpflanzen in Nord-Europa. 

Acta Phytogeogr. Suec. 6, 1-211. 
Savile, D. B. O. (1961). The Botany of the Northwestern Queen Elizabeth Islands. Canad. 

J. Bot. 39, 909-942. 
ScHAANiNG, H. L. T. (1933a). A contribution to the bird fauna of East-Greenland. Norges 

Svalbard og Ishavsundersokelser. Skr. Svalbard of Ishavet 49, 25-39. 
SCHOLANDER, P. F. (1934). Vascular plants from northern Svalbard. Norges Svalbard og 

Ishavsundersokelser. Skr. Svalbard og Ishavet 62, 1-153. 
Sladen, W. J. (1960). The flora of a breeding area of pink-footed geese in central Iceland. 

Proc. Linn. Soc. London 171, 30-52. 
SoRENSEN, T. (1933). The vascular plants of East Greenland from 71^ to 73° 30' N. Lat. 

Medd. om Groenl. 101(3), 1-177. 
SvERDRUP, H. (1957). The stress of the wind on the ice of the polar sea. Norsk Polarinstitutt 

Skrifter 111, 1-11. 
Swan, L. W. (1961). The ecology of the High Himalayas. Scientific American 205, 68-90. 
TiKHOMiROV, B. A. (1951). Ontheroleof the wind in the plant distribution in the Far North 

(in Russian). Priroda 1951 23-25. 
TiKHOMiROV, B. A., Shamurin, V. F., and Shtepa, V. S. (1960). Temperature of Arctic 

plants (in Russian). Isvestiya Akademii Nauk SSSR, Ser. Biol. 3, 429-442. 
WiTHERBY, H. F., Jourdain, F. C. R., Ticehurst, N. F., and Tucker, B. W. (1939). The 

Handbook of British Birds. H. F. and C. Witherby, London. 
YocoM, C. F. (1947). Observations on bird life in the Pacific Ocean off the North American 

shores. Condor 49, 204-208. 



Emil Hadac 

Department of Botany, Czechoslovakian Academy of Sciences. Pruhonice, near Praha, 


The flora of the Arctic is very variable. We find there plant species which are 
,^_^ broadly Circumpolar and others which are confined to some part of the Arctic 
ticrritory. There are also species of American, Asiatic, or European origin 
— ^nixed with the Circumpolar High Arctic element. 

If we want to know how old the Arctic flora really is, we must first answer 
several questions: How long have conditions favoring the tundra vegetation 
existed in the Arctic? Was there any tundra vegetation during the Tertiary 
epoch, or was it formed during the Ice Ages? If they already existed in 
the Tertiary epoch, could they have survived all the glaciations? Is there 
really any indigenous Arctic flora, or are plants now living in the Arctic all 
immigrants from the mountains of the adjacent continents? 

Many questions and no easy answers! 

We have as far as I know no direct proof for a tundra vegetation earlier 
than the Quaternary period. On the other hand, we know that the Taiga 
formation (Piceetea) was already widely distributed in a large part of Siberia 
in the second half of the Miocene and in the Pliocene. If Taiga existed as far 
south as Baikal Lake, then there must have been conditions for a tundra 
vegetation north of this vegetation type, if not in lowlands, then at least in the 
Arctic mountains. 

If Lindquist's (1947) identification of Betula callosa in the Icelandic Miocene 
is correct, we can expect true Arctic conditions not far to the north of the 
Arctic Circle during that time. 

Now it is well known that in the Neogene the continents, at least in the 
Northern Hemisphere, were raised perhaps several hundreds of meters 
(cf., for example, Strachov, 1948). By this movement a land connection 
between America, Iceland, and Scandinavia could be and most probably 
was formed. Terrestrial sediments are found in this region originating mainly 
from the Eocene (Kjartansson, 1940; cf. also Dahl, 1958), and the submarine 
relief of the northern Atlantic shows us that there was a land connection not 
only between Greenland and Scandinavia, but also between Northeast 
Greenland and Spitsbergen (Fig. I). 




The influence of such a land connection upon the climate and vegetation 
must have been enormous, especially because in the Neogene the Arctic Sea 
was cut off" from its broad connection with the warm southern part of the 
Ocean; in the Paleogene this connection was realized through Siberia (cf. 
Strachov, 1948). If there was any warm current hke the Gulf Stream, it must 

Fig. 1. The submarine relief of the Arctic basin and the northern Atlantic. 

have stopped on a line from southern Greenland over Iceland to Scotland. 
Most of its heat was given off" on this coast and a mild Atlantic climate must 
have resulted from this change. But how was it north of this land connection? 
When no warm waters could enter the Arctic Basin, a real High Arctic 
climate could be felt already in the central parts of the present Arctic. In the 
contact between the cold Arctic air and the mild, wet Atlantic winds, there 
was probably heavy precipitation and large glaciers could be formed in the 
contact zone, whereas in the Arctic itself a continental climate could have 
favored continental refugia. 

This hypothesis is of course pure fantasy based on a few sohd facts only; 
the paleogeographers and the paleoclimatologists will have the last words on 
these problems. But if we accept such suppositions as a working hypothesis we 
can arrange our biological data concerning the Arctic and Atlantic biota in 
a relatively satisfactory and logical way. 

If the Arctic climate already existed in the Arctic in the Neogene during 


ca. 7,000,000-10,000,000 years, this period was certainly sufficient for the 
formation of the bulk of the Eoarctic flora from the Temperate or Sub- Arctic 
flora growing there in the Tertiary. 

But previously to this, a very similar flora was formed in the high mountains 
of Eurasia and America. This flora was isolated for a long time and could 
develop well-characterized taxa of even higher ranks than species. At the 
end of the Tertiary new possibihties arose for such floras — the Arctic vegeta- 
tion reached as far as some mountain ridges and the mountain flora of North 
America and northeastern Asia could migrate into the Arctic and vice 
versa-Arctic elements could penetrate into the interior of the continents or 
far to the south. 

In the beginning the conditions were somewhat different in Europe itself. 
Although certainly the Alps, the Pyrenees, the Caucasus, etc., had their own 
mountain floras, they had no connection with the Arctic at first. Even during 
the first Ice Age the contact between the Arctic and the Alpine floras was 
probably minimal. The Arctic flora existed mainly north of the continental 
ice-shield. Now the migration of plants occurs differently if these have to 
penetrate into regions already occupied by other plants or if they have to 
colonize land laid bare by sea or ice regression. We must therefore suppose that 
the penetration of the Arctic element southwards was much slower, more diffi- 
cult and less effective than the advance of the mountain flora northwards. The 
First Ice Age could therefore hardly bring any Arctic elements to central 
Europe and the central European mountain flora could scarcely give more 
than a small contribution to the Arctic flora because the ice shield did not 
reach so far south at that time (Fig. 2). 

Only during the Mindel and Riss Glacials, and partly also during the WUrm 
period, could a greater exchange of plant species take place between the 
central European mountains and the Arctic as well as between the Alpine and 
Siberian elements. 

The exchange of the flora of the Arctic and of the mountains of America 
and Eurasia was only partial; we know well enough that only certain mountain 
species migrated to the north and that not aU Arctic species migrated to the 
south although chmatic conditions seemed favorable to both. We must take 
into consideration the ability of the mountain plants to five in the tundra 
communities; there are sharp differences in the life conditions of the steep 
mountain slopes and ridges and the mostly flat tundra. The capacity of seed 
transport was also of some importance. These two factors combined brought 
about a situation where certainly old oreophytic species of the European 
mountains, as, for example, Carex firma (forming extensive and very charac- 
teristic plant communities in the Alps, the Pyrenees, the Carpathians, and 
the Karavanken, and elsewhere) did not reach north of the Tatra Mountains, 
whereas Salix herbacea and other similar species migrated towards the north. 

It seems that there must also have been other factors influencing this 



migration: e.g. the physiological qualities of the plants themselves. There are 
probably several stages in the development of every plant species. Young 
ones are plastic ecologically as well as morphologically. At this stage the 
species are able to migrate long distances if climatic and geographic factors 
allow it, and they easily form new taxa under the influence of different 

(a) (b) 

Fig. 2 (a). The extent of ice-shields during the Great glaciation. (Compiled after 

several authors.) 

(b). The extent of ice-shields during the Last glaciation. (Compiled after 
several authors.) 

"Mature species" have a stabilized morphology, but they are still plastic 
enough ecologically. Thus, if they get the opportunity to migrate, they may 
reach distant regions, but do not change in subsequent periods even if they are 
isolated for a long time. These are the species with disjunctive areas, which 
mostly are relics from former periods. "Old species" lose even their ecological 
plasticity, they cannot spread even over short distances, their energy seems to 
suffice only for holding their former area. If their life conditions change, they 
die out. 

The question of the survival of the Arctic flora in situ is, 1 think, positively 
established by biological as well as by geological arguments. An example is 
Bear Island, which, though small and situated far to the north, was only 


partly glaciated. I have shown (Hadac, 1941, 1960) that practically all its 
rare flora, all its relics, grow in places never glaciated during the Last Ice Age. 

Spitsbergen, Siberia, Alaska, eastern North America and Greenland have 
vast areas of never glaciated territories where, by adaptation and isolation 
during the Ice Age or Ages, new species could be formed, as, for example, 
Coptidiiim spitsbergense, Pediciilaris dasyantha, PuccineUia vilfoidea, etc. 

Meanwhile, during the Mindel and Riss Ice Ages, many Arctic elements 
penetrated southwards. I must say that I was very sceptical towards reports 
of Arctic plant species from far distant mountains in the south until I myself 
collected Oxyria digyna in the mountains of the Iraqi Kurdistan. 

In discussions on the Arctic flora and its migrations several authors have 
emphasized that the migration followed step by step (i.e. not, or very seldom, 
as single plants but rather in whole plant communities) which is possible only 
over land connections. Dahl (1958) has shown (by statistical analysis of the 
Scandinavian Amphi-Atlantic element and its adaptation to, or incapability 
of, long-distance dispersal) that the occurrence of "western" elements in the 
Scandinavian flora is not due to long-distance dispersal. This is a negative 
proof. But we can test this problem also in another way. 

Let us suppose, for example, that the flora of Spitsbergen was supplied by 
long-distance dispersal from the south, east, and west, and that its plant 
communities were formed by free combination of all these geographical 
elements according to their requirements. In such a case the different geo- 
graphical elements of the flora should be represented in plant communities of 
Spitsbergen according to the law of probabihty, and not all grouped together. 

The most common plant association of the Spitsbergen tundra in the Inner 
Fjord zone is Tomentohypnetiim involuti (cf. Hadac, 1946, p. 155). The 
climax community seems to be Cassiopetum tetragonae. The most luxuriant 
community of this region is Trisetetum spicati. Let us now compare the 
percentage occurrence of different geographical groups (which will be dis- 
cussed later on) in the associations mentioned with the average of the whole 
flora, i.e. with the "probable" occurrence: "probable" under the supposition 
that their species came to Spitsbergen independently by long-distance dis- 
persal (Table 1). 

More instructive may be the actual and "expected" numbers of species in 
geographical groups (Table 2). 

We can see that the actual numbers of plant species in individual groups 
very seldom agree with the "supposed" numbers and that the differences in 
several cases (designated with*) exceed the standard deviation ; the differences 
are thus statistically significant. 

It is also evident that in most communities plants of the same origin 
remain together; they came to Spitsbergen not as single species but as 
members of the same plant community. This can be proved as well for Iceland 
and Greenland, as elsewhere. 



I would like to bring in a striking example of this nature from a region far 
from the Arctic. In high mountains of the Iraqi Kurdistan, A.D.Q. Agnew and 
I found two well-defined plant communities side by side: a community of 
Prurigos feru/acea and Astragalus kurdicus (sect. Tragacanthd) and other more 
or less endemic species, which have survived there or in the neighborhood 
from the Tertiary. We also noted some other hygrophytic communities of an 
alliance which we call Primuleto-Blysmion compressi, with Primula auriculata 
and not few European or Eurasiatic elements like Eleocharis quinqueflora. 
Car ex panicea, C. distans, Juncus Gerardii, Juncus inflexus, Trigloc/iin palustre, 
Deschampsia ccespitosa, Sagina saginoides, Cerastium cerastoides, etc., 
practically without any local endemic element. This last community immi- 
grated into this area very probably during the Second or Third Pluvial. 

Table 1 

The Occurrence (in %) in Certain Plant Communities of Plants belonging to Various 

Geographical Groups 

% in the 

% in the 

% in the 

% in the 

Geographical groups 









1. Circumpola Oreo-Arctic 





2. Circumpola Arctic-Sub-Arctic 





3. Circumpola Arctic 





4. Circumpola High Arctic 




5. Amphi- Atlantic- Arctic-Sub- 





6. Amphi-Atlantic-Arctic 




7. European-Alpine-Atlantic 




8. European-Asiatic 


9. Beringian 


10. East Siberian 




11. Greenland-Spitsbergen 


12. Novaya Zemlya-Spitsbergen 



13. Spitsbergen endemics 


14. Recent immigrants 


But this holds true not only for plants : in the communities of the order 
Prangetalia live reptiles and insects of an Irano-Turanian type of distribution, 
whereas in the alliance Primuleto-Blysmion we meet frogs like Raua ridibunda 
and Hyla arborea — typically European or Eurasiatic — or "European" insects 
like Tipula maxima, T. lunata, T. lateralis. Pales crocata, and P. pratensis (J. 
Slipka, unpubl.). We can thus state that not only plant communities, but 
whole biocoenoses migrate step by step, even if their individual members 
could migrate over long distances. 

Obviously, the history of the Arctic flora is very intricate. It is not easy to 








vo r-- 

>j-i ir-) 

'^ ^ 

d d 

d d 

d d 


ON o 

O^ CN 

r- 00 

d d 

^ d d 

d d 


lO 00 

-* t^ 

d d 

d d 

d d 


<0 <N 

^ -: 

p p 


OO 00 

r- 00 

>/-) r~-- 

d d 

d d 

d d 




Tt ^o 

Tt \0 

(^1 IT) 


d d 

d d 

d d 



0\ r<^ 

r-; m 

r-; -H 


<N -^' — " 

1— I 1—1 

*— — « ^^ 






<N p 

q p 

00 o 

-^ — 1 .— ( 

r-t ■^ ■^ 

-^ d d 


^o r- 

m r~- 

Tt ^ 

„ d — ■ 

d d 

d d 



* "* 


as 00 

^ NO r~; 

*i 1^ "/^ 

m r^ — ^ 

-H rW — ■ 

1 <N — <■ 





* "^ f^ 

^ ro \C 

* '^'^ 

00 en -4 

00 ro '-i 

^ (N — 1 





r- fN 

(N — 

t^ OO 

>r) >d <N 

V-) vd r4 

VO Tf — ' 



* -". °o 

^ 00 00 

^ OO >o 

\o -sf -^ 

OV ro — * 

vc <^i — .' 







!- O 

>- o 

>- o 

1) — 

W -r; 

(U .— 

s 1- -C Ki 

>- -^ CS 

>- -^ rt 

5; <U C -2 

X3 S > 

^ if 

^ c ^ "1^ 
~ n c -o 

r- 3 <U 

__ c c -o 

§ c^ 

^ c "^ "^ 

S C "^ "O 

- C "O "^ 

C' oj s- 

~ OJ •- 

=: '- !L) I- 

-^ — *j c3 

2 "5 ^ -a 

.11 1| 


^ O X ^ 

K U X S 

;ij O X _2 

1 < W iy2 

g < W C/3 

■S < W (/: 


define its different components, especially the old ones. Only detailed world 
monographs of whole genera, their cytotaxonomy and ecology combined with 
a solid paleogeographic knowledge will throw a clear light onto the origin of 
the Arctic species. 

To get at least some information I have analysed here the flora of Spits- 
bergen by mapping the distribution of all its vascular plant species. I found 
several types of distribution, which can be applied also for other parts of the 
Arctic, but these cover only a part of the whole Arctic flora components. 
Since there are some serious changes from some of the categories which I 
published in 1960, I shah repeat here all the groups of the Spitsbergen flora 
(used in our considerations above) and try to explain their characteristic 
distribution (cf. Hadac, 1960). 


(I prefer to use the term oreophytic, i.e. mountain type, rather than "alpine", 
to avoid confusion with plants growing in the Alps.) The species of this 
group occur in all the Arctic as well as in the mountains of Eurasia and 
America. Some of them reach even the mountains of the Southern 

They certainly belong to the "Eoarctic" component, but it is difficult to say 
where their place of origin is found. It could be in some of the mountain 
ranges of Asia or North America, but also some place in the Arctic itself. 
Anyway, they must have been relatively old, "mature" species already at the 
end of the Pliocene. During the Pleistocene they had good opportunities of 
spreading over all the Arctic as well as into very distant mountain ridges 
They have persisted in different refugia during the whole of the Pleistocene. 

To this group belong, among others, Equisetum variegatum, Cystopteris 
Dickieana, Woodsia glabella, Eriophorum Scheuchzeri, Carex rupestris, 
Kobresia can'ciiia, Juncus triglumis, Oxyria digyria, Bistorta vivipara, 
Saxijraga oppositifolia and several others. The last three species were found 
in Rissian deposits in Poland by Srodoii (1954). 

Some of the species which I previously grouped in this element, due to a 
more accurate taxonomic investigation now appear to belong elsewhere. 
Thus, Empetnim hermaphroditum Hagerup (or E. Eamesii Fern. & Wiegand 
ssp. hermaphroditum D. Love) appears to have a distribution similar to that 
of Arabis alpina, Sallx herbacea, or Saxifraga aizoides and belongs therefore 
rather to my European-Alpine- Atlantic group (cf. Vassiljev, 1961 ; D. Love, 
1960). Also Lycopodium (or Huperzia) selago seems to be a heterogeneous 
taxon deserving an accurate monograph. 


The origin of this group seems to be somewhere in the Arctic during or 
before the early Pleistocene; during the Ice Ages it attained a broad circum- 


polar Arctic-Sub-Arctic distribution. Part of it reached even some relatively 
adjacent mountain ridges in the nearest continents. There it has a pronounced 
relic character with species like Ranunculus pygmaeus and Saxifrago nivalis. 
The whole group is represented in Scandinavia and Iceland. 

In Spitsbergen the following species belong to this group: Phippsia algida, 
Festuca Richardsonii ssp. cryophila. Carex subspathacea, Carex saxatilis, 
Juncus higlumis. Koeniga islandica. s.l., Tofieldia pusilla. Sfel/aria humifusa. 
Ranunculus hyperboreus. Cardamine bellidifolia, Draba alpina, Saxifraga 
hirculus. Cassiope tetragona, and others. 

Some of them, like Cassiope tetragona, seem to be Preglacial. 


This is another group of the Eoarctic species adjusted to Arctic conditions 
but not occurring in the Sub- Arctic regions. In Scandinavia it has a centric 
distribution, but it is more or less lacking in Iceland. 

Its age may be Late Pliocene or Early Pleistocene. To this group belong, 
for example. Equisetuni scirpoides. Hierocliloe alpina. Arctagrostis latifolia, 
Poa arctica. Carex misandra, Luzula nivalis, Melandrium apetalum. Braya 
purpurascens. Draba lactea, Pedicularis hirsuta. and other species. 


The Eoarctic species of this group are very old — probably from the 
Pliocene. Dupontia Fisheri and Pleuropogon Sabinei belong to the endemic 
genera in the Arctic. I think that the occurrence o^ Pleuropogon in Altai is due 
to migration from the Arctic. Whether Pleuropogon originated in Beringia 
(as assumed by Roshevizh, 1952) or in another part of the Arctic cannot be 
proved at present. 

Some of the plants have their nearest relatives in the Eurasiatic or American 
mountains. Thus, for example. Saxifraga setigera has a related species in 
central Asia; Eutrema Edwardsii has one in Altai and Sayan mountains; 
Festuca brachyphylla one in the mountains of North America. They came 
very early to the Arctic and completed their circumpolar distribution during 
the Ice Ages. 

Alopecurus alpinus must be an old species. It occurs in Scotland but not in 
Scandinavia; it has been found in the stomach of the Siberian mammoth 
(Tikhomirov and Kupriyanova, 1954). In Scandinavia, some of the species of 
this group have formed related species or subspecies probably during the 
Last Glaciation: Arnica angustifolia has a very closely related species in 
Scandinavia, Arnica alpina: Melandrium furcatum has a near relative, M. 
angustiflorunu in Scandinavia and western Siberia. 

Besides the species mentioned we can include in this group, e.g. Deschampsia 
breiifolia, Poa abbreriata, Puccinellia angustafa, Draba subcapitata, and 
several others. 



The "Amphi- Atlantic element" is very heterogeneous and should be divided 
into more homogeneous sub-groups. This group is at home in the Arctic and 
Sub-Arctic, but not in central Europe. It has crossed the Atlantic either via 
northeast Greenland-Spitsbergen, or via south Greenland-Iceland (or vice 
versa). In Spitsbergen only a few species belong to this group: Harrimanella 
hypnoides is probably Preglacial, Deschampsia alpina and Draba rupestris are 
probably of Early Pleistocene age. 


Plant species of this group have their distribution in the Arctic on both 
sides of the Atlantic. It is probable that they migrated by a northern land 
connection, i.e. via Greenland-Spitsbergen-Novaya Zemlya or vice versa. 
This land connection is supposed by Sorensen (1945) to have existed as late as 
during the Mindel-Riss Interglacial. If this is right, the species of this 
group must be of Early Pleistocene age. 

Here belong Car ex ursina, extending from Taimyr to the American Arctic, 
Carex uardina growing from Alaska to Novaya Zemlya and Scandinavia; 
Campanula imiflora and Papaver Dahliamim ssp. Dah/iaiium have a similar 
distribution. Minuartia Rossii reaches from its Arctic American area to 
Spitsbergen but not farther. 

In this group I have also included Eriophorum triste, but recent records 
seem to speak for its inclusion in an Arctic Circumpolar group (No. 3). On 
the other hand, Cerastium Regelii probably belongs here rather than to the 
third group. 


The origin of these species lies in European mountains, in the Alps, the 
Pyrenees, or the Carpathians perhaps as early as the Pliocene or even before. 
Anyway, in early Pleistocene they were well established, "mature" species. 
During the Ice Age they descended from the mountains, forming a part of the 
leading vegetation type of this time, the tundra. In the "Dryas flora" of the 
Riss deposits in, for example, Poland (cf. Srodoh, 1954), were found 
Arahis alpina, Betula nana, Salix herbacea, S. reticulata, together with several 
species of the Circumpolar Oreo-Arctic group. Following the retreating ice- 
shield they came to Scandinavia, Iceland, Greenland, and Arctic America, 
to western Siberia, Novaya Zemlya, and Spitsbergen. 

To this group in Spitsbergen belong: Salix reticulata, S. herbacea, Betula 
nana, Beckwithia glacialis, Arabis alpina, Saxifraga aizoides, Potentilla 
Crantzii. Poa alpina vivipara, and perhaps Festuca vivipara. Empetrum 
hermaphroditum seems also to belong to this interesting group as may be seen 
from the map in Vassil'ev's (1961) recent monograph. 



This group is represented in Spitsbergen by two species only: Potentilla 
midtifida and Comastoma tenella. Both have a very broad distribution, the 
first one growing in Himalaya, Altai, and northeast Scandinavia, the second 
one in the central Asiatic mountains, Siberia, Scandinavia, Iceland, reaching 
even Greenland. They are certainly of Preglacial age. 


This group, formed in Late Pliocene or Early Pleistocene in Beringia and 
spreading since to east and west, is also poorly represented in the Spitsbergen 
flora. To it belong: Arctophila fulva, which attained Scandinavia only in Late 
Glacial time when the Bothnian Bay communicated with the Barents 
Sea: Coptidium lapponicum, Rubus chamaemorus, and ChrysospJenium 

Previously I included in this group also PuccineUia phryganodes in the 
sense of Sorensen (1953) in his recent monograph. But now I agree with 
Love and Love (1961b), who distinguish two species, PuccineUia vilfoidea and 
P. phryganodes. 

The geographic distribution and cytology of taxa, belonging to this complex, 
can perhaps be explained in the following manner : the mother species of this 
complex already existed at the end of the Pliocene somewhere in the Beringian 
region. By geographic changes caused by glaciation and submergence the 
Asiatic and American populations were split and gave rise to PuccineUia 
phryganodes and P. vilfoidea. both sensu lata. During the Late Pleistocene both 
evolved two subspecies in isolated refugia: P. phryganodes formed one 
subspecies, now distributed in Greenland and northeastern America, and 
another in the Beringian area. In the large ice-free area of eastern Siberia 
PuccineUia vilfoidea formed the subspecies sibirica, occurring also in Scandi- 
navia, whereas in Spitsbergen and Novaya Zemlya may be found ssp. vilfoidea 
itself. This last subspecies must have been developed after the land connection 
between Spitsbergen and Greenland was destroyed, i.e. probably after the 
Mindel-Riss Interglacial. It must have occurred after the time when the 
connection between the Novaya Zemlya-Spitsbergen complex was isolated 
from the mainland, but before the isolation of Spitsbergen from Novaya 
Zemlya was realized. 

PuccineUia vilfoidea spreads mostly by vegetative means. The relatively 
short distance between Spitsbergen and Greenland seems to be insurpassable 
for it; even the channel between Novaya Zemlya and the mainland is a 
barrier to it. The occurrence of its ssp. sibirica in the Bothnian Bay indicates 
that it was already present on the Arctic coast of Scandinavia in the early 
Post-glacial, when the direct connection between the Bothnian Bay and the 
Barents Sea still existed. 



This group originated probably in the Early Pleistocene in the ice-free parts 
of eastern Siberia. Salix polaris is known from the "Dryas flora"' of the 
Rissian Tee Age in Poland, and its arrival in Spitsbergen must have been after 
the Penultimate Jnterglacial, because it has not reached Greenland. This 
fits also for most of the other members of this group with the exception of 
Polemonium boreale and Taraxacum arcticum. 

To this element belong: Phippsia conciniia, Liizula Wahlenbergil, Petasites 
frigidus. Parrya luidicauJis, Polemonium horeale, and Taraxacum arcticum. 

This group is problematic. It may perhaps be included in the Arctic-Amphi- 
Atlantic group. Ranunculus Wilanderi and Poa Hartzii might be of Early 
Pleistocene age. Festuca hyperborea was described recently and therefore its 
place in this group is provisional. Carex pseudolagopina belongs to C ambly- 
orhyncha Krecz.. which has a broader distribution and therefore must be 
removed from this group. 


The center of origin of this group seems to be in a previous land-mass 
connecting Novaya Zemlya and Spitsbergen. Some of the species have 
reached Greenland and Scandinavia {Carex paral/ela, Arenaria pseudo- 
frigida). Taraxacum brachyceras extends from Vaygach Island to Greenland, 
but Pedicularis dasyantha grows only in Arctic Siberia, Novaya Zemlya and 
Spitsbergen and forms thus a transition to the next group. 

This group originated probably in the Late Pleistocene and is therefore the 
youngest of the Spitsbergen flora. Puccinellia vaciJIans has several locaHties in 
Spitsbergen and one in Novaya Zemlya. Coptidium spitsbergense, comb, nova 
(based on Ranunculus spitsbergensis Hadac 1944 in Norges Svalbard og 
Ishavsunders. Skr. 87, p. 36) occurs in several distant places of Spitsbergen. 
Its distribution can be explained only by supposing that it already existed in 
the last Interglacial period. Papaver Dahlianum ssp. Hadacianum is hitherto 
known only from Spitsbergen. 


The only recent natural immigrant seems to be Cakile maritima (or C. 
cdentulal — there were no fruits!). There are plenty of other species introduced 
by man. 

The analysis of the Spitsbergen flora shows that most of its species are 
Preglacial or from the early Pleistocene; only 2.8 per cent are "Neo-Arctic" 
species from the Late Pleistocene. 


From all 1 have said above, it can be concluded that the Arctic flora is in a 
great part Preglacial and only a small portion of it can be dated to the Late 
Pleistocene. It was formed during the Neogene and Pleistocene partly from 
plant species occurring in the Arctic itself, partly from plants of American 
and Eurasiatic mountain ridges. 

1 am indebted to Miss Jaroslava Riedlova for help with the statistical 
problems, and to Dr. Lumir Klimes for correcting my English. The nomen- 
clature followed in this paper may be found in Hadac (1942, 1960) and Love 
and Love (1961b). 


Dahl, E. (1946). On different types of unglaciated areas during the Ice Ages and their 

significance to phytogeography. New. Phytol. 45, 225-242. 
Dahl, E. (1958). Amfiatlantiske planter. Blyttia 16, 93-121. 
Hadac, E. (1941). Et bidrag til historien om Bjornoyas flora. Natuien 1941 (5). 
Hadac, E. (1944). Die Gefasspflanzen des "Sassengebietes" Vest-Spitzbergen. Noiges 

SvalhanI og Ishavsumlersokelser, Skrifter No. 87, 1-71. 
Hadac, E. (1948). On the history of the flora of Iceland. Studio Botanica Czechoslovcika9, 

Hadac, E. (1960). The history of the flora of Spitsbergen and Bear Island and the age of 

some Arctic plant species. Pieslia 32, 225-253. 
Horn, G. and Orvin, A. K. (1928). Geology of Bear Island. Noiges Svalhaid og Islwvs- 

iindersokelser, Skrifter No. 15, 1-152. 
Knaben, Gunvor (1959). Papa per-studier, med et forsvar for P. radicatiini Rottb. som en 

islandsk-skandinavisk art. Blyttia 16, 61-80. 
Kjartansson, G. (1940). Uni aidur tertieru basaitspildnanna i nordanvcrdu Atlantshafi. 

Ndltiirufr. 10, 118-128. 
Kupriyanova, L. a. and Tikhomirov, B. A. (1954). Issledovanie pyltzy is rastitei"nych 

ostatkov beresovskovo mamonta. DAN 45 (6). 
LiNDQUiST, B. (1947). Two species of Betitia from the Icelandic Miocene. Svensk Bot. 

Tidskr. 41, 339-353. 
Love, A. and Love, D. (1959). Biosystematics of the black crowberries of America. 

Canad. J. of Gen. et Cyt. 1, 34-38. 
Love, A. and Love, D. (1961a) Some nomenclatural changes in the European Flora. 

I. Species and supraspecific categories. Bot. Notiser 114, 33-47. 
Love, A. and Love. D. (1961b). Chromosome numbers of Central and Northwest European 

plant species. Opera Botanica 5, I-VIII, 1-581. 
Love, D. (1960). The red-fruited crowberries in North America. Rhodora 62, 265-292. 
Markov, K. K. (1951). Paleogeografiya. Moscow 1951. 
RosHEVizH, R. Y. (1952). Analis arealov nekotorich kharakternich dlyaA rktiki slakov 

(eoarktikov). Areal I. 
SoRENSEN, Th. (1945). Summary of the botanical investigations in N.E. Greenland. Medd. 

om Groenl. 136 (3), 1-179. 
SoRENSEN, Th. (1953). A revision of the Greenland species of Piiccinellia Pari. Medd. om 

Groenl. 136(3), 1-179. 
Strachov, N. M. (1948). Osnovy Istoricheskoy Geologii I-II. Moscow-Leningrad. 
Srodon, a. (1954). Flory Plejstocenskie z Tarzymiecltow nad Wieprzem. Warsaw. 
Tikhomirov, B. A. (1946). K filotzenogenesu nekotorich rastitelnych formatziy arkticheskoy 

Evrasii. Bot. Zlwrn. 31 (6), 27-41. 
ToLMACHEV, A. I. (1952). K istorii rasvitiya flor sovetskoy Arktiki. Areal I. 
Vassil'ev, V. N. (1961). Rod Empetnim. Akad. Nauk SSSR, Leningrad 1961. 


Knut F^gri 

Botanical Museum, University of Bergen, Bergen, Norway 

In discussing the characteristics of the Scandinavian ilora, botanists have, since 
Blytt's days at least, been inclined to speak of and think in "flora elements". 

The concept of a flora element presumes a high degree of generahzation, 
which is not always unambiguous. If Scandinavian phytogeographers speak 
about the West-Arctic flora element, they think of a group of plants with a 
certain total geographical distribution outside of Scandinavia. If they speak 
of an oceanic element, distribution within Scandinavia is meant. A flora 
element may even be defined by temporal relations, or by ecological ones. 
Even if these ambiguities are under control the generalization as such is 
dangerous. A common distribution area does not in itself explain anything. 
On the contrary, it is a problem in itself, and what is generally done, is to 
consider the hypothetical explanation of the reason why a group of plants 
occur together as a fact to be used in the analysis of other problems as well. 
But in nature there is nothing like a flora element behaving as a collective 
unit. There are only individual plants (not plant species!) reacting each in an 
individual way. And the more or less fortuitous occurring together may be 
the result of widely differing histories and ecologic demands. 

Like that of any area, the flora of Scandinavia comprises elements of 
different age. Leaving apart the endemic element — which is in our case 
insignificant — we may summarize the history of any plant by the following 
main points : 

1 . Origin and road of immigration. 

2. Dispersal during immigration. 

3. Conditions in Scandinavia at immigration. 

4. History of the plant in Scandinavia after immigration. 

In the current discussion these points have been taken into consideration in 
varying degree, as they are also of diff"erent importance for different plants. 

The simplest group to explain is that of the most recent immigrants, the 
anthropochorous apophytes. The importance of this group of plants increases 
with increasing intensity of traffic, whereas increasing "cleanliness" and 
efficiency of modern transport counteracts the spread of such plants. Ouren 
has recently (1959) shown that a maximum of "transport efficiency" of 



accidental diaspores was reached during the latter part of the nineteenth cen- 
tury with gravel ballast. Many plants, which were then more or less regularly 
imported to Scandinavia and were considered at least transient members of 
the flora, are now rarely seen, or not seen at all except as old herbarium 
specimens, and might perhaps just as well be omitted from future handbooks 
of the flora. Similarly, the recent habit of many mills to grind or grit also the 
sorted-oiit weed seeds before selling them for fodder deprives the avid 
collector of many interesting (but insignificant) finds. 

Nevertheless, there are still many sources of introduction left even today : 
packing material, wool rejects, etc. Most of the plants introduced this way 
either (1) are already represented in our flora, or (2) cannot estabUsh them- 
selves; but sometimes an immigrant secures a foothold. Buuias orient alls or 
Matricaria matricarioides may be quoted among the classical examples, the 
immigration of which dates back, respectively 200 and 100 years (Holmboe, 
1900). And from today Epilobium adenocaulon may be mentioned, which 
evidently has spread very rapidly since World War II, although it had been 
found in Scandinavia before that time. 

With anthropochorous dispersal being overwhelmingly important in 
our days, one is likely to overlook the contemporaneous "natural" long- 
distance dispersal. It is very difficult to decide if man has had nothing to do 
with the case — most of the cases in which man's activity can be ruled out, are 
undateable. The case of Coleantlms subtilis is one of those that can probably 
be comparatively well dated (Lid, 1948). Unfortunately, the plant failed to 
establish itself. Another most probably recent and non-anthropogeneous 
(Holmboe, 1930) case is that of Elisma natans (Luronium natans). The plants 
that introduce themselves by natural means are generally more successful 
immigrants than the anthropogeneous ones, within which group the percen- 
tage of failures is very great. Unsuccessful immigrants of the former group are 
usually not observed, but there is no reason to believe that they do not occur, 
cf. the Mucuna seeds found on our shores. It is of importance that the two 
cases mentioned above presume modern, spontaneous, long-distance dispersal 
of the magnitude of at least some 300-400 and 1000 km. Other instances of 
spontaneous, long-distance diaspore transport readily come to mind. Although 
one cannot exclude the possibility that disjunct finds may represent relict 
occurrences, many of them also must represent long-distance dispersal, not 
only where the dispersal mode is more or less evident (sea transport, waders, 
ducks, etc.), but also in plants with a less obvious dispersal ecology. 

One may discuss how far back the "modern" era of dispersal should stretch. 
The great discoveries of the fifteenth and sixteenth centuries seem to furnish 
a good delimitation. Before that time whatever transport there was, was 
confined to the relatively small area of the Medieval World. The increasing 
intensity of traffic could not bring anything new to Scandinavia that had not 
had a chance of immigration hundreds of years before. 


The importance of the opening up of new routes is vividly realized by any 
European botanist visiting the U.S.A. for the first time. Along all the roads, 
in all waste places in the northeast, the ruderal flora is completely "European" 
in appearance, and a botanist could not tell from the flora on which side of 
the ocean he was. The few non-European plants are such ones that also 
occur in similar places in Europe: Erigeron canadense, Oenothera biennis, etc. 

Obviously, the reason for the dominance of European weeds is the fact 
that these, coming from an area of more intensive agricultural practice and 
human activity as a whole, had been more rigidly selected and adapted to 
ruderal conditions. European agricultural practice and settlement density 
created conditions under which the local flora could not compete — except for 
a few species, which, on the contrary, managed to establish themselves in 
Europe as well. This example is important because it shows the significance of 
ecological changes taking place within the area concomitant with or at least 
more or less contemporaneous with the introduction of alien plants. Due to 
the changes, new ecologic niches are created which may open up room for 
immigrants. What has happened in North America during the last few 
hundred years is important as a picture of what must have happened when 
Neolithic man first came to northwestern Europe with his husbandry and 

Whereas ecologic conditions in America have changed radically during 
the "modern" period, due to the introduction of European agriculture, those 
in Europe have not changed similarly during the same period. The changes in 
agricultural methods have altered the aspects of the weed flora, but not 
fundamentally. We may therefore, in the discussion of this group, neglect 
point 4. Coming now to the next group we shall see how the historical develop- 
ment is of much greater importance. 

The "middle" period is introduced by two overlapping events, the advent 
of agriculture and the deterioration of climate. In some places the first, in other 
places the latter, is the older of the two. In southern coastal districts agriculture 
arrived during the Post-Glacial Hypsithermal period; in central and northern 
districts it came (much) later. The effects of agriculture thus intergrade with 
those of a major cHmatic change. 

Before the introduction of agriculture, vegetation must have proliforated 
itself to an extent inconceivable today. The picture drawn by Iversen (1949) 
is based upon the flat Danish landscape; in Norway it must have been more 
varied, but, even so, there is no doubt that the forested pre-agricultural 
landscape was a relatively monotonous one. Within the conifer region we 
still have relatively untouched areas which in their floristic composition are 
even more monotonous than the oceanic heaths in their terrible poverty. Even 
the smallest clearing around a dairy chalet in the forest looks like an oasis in 
these green deserts. Just as our ruderal and weed flora had been subjected to a 
very strong selection before being spread to America, so our pre-agricultural 


flora had been subject to a very strong forest selection; only those biotypes 
able to survive under the conditions of dense forest could persist unless they 
found refugium in an "atypical" habitat. Such habitats were found above the 
(dense) forest Umit in the mountains or in places covered by some non-climax 
vegetation. That would mean bogs, beaches (of the sea, lakes or rivers), 
screes, or fissures, cracks, or shelves in steep rocks. This again means that 
species with ecologically special requirements (chamaephytes, halo- or 
hydrophytes, etc.) would have a certain chance of survival, and so would plants 
that could grow in the upper, open mountain forests. On the other hand, the 
ordinary open-ground species, above all those of the meadows, would have a 
very difficult time, although we should not underrate the effect of grazing by 
the spontaneous (deer) fauna. To what extent spontaneous forest fires or 
catastrophic gales (Sernander, 1936) influenced the picture, remains con- 

What happened when land was originally cleared for agricultural purposes 
is a phenomenon whose nature cannot only be inferred but also confirmed, 
on the whole, by information from pollen analyses. The effects may be 
summarized in the following statements: 

1 . Certain plants were directly used for different purposes and their number 
reduced by selective utilization. 

2. Certain plants occur in soil which was preferred for agriculture, for 
which reason they were more or less completely exterminated together 
with the corresponding vegetation types. 

3. On the soil thus vacated there were possibilities for the estabhshment of 
new vegetation types composed of (a) archaeophytes — many of them 
previously suppressed or restricted by and in the chmax vegetation, 
(b) cultivated plants, and (c) immigrant weeds and ruderals. 

The difficulty here consists in distinguishing between Groups 3 (a) and (c). 
To cite an example, pollen evidence shows unmistakenly that Plantago 
ianceolata expanded enormously with the introduction of agriculture, but are 
the very few P. Ianceolata grains observed before the land-clearing phase due 
to long-distance dispersal and or contamination, or do they mean that some 
few P. Ianceolata specimens existed even before agricultural land was cleared ? 
And what about the many subspecific taxa within this species? A similar 
question may be raised for P. major, but whereas it is easy to imagine a 
possible refugium for this species on beaches (the beach-form may be second- 
ary!), it is more difficult to visualize where P. Ianceolata could have taken 
refuge, except on those shelves in rock walls where I am afraid we shall feel 
obliged to put up a major part of our flora if we do not accept the possibilities 
of its anthropogeneous origin. Where else can we place a plant like Digitalis 
purpurea ? 

It is very difficult to give any estimate of the number of species in these 


three groups. We know of very few plants belonging to Category 1 being 
utilized to such an extent as to be exterminated locally (Archangelica officinalis, 
Gentiana purpurea) or very much reduced in number {Ulmus scabra). We 
must assume that under Category 2 we have completely lost important plant 
communities of easily arable soil and that vegetation types known today are a 
one-sided selection of those that would have existed in the absence of agricul- 
ture. How many species are involved in the third principle will forever 
remain conjectural, but I should personally be inchned to assign to this group 
(3c) a rather high percentage of our flora — some 20 per cent perhaps. 

The development of climate and vegetation since the advent of agriculture 
must be taken into account. For those parts of the country where land 
occupation took place during the Iron Age, the remaining spontaneous 
vegetation types are probably not too different from those existing before the 
land was cleared. For the old agricultural regions we may presume that 
changes due to climate — in addition to those due to agriculture — have 
completely altered the face of the land, and that nothing remains of what was 
the original vegetation. When discussing these problems, we should keep in 
mind that much good soil, i.e. soil that is considered good today, was unsuit- 
able for farming with the primitive tools of the pre-Iron Age man. 

Wendelbo (1957) has suggested a very interesting recent replacement of 
species, viz. the disappearance of Centaurea pseudophrygia and its gradual 
replacement by C. nigra. The former is a species of open deciduous forests, 
such as are now gradually disappearing with the abandonment of grazing in 
forested areas. C. nigra, however, grows in meadows, on roadsides, etc., and 
is better adapted to conditions of modern agricultural practices. Interbreeding 
and introgressive hybridization may contribute to speeding up the replace- 
ment. One must assume that similar processes took place on a grand scale 
during the more radical transformation of the landscape concomitant with 
the primary introduction of agriculture. 

During the very long time between the disappearance of the Pleistocene 
ice and the advent of agriculture, vegetation could develop under the influence 
of two major factors: the changing climate, and the immigration of species. 
The latter was again a function of the changing climate, but also of two other 
factors: the location of glacial refugia, and the speed and possibilities of 
dispersal. There is no doubt that in the discussion of the immigration of the 
Scandinavian flora, these conditions have not been adequately treated. 

It has become increasingly clear that the sequence of forest types met with 
in pollen-analytical investigations in Scandinavia cannot be interpreted in 
such a one-sided climatological manner as has been the case. Registered 
regional reversals must have a background in climatic shifts : thus the vegeta- 
tional changes indicating the onset of the Younger Dryas and the Sub- 
Atlantic periods are generally due to chmatic deterioration. But we have no 
guarantee that the immigration of, for example, Corylus came just at the 

226 KNUT FAiGRl 

moment when the "Corylus cUmate" first estabhshed itself. The sequence 
{PopuIus~)Betula-PhiuslCory/us~broadlesif forest is very reminiscent of what 
one would expect from an ordinary succession under relatively uniform 
climatic conditions. We possess at least one positive indication that differen- 
tials of immigration speed are of importance: the gradual delay of Pinus in 
relation to Corylus along a line from Denmark to western Norway. One 
might conceive of climate first running through a Pinus-non-Corylus type in 
Denmark and through a Corylus-non-Pinus type in west Norway, but the 
assumption is not very probable. Different rates of dispersal give a better 
explanation. It should be noted that, of the two, Piinis diaspores are better 
adapted for long-distance dispersal, but Corylus nevertheless advances faster. 

Given the disadvantageous climate of the Younger Dryas period, we 
must conceive of a comparatively rapid amelioration most probably bringing 
climate up to at least present-day level in a rather short time. The omni- 
presence of a (sub-)pioneer Betula phase at the beginning of the Post-Arctic 
forest period indicates that the present laws of plant succession were valid 
then also. 

The succession of forest types in northern Europe on the whole matches the 
one to be expected under a static climate (which does, of course, not prove 
anything). But there are some remarkable exceptions, the main one being the 
history of Alnus. There is hardly any doubt that the species generally registered 
in the lowlands is A. glutinosa. Its appearance has been seen in connection 
with the old conception of a change from a drier to a more humid climate at 
the Boreal-Atlantic transition, and it is difficult to suggest a better alternative 
explanation. There has been a tendency on the part of Quaternary geologists 
to operate with precipitation and temperature as if they were independent 
entities instead of being tied together in the same circulation system. However, 
it must be admitted that till today meteorologists have not been able to give 
much help in explaining the changes inferred from geologic-biologic evidence, 
and even less the astonishing parallelism in alternations between humid and 
dry periods according to the Blytt-Sernander concept in the cyclonic belts of 
north Europe, and of the Sahara desert as shown by recent finds (Meriel, 

The occurrence during the Hypsithermal of exigent species which are today 
much rarer (Cladium) in or even absent (Trapa) from Scandinavia undoubtedly 
indicates that climate deteriorated, and so does the general recession of 
vegetation boundaries both from the mountains and from the north. It is 
very doubtful if new competition can completely exterminate a species within 
its proper climatic area, but it is obvious that every new immigrant that 
reaches domination must do so at the expense of the plants previously extant. 
The changes from, for example, Qiiercetum mixtum to Picea in Scandinavian 
pollen diagrams at the last zone transition (Sub-Boreal-Sub-Atlantic) first 
and foremost indicates that the newcomer has conquered its area at the 


expense of the broadleaf forests. That might have happened even under a 
uniform cHmate. On the other hand, the broadleaf forest also disappears 
where no Picea comes in, and one may also ask why the spruce has started its 
immigration at all. But conditions are by no means as simple as they are 
frequently described. In other cases, e.g. that of Cladium, the disappearance 
may be due to a change of environment. Again, one may raise the question 
whether menotrophication came as a result of a climatic change. The appear- 
ance of similar changes during the climatic decline at the end of the Last 
Interglacial is indicative. 

The Post-Arctic climatic amelioration has a double aspect: the positive, 
viz. the immigration and settling of new plants and communities, and also 
the negative, viz. the crowding out of earlier types. The Late-Glacial cHmate is 
still enigmatic, and again we may have been deluded by overlooking ecological 
problems. The open, fresh soil left by the Pleistocene glaciers had a selective 
influence on the germination of the few diaspores that initially landed on it. 
To be able to establish themselves, the species have to be such plants that today 
also occur on open, humus-free soil, a soil type that is primarily found in the 
semi-deserts. It maybe that we have been too much inclined to interpret the 
finding of steppe plants (and animals) as witnesses of a steppe climate instead 
of a steppe soil. Even from a very early period pollen grains have been found 
of plants which are decidedly not Arctic in their requirements. 

This has certain implications for our understanding of living conditions 
during the Late-Glacial, but even more so for our understanding of the 
disappearance of the Late-Glacial biota. Did animals and plants vanish 
because of a change from a steppe climate to a more humid type, or did they 
do so because the soil evolved towards a more humic type without change 
of climatic humidity? Or both? Or because of the onset of competition? 
The role played by mycorhiza should not be left out, but mycorhizal plants 
do appear rather early, at least in humid regions. 

The absence of competition is another important factor not to be discounted. 
Where "Late-Glacial plants", e.g. Ephedra, can maintain themselves well 
into the Atlantic (at least) in Norway (Hafsten, 1953), this is certainly due not 
only to the fact that small crags and shelves in steep rock faces were not 
shaded by forest trees, but also to the fact that in such habitats competition 
never became very severe. Just as such localities have been proposed as 
possible stations for Digitalis and other species of the same ecology, they 
most probably have also furnished havens for the Late-Glacial flora long 
after it had been crowded out in soil progressing towards a climax condition. 

We know that the Late-Glacial flora of southern Scandinavia contained a 
number of Alpine plants, i.e. plants that are today chiefly found in the Alpine 
zone of our mountains. Dryas octopetala, Salix herbacea, Saxifraga oppositi- 
folia, and many others have been identified among pollen and macro-fossils. 
The fate of this flora has for a very long time been controversial. That these 


plants were crowded out from the flat lands in Denmark and a large part of 
south Sweden is generally accepted. But one should not forget the persistence 
of Glacial-Alpine elements in the flora both of the steep mountains in 
south Sweden (Omberg. Hesselman, 1935; Kinnekulle, Albertson, 1940) 
and the flat, thin, poor limestone soils of Oeland (Sterner, 1938). On the 
whole, however, very little is left, and the problem has been raised whether 
the peri-glacial Ice Age flora had any chance to reach the Scandinavian 
mountains this way. The absence of fossils of Arctic plants between south 
Sweden and Jamtland has been mentioned although the argument is perhaps 
not too convincing. Its value depends on how much has been done in order to 
find such fossils. There is no doubt that the climate was sufficiently con- 
genial for a forest vegetation to follow very close to the ice edge. But dimen- 
sions are diff'erent in biology and geology: "very close" and "very fast" in a 
geologic sense may still leave plenty of place and time for Arctic plants to 
colonize. And besides, the bee-line route from Scania to the central Scandin- 
avian mountains was not the only possible route: if Arctic plants were able to 
reach the outer part of the Oslo district (outside the Ra end-moraine line), 
they would also have very good possibilities of spreading towards the moun- 
tains this way. It should not be forgotten that Alpines still grow at low 
altitudes in the Oslo region (Dryas, Carex nipestris in Solbjergfjell, Lid, 1958; 
Arctic species in Krokkleiva, etc.). Without insisting that these plants must be 
true relics from the Late-Glacial (or rather pre-Boreal, as the areas were 
ice-covered until then), we may safely state that no other explanation of their 
presence is more convincing. 

Before proceeding with the Arctic-Alpine species I should like to draw 
attention to the so-called xerothermic element of the flora of southeastern 
Scandinavia. In Oeland it grows intermingled with Arctic-Alpine plants, and 
the problem is how great a part of this element immigrated during the Late- 
Glacial, and how great a part during the presumed xerothermic later period. 
A plant like Coronilla emerus might well be a Late-Glacial relic, occurring 
today in Oeland and at one station on the Norwegian southeast coast. 

Returning now to the Arctic-Alpine plants of south Norway, my main 
thesis would be that unless immigration via the Oslo area can be definitely 
ruled out, there is nothing in dispersal to prevent the mountain flora of 
south Norway from having been recruited from the flora surviving at the 
southern edge of the ice. Which, of course, is no proof that it has been 
furnished that way. 

Whatever are the concepts about ice limits farther north, there can be no 
doubt that Norway south of 61° N. Lat. was completely ice-covered, cf the 
date 13000 B.P. for a chionophilous ice-edge community at Jaeren or 12500 
B.P. for a sub-moraine deposit at Blomvaag (Nydal, 1960) at the extreme 
edge of the land towards the sea. The ice edge must have gone far out in the 
sea. Nevertheless, the Jaeren deposit exhibits a typical Arctic- Alpine flora 


which must have immigrated across very great distances. True, tlie number of 
species known from this flora is low, and they are well equipped for dispersal. 
Nevertheless, their early appearance at Jaeren — isolated by ice and sea — 
can hardly be explained unless iceberg transport from Denmark or some 
other place is resorted to. Even today icefloes transport great quantities of 
Danish rocks to the Oslo area. There is nothing inconceivable in iceberg (or 
-floe) transport of diaspores as weU as of the flint and amber (Holmboe, 1912; 
Johansen, 1957) found in west Norway. 

Whereas the possibility of immigration from Denmark has been recognized 
for a long time, other possibilities have opened up later. The role played by 
the North Sea continent is certainly very difficult to evaluate, but in Britain 
more and more remains of an Arctic Ice Age flora are coming to hght. 

The recent finds in the British Isles of such exclusively Arctic species as 
Koenigia, Artemisia norvegica, Diapensia lapponica, and others throw a new 
and unexpected light on the problems of survival of our mountain flora. 
Mention should also be made of the Scapiflora Papaver seeds found by 
Conolly (1958) although the possibihty cannot be ruled out that they were not 
identical with the Scandinavian Scapiflora. 

If Pleistocene Britain has been a refugium of our mountain flora it is no 
wonder that we find no traces left of it there: in the humid climate of the 
British Isles the ensuing humus cover must have been almost fatal for 
the Post-Glacial survival of these plants as were the salt waters covering 
the North Sea continent. It is remarkable that anything is left at all. 

The restriction of the Scandinavian mountain flora to the two well-known 
centers represents a challenge. So does the apparent isolation of the West 
Arctic species. A solution to the problem has been off'ered by biologists : the 
hypothesis of Ice Age survival within Scandinavia. Incidentally, if survival is 
accepted for the Last Quaternary Glaciation, it is very difficult to get away 
from the same argument for earlier ones. 

The idea of Pleistocene survival, so brilliantly presented and defended by 
Nordhagen (1933) has met with very strong, almost passionate resistance from 
geologists. Each argument has been met with a counter-argument, and for 
many years now httle new has been presented : the protagonists of the two 
opinions have mostly been riding their rather age-worn hobby-horses. 

It is easy to see that some of the consequences drawn from the Ice Age 
refugium hypothesis are incongruous, like Lindroth's (1949, p. 775) series of 
refugia in places where we know positively that there cannot have been any, 
or Lindquist's (1948) idea of the survival of Picea abies (cf. F^egri, 1950). 
However, it does not disprove a theory that erroneous conclusions have been 
drawn from it. 

It should also be said that geologists have been completely negative in their 
argumentation: the two main problems of centricity and West-Arctic distribu- 
tion exist, and they demand an explanation. 


On the other hand, it must also be admitted that biologists should take up 
the problems for some fresh thinking. Is centricity anything more than one 
might expect from actual ecologic conditions? Nobody has tried to give an 
objective answer to this, but it is a fact that very few, if any, other areas in 
Scandinavia offer the same combination of favorable bed-rock, climate, and 
topography. If this is enough to explain the distribution without recourse to 
other factors, centricity loses much of its argumentative value. 

Some of the "rare" mountain plants possess good means of dispersal; 
others seem to be in a rather unfavorable situation. The example of the Post- 
glacial immigration of Phms and Corylus shows that our evaluation of dis- 
persal potentialities does not always give the right result and should be an 
admonition of some prudence. But it must be admitted that a species like 
Stellaria crassipes certainly represents some problems, although an auxiliary 
hypothesis of a Late Quaternary development towards reproductive sterility 
is not in principle more — or less — improbable than the hypothesis of survival. 
With its present degree of sterility it is even difficult to think how the species 
could migrate from coastal refugia to its present stations. 

Another species that has presented some difficulties is Carex scirpoidea, 
known in Europe from two stations at Solvaagtind only. The rest of its area is 
American, and the species is the most extreme West-Arctic of all. Carex 
scirpoidea is dioecious, and any attempt at explanation involving long- 
distance dispersal would have to account for simultaneous dispersal of a 
"male" and a "female" seed, which would seem rather unlikely. In itself 
that may be used as an argument for rarity of the species in Europe. However, 
a monoecious form, f. isogyna Dyring, has been described from Solvaagtind. 
This suggests another solution of that particular problem: the Solvaagtind 
occurrence may derive from accidental long-distance dispersal of a seed with 
a "monoecious" tendency. Similarly, monoecious plants are found in at least 

Conditions in north Scandinavia are more complicated than in the south. 
In the eastern part there is a separate flora element with Crepis multicauUs and 
Oxytropis deflexa as its most famous representatives. The nearest stations are 
far east in Asia, and none are known to occur in the intermediate mountains, 
especially not in the Urals. I would consider it very possible that these species 
have survived the last glaciation in or near Scandinavia, but can see no 
reason compelhng us to reject the hypothesis that they survived at the edge of 
the glaciated area in the U.S.S.R. The absence of these species there now is 
easily explained by reference to their ecologic demands and the Post-Glacial 
Hypsithermal period forests that would have crowded out these and many 
other species from their former stations. As survival during a glaciation and 
re-immigration must entail great losses, the absence of these species in the 
Urals is not remarkable, although the Ice Age should have made itself less 
felt in those eastern areas. 


Attempts have been made to elucidate the problem of Pleistocene survival by 
cytogenetic investigations. Unfortunately, the results are far from being 
unambiguous. Favarger has recently (1960) pointed out that polyploidy may 
indicate both youth and great age in a taxon, and moreover these terms are 
relative only, and cannot be adequately translated into geologic or absolute 
chronology. As long as we know nothing about the speed of evolution, 
cytogenetic data cannot contribute very much, nor can the presence or 
absence of endemic taxa. We should not forget that terms like "old" and 
"young" refer to a very local time-scale only, in biology each taxon has its 
own time-scale, and we have no possibility for translating these into terms of 
each other. To use a word like "young" derived from a biologic time-scale in 
a geologic argument is not admissible. 

The current situation is extremely unsatisfactory. The present-day distribu- 
tion pattern of the Scandinavian mountain flora has been referred to ecologic 
and to historical reasons: Ice Age survival has been referred to marginal 
areas in the east, south and west of the ice sheet, and, in addition, to the 
coastal refugia as well as to nunataks. None of the ideas has been properly 
proved, none of them definitely disproved. Biologists have a problem and an 
attempt at an explanation. Geologists refuse to accept the explanation and 
cannot offer any solution to the problem. And no new argument has been 
presented in this tug-of-war, only new versions of the old ones. 


Albertson, N. (1940). Scorpidiitm tuigescens (Th. Jens.) Moenkem. En senglacial relikt i 

nordisk alvar-vegetation. Acta Phytogeogr. Stiecica 13. 7-26. 
CoNOLLY, A. (1958). The occurrence of seeds of Papaver sect. Scapifloia in a Scottish 

late glacial site. Verdff\ Geobot. Inst. Riibel Zurich 34, 27-29. 
Favarger, C. (1961). Sur Temploi des nombres des chromosomes en geographic botanique 

historiques. Ber. Geobot. Inst. E.T.H. Stiftung Riibel 32, 119-146 (1960). 
F/EGRL K. (1950). Studies on the Pleistocene of western Norway. IV. On the immigration of 

Picea Abies (L.) Karst. Univ. Bergen Arbok 1949, Naturv. rekke 1, 1-52. 
Hafsten, U. (1953). Nyopdagede pionerplanter i Norge etter istiden. Natiiren 11, 501-505. 
Hesselman, B. (1938). Ombergs karlvaxtflora. Svensk Bat. Tidskr. 32, 1-88. 
HoLMBOE, J. (1900). Nogle ugrjEsplanters indvandring i Norge. Nvtt. Mag. Naturv. 38, 

HoLMBOE, J. (1912). Naturlig forekommende rav paa Karmoen. Naturen 36, 381-383. 
HoLMBOE, J. (1930). Spredte bidrag til Norges flora I. Nytt Mag. Naturv. 68, 119-151. 
IvERSEN, J. (1949). The influence of prehistoric man on vegetation. Danmarks Geo/. Unclers. 

IV Rtrkke 3 (6), 1-25. 
JoHANSEN, E. (1957). Norsk og svensk boplasflint — er den hentet i Danmark-Skane eller i 

norske strande? Medd. Dansk. Geol. For. 13, 257-258. 
Lid, J. (1948). Eingong veks Coleantltus subtilis i Noreg. BIyttia 6, 33-36. 
Lid, J. (1958). Two glacial relics of Dryas octopetala and Carex rupestris in the forests of 

southeastern Norway. Nytt. Mag. Sot. 6, 5-9. 
LiNDQUiST, B. (1948). The main varieties oi Picea abies (L.) Karst. in Europe. Acta Hort. 

Bergiani 14, 249-342. 
LiNDROTH, C. H. (1949). Die fennoscandischen Carabidae. Goteborgs Kgl. Vetensk. 

Vitterh. Sani/i. Handl. 6 B 4 (3). 

232 KNUr tMGRl 

Meriel, Y. (1962). Les oscillations des climats de la zone aride dans le dernier million 

d'annees. La Nature 1962, 67-74. 
NoRDHAGEN, R. (1933). De senkvartoere klimavekslinger og deres betydning for kultur- 

forskningen. Inst. SammenJ. Kulturforsk. A. 2. Oslo. 
Nydal, R. (1960). Trondheim natural radiocarbon measurements II. Amer. J. Sc. Radioc. 

Suppl. 2, 82-96. 
OuREN, T. (1959). Om skipsfartens betydning for Norges flora. Blyttia 17, 97-118. 
Sernander, R. (1936). Granskar och Fiby urskog. Acta Phytogeogr. Suec. 8, 1-232. 
Sterner, R. (1938). Flora der Insel Oland. Acta Phytogeogr. Suec. 9, 1-170. 
Wendelbo, p. (1957). Arter og hybrider av Centaurea underslekt Jacea i Norge. Univ. 

Bergen Arbok 1957, Natiirv. rekke 5, 1-29. 

After this paper was set, the following publication has come to my knowledge: 
Tolmacev, a. I. and O. Rebristafya. O geograficeskom rasprostranenii Crepis multisaulis 
Ledeb. i o zabytom vide Crepis gnielini (L.) Tausch. Bot. mater, gerb. bot. inst. B. L. 
Koiuarova Ak. N. SSSR 21, 402-415, which shows that the subspecies of C. multicaulis 
previously found in Norway has actually been found at the former border of the Ice Age 
ice sheet in Siberia. 


Zdenek Cernohorsky 
Botanical Institute, Charles University, Praha, Czechoslovakia 

There have been several students who have contributed to our knowledge of 
the lichen flora of the North Atlantic basin. The greatest merits in this 
respect go to the Scandinavian authors. Apart from monographs written on 
some genera, collective pubUcations of the last 30 years are especially impor- 
tant for our consideration (e.g. Dahl, 1950; Dahl, Lynge and Scholander, 
1937; Lynge, 1928, 1934, 1937, 1938, 1940a; Lynge and Scholander, 1932). 
As these pubUcations contain historical chapters as well, they reveal names of 
men who, in the past, devoted their study to the exploration of the lichen flora 
of the North Atlantic basin. 

Owing to these explorers we have nowadays quite a sound knowledge 
concerning macrolichens and their distribution in the said space. This is 
especially true of the conspicuous species, collected even by non-lichenologists. 
As far as microlichens are concerned, their systematic study has only begun 
(e.g. Lynge, 1940b). Furthermore, macrolichens in different regions are 
treated variously. Apart from this, we are lacking in paleontological records 
and this is the reason why I can contribute but partly to the solution of our 
general problem. 

The knowledge of macroUchens led some authors to compare the lichen 
flora of various parts of the North Atlantic basin. We must mention especiafly 
the Norwegian lichenologist Lynge (1934, 1938, 1940a), who compared 
macrolichen floras, e.g. in Iceland, northeast Greenland, Svalbard, and Novaya 
Zemlya. He states that these islands have many species in common and that 
their floras are like that of Scandinavia. He mentions also that the most 
striking of these floras is that of Novaya Zemlya, due to the presence there of 
several eastern species not found farther to the west. The microhchen species, 
e.g. of the genus Rhizocarpon, indicate even greater diff'erences. 

The aforementioned similarity and the widening knowledge about distribu- 
tion of the macrolichen species aroused among lichenologists a question 
concerning the history of these North Atlantic species. It was again Lynge 
who constantly approached this problem, and he had tried to summarize it 
before his death (Lynge, 1938, 1939a). He ascertained that in Svalbard 
(Spitsbergen and North East Land) grows a relatively considerable amount of 



conspicuous macrolichen species, known either from the northern coasts of 
these islands or concentrated there in a remarkable manner. It is to be stressed 
that the climate of the northern coast of Svalbard is highly arctic. One of the 
most interesting cases is the genus Dactylina, represented there by three 
species: Dactylina arctica (Richards.) Nyl., D. madreporiformis (Wulf.) Tuck., 
and D. ramuJosa (Hook.) Tuck. From the maps (Lynge, 1933) we perceive 
that they have an enormous distribution in the world, but that this distribu- 
tion is not continuous. They are unknown from large areas within their 
limits, e.g. there is a wide gap between the occurrence of Dactylina madre- 
poriformis in Svalbard and in the high mountains of central Europe. Lynge 
(1938) concludes that they are very old relict plants which no doubt formerly 
had a wider distribution. Later, their area was reduced, resulting in these 
great gaps, although we have no fossil records to support this hypothesis. 
On the other hand, the comparison with some higher plants, known as 
fossils from previous geological periods, shows that this conclusion is reason- 

From other north coast lichens Lynge (1938) indicates some conspicuous 
species of the genera Parmelia and Cladonia. To explain their existence on the 
north coast of Spitsbergen and North East Land he undertakes an analysis 
of their distribution in Norway. This analysis shows that only Parmelia 
intestiniformis (Vill.) Ach. is High Alpine. Most of the species are Sub- 
Alpine, with the weight of their distribution in the more continental Norwegian 
forest zone. Judging from their distribution in Svalbard and in Norway it can 
be assumed that they have great amplitude in their demands upon Hfe and that 
time is the main factor in explaining their presence on the north coast of 
Spitsbergen and North East Land. As far as the existence of these species in 
Svalbard is concerned, Lynge draws the following conclusion: "... their 
area, or part of it, should have been ice-free refugia during the last glaciation, 
perhaps all through the time subsequent to the Tertiary Age, and that these 
lichens should be relics which have persisted, at least, from the Last 
Interglacial down to the present time". 

Some of the indicated north coast lichens of Svalbard grow in central 
Europe as well, e.g. Parmelia centrifuga (L.) Ach. It is very rare, having its 
southern distribution limit here. Parmelia centrifuga is to be found on summits 
and ridges of mountains (Harz; a broad area of Riesengebirge, mostly at an 
ahitude of 1000-1500 m; Bohmerwald at 820-850 m; the Alps at 1900-2000 
m). As a rule it is sterile and does not reproduce here. l\\ case it does re- 
produce, it is by means of thallus fragments. Parmelia centrifuga is fertile only 
in Harz and in Bohmerwald. No doubt it reproduces in Bohmerwald but its 
distribution there is limited. In central Europe its locahties are scattered 
around the 50th parallel ; they indicate a possible, more or less close relation 
to the border of the greatest glaciation. As the last glaciation was relatively 
small, Parmelia centrifuga hardly penetrated into central Europe from the 


north during the last Glacial Age only. The dispersal of our species over longer 
distances is limited. Apart from that, there were, after previous heavier glacia- 
tions, between Scandinavia and the Riesengebirge probably few suitable 
habitats for saxicolous species, so it could be assumed that its immigration into 
Czechoslovakia had already occurred during Mindel time (Cernohorsky, 1961). 

Obviously Parmelia centrifuga is a very old type which has great ecological 
amplitude. In spite of the fact that the climatic conditions during the Glacial 
Ages in arctic and central Europe (especially below the firn line) were sub- 
stantially different, I suppose that the inability of this species to occupy new 
localities, particularly on the northern and southern limits of its area, con- 
firms Lynge's conclusion. 

Later Dahl (1946, 1950), a close collaborator of the late Professor Lynge, 
tried to solve the survival question of the North Atlantic lichens during the 
Glacial Ages. His method was based on a much broader scale, which he 
further extended in his later works (Dahl, 1954, 1955, 1958). 

In his first paper Dahl (1946) sketches the history of the Arctic-Alpine 
flora in Scandinavia from the Last Interglacial. During the Last Glacial Age 
the central part of Scandinavia was covered by ice, but unglaciated refugia 
existed in limited spots on the western and northern coasts. Some of the 
species survived the Last Glacial Age there, and after the retreat of the ice 
they migrated up into the mountains. Later, in the mountains, they met with 
other species which had penetrated to this region following the retreating ice 
from the south and east. The present Arctic-Alpine flora in Scandinavia is 
therefore a mixture of these two elements. Further, he describes the history 
even more precisely. He collects proofs to show the unglaciated refugia in the 
Northern Hemisphere during the Last Glacial Age. Then he distinguishes 
these refugia into two types: (a) the coastal mountain type (with high 
mountains near the border of deep oceans) that is further divided into two 
subtypes: (aa) the Scandinavian subtype — firn line never descends to sea 
level, Atlantic chmate, rich vegetation of vascular plants, mosses and lichens; 
(ab) the Antarctic subtype — firn fine descends to the level of the sea during the 
severest period, Antarctic cUmate, few or no vascular plants, few mosses, rich 
vegetation of lichens, especially microlichens ; (b) the tundra type (with a 
continental climate, firn line never descends to sea level, rich vegetation of 
vascular plants, mosses and lichens). Refugia of the first type (Scandinavian 
subtype) are found in western and northwestern Scandinavia, probably in 
Scotland, Iceland, southern Greenland, and possibly Labrador; refugia of 
the second type in Siberia, possibly in the Kola peninsula, in Novaya Zemlya, 
northern Norway, Bear Island, Spitsbergen, northern Greenland, and Arctic 
Canada. The following Arctic-Alpine Hchens are characteristic for the coastal 
mountain refugia (Scandinavian subtype): Thyrea radiata (Smft.) Zahlbr., 
Agyrophora rigida (Du Rietz) Llano, UmbUicaria havaasii Llano, and Alec- 
toria nitidula (Th. Fr.) Vain. Dahl regards these macrolichen species as typical 


for the tundra refugia: Dactylina arctica (Richards.) Nyl., D. ramulosa 
(Hook.) Tuck., D. madreporiformis (Wulf.) Tuck., Parmelia subobscura Vain., 
Omphalodiscus krascheninnikovii (Sav.) Schol., and Cetraria chrysantha 
Tuck. Before he draws a conclusion, he shows also that some southern species 
could survive the Last Glacial Age on unglaciated coastal mountain refugia, 
and the more so the more Atlantic the climate was, for example on the west 
coast of Scandinavia and especially in the British Isles. 

But Dahl does not enumerate any hchens of this latter type. No doubt, as 
an example there can be mentioned some of the southern species from 
southwest Greenland (Dahl, 1950, p. 157-158) and probably even some of the 
north coast hchens from Svalbard. Placopsis gelicia (L.) Nyl., which forms a 
transition between macrolichens and microhchens, could probably also be 
named. This is a species of a more or less pronounced Atlantic (oceanic) 
nature, the distribution of which was mapped by Lamb (1947). Because it 
was previously mentioned even from Czechoslovakia, I carefully noted its 
occurrence in the southwestern part of Iceland in 1948. I collected it there in 
several low altitude localities, but also on the nunatak of the glacier Thoris- 
jokuh, 1060 m (with the firn hne above the foot of the nunatak). Therefore 
we may conclude that it could have survived the Last Glacial Age even in the 
unglaciated refugia in Iceland. 

Lamb (1947) writes that only the coastal mountain refugia would have 
been available for it during the Pleistocene age. He supposes that its present 
extention to Spitsbergen, Bear Island, northernmost Norway, and Novaya 
Zemlya in the east, and to the Boothia Peninsula of Arctic Canada in the 
west, must therefore be the result of migration in the Late Glacial and Post- 
glacial times. This statement may be supported by the fact that soredia of 
Placopsis gelida are usually present. In this connection I would hke to mention 
the said specimens from the nunatak Thorisjokull, which, compared with 
those from the lower locahties of southwestern Iceland, have the whole 
thallus sorediate. It seems that this phenomenon is a reaction of the plant to 
external factors. Also Lynge (1939b) states that Placopsis gelida is sorediate 
in the eastern Spitsbergen islands. For our consideration this phenomenon 
is of great importance since Placopsis gelida is often sterile and without 
apothecia in the Arctic. Besides, reproduction by spores is relatively com- 
plicated, more so than we used to imagine not long ago (cf. Ahmadjian, 1960), 
and therefore not always effective in the Arctic. 

On the other hand, the antiquity of this species must be stressed. This age 
derives from the fact that it is spread not only in the Northern Hemisphere, 
but also in the southern one, with a great gap between its distribution in the 
north Pacific and in the Southern Hemisphere. It is not unlikely, therefore, 
that it could have survived the Last Glacial period even in more northern 
localities than Lamb (loc. cit.) supposes. 

Soredia otherwise are found only in a few Arctic macrohchens, and if 


seen, they are often poorly developed and corticated. The dispersal by means 
of these diaspores may not always be effective : the circumpolar sorediate 
lichens are usually common, but all the non-circumpolar sorediate species 
are rare in the Arctic (Lynge, 1934). Because the formation of apothecia (and 
spores) in many Arctic macrolichens is also limited, they can reproduce only 
by means of thallus fragments. These latter of course do not permit dispersal 
over long distances. 

Now I return to the conclusion in Dahl's paper (1946). In it, Dahl solves the 
problem of the southwest Greenland macrolichen flora and its relation to the 
Scandinavian one, which it approaches in identity. In his later paper (Dahl, 
1950) he goes into the same problem, using a rich material. Here he compares 
the macrolichen flora of southwest Greenland and the macrolichen species in 
the floras of Scandinavia and central Europe. He states that 24 species are 
found in southwest Greenland and in Scandinavia, but not in the mountains 
of central Europe. Analogically there are in Scandinavia 36 species of Alpine 
lichens which have not been discovered in the mountains of central Europe. 
The Scandinavian and the central European mountains possess 15 common 
Alpine macrolichen species. Only 1 1 Alpine species found in the central Euro- 
pean mountains occur neither in Scandinavia nor in southwest Greenland. 
Consequently the relation between the macrolichen flora of southwest Green- 
land and the Alpine macrolichen flora of Scandinavian is closer than the 
relation between the Alpine macrolichen flora of Scandinavia and that of the 
central European mountains. Dahl explains the cause of this phenomenon: 
he supposes, that "(1) the lichen floras of Southwest Greenland and of 
Scandinavia date from a period when the correspondence in the whole flora 
between Europe and east America was closer than today. (2) The conditions, 
under which the lichens in Scandinavia and southwest Greenland had to live 
during the Last Ice Age, were almost the same in the two districts, but 
diff'erent from those in other areas like Novaya Zemlya or the Alps" (refugia 
of the coastal mountain type, the Scandinavian subtype). 

Dahl (he. cit.) came to the first point of his explanation by comparing 
vascular plants known in Europe only as fossils from Interglacial layers, but 
still growing today in America. From this he concludes that the flora of 
Europe during the Last Interglacial probably had a more American character. 
If this was really the case, then the Arctic-Alpine flora could either survive 
even the largest glaciation on some refugia in northwest Europe (when 
several of them in the Northern Hemisphere probably were of the Antarctic 
type), or else a closer connection existed between northwest Europe and 
northeast America during the Last Interglacial. 

Later Dahl (1955) returns again to the question of unglaciated refugia in 
Scandinavia. He devotes his attention exclusively to vascular plants and 
underlines here the presence of West Arctic species which otherwise are 
lacking in Europe and west Asia. Further, he points to the endemism in the 


Scandinavian flora and to the bicentric distribution of some rare plants in the 
Scandinavian mountains. Two facts are of the utmost importance: "Fossil 
records indicate that conditions in South and Central Sweden were not 
favorable for immigration from the south of much of this flora during the 
deglaciation period. Other fossil records indicate that Arctic-Alpine plants 
grew in the mountains immediately after deglaciation.'" By these facts, he 
gives additional evidence for the survival of the present Arctic-Alpine flora 
during the Last Glacial Age in refugia along the Scandinavian coast and 
adjacent mountains. 

Among the Scandinavian lichens are also found West Arctic species, e.g. 
the previously mentioned Agyrophora rigida (Du Rietz) Llano and Umbilicaria 
havaasii Llano (Llano, 1950), and endemic types, e.g. Rhizocarpon superficiale 
(Schaer.) Vain. ssp. spleudidiim (Malme) Runemark (Runemark, 1956), as 
well as representatives of the bicentric distribution, e.g. Philophoron robustum 
(Th. Fr.) Nyl. (Dahl, 1950). 

We have not mentioned, as yet, the epiphytic lichens. Degelius (1957) is 
applying himself to the question of epiphytic lichens and their survival during 
the Glacial Ages in Iceland. He points especially to ParmeVia aspera Mass. 
Its habitats are situated near presumed refugia of vascular relict plants during 
the Last Glaciation. At the same time this obligately epiphytic species has a 
dominant place in the epiphytic vegetation of birch stands in Iceland (but not 
in Scandinavia). It is known from Scandinavia, where it shows a wide ecologic 
amplitude, and from North America, but not from Greenland. From these 
facts, as well as from others, DegeHus presumes that it is a very old type, of 
Interglacial or perhaps Preglacial age. Supporting this view is also the fact 
that remnants of a birch closely related to recent forms growing in Iceland 
have been found in Tertiary layers, and remains of a birch in an old Interglacial 
layer have also been collected there. Dermatina major (Nyl.) Lettau may be of 
the same age, too. Other epiphytic lichens of Iceland are probably younger, 
some of them being introduced by man in recent times (e.g. Xanthoria 
lobulata (Flk.) B. de Lesd.). 

In the above outline I have restricted myself to the probability of lichen 
survival in refugia in the North Atlantic basin during the Last Glacial Age. 
I did this mainly for the reason that we have no paleontological records of 
lichens from earlier geological periods. In spite of this fact, it is evident from 
this outline that the lichen flora in the described space includes various ele- 
ments which of course are of a different age there. Various authors evaluate 
differently the age of the same type, when even the same author rates the age 
of a given type very vaguely. It can be said, in spite of this, that the oldest are 
the Arctic-Alpine species with immense, but not continuous distribution 
(i.e. showing great gaps) in the world. These species have persisted there 
probably since the Tertiary age. What a pity that we do not know better the 
distribution of Arctic-Alpine microMchens which undoubtedly survived 


even the Greatest Glaciation ! Among them are some especially old types, as 
for example, Rhizocarpon inarense (Vain.) Vain, (mapped by Runemark, 
1956). The species with a broad ecological amplitude (or with different 
ecotypes. unknown to us?) or with new resistent mutations probably also 
have survived the Greatest Glaciation. In any case, the glaciation caused a 
limitation of convenient habitats and a strong natural selection, which resulted 
in decreasing the number of species. On the other hand, it of course did not 
prevent new taxa from arising. The number of species increased again during 
the Interglacial ages. In the relatively short Postglacial age a small number of 
species have penetrated into the North Atlantic zone. Evidently the species 
have immigrated during the different periods and mainly over land. 

In this connection I would like to finish with the conclusion of one of 
Dahl's last papers (1958): 

". . . there is little to indicate that species unable to grow above 800 m in South Norway 
migrated across the Atlantic as more hardy species probably did. 

These observations suggest a climate on the hypothetical land bridge similar to 
that found in the birch belt in South Norway. Such climate probably did not occur in 
the region concerned until the Pliocene period and it is concluded that the land 
bridge existed in the Pliocene. 

Independent geological evidence suggests that a basalt plateau was formed stretching 
from Scotland to East Greenland in the early Tertiary period and existed as a land 
area until it was broken up in the Pliocene or perhaps as late as the Pleistocene age." 


Ahmadjian, v. (1960). The Lichen association. The Bryologist 63, 250-254. 
Cernohorsky, Z. (1961). Die Flechte Parmelia centrifuga (L.) Ach. im Bohmerwald. 

PresUa 33, 359-364. 
Dahl, E. (1946). On different types of unglaciated areas during the Ice Ages and their 

significance to phytogeography. New Phytologist 45, 225-242. 
Dahl, E. (1950). Studies in the macrolichen flora of South West Greenland. Meckl. oni 

Groenl. 150(2), 1-176. 
Dahl, E. (1954) Weathered gneisses at the island of Runde, Sunnmore, Western Norway, 

and their geological interpretation. Nytt Mag. Bot. 3, 5-23. 
Dahl, E. (1955). Biogeographic and geologic indications of unglaciated areas in Scandi- 
navia during the Glacial Age. Bull. Geol. Soc. Amer. 66, 1499-1519. 
Dahl, E. (1958). Amfiatlantiske planter. Problems of Amphiatlantic plant distribution. 

Blyttia 16, 93-121. 
Dahl, E., Lynge B. and Scholander, P. F. (1937). Lichens from Southeast Greenland. 

Skr Svalbard og Ishavet, No. 70, 1-76. 
Degelius, G. (1957). The epiphytic lichen flora of the birch stands in Iceland. Acta Horti 

Gotoburg. 22 (l), 1-51. 
Lamb, I. M. (1947). A monograph of the lichen genus Placopsis Nyl. Lilloa 13, 151-288. 
Llano, G. A. (1950). A Monograph of the Lichen Family Umbilicariaceae in the Western 

Hemisphere. Washington. 
Lynge, B. (1928). Lichens from Novaya Zemlya. Rep. Sci. Res. Norweg. Exped. Novaya 

Zemlya 192h Oslo, No. 43, 1-229. 
Lynge, B. (1933). On Dufourea and Dactylina. Three Arctic lichens. Skr. Svalbard og 

Ishavet, Oslo, No. 59, 1-62. 
Lynge, B. (1934). General results of recent Norwegian research work on Arctic lichens. 

Rhodora 36, 133-171. 


Lynge, B. (1937). Lichens from West Greenland collected chiefly by Th. M. Fries. Medd. om 

Groenl. 118 (8), 1-193. 
Lynge, B. (1938). Lichens from the west and north coast of Spitsbergen and the North East 

Land. L Macrolichens. Skr. Norske Vid.-Akad. Oslo, cl. math.-naturv. 1938 (6), 1-136. 
Lynge, B. (1939a). On the survival of plants in the Arctic. Norsk Geogr. Tidsskr. 7, 489-497. 
Lynge, B. (1939b). A small contribution to the lichen flora of the eastern Svalbard Islands. 

Norges Svalbard Ishavs Unders. Medd. No. 44, 1. 
Lynge, B. (1940a). Lichens from Iceland. 1. Macrolichens. Skr. Norske Vid.-Akad. Oslo, 

cl. math.-naturv. 1940 (7), 1-56. 
Lynge, B. (1940b). Lichens from North East Greenland collected on the Norwegian 

scientific expeditions in 1929 and 1930. II. Microlichens. Skr. Svalbard og Ishavet, Oslo, 

No. 81, 1-143. 
Lynge, B. and Scholander, P. F. (1932). Lichens from North East Greenland collected on 

the Norwegian scientific expeditions in 1929 and 1930. I. Macrolichens. Skr. Svalbard 

og Ishavet, Oslo, No. 41, 1-116. 
Runemark, H. (1956). Studies in Rhizocarpon. Opera Bot., Lund, 2(1/2), 1-152, 1-150. 







Rolf Nordhagen 

Botanical Museum, University of Oslo, Oslo, Norway 

The mountain flora of Scandinavia, and particularly that of Norway, contains 
types of plant distribution which by no means can be explained according to 
the theory that the Scandinavian Peninsula was completely buried under the 
ice during the Last Glaciation. This idea has received the epithet of tabula 
rasa theory because according to it all life existing in this part of Europe 
during the Last Interglacial was completely eradicated during the Last 
Ice Age. After this calamity all the present flora and fauna of Scandinavia had 
to immigrate anew, preferably from south, east and northeast. 

One of the first Scandinavian botanists to oppose the tabula rasa theory was 
the Norwegian Axel Blytt (1881, 1882). Judging from his papers in the 
1880's he evidently realized that a number of hardy species in the Norwegian 
flora had survived the Last Glaciation inside Norway. In 1882 he wrote: 
"Es ist moghch, jawohl sogar warscheinlich dass jene gronlandischen Ele- 
mente in unserer Flora Reste aus den interglazialen Zeiten sind" (It is possible, 
yes, even probable that this Greenlandic element in our flora is a remnant 
from the Interglacial times). Other statements by Blytt in his papers from the 
1880's are, however, on the whole rather haltering. 

However, in a later paper Blytt (1893) contested the idea put forward by the 
Swedish botanist F, W. C. Areschoug (1869) that the so-called "Arctic flora" 
n Scandinavia should have a north Siberian origin. Regarding the Norwegian 
mountain flora, Blytt states that it has rather a Greenlandic or Greenlandic- 
American origin. Here he points his finger to something of utmost importance, 
to stiU actual problems of a historical-phytogeographical nature that Nordic 
natural science will never be able to by-pass. 

Whereas most of the present plant species on the Scandinavian Peninsula 
are to be found in Europe — though in many cases in distinctly disjunct areas — 




the Scandinavian and particularly the Norwegian mountain flora contains a 
group of species with its main center in Greenland and North America. In 
Norway and Sweden there are a number of these very rare and exclusive 
species such as the little yellow-flowered Draba crass if olia. A still more 
striking example is offered by Car ex scirpoidea (Fig. 1). This species was 

Fig. 1. The only European localities of Carex scirpoidea (asterisk); Oxytropis 
deflexa (circle with cross), and Crepis midticaulis (square) in north Norway. 

discovered in 1854 on the Soivagtind Mt. (66° 55' N. Lat.) in Nordland 
Fylke. During the past 100 years, numerous Norwegian and Swedish 
botanists have searched in vain for Carex scirpoidea in other localities. The 
nearest occurrences of this species in relation to its Norwegian locality are 


found in western Greenland. As a third and a fourth example in this connec- 
tion, I want to mention the mountain species Pedicidaris flammea and 
Arenaria humifusa. Both occur at present in northern Norway and northern 
Sweden, but are rare. The nearest localities for Pedicidaris flammea are in 
Iceland and Greenland; Arenaria humifusa has a single locality on Spits- 
bergen but is not very rare in west Greenland and occurs frequently over 
extensive areas in northern North America. It has never been found in east 
Greenland, a fact which is of considerable interest (cf Nordhagen, 1935, 

In order to explain the occurrence of the Greenlandic-American element in 
the mountain flora of Norway (and of Scandinavia), Blytt postulated a land- 
bridge during the Quaternary time joining Greenland with western Norway 
via Iceland and the Faeroes. This land-bridge should have been glaciated 
only in part, and never throughout simultaneously, thus permitting dispersal 
of plants in both directions. At present it is submerged but is often referred to 
as the "Iceland-Faeroe ridge". 

In other respects, admittedly, Blytt was "skating on thin ice". He was wrong 
at least in part in his uncritical characterization of the Scandinavian mountain 
flora as emanating from Greenland and North America and not from 
northern Siberia. If we include the Sub-Arctic flora from the northernmost 
counties of Norway— a flora which is found also in the mountains — we will 
find just as remarkable examples of disjunct distribution as those mentioned 
above. One is represented by Oxytropis deflexa (Fig. 1), discovered in 1879 on 
a low mountain at Masi in west Finnmark. Norwegian, Swedish, Finnish, and 
Russian botanists have failed ever since to locate this species in any other 
place in Europe. Its nearest locahties occur in Altai and along the River 
Olenek in northeastern Siberia (cf. Hulten, 1950). Another species from 
northern Norway worth mentioning because of its highly disjunct area is 
Crepis multicaulis (Fig. 1). It was discovered as early as in 1851 in a single 
locality on the south side of the Varanger Peninsula in east Finnmark. For 
more than 100 years botanists from northern Europe have looked in vain for 
it in other areas. Altai, Mongolia, and northeastern Siberia provide the 
nearest localities (cf. Nordhagen 1935). 

I also want to draw the attention here to Scirpus pumilus (syn. S. alpinus, 
S. emergens) (Fig. 2) with its highly paradoxical distribution area. In northern 
Europe it was originally known only from the Eocambrian dolomite outcrop 
at Porsanger Fjord in Finnmark where it was discovered in 1864. It grows here 
both at sea level and on a low mountain with a rich flora (it was found by the 
Xlllth International phyto-geographical excursion close to the top of the 
dolomite mountain depicted on page 8 in the paper by Gjasrevoll, 1961). 
In 1939 this species was found in the low-alpine region of a mountain at 
Nordreisa in Troms Fylke and during the summer of 1961 in two locahties 
on a small island in the Kvaenangen Fjord, about 40 km north of Nordreisa. 



The islands in the Kvfenangen Fjord are all low and consist of schists with 
impressive layers of limestone, probably of Cambro-Silurian age. Scirpus 
pumilus is easily recognized, but it has never been seen outside of the counties 
Finnmark and Troms, thus never in the mountains of southern Norway, 

Fig. 2. The north European distribution of Scirpus pumilus, only in north Norway. 
The nearest localities are in the Carpathians and the Alps. 

where there are calciferous mica schist and hmestone in many places. The 
nearest Scirpus pumila localities outside of Troms and Finnmark are found in 
the Alps and in the Carpathians. 

No phytogeographer can possibly accept the tabula rasa theory in view of 
the examples from Scandinavia which I have enumerated above: on one 
hand Draba crassifoUa, Carex scirpoidea, Pedicularis flammea, and Arenaria 


humifusa, so-called "West Arctic" species (from a Scandinavian point of 
view), and on the other Oxytropis deflexa, Crepis multicaulis, and Scirpus 

It is worth mentioning here what is written in a paper by the Swedish 
botanist Rutger Sernander (1896): ". . . that in the Norwegian mountains 
have been preserved a not negligible number of remnants from the Inter- 
glacial flora of Scandinavia, especially in Dovre, Nordland and Finnmark, 
which were not over-run by the second Inland Ice. From this time we have 
especially the American-Greenlandic element in the Scandinavian flora" 
(translated from Swedish). In the same treatise Sernander goes as far as to 
state: ", . . into our South-Swedish mountain regions in Jamtland and Harje- 
dalen the most important elements have arrived from that western flora which 
was not destroyed by the second glaciation, and not, as has been suggested, 
from the glacial flora which dispersed from the south, foflowing the rim of 
the regressing Inland Ice up through Norway and Sweden" (translated from 

The Norwegian geologist and prehistorian, Andreas M. Hansen tried 
(1904 a, b) to solve all the problems pertaining to the Norwegian mountain 
flora (above I have touched upon only a few of them) by launching a theory 
according to which there existed along the Norwegian Atlantic coast a broad, 
ice-free margin on which a large number of plant species could have "over- 
wintered" the Last Glacial Age. In 1905 a thesis was published by the Nor- 
wegian botanist N. Wille in which he pointed out that Hansen's theory 
contains much exaggeration but holds nevertheless also a kernel of truth. 
Wille's thesis does not give much new information but he demonstrates by the 
aid of statistics that the number of "rare Arctic plants" in southern Norway 
decreases from the area of Dovre- Vaga-Lom southwards to the Valdres and 
Aurland Mountains and still more so towards the Hardangervidda. This 
gradation from north to south, in Wille's opinion, corresponds badly with 
the view that the south Norwegian mountain flora should have immigrated 
from the south after the Last Glacial. 

Unfortunately, neither Blytt, Sernander, Hansen nor Wille tackled these 
difficult phytogeographical problems by a cartographic presentation of the 
distribution of the species concerned inside Scandinavia. 

The theory of glacial survival must of course also be considered from the 
point of view of Quaternary geology. Hansen (1904) did so and believed he 
could trace a more or less continuous Une of moraines in the fjord-districts of 
Norway. He supposed that this line marked the outer border of the last large 
ice-shield along the Norwegian Atlantic coast (Fig. 3). More recent investiga- 
tions (Undas, 1942; Holtedahl, 1953) have shown, however, that this Une, 
the so-called Ra-line (corresponding to the Finnish Salpaussalka-line) marks 
a certain stage during the deglaciation of Norway and not the outer boundary 
of the last ice-shield. Nevertheless, there exist fragments of end-moraines in 



the west- and north-Norwegian coastal areas which are older than the Ra- 
stage. As a matter of fact, the older stages in the deglaciation of western and 
northern Norway have not been definitely estabhshed yet. 

In a later paper, Wille (1915) drew attention to the conditions — even from 
a Norwegian point of view- — of the literally brutal relief which characterizes 
certain parts of western Norway (especially the coast of More) and all the 

Fig. 3. Map of the Salpausselka-Ra stage of end-moraines in Fennoscandia 
(after Martninussen, 1961). 

coast of northern Norway (from the northernmost part of Nordland County 
northeast towards west Finnmark). He could not imagine that this coast had 
ever been totally covered by ice during the Last Ice Age and proposed that 
there had been refugia with plant life. (cf. Figs. 9 and 10). 

Already in 1912 the late Norwegian geologist Th. Vogt had pubhshed a 
paper about the outermost (southern) islands of Lofoten, Vsero and Rost, 
which he claimed to have been ice-free during the Last Glacial period. 

It was, however, the Swedish botanist Th. C. E. Fries, who in his doctoral 
thesis from 1913 succeeded in pulling down the "survival theory" from its 
lofty heaven of speculations to terra firma. Inspired by Vogt's (1912) work, 
Fries made dot-maps of the present distribution of the mountain plants on 
the Scandinavian Peninsula. He adhered also enthusiastically to the theory of 
survival. Fries divided the Scandinavian mountain flora into four groups : 

(a) The ubiquist group, or plants covering all of the Norwegian-Swedish 
mountain chain. 

(b) The bicentric group, or plants concentrated — though without absolute 
congruence — in two distant "islands" in the mountain chain (Fig. 4). This 



condition is especially evident in Norway, where we find a concentration of 
interesting species partly in the area north of Jotunheimen over Mt. Dovre to 
TroUheimen and the Sunndal Mts., partly from Saltdal in Nordland County 
to west Finninark (Fig. 5). 

Fig. 4. Map of the two "island"-like areas in the Scandinavian Mts. which have an 
unusually rich mountain flora (after Nordhagen, 1936). 

(c) The northern imicentric group, or plants confined to the northern 
"island", where it occurs together with a number of "bicentric" species 
(Fig. 5). 

(d) The southern unicentric group, or plants connected exclusively to the 
"island" area in the southern Norwegian mountains (Fig. 5). 

I was myself for a long while very sceptical towards the "survival" theory, 
which is evident in my thesis for the doctorate from 1921. But after an exten- 
sive journey in 1930 through Norway from the Valdres Mts. in the south all 
the way to the Varanger Peninsula in the northeast in order especially to 
study in nature the populations of Papaier sect. Scapiflora which appear in 
Scandinavia, I became a convert to this theory (Nordhagen, 1931, 1936). 



The sequence of Papaver taxa (in part with 2« = 70 chromosomes, and in 
part with 2« = 56) (Fig. 6) occurring in Scandinavia from north to south, of 
which the northernmost populations on the Varanger Peninsula and the 
southernmost ones in the mountains of Valdres and Sogn have white latex, 
whereas all the others have yellow latex, is so remarkable from a 
phytogeographical point of view that it cannot possibly be explained as a 
result of Late Glacial immigration with a consecutive species formation. Here 
the ""tabula rasa''' theory Uterally crumbles. However, as I still do not consider 
the problem of Papaver taxonomy and nomenclature finally settled, I do not 
want to elaborate on it here. 

unlcenlric specleS' ./ ^y^ ' ...^ 


-i - f%<v." ;,•■• 


V X--J.-- ■■:■■ 


bicentric speciesj ^ ,j^^/,. X"^' — >^'/i5l^' < ~\ ' ' 


f Pi:#^ P#IP P4W ' 
^^--OV fe'^;-%^ fe^^ft^ '. 

Fig. 5. Examples of uni-and bicentric distribution of mountain plants in Norway 
and Scandinavia (after Nordhagen, 1936). 

Fries (1921) made it clear that the Scandinavian mountain flora possesses 
an endemic taxon, Euphrasia lapponica, belonging to the salisburgensis group 
(Fig. 7). It is a bicentric species, absent between South Trondelag and the 
Polar Circle. Hugo Dahlstedt, the late, famous'Swedish specialist on Hieracium 
and Taraxacum, demonstrated in 1928 that Taraxacum Reichenbachii 
Huter ssp. dovrense Dt. ought to be regarded as a species of its own, T. 
dovrense Dt. It is found exclusively in the south Norwegian mountains from 
Opdal in the north to Lom in the south. It is closely related to the extremely 
rare T. Reichenbachii, a taxon of the Tirolian Alps east of the Brenner Pasa 



Fig. 6. Map of the occurrences of Papaver sect. Scapiflora in Scandinavia: 
P. dahliamiin Nordh. (squares); P. r^//c/»/» (Lundstr.) Nordh. (crosses);/', lapponi- 
cum (Tolm.) Nordh. (triangles) ; P. laestadianum Nordh. (circles). Other signs mark 
the occurrences of taxa belonging to P. radicatum Rottb. sensu Nordhagen. 



(cf. Dahlstedt, 1908). Besides this, several endemic species of Taraxacum 
sect. Ceratophora are found both in Norway and in the rest of Scandinavia. 

Tig. 7. Map of the distribution of Euphrasia lapponica in south Norway. For 
the total distribution in Fennoscandia. see Hulten, 1950. 

One of them, T. aleurophorum Hagl. (syn. T. aleurodes Hagl., T. cornutum Dt. 
var. pruinstumDt.) is of special interest because a closely related species, the 
single representative of the Ceratophora group south of Scandinavia, is found 
in Engadin, Switzerland (cf. Dahlstedt, 1906). 


The "survival" — or "refugium" — theory has lately been subjected to a 
rather rough criticism from several Norwegian and Swedish Quaternary 
geologists. They clamor distinct proofs for the existence of ice-free areas 
along the Norwegian coast as well as for nunataks farther inland. Other 
geologists do not totally deny the "refugium theory"' but prefer to remain 
passive. 1 will here stick to the plants and their distribution, and especially to 
the distribution of the bicentric mountain species. Norwegian and Swedish 
botanists have cooperated conscientiously during the last generation, I dare 
say even to their utmost, in order to investigate in full the geographical 
distribution of the bicentric species. The result is that the bicentricity is a fact 
for a number of species. On the whole, it is possible to state that the botanical 
indicia in favor of the survival theory have increased, and many previously 
unknown conditions have been brought into light (cf. Elfstrand, 1927; 
Ekman, 1927; Nordhagen, 1930, 1936, 1940, 1952, 1954, and others). 

In order to explain the bicentricity some authors have resorted to the 
Post-glacial Hypsithermal period. It caused a strong displacement north- 
wards and upwards of the Piniis silvestris forests, and the authors in question 
maintain that the presently bicentric mountain species had a continuous 
distribution in the Norwegian-Swedish mountains before this warm time. 
The cause of the present bicentricity should be that precisely the area from 
north Trondelag to Saltdal in Nordland County consists of mountains reach- 
ing a lower altitude than those to the south and the north of it, and thus 
largely were covered in the Hypsithermal by Pinus silvestris forest which due 
to its shading effect obliterated the light requiring bicentric species from this 
area. Such an explanation is, however, unacceptable. A great many of the 
bicentric species are today found right inside the upper reaches of the Pinus 
silvestris belt. Saxifraga hieraciifolia offers a good example. It was found 
already in the 1870's at an altitude as low as ca. 600 m above sea level at 
Lake Olstappen in the eastern part of Jotunheimen. One of the botany 
students at the University of Oslo, Sverre Lokken, has recently discovered it 
at only 450 m altitude in a river gorge at Garmo, inside the Piims-hQ\i, in the 
northern part of Jotunheimen. The finds from Olstappen and Garmo are 
very important. It is evident that Norway had also during the Hypsithermal a 
very broken relief with eroding rivers, gorges and lakes. Thus, if in recent 
times a plant like Saxifraga hieraciifolia can occur not only above the timber- 
line, but also here and there at a lower altitude, down to 450 m, inside the 
Pinus region, a displacement upwards during the Hypsithermal of the Pinus 
silvestris forests cannot possibly be used to explain the bicentricity of this 
species. Below I will discuss the relationship between the bicentric Rhodo- 
dendron lapponicum and the Pinus forests. 

The survival theory postulates first and foremost that there were ice-free 
refugia along the coast of More in southern Norway (Undas, 1941), and from 
Saltdal and Lofoten to west Finnmark in northern Norway (Holtedahl, 1929; 


Undas, 1938; Nordhagen, 1936; Marthinussen, 1961). Further it postulates 
that during the deglaciation of Norway the "over-wintering" species dispersed 
eastwards and inland, taking hold in the nearest mountainous regions at the 
same time as they lost foothold in the coastal areas, among others because 
other flora elements immigrated and occupied the coast as the cUmate under- 
went changes in Post-glacial time. 

Since 1940 the plan for my scientific work has been to investigate especially 
the coastal mountains of southern Norway in order to find "traces" left by the 
bicentric mountain plants there. These species at present have their center 
far inland from the coastal districts where they must have survived the Last 
Glacial period if the survival theory holds true. To use a poetic expression 
from Macpherson (ahas "Ossian"), I have been hunting for "fragments of 
ancient poetry" ever since 1940. 

Already in 1941 a young geologist, Anders Heltzen, rendered me unex- 
pected help. While investigating the massif of Lauparen between Romsdals- 
f jord and Storf jord in south More about 45 km east of the city of Alesund, 
Heltzen discovered Saxifraga hieraciifolia at an altitude of 1200-1300 m 
above sea level in weathered gneissic soil on one of the peaks. The summer of 
1944 I made an excursion to the said peak, and after a frightening steep 
ascent I rediscovered the locality (Heltzen and Nordhagen, 1944). Saxifraga 
hieraciifolia occurs very sparingly here together with among others Pedicidarls 
Oederi which also has its western limit in Norway on this very peak. In the 
Sub-alpine region at the foot of the Lauparen Massif there are a number of 
typical coastal species, such as, for example. Digitalis purpurea, Dryopteris 
Oreopteris, and slightly farther to the north, Luzula maxima. In 1946, Saxi- 
fraga hieraciifolia was discovered also on a mountain northeast of Lauparen 
not far from the Romsdalsfjord itself. 

Up to 1930 Saxifraga hieraciifolia had been regarded as something of a 
prime example of Axel Blytt's idea that at present time the "arctic flora" of 
South Norway is confined to those areas which lie in the rain shadow of the 
western mountains in south Norway. The species was considered a typical 
inland plant, avoiding the mild, humid climate along the west coast. But 
already in 1930 I was sceptical towards this dogma since I had discoveree 
the species in Eikisdalen, a valley east of the Romsdalsfjord (Nordhagen, 
1930). Thus, during the years 1940-44 1 saw this dogma foiled, and my dream 
to find the "fragments of ancient poetry" in the coastal mountains had 
become true. 

On the whole, the survival theory cannot get a better support than the 
geographical distribution in southern Norway of Saxifraga hieraciifolia: it is 
now found to exist sporadically all the way to the coast of More. The remark- 
able distribution pattern of this species can definitely not be explained in 
accordance with any tabula rasa theory. As S. hieraciifolia at present occurs 
as a rarity in the Carpathians, the Steyermark and Auvergne, it was easy to 


propose that it could have immigrated to Norway from the south as late as 
during the deglaciation. But then, its distribution pattern should have been 
entirely different. Since the species is totally lacking south of the Jotunheimen 
Massif (in spite of the fact that there are innumerable suitable habitats for 
it, for example, on the Hardangervidda, the mountains in ValdresandSogn, 
all botanically very well investigated areas), but appears far out in the coastal 
mountains of More, it is virtually impossible to accept the hypothesis that it 
should have immigrated from the south during the deglaciation. 

Another important example is given us in the distribution of Euphrasia 
lapponica. Until 1947 this bicentric species had in southern Norway been 
found only inland, from Hardangervidda in the south to the Opdal Mts. in 
the north. But in the summer of 1948, I discovered E. lapponica growing on 
some relatively low mountains far out north of the Romsdalsf jord close to the 
sea but above the birch region. Besides the ordinary form with a white 
corolla I found a new race of the species with a deeply blue-violet corolla 
{Euphrasia lapponica Th. Fr. fil. var. purpureocoerulea Nordh. ; cf. Nordhagen, 

The last example of bicentric species to be mentioned in this paper is 
Rhododendron lapponicum, a dwarf shrub and the only European representa- 
tive of the genus Rhododendron indigenous to northern Europe. In Scandin- 
avia it has a typically bicentric area. A previous report of its occurrence on 
Bear Island has proved erroneous. The nearest localities of the species outside 
of Scandinavia is Greenland to the west and Lake Baikal and River Lena to 
the east. 

The south Norwegian area of this species was very hard to interpret 
until 1957. Already during the last century it had been located in several 
places in the northernmost part of Jotunheimen, above the timberhne, in 
the cantons Vaga and Lom and early in this century somewhat farther west 
in Skjak but as a real rarity. To Axel Blytt, Rhododendron lapponicum 
functioned as a cornerstone in his scholarly construction concerning the 
continental character of the "Arctic flora" element in Norway since it seemed 
to be confined to areas situated in the rain shadow of the western mountains 
with their glaciers. 

During the summer of 1957 I started, perhaps by instinct, to investigate the 
northernmost canton Lesja, in the valley of Gudbrandsdalen. It borders to 
Romsdalen. Perhaps it was the above-mentioned discoveries of Saxifraga 
hieraciifolia (Fig. 8) which stirred in my subconscious. Once more fate was 
kind to me. In Lesja a lady told me that in 1945 an Enghsh salmon fisherman, 
Mr. Jack, had divulged to her the precise locality of Rhododendron 
lapponicum southwest of Lesjaverk. I was accompanied on this occasion by 
Curator R. Berg, and thanks to our combined efforts we succeeded in re- 
locating Mr. Jack's locaUty, and even found a new one near by. 

Berg and I returned later to the mountains of western Lesja in the summer 



of 1960. In the meantime a dentist from the town of Molde had written to 
Professor O. GjaerevoU about a plant which he beheved to be Rhododendron 
lapponicum and which he had found farthest to the west in Lesja on the 
Ranakollen Mt. But the dentist had not taken a voucher specimen, neither 

Fig. 8. The distribution in south Norway of Saxifraga hieraciifolia. For its 
occurrence in north Norway, see Hulten, 1950. 

had he told from what side he had ascended RanaskoUen. Towards west- 
northwest this mountain precipitates almost vertically down to the River 
Rauma (after which Romsdalen, the valley of Rauma, is named) After a 
furtive ascent of the mountain from southeast, I decided to risk climbing its 
west-northwest face, and first Berg and later I myself succeeded in locating 
some very sparse occurrences of Rhododendron lapponicum at a little more 
than 1000 m altitude. 

The most interesting facts concerning the occurrences of the individuals 
here were first and foremost their depauperated appearance and then the 
habitat, an ericaceous heath rich in hchens on the very rim of a rock ledge 


with no trace of accompanying calciphilous plant species. Ranakollen is 
strongly humified above the timberline and has a very trivial vegetation; all 
we found were a few specimens of Carex rupestris along some rills and in a 
few rocky cracks a little of Saxifraga oppositifoUa. 

The occurrence of Rhododendron lapponicum on Ranakollen must be 
interpreted as a typically relict habitat — the plant literally "hangs on by its 
fingernails" to the very rim of an acid humus surface. Since in Vaga and 
Lorn as well as in northern Scandinavia this species is confined to calcareous 

Fig. 9. Trolltindene in Romsdalen, seen from Storgravbotn. Norwegian botanists 
interpret these geological formations as "nunataks""; Norwegian geologists prefer 

another explanation. 

rock, it is impossible to interpret its occurrence in western Lesja as a result 
of recent, or even Sub-Atlantic, dispersal westwards from an inland locality 
farther east. Personally, I am convinced that some energetic field botanist may 
succeed in finding more Rhododendron lapponicum (Fig. 1 1) still farther out 
towards the coast, especially if he decides to use a "fine toothed comb" on 
the mountains between western Lesja and Tafjord in south More. In the 
1890's a comparatively rich mountain flora was discovered there in several 
localities. As Rhododendron lapponicum flowers early and is easily over- 
looked in its depauperate form during July and August, the species could 
possibly have been missed on the said mountains by collectors. 

The results of my investigations in western Lesja are, thus, the following : 
from one or more refugia along the coast of Romsdal — More, Rhododendron 
lapponicum has dispersed inland during the deglaciation and arrived in an 



area where the rock substrate fully suited its ecological requirements, i.e. 
the area of Vaga-Lom-Skjak. 

Precisely in this district the relationship between Rhododendron lapponicum 
and Pinus silvestris is very instructive. Already in the last century R. lapponi- 


10. Kvandalstind with "Thor's Hammer", Romsdalen, Norway. This 
formation is considered a nunatak by Norwegian botanists. 

cum was discovered in pine forests, situated at a relatively high altitude in 
Lom. Recently the above-mentioned young Norwegian botanist Sverre 
Lokken found a mass-occurrence of this species at 650 m above sea level in 
an open, mixed Pinus silvestris-BetuJa tortuosa forest at Julian in Lom. On 
this locality R. lapponicum descends even as far as to 500 m above sea level, 
but there it is, of course, very sparcely represented. 

At last a few words about Pedicularis Oederi in Scandinavia (Fig. 12). Its 
present chief area in the Norwegian mountains reaches from Jotunheimen in 
the south to the southernmost part of the Trondelag in the north with a western 



limit in the districts south of Romsdalsf jord and on north More. (cf. Fig. 12). 
Towards east this species follows the Dovre Ridge and its continuation 
northeastwards to the borderhne to Sweden. In our neighborland Sweden it 
occurs in several locaUties in the provinces of Harjedalen and Jamtland. Here 
it gradually peters out, but the species has been discovered in a few, isolated 

Fig. 1 1 . The distribution in south Norway of Rhododendron lapponicum. For its 
occurrence in north Scandinavia, see Hulten, 1950. (cf. also Addendum). 

localities as far north as to ca. 65° N. Lat. It is, however, its southernmost 
limit, in northernmost Hardangervidda, Norway, that attracts most interest. 
Among Scandinavian field botanists it is a well-known fact that Pedicularis 
Oederi is favored by a calcareous substrate. Hardangervidda is an enormous 
massif which for a long time especially east of Sorfjord in Hardanger and 
north of the remarkable Mt. Harteigen has been considered a botanist's 
"paradise", with a rock substrate rich in lime and schists. Here occur Dryas 
octopetala, Arenan'a norvegica, Euphrasia lapponica, and many other "de- 
manding" plant species. But precisely here, in this "paradise", Pedicularis 



'A\ \ 


'^'#§ -i"-"^-^:; 





' i„>*^fi 






















































Oederi is totally lacking. The same is true for all the Hardangervidda which 
has been thoroughly investigated by Curator Johannes Lid. 

UPedicularis Oederi, which does occur in the Alps, had dispersed to Norway 
from the south during the deglaciation, the species ought to have gained a 
foothold just in the petrographically so favorable area which is mentiond 
above. But instead it is really conspicuous by its absence there. The southern 
limit of this species shows with all desirable clarity that it has reached the 
northernmost part of Hardangervidda/row the north, not from the south. 

In a paper written already in 1930, I drew attention to this fact. Pedicularis 
Oederi must have survived the Last Glaciation on the More coast and from 
there spread towards east, northeast and south during the deglaciation. 
Investigations by other Norwegian botanists as well as by myself have also 
demonstrated that in those parts of Norway where the species has its main 
center it possesses biotypes which are able to grow on rock poor in lime, and 
also biotypes which can descend into low regions, as far down as to 400 m above 
sea level. The species has been observed both in south Trondelag and in 
north More in hillside bogs in the Piims sihestris region. During its dispersal 
southwards in Norway, the number of biotypes inside this species seems to 
have diminished strongly because at its southern hmit the plant is confined 
exclusively to more or less calcareous habitats above the upper limit of the 
birch forest. 

The peculiar geographical distribution in Norway of Pedicularis Oederi 
seems to be, really, the strongest botanical argument that we have in favor of 
the survival theory. 


Areschoug, F. W. C. (1869). Om den europeiska vegetationens ursprung. Skand. Natur- 

forsk. Mode Forhandl. 10, 54-80. 
Blytt, a. (1881). Theorien om vexlende kontinentale og insulcere klimater anvendt pa 

Norges stigning. Forhandl. Vidensk. Selsk. Christiania 1881, No. 4. 
Blytt, A. (1882). Die Theorie der wechselnden kontinentalen und insularen Klimate. 

Englers Bot. Jalirb. 2, 1-50, 177-184. 
Blytt, A. (1893). Zur Geschichte der nordeuropaischen, besonders der norwegischen 

Flora. Englers Bot. JcJvb. 17, Beibl. No. 41. 
Dahlstedt, H. (1906). Arktiska och alpina arter inom formgruppen Taraxacum cerato- 

plwrum Led. Arkivf. Bot. 5 (9), 1-44. 
Dahlstedt, H. (1908). Taraxacum Reichenbachii Huter subsp. dovrense. Arkivf. Bot. 7 (1), 

Dahlstedt, H. (1928). De svenska arterna av slaktet Taraxacum. III. Dissinnlia, IV. 

Palustria, V. Ceratophora, Ml. Arctica, VII. Glabra. Kgl. Sv. Vetensk. Akad. Handl.6{7>), 

Elfstrand, M. (1927). Var hava fanerogama vaxter overlevat istiden i Skandinavien ? Sv. 

Bot. Tidskr. 21, 269-284. 
Ekman, E. (1927). Three new bicentric plants in the south of Norway. Nyt. Mag.f. Naturv. 

66, 93-95. 
Fries, Th. C. E. (1913). Botanische Untersuchungen im nordlichsten Schweden. Akad. 

Abhandl. Uppsala and Stockholm. 


Fries, Th. C. E. (1921). Die skandinavischen Formen der Euphrasia salisburgensis. Arkiv 

f. Bot. 17 (6), 1-17. 
Gj/Erevoll, O. (1961). XIJI. International Phytogeographi al Excursioi to Finnmark and 

Troms 26. 7-5. >^ 1961. Trondheim. 
Hansen, A. M. (1904a). Hvorledes har Norge faat sit plantedekke? Natiiren 28, 143-156, 

Hansen, A M. (1904b). Landnam i Norge. Kristiania. 
Heltzen, a. M. and Nordhagen, R. (1944). En vestlig utpost av Saxifraga hieracii folia. 

Nat wen 1944. 
Holtedahl, O. (1929). Some remarkable features of the Sub-marine relief on the north 

coast of the Varanger Peninsula, Northern Norway. Norske Vidensk. Akad. Avhandl. 

I. Mat.-naturv. klasse 1929, No. 12, 1-14. 
Holtedahl, O. (1953). Norges Geolcgi II. Norges Geol. Unders. No. 164. 
Hulten, E. (1950). Atlas over vcixternas utbredning i Norden. Stockholm. 
Marthinussen, M. (1961). Brerandstadier og avsmeltningsforhold i Repparfjord — 

Stabbursdal-omradet, Finnmark. Et deglaciasjonsprofil fra fjcrd til vidde. Norges 

Geo!. Unders. Arbok 1960, No. 213. 
Nordhagen, R. (1921). Kalktufstudier i Gudbrandsdalen. Norske Vidensk. Selsk. Skrifter 

I. Mat.-naturv. klasse 1921 (9), 1-155. 
Nordhagen, R. (1930). En botanisk ekskursjon i Eikisdalen. Bergens Mi:s. Arbrok 1930. 

Naturv. rekke 8. 
Nordhagen, R. (1931). Studien iiber die skandinavischen Rassen des Papaver radicatum 

Rottb., sowie einige mit denselben verwechselte neue Arten. Bergens Mus. Arbok 1931, 

Naturv. rekke 2, 1-50. 
Nordhagen, R. (1935). Om Arenaria humifusa Wg. eg dens betydning for utforskningen av 

Skandinavias eldste floraelement. Bergens Mus. A bok 1935, Naturv. rekke 1, 1-185. 
Nordhagen, R. (1936). Skandinavias fjellflora eg dens relasjon til den siste istid. Nord. 

{19. ska d.) naturforskarnuitet i Helsingfors 1936, 93-124. 
Nordhagen, R. (1940). Staurene ved Ofjordnaeringen pa Soroya. Et eldgammelt strandniva 

i Vest-Finnmark. Norsk Geogr. Tidsskrift 8 (4), 124-155. 
Nordhagen, R. (1952). Bidrag til Norges flora II. Om nyere funn av Euphrasia lapponica 

Th. Fr. fil. i Norge. Blyttia 10, 29-50. 
Nordhagen, R. (1954). Some new observations concerning the geographic distribution and 

the ecolcgy of Arenaria humifusa Wg. in Norway as compared with Arenaria norvegica 

Gunn. Bot. Tidsskr. 51, 248-262. 
Sernander, R. (1896). Nagra ord med aniedning av Gunnar Andersson : Svenska Vaxtvarl- 

dens Historia. Bot. Notiser 1896. 
Undas, I. (1938). Kvartarstudier i Vest-Finnmark og Vesteralen. Norsk Geol. Tidsskr. 

18(2), 81-217. 
Undas, I. (1942). On the Late-Quaternary history of More and Trondelag. Kgl. Norske 

Vidensk. Selsk. Skrifter 1942 (2), 1-92. 
VoGT, Til (1912). Landskapsformene i det ytterste av Lofoten. Norsk Geogr. Selsk. Arbok 

WiLLE, N. (1905). Om invandringen af det arktiske floraelement til Ncrge. Nyt Mag. f. 

Naturv. 43 (4). 
WiLLE, N. (19 1 5). The flora cf Norway and its immigration. St. Louis, Missouri, Bot. Gard. 

Annals 1915 (2), 59-108. 


In August 1962, Nordhagen and S erre Lokken succeeded in finding Rhododendron 
lapponicum on still another mountain in western Lesja (Kvaerhusho, southwest of 
RaLakollen; cf. p. 255 and Fig. 11). 



Royal Norwegian Society of Sciences^ Trondheim, Norway 

During the 80 years that have passed since Axel Blytt (1876) set up his 
theory about an Interglacial element in the mountain flora of Scandinavia, a 
very comprehensive record of supporting biological arguments has been ac- 
quired. From the point of view of a botanist, the distribution of numerous 
species cannot be satisfactorily explained without assuming their survival 
during at least the Last Glaciation on refugia along the west coast of southern 
as well as of northern Norway. I will not here resume the arguments, nor the 
discussion, that are still going on between biologists and disbelieving geologists. 
I take it for granted that a great part of our alpine vegetation is of an Inter- 
glacial age. 

The question of where the plants might have survived has been subject to 
much discussion. The main view has been that the plants persisted on refugia 
close to the coast, partly outside the present coast Hne on the continental 
shelf not submerged during the Ice Age. The shelf is narrowest outside More 
in southern Norway and Troms-Finnmark in northern Norway. From Green- 
land and other Arctic areas we know that the alpine plants are descending 
toward sea level. 

At the same time it is evident — to judge from the present situation in 
Greenland — that if coastal refugia have existed in areas adjacent to high 
mountains in the neighborhood, these mountains must have protruded from 
the ice-shield as nunataks. Would it have been possible for alpine plants to 
persist in the nunatak areas? 

In his epoch-making paper "Essay on the immigration of the Norwegian 
flora during alternating rainy and dry periods", Blytt (1876) pointed out a 
group which he called "Arctic plants" occurring in northern as well as in 
southern Norway, particularly in the continental mountains. 

From the work of Th. C. E. Fries (1913), "Botanische Untersuchungen in 
nordlichsten Schweden", the theory of survival got its cartographical founda- 
tion. He showed that many important species in the Scandinavian alpine 
flora have a centric distribution (Fig. 1). About 25 species display a biceutric 
distribution occupying one area in southern Norway and one in northern 
Scandinavia. The unicentric ones are found only in one of these areas, about 










10 (species and subspecies) in the southern one and about 30 in the northern 
one. In numerous papers (Nordhagen, 1931, 1935; Nannfeldt, 1940) strong 
evidence is given for the probability that this centricity must have a historical 
explanation. Briefly recapitulated, this explanation is to the effect that there 
must have been two areas where the glaciation has not been total: one area of 
ice-free refugia in the district of More, another from about the Polar Circle to 
western Finnmark. Owing to the reduced ability and possibility of dispersal of 
some of the species a centric distribution has been formed. By far the greater 
part of the problematic species of the Scandinavian alpine flora is found in 
these two areas. 

Fig. 2. The total distribution of Rhododendron lapponicinn (inch R. parvifolium) 

(after Hulten, 1958). 

Blytt also indicated an American-Greenlandic element in the Scandina- 
vian flora. Today the species belonging to this element in Scandinavia are 
called West Arctic. They total about 30 taxa most of which belong to the 
centric groups. Nearly all of them are found in the northern area, but only 
one-third in the southern one. 

Rhododendron lappouicum in Europe is only known from Scandinavia 
where it occurs — often predominating — within a large area in the north, and 
a small one in the south, thus it belongs to the bicentric species (Fig. 2). 

Pedicidaris flammea is, in the same manner, very distinctly a West Arctic 
species, but it is restricted to the northern area (Fig. 3). 

There are also some eastern species displaying the same distribution. 



Fig. 3. The total distribution o^ Pedicnlaiis flammea (after Hulten, 1958). 

Artemisia norvegica 

Fig. 4. The total distribiitiovn of Artemisia norvegica (after Hulten, 1954). 



Artemisia norvegica is one of them (Fig. 4). It is known from the Urals, 
southern Norway, and a single locality in Scotland. This species is closely 
related to Artemisia arctica of eastern Siberia and North America. Platanthera 
oligantha occurs in the northern area only. As seen from the map the nearest 
locality is found in the Yeniseysk area (Fig. 5). 

Fig. 5. The total distribution o^ Platanthera oligantha (after Hulten, 1958). 

It should furthermore be mentioned that an Arctic element is represented 
in the southern area by Phippsia concinna without a single locality in northern 
Scandinavia. On the other hand there are representatives in the northern area 
of a European (or Eurasian) alpine element lacking locahties in the southern 
area. Scirpus pumilus (Fig. 6) and Antennaria carpatica have this distribu- 

The Scandinavian Alpine flora is poor in endemic taxa, and this has been 
taken as an argument against the theory of survival. Detailed taxonomic 
investigations have brought to light more endemics than previously assumed. 
In this connection we should notice that practically all the endemics belong to 
the centric groups. Nannfeldt (1940) has shown that within the polymorphic 
Poa arctica complex there are several endemic races, some of them southern 
unicentric, others northern unicentric. 

The alpine poppies belonging to the Papaver radicatum complex have on 
the whole a bicentric distribution. From the investigations by Nordhagen 
(1931) it is elucidated that this complex is chiefly represented by southern and 
northern unicentric endemics. I will return later to the problem of the southern 



Norwegian Papa ver-populations as seen in the light of recent cytological 
investigations (Fig. 7). 

This concentration of rare species in two widely separated areas cannot be 
accidental. As we have seen, the species belong to entirely different geographi- 
cal elements, partly they are West Arctic, partly Arctic-Alpine, partly Asiatic, 
partly endemic. 

Fig. 6. The total distribution of Scirpus pumilits (after Hulten, 1958). 

With regard to the poppies, Nordhagen (1931) has drawn the conclusion 
that their peculiar distribution cannot be explained by Post-glacial migration, 
but more likely is the result of survival in separate refugia along the west 
coast of Norway. The present distribution of the poppies indicates that there 
must have been several refugia. This leads to still another question: is it 
possible from the present-day distribution of the species to delimit the refugia 
more exactly? We know that plants are able to disperse — in some cases very 
rapidly — but we also know, for instance from the Alps, that relic species may 
be so depauperated that their ability to disperse is strongly reduced. 

In 1935, Nordhagen gave a very striking example indicating a refugium in 
the district of Nordland. I want to recall this example, supplemented by some 
recent observations. The map (Fig. 8) shows the concentration of a number of 
rare species east of the Salten Fjord in the area of the Swedish-Norwegian 

1. A short distance south of this area there is a peculiar Papaver locality, a 
local endemic, Papaver radicaUnn ssp. siihg/ohosiini. 



2. The West Arctic species Areimria humifusa is known from several locali- 
ties in the Sarek Mts. on the Swedish side of the border and, so far, from a 
single, isolated locality farther south. It is a northern unicentric species. 
According to Polunin (1943) the nearest locality is on Spitsbergen, and beyond 
that no closer than east Greenland. 

Fig. 7. The distribution in southern Norway of Papaver ladicatum ssp. ovatilobwn 
(dots) and P. relictuin (crosses) (after Nordhagen 1936). 

3. Another West Arctic and northern unicentric species, Draba crassifolia, 
is also found in several localities in the same areas on both sides of the 

4. At Lake Balvatnet is found Saxifraga aizoon. In Norway it is found 
furthermore in a restricted area in the southernmost mountains (Ryfylke). 
Nordhagen (1936) has shown that the southern population is identical with 
the central European S. ai-oon, whereas the population at Lake Balvatnet is 
so different that it must be given the rank of a subspecies ; thus it is endemic 
to this area. The occurrence and the taxonomical difference can only be 
explained by the theory of survival. 

5. In 1942 the Swedish botanist Sten Selander discovered in the Sarek Mts., 



another West Arctic species, Potent ilia hyparctica, new to Scandinavia. In 
Eurasia this species was previously known from Kanin, Vaygach, Novaya 
Zemlya and Spitsbergen. 

Fig. 8. Important botanical indications of refugia in the district of Nordland, 
Norway. Arenaria huinifusa (dots); Papaoer radicatum ssp. subglobosiim (asterisks); 
Carex scirpoiclea (triangles); Saxifraga aizoon ssp. laestadii (squares); Draba 
crassifolia (crosses); and Potent ilia hyparctica (circles with cross) (partly after 

Nordhagen, 1935). 

6. Another interesting West Arctic species is the dioecious Carex scirpoidea, 
known from a single mountain in Saltdal, the only locality in Europe. It is 
widely distributed in Greenland and North America (Fig. 9). According to 
my own experience in Alaska it is a species with a very wide ecological ampli- 
tude. In the Norwegian locality Mt. Solvagtind, it grows in abundance, but 
within a very narrow ecological limit. It has not been able to disperse to the 
neighboring mountains where the conditions seem to be very suitable. It 
gives the impression of a depauperated relic, and I am inclined to believe 
that C. scirpoidea inhabits just the very area where it survived the glaciation, 
i.e. on the south-facing slope of the nunatak Mt. Solvagtind. According to the 
geologist Gunnar Holmsen, Mt. Solvagtind is supposed to have been a 



This strange concentration of rare species cannot be explained as casual. 
The geologist O. T. Gronlie (1927) has published numerous observations 
from northern Norway showing that there must have been quite extensive 
ice-free areas. He is of the opinion, among others, that ice-free areas have 
existed as far east as the boundary between Norway and Sweden, exactly in 
this area where the concentration of important species is found. 

Fig. 9. The total distribution of Carex sciipoidea (after Hulten, 1958). 

Let US return to southern Norway. As already mentioned, Blytt (1876) 
pointed out that his Arctic element was encountered exclusively within 
continental mountains consisting of calcareous rocks and schist. Our know- 
ledge of the distribution of the plants has increased gradually so that we now 
have a very detailed picture of it. A sensational observation might still be 
made, of course, but most likely the present picture of distribution will be 
but slightly altered. 

If we look at the distribution of the bicentric and the southern unicentric 
species, we see that they are restricted to the interior mountains, thus con- 
firming the theory of Blytt (1876). He maintained, indeed, that they were 
continental, and undoubtedly he had in mind the mountains of Dovre and 
Jotunheimen. Since then it has been proved that the mountains south of 
Sunndalen and the TroUheimen Mts. should be included in the "species-rich 
area". Particularly the mountains of TroUheimen are strongly influenced by 
oceanic weather conditions (high precipitation). 

Let us again look at Artemisia norvegica and its Norwegian distribution 



which is very concentrated, comprising the mountains of Dovre, Sunndalen, 
and Trollheimen (Fig. 10). Within this area, it is a very common species. 
This picture of distribution has not changed in the time of botanical explora- 
tions. One might infer that its delimitation is a geological one. Artemisia 
norvegica is undoubtedly most common on easily disintegrating schist, but is 
not merely restricted to that kind of soil. It is also commonly encountered on 
mineral soil, poor in lime. At the eastern and southern borders of this area 
there are schisty rocks everywhere, and accordingly a geological obstruction 

Fig. 10. The distribution o^ Artemisia norvegica in Norway. 

cannot be the delimiting agent. Though the geological situation is different at 
the western border, there are, according to my own experience, plenty of 
suitable localities. Nor may the altitudinal conditions, bearing in mind the 
Post-glacial Hypsithermal period, hamper the dispersal of the species. 

If we consider the centric species one by one, we will find the same distribu- 
tion. Among the West Arctic, bicentric species Campanula wiiflora is found 
in the area of Jotunheimen — Trollheimen, and so are also Sagina caespitosa. 



Caiex arctogena, and C parallela. The circumpolar, bicentric species Miim- 
artia rubella, Luzula arctica, Carex misandra, and Rammcuhis nivalis all have a 
similar distribution (Figs. 11-15). 

Furthermore, the same type of distribution is met with for some endemic 
species. Thus Taraxacum dovrense occurs in the area of Jotunheimen- 
Trollheimen (Fig. 16). Ecologically, if differs from all other alpine Taraxaca by 

Fig. 11. The distribution of Saginci caespitosa in southern Norway. 

growing in exposed localities, frequently together with Campanula uniflora. It 
belongs to the cf/T^/ca-group, being its only representative in Scandinavia. Its 
closest relative [T. reichenbachii) occurs in the Alps, and Dahlstedt, who 
described the species, expressed the opinion as early as 1928 that these two 
species originally were identical, occupying a common area before the Ice Age. 
Draba dovrensis has a similar, but somewhat wider distribution. Stellaria 
crassipes belongs to the bicentric element, and in the southern area it is 
represented by an endemic race, var. dovrense Hult. So far it is known from 
three localities only (Fig. 17). Normally, it seems to reproduce vegetatively. 





.2 fl 

■ - 3 

t o 



g %^ 

re 5 

:=r i J f n\ 





u b 

.2 c 

« o 
•— y) 









Regarding Nannfeldt's (1940) map of the different Poa arctica races, we 
find that ssp. depauperata occurs in the region from Jotunheimen to Troll- 
heimen, and outside this area it is known only from northwestern Iceland. 
In southern Norway, ssp. elongata is endemic, showing a somewhat wider 
distribution than the former (Fig. 18). 

Fig. 18. The distribution in southern Norway of the different races of Poa arctica 

(after Nannfeldt, 1940). Area of ssp. depauperata (solid line); area of ssp. elongata 

(broken line); localities of Poa stricta (dots). 


Nannfeldt {he. cit.) also includes the endemic, viviparous Poa stricta in 
the same complex. This race has an abundant and very concentrated occur- 
rence in the Dovre Mts. (Fig. 18). 

I have tried to show that the greater part of the problematic plant species 
of the south Scandinavian alpine flora is confined to the mountains some 
distance from the coast. Many of the plants are fairly evenly distributed 
within this southern area, but there are also striking irregularities. Rhododen- 
dron lapponicum and Braya linearis are known only from the northern part of 
Jotunheimen, Stellaria crassipes and Carex bicolor only from Dovre. Poa 
stricta has a small area in Trollheimen, but its main distribution is in Dovre, 
and it is not found at all in Jotunheimen. Ranunculus nivalis, Carex arctogena, 
and Draba nivalis do not occur in Trollheimen. 

Saxifraga hieraciifolia with a single coastal locality in the Romsdal Mts. 
has its main distribution in Sunndalen, Jotunheimen, and the southern part of 
Dovre. It is not known from the rich areas of central Dovre and Trollheimen. 

How is it then possible to explain this pecuUar distribution and the dis- 
junctions within the southern area? 

It has been assumed commonly that the species must have survived in the 
coastal districts of More. When the ice began to retreat, the plants followed 
the retreating ice to the mountains of Sunndalen, Trollheimen, Dovre, and 
Jotunheimen where they are met with today. The humid, oceanic cUmate is 
unfavorable to alpine species, due for instance to peat formation. Therefore, 
these species have suffered competition and disappeared. 

I myself have not been content with the assumption of only a coastal 
survival area. It is not possible to explain the total absence in the coastal 
mountains of all the species mentioned by referring to peat formation and 
soil conditions. It is a fact that there are high mountains close to the coast 
where neither peat formationn or climate may represent any danger to these 
species. I have had the opportunity to make some investigations in the Roms- 
dal Mts., and there are plenty of suitable localities, e.g. for Artemisia norvegica. 

There are also coastal mountains consisting of calcareous rocks and schist, 
where the calciphilous Dryas octopetala grows abundantly. This is usually the 
best indicator of "rich areas". In this connection Mt. Talstadhesten in outer 
Romsdal is very interesting. It is a limestone area and harbours a flora 
comparatively rich in species, but none of those mentioned above are present 
even though the geological conditions ought to ht suitable. Large areas are 
not at all subject to critical humus production. Most likely this mountain 
has been a nunatak, and, according to Nordhagen (1952), Euphrasia lapponica 
is represented here by a variety known only from this locality and the moun- 
tains of Hardangervidda. Furthermore, according to Porsild (1958) the 
Dryas growing very abundantly here is not D. octopetala s.str., but D. 
Babingtoniana, otherwise known only from the British Isles. 

I should like also to call attention to the fact that a snow-bed species like 


Phippsia algida is not found in the outer mountains either. Through my 
investigations of the snow-bed vegetation of the Scandinavian mountains I 
have observed that the extension of some communities varies from the 
continental mountains to the coastal ones. The sociological structure, how- 
ever, is strikingly identical. This means that the deciding factors are the dura- 
tion of the snow cover and the soil conditions. In view of this, the absence of 
Phippsia algida is remarkable. 

From Greenland we know that under certain conditions even nunataks 
may harbor a comparatively rich flora. From this fact the following question 
appears self-evident: Did inland nunataks also exist in Norway during the 
Ice Age ? 

Dr. N. A. Sorensen treated this possibility in a paper published in 1949. 
There is some disagreement regarding the position of the rim of the ice sheet 
during the maximum of the glaciations. Thorough investigations have been 
performed by the geologist Undas (1938, 1942). Taking into account his 
opinion of the ice border and further on the known normal gradient of a 
large ice-mass, a number of mountains in the district of More must evidently 
have been nunataks. Furthermore, there is good reason to believe that some 
of these nunataks offered the plants very tolerable living conditions. Ex- 
periences from Greenland have revealed that a "suitable" nunatak must 
consist of loose rocks and moreover there must be south-facing slopes 
strongly exposed to the heat of the sun. 

I want to draw attention to a certain area in the Trollheimen where I 
(at the start together with Dr. N. A. Sorensen) have had the opportunity to 
perform some investigations. The area in question is Gjevilvasskammene 
Mts. on the north side of Lake Gjevilvatnet (Fig. 19). From the altitude at 
this lake of 663 m, the mountains ascend with steep south-facing slopes and 
precipices to an elevation of 1640 m. The effect of this exposure is a growth 
period about one month longer than is normal for this area. 

The mountains are built up by loose schists, mainly mica schist. West of the 
Gjevilvasskammene Mts. the elevation of the lowest passes to the nearest 
fjords varies from 800 to 950 m. With an ice border as proposed by Undas 
{he. cit.) and a normal gradient of the ice, the height of the ice front and the 
rise of the land cannot have permitted an ice level in the Gjevilvasskammene 
Mts. higher than about 1350 m. This agrees very well with the observations 
made by the geologist A. Gronlie (1950) indicating an ice level of 1300-1350 m 
for this area. Other geologists are of a different opinion, maintaining that ice 
covered all the mountains in the Trollheimen area. 

Some peculiar soil layers constitute a special phenomenon to be found 
situated on the tops and highest ridges of these mountains. Apparently 
these layers are now subject to heavy wind erosion (cf. Figs. 20, 21). Several 
geologists have examined the layers, and it has now been sufficiently demon- 
strated that they consist of soil disintegrated in situ. By means of X-ray 



Fig. 19. The position of the Gjevilvasskammene Mts. in southern Norwav. 

Fig. 20. From Gjevilvasskammene Mts. The soil layer is situated on the Hat top 

of the mountain. 



examinations Dahl (1961) has proved that the soil contains the minerals 
vermiculite and hydrobiotite formed by an advanced disintegration of 
biotite. Dahl is of the opinion that the degree of disintegration is much higher 
than that hitherto known for the biotite in the lowland of southern and central 
Scandinavia. As this disintegrated soil rests in such a strongly exposed 
position it does not seem likely that it could have been able to resist the 
erosion of a thick ice sheet. Accordingly, the soil layers also indicate nunataks 
in this area. 

■ kjiiiiii* 

Fig. 21. Photo of the soil layer on the highest peak of the Gjevilvasskammene 
Mts. at 1640 m altitude. 

Botanical investigations have revealed a very rich and interesting flora. 
Artemisia norvegica is extremely common. Among the bicentric species 
should be mentioned Luzula arctica, Nigritella nigra, Carex paral/ela. C. 
misandra, Sagina caespitosa, Papaver radicatimu Draba fladnizensis, D. 
lactea, Miimartia rubella, Poteutilla nivea, and Euphrasia lapponica. Besides 
Artemisia norvegica there are several southern unicentric species such as 
Taraxacum dovrense, Poa arctica ssp. elongata and ssp. depauperata, and 
Pedicularis oederi. Another important species is Arenaria norvegica which has 
played an important part in the survival theory because of its strange distribu- 
tion. Outside of Norway, it is known from northern Scotland, the Orkney 
Islands and Iceland. The same distribution, with the addition of southern 
Greenland and Labrador, is typical of Poa flexuosa which occurs also in the 
Gjevilvasskammene Mts. 

Thus, there is again a remarkable concentration of species which have 
played a great part in the survival theory. Tt is necessary to be very careful 


when drawing parallels with the present situation in Greenland, but it is by 
no means an impossible hypothesis to postulate an inland nunatak refugium 
in the Gjevilvasskammene Mts. The steep south-facing slopes might have 
offered sufficiently favorable living conditions for a hardy flora. If this 
theory be correct, it gives us a much more likely explanation of the distribu- 
tion and migration of many species than does a theory confining the plants to 
a coastal survival area only. 

It should be added that the geologist H. Reusch proposed as early as 
1884 that mountain areas in the central and western parts of southern Norway 
must have protruded from the ice sheet as nunataks. Recently Dahl (1961) has 
drawn attention to the "felsenmeers" (= mountain top detritus) of the 
mountain tops. He is of the opinion that on the basis of the degree of disinte- 
gration, the lower "felsenmeer" border corresponds to the surface of the in- 
land ice. If this hypothesis proves correct, many and large nunatak areas 
have existed in southern Norway during the Ice Age. 

Concerning the possibiUties of higher plants persisting in a nunatak area, 
several geologists have expressed doubt that the climatic conditions would 
permit any kind of plant life. Observations from Greenland nunataks confirm 
that higher plants are able to persist there. Furthermore we know of a fairly 
rich flora from the tundra district of Peary Land. Holmen (1957) reported 
96 phanerogams from this area. According to Koch (1928), the mean annual 
temperature is 20°C below zero. For about 100 days the temperature hovers 
around — 40°C. The precipitation is mainly in the form of snow, and amounts 
to less than 100 mm. In this connection it is important to state that the sum- 
mer temperature rises only to some few degrees above freezing (0°C); in 
July it has an average of ca. 6°C. 

In North America it is now generally accepted that plants as well as 
animals found a refugium in the Canadian Arctic Archipelago. It would be 
very unlikely that more severe conditions existed on coastal mountain 
refugia in Norway than in the tundra refugia of Arctic Canada. 

I have mentioned some of the detailed taxonomical investigations per- 
formed by Nordhagen (1931, 1935, 1952) and Nannfeldt (1940) and their 
great significance for the survival theory. In this connection I also want to 
recapitulate the recent cytological studies of Papaver radicatum performed by 
Dr. Gunvor Knaben (1959). During a number of years she crossed poppies in 
order to investigate variation within the Papaver radicatum complex. The 
evolution within many genera most certainly corresponds to structural 
chromosomal alterations. If the parent plants have the same chromosome 
number but a different chromosome structure the result of the cross gives an 
irregular meiosis. 

Dr. Knaben crossed poppies from a number of localities in southern 
Norway and she reached the following conclusion: in southern Norway there 
are six different races of Papaver radicatum. They all have a ditTerenl 



chromosome structure and at the same time they are also phenotypically 
distinct. As seen from Fig. 22, their areas as well are separated : 

Fig. 22. The distribution in Norway southern of the different races of Papaver 
radicatiim, based on the recent investigations by G. Knaben. Explanations in the 


Papaver radicatum ssp. 

(1) relictum, Valdres-Sogn 

(2) intermedium, Jotunheimen 

(3) ovatilobum. Central Dovre 

(4) gjaerevollii, Trollheimen 

(5) groevudalense, Sunndal Mts. 

(6) oeksendalense, Oksendalen 

Within the different races, the populations display shght alterations 
in their chromosome structure, but Dr. Knaben is of the opinion that this is 
due to recent changes. Furthermore, she has shown that the differences in 


chromosomal structure between the various south Norwegian races equal 
the differences between races of southern Norway, northern Norway, and 
the Faeroes as well as Iceland. Consequently, the south Norwegian races 
must have been separated during a period of about the same duration as 
that which divided the different races in the area surrounding the North 
Atlantic Ocean, i.e. they must have been separated for a period much longer 
than that of the Post-glacial time. 

The studies performed by Dr. Knaben as well as her conclusion strongly 
support the hypothesis of nunatak survival. The distribution of the different 
races cannot be explained by referring them to coastal survival areas with a 
subsequent group migration to their present locahties. 


Arwidsson, T. (1943). Studien iiber die Gefasspflanzen in den Hochgebirgen der Pite 

Lappmark. Acta Phytogeogr. Siiec. 17, 1-274. 
Blytt, a. (1876). Essay on the Immigration of the Norwegian Flora during Alternating 

Rainy and Dry Periods. Christ iania. 
Dahl, E. (1946). On diflferent types of unglaciated areas during the Ice Ages and their 

significance to plant geography. New Phytol. 45, 225-242. 
Dahl, E. (1954). Weathered gneisses at the island of Runde, Sunnmore, western Norway, 

and their geological interpretation. Nytt Mag. for Bot. 3, 5-23. 
Dahl, E. (1961). Refugieproblemet og de kvartoergeologiske metodene. Svensk Naturvet- 

enskap 14, 81-96. 
Dahlstedt, H. (1928). De svenska arterna av slaktet Tara.xacum. Kgl. Sv. Vetensk. Akad. 

Handl. Ser. 3, Bd 6 (3), 1-66. 
Fries, Th. C. E. (1913). Botanische Untersuclmngen in nordlichsten Schweden. Akad. 

Abhandl. Uppsala & Stockholm. 
Gronlie, a. (1950). Pa kvartaergeologisk ekskursjon i Trollheimen. Trondhjems Turist- 

forenings Arbok 1950 
Gronlie, O. T. (1927). The Folden Fiord. Quaternary Geology. Tromso Museums Skr. 

I (II), 1-73. 
HoLMEN, K. (1957). The vascular plants of Peary Land, North Greenland. Medd. om 

Grord. 124 (9), 1-149. 
HuLTEN, E. (1954). Artemisia norvegica Fr. and its aUies. Nytt Mag. for Bot. 3, 63-82. 
HuLTEN, E. (1958). The Amphi-Atlantic plants and their phytogeographical connections. 

Kgl. Sv. Vetensk. Handl. Ser. 4, 7(1), 1-340. 
Knaben, G. (1959). On the evolution of the radicatum-group of the Scapiflora Papavers as 

studied in 70 and 56 chromosome species. Part A. Opera Botanica 2 (3), 1-74. 
Koch, L. (1928). Contribution to the glacioiogy of North Greenland. Medd. om Grunl. 

65, 181-464. 
Nannfeldt, J. A. (1940). On the polymorphy of Poa arctica R.Br, with special reference 

to its Scandinavian forms. Symb. Bjt. Upsaliensis Jy{4), 1-86. 
NoRDHAGEN, R. (1931). Studien iiber die skandinavischen Rassen des Papaver radicatum 

Rottb. sowie einige mit denselben verwechsehe neue Arten. Bergens Mus. Abok 1931. 

Naturv. rekke 2, 1-50. 
Nordhagen, R. (1935). Om Arenaria hiimifusa Wg. og dens betydning for utforskningen 
av Skandinaviens eldste floraelement. Bergens Mus. Arbok 1935. Naturv. rekke 1, 

Nordhagen, R. (1936). Skandinavias fjeliflora og dens relasjoner til den sisle istid. Nord. 

(19. Skand.) Naturforskarm. i Helsingfors 1936, 93 124. 
Nordhagen, R. (1952). Bidrag til Norges flora. II. O.n nyere fund av Fiiplrrasia lapponica 

Th. Fr. hi. i Norge. BIyttia 10, 29-50. 


PoLUNiN, N. (1943). Geographical distribution oi Arenaria huinifiisa Wahlenb., new to the 

flora of Spitsbergen. Nature 152, 451-452. 
PoRSiLD, A. E. (1958). Divas Babingtoniana, nom. nov. An overlooked species of the British 

Isles and Western Norway. Dept. Northern Affairs and Nat. Resoinc. Bull. 160, 133-145. 
Reusch, H. (1887). Tstiden i det vestenfieldski Norge. Nyt Mag for Naturv.lS, 161-170. 
Selander, S. (1942). Potentilla emarginata Pursh i Sverige. Bot. Notiser 1942, 69-74. 
SoRENSEN, N. A. (1949). Gjevilvasskammene — nunatakker i Trollheimens midte? Nat men 3. 
Undas, I. (1938). Kvartccrstudier i Vestfinmark og Vesteralen. Norsk Geol. Tidsskr. 18, 

Undas, I. (1942). On the Late Quaternary history of More and Trondelag. Kgl. Norske 

Vidensk.-Selsk. Skr. 1942 (2), 1-92. 


Tyge W. Bocher 

Institute of Plant Anatomy and Cytology, University of Copenhagen, Copenhagen, 


Greenland is undoubtedly the most important working field for Danish 
botanists. Its phytogeography has always been of international importance. 
Although our present knowledge is comparatively large, new problems seem 
continuously to arise. The phytogeographical work going on is partly 
floristic, partly ecological or palynological (Bocher, 1954, 1963; Bocher and 
Laegaard, 1962; Fredskild, 1961; Holmen, 1957; Iversen, 1953; Schwarzen- 
bach, 1961; Sorensen, 1953). Parallel to the field investigations, the species 
that are the bases of all phytogeographical discussion are examined separately 
by means of experimental cultivation and cytological studies. The survey 
given below of the phytogeography of Greenland deals primarily with the 
floristic problems; a few particularly important questions about the vegeta- 
tion will also be touched. 

In Greenland there are now known about 500 native vascular plant species 
of which about 50 per cent are circumpolar; about 1 14 species have their main 
area west of Greenland and 82 east of Greenland, whereas 35 (mostly apo- 
micts) are endemic. Southwest Greenland is richest in species, having 335 
native taxa, but even in northernmost Greenland the flora comprises 101 
species (for further details see Bocher, Holmen and Jakobsen, 1957, 1959; 
Holmen, 1957). 

It has been possible to divide Greenland into a number of natural floristic 
provinces and districts which are shown in Fig. 1 . Their boundaries are placed 
along the area-limits of 282 species whose distribution has been mapped 
(see Bocher, 1938, 1963). The most important floristic boundary is undoubted- 
ly that cutting off the provinces SW.-S.-SE. from the rest. This boundary 
has been studied since 1932, and as it seems to have more than local importance, 
it will be dealt with in some detail. 

One of the main results of the early studies in middle east Greenland 
(Bocher, 1933, 1938) was a demonstration of a distinct floristic boundary 
on the Blosseville Coast about 68° 30-69° N. Lat. This delimitation is 
caused by the disappearance northwards of a flora connected with oceanic 
climatic conditions and a similar diminution of true Arctic species in a 




southerly direction. Mapping of the species involved showed that in west 
Greenland they behaved similarly, but at the same time it became clear that 
full understanding of the nature of the boundary in east Greenland and a 
corresponding one in west Greenland could not be obtained until the inland 

Fig. 1. On the left: floristic provinces and districts of Greenland. On the right: 
frequencies of western and eastern species indicated as per cent of the total native 
flora of each district. The first figure indicates western, the second figure eastern 
species. (From Bocher, Holmen and Jakobsen 1959.) 

areas of middle west Greenland had been explored adequately and until the 
world ranges of the species were better known. Now both requirements are 
fulfilled. The interior of west Greenland has been studied during several 
expeditions and the world ranges of the Greenland plants are now well 
illustrated, thanks to the works of Hulten (1950, 1958), Porsild (1955, 1957) 
Raup (1947), Tolmachev (1960) and others. 

Unglaciated middle west Greenland, which is broad enough to include 
areas of continental and maritime cHmatic types, is particularly well suited to 
phytogeographical studies. In fact, the detailed treatment of its flora gives us 
one key to a main phytogeographical division of the North Atlantic area. 

Climatic curves for precipitation and annual temperature range bend from 
the interior of southwest Greenland towards the northwest and sooner or 



later reach the coast. Those expressing a high degree of oceanity veer away 
from the land in south Greenland where a species like Jiincits squarrosus is 
present. Curves for a lesser degree of oceanity leave Greenland in middle 
west Greenland, but the northernmost ones cut off south Disko or penetrate 
into Disko Bay, reaching the mainland at Jacobshavn and Sargag. 


Fig. 2. On the left: eastern limits in middle west Greenland for southern species 
requiring humidity. On the right: percentage occurrence of Continental species 
(black), Oceanic-sylvicolous species (dotted), and arctic species (cross-hatched). 
Lines connecting 25 per cent Oceanic-sylvicolous and 25 per cent Continental 
species. (From Bocher, 1963.) 

It has been shown that the majority of southern species follow the oceanity 
curves. This means that most such taxa in Greenland have high humidity, 
much snow, and perhaps high winter temperature as their main requirements. 
They are not primarily dependent on high summer temperatures. This fact 
may be of great importance in the discussion of their survival in Greenland 
during the Last Glaciation. Fig. 2 on the left shows the position of the 
eastern limits of such species in middle west Greenland. Line No. 4 indicates 
the limit for many species, e.g. Thymus dnicei, Alchemilla alpina, A.filicauUs, 
Hieracium hyparcticum, H. sect, foliolosa, whereas No. 10 delimits another 
large group, e.g. Polyslichum louchitis, Leuchorchis albida. Phleum cowmiita- 
limi, and St ell aria calvcantha. 


A number of low-arctic snowbed-species, as well as species from the snow- 
protected heath vegetation are able to reach farther north, and they turn 
eastwards through the inland area at Nordre Stromfjord-Arfersiorfikjord to 
the Disko Bay mainland (Lines 11-19 in Fig. 2). They are absent from the 
extremely continental interior at the head of Sondre Stromfjord or, if present 
here, are montane and rare because they are restricted to certain rare types of 
habitat. In this group we have Phyllodoce coemlea, Harrimanella hypnoides, 
Sibbaldia procumbens, Salix herbacea ; among dry soil species Juncus trifidus 
joins this group, and a species Uke Luzula spicata shows a marked decrease 
in frequency in the continental pocket at the head of Sondre Stromfjord 
(Bocher, 1963). 

The world ranges of the species which follow the curves in Fig. 2 (left) are 
either North Atlantic montane species or sylvicolous boreal species. No true 
Arctic species behaves in this way. There is, however, one type of Arctic 
distribution which shows some relation to the southern oceanic type. Some 
few true Arctic species, near their southern limits in middle west Greenland, 
are found only in the coastal mountains. Being dependent on low temperatures 
and some humidity they are exclusively montane in the south though 
only in the coastal mountains where there is enough snow. This applies 
to Potentilla hyparctica, Erigeron erioceplialus, and Melandrium apetaJum 
ssp. arcticum. Finally, some medium Arctic species are distributed in a 
similar way, being confined to the mountains surrounding the driest 
inland areas. As examples may be mentioned Ranunculus nivalis, Antennaria 
intermedia, A. glabrata, Draba crassifolia, and Erigeron Immilis. 

A distributional pattern opposite to that of the southern, humidity requiring 
plants is found in a number of Arctic-Continental species, some southern 
species which are dependent on high summer temperature, and some which 
require certain edapliic conditions connected with a dry climate. A map show- 
ing the western Umits of these species is shown on the left in Fig. 3. On the 
right in the same figure is given the distribution of the most important mem- 
ber of the Arctic-Continental species, Care.x supina ssp. spaniocarpa. There is 
a concentration of species at the head of Sondre Stromfjord. This is mainly 
the result of the edaphic conditions here. The chmate causes the formation of 
ultrabasic soils on dry slopes with soil-water evaporation from the surface or 
along salt lakes formed in bowl-shaped depressions or valleys with no out- 
flow to the sea. The most important species here may be Braya novae-angliae, 
B. linearis, Primula stricta, Gentiana detonsa var. groenlandica. Ranunculus 
pedatifidus (not the Arctic subspecies). Among species extending their 
ranges to the larger part of the cross-hatched area in Fig. 2 (left) there are, 
for example, Draba lanceolata, Pedicularis labradorica, Antennaria affinis. 
Car ex boecheriana, and related forms (see Bocher, 1 963 ; Bocher and Laegaard, 

The southern Boreal plants which in the north are confined to this inland 



area (particularly the pocket at the head of Sondre Stromfjord) include not 
only many helo- and hydrophytes, e.g. Potamogeton gramineus, Scirpus 
pauciflorus, Jwicus alpinus, J. ranarius, but also a species like Arctostaphylos 
uia-wsi (American variety). These southern species are with few exceptions 
absent from east Greenland but occur — again with few exceptions — in the 
interior of south Greenland. 

Fig. 3. On the left: western limits for a number of Continental species in middle 

west Greenland. On the right: one of the most important members of this group: 

Caiex supina ssp. spaniocarpa. (From Bocher, 1963.) 

In middle east Greenland almost the same species as shown in Fig. I (on 
the left) are found northwards up to the boundary on the Blosseville Coast or 
in the northern part of the Angmagssalik district. Many of the Arctic species 
occur north of these limits and inland at Kangerdlugssuaq (68° 10'-68'' 30') 
or inland at Angmagssahk (Bocher, 1938). 

Calculations of the percentage occurrence of distributional types in a 
number of local floras in middle west Greenland are summarized in Fig. 2 
(on the right). The most important groups of distributional types, viz. the 
Arctic, the Continental, and the Oceanic-sylvicolous, are shown. Only north 
of line 10 in Fig. 2 (on the left) and not south of Disko Bay are there floras 
with a majority of Arctic species. Middle west Greenland constitutes a 
transitional zone between a flora which is Arctic (Low-medium Arctic) and a 
flora which greatly approaches a Boreal one. The lack of forest in southwest 


Greenland is perhaps more a result of climatic oceanity which prevents tree 
growth but does not present any obstacles for such forest floor species as 
Coptis trifolia, Pyrola minor, Orthilia secunda ssp. obtusata, Linnaea borealis 
ssp. americana, or a Boreal muskeg species such as Ledum groenlandicimi. 

The two hues in Fig. 2 (on the right) connect points (local floras) with more 
than 25 per cent Oceanic-sylvicolous species (western line) and more than 25 
per cent Continental species (eastern Hnes). South of Holsteinsborg these two 
curves overlap but they diverge northwards in the inland archipelago of 
Nordre Stromfjord where the mountains are lower and the oceanic air 
masses sometimes are able to penetrate into the inland areas (Bocher and 
Loegaard, 1962). 

The old question about the occurrence of American and Eurasiatic flora 
elements in Greenland was recently considered (Bocher, Holmen and Jacobsen, 
1959; Lindroth, 1960; Bocher, 1963). How to classify eastern and western 
species is always a matter of opinion because some of them are Amphi- 
Atlantic but with almost equally large areas on both sides. In such cases, 
however, an increased variability on one of the sides may give additional 
valuable criteria. But, although some of the species tabulated as eastern or 
western were removed, the material is sufficient to show that European 
plants play a subordinate role in all parts of Greenland except southeast and 
middle east Greenland where western and eastern species are present in 
almost equal numbers, although not with the same frequency. The eastern 
species are obviously more abundant (Bocher, 1938), a fact which has a 
connection to their climatic requirements. Most of the eastern plants are 
clearly montane North Atlantic (sometimes also North Pacific) and absent 
from the driest tundra areas of northernmost Asia and Canada. Ecologically 
they demand snow in great quantities and high humidity. Such conditions 
are found in the Alps, Norway, Scotland, the Faeroes, Iceland, and the 
southern part of Greenland (except for certain inland areas). These species 
have been able to migrate from Europe to North America via the Atlantic 
islands which are washed or influenced by the Gulf Stream and therefore 
have a suitable climate. On the other hand, the majority of American species 
require a continental chmate. They occur in north Greenland and southwards, 
mainly inland. The American species came mostly from Arctic-Continental 
areas and hence many of them were unable to penetrate to south Greenland 
with its maritime chmate. Some of them (11 per cent) were High Arctic and 
were able to reach east Greenland through the north Greenland barrens. 

This diff'erent behavior of western and eastern species in Greenland is 
evidenced through the calculations summarized in Table 1. Most important is 
the fact that U-distributions (W.-S.-E.) are most frequent among the European 
species and (l-distributions (W.-N.-E.) among the American species. 

There are, of course, deviations. Thus, Braya linearis, Draba sibirica, and 
Fotentilla stipularis exemplify eastern species which have reached continental 



Table 1 . Greenland distribution of Western and Eastern species 
Data from Bocher, Hoi men, and Jakobsen (1959) 

Area type within Greenland 

W + 

W + 

E + 

E + 













Per cent of "] 
Western , 

species f 












(114) J 

'Including 1 

Per cent 

Taraxa- > 










cum J 

Eastern -; 




Taraxa- V 









L ^ 

1) U: Distribution W.-S.-E.; (1: Distribution W.-N.-E.; C : With a gap in East Green- 
land; ( ): W. and E., not N. and S.; O: Circum Greenlandic distribution. 

areas in Greenland, whereas western species such as Hiewchloe orthantha, 
Carex stylaris, C. deflexa, and Veronica wormskjoldii reach areas of the 
maritime type in Greenland. 

There has been much discussion about the time of these migrations (Iversen, 
1953; Bocher, 1956). For the solution of the questions it is evidently of 
importance to realize that very many species of both elements, as well as the 
Circumpolar species in Greenland, have reached areas which are explained 
ecologically. Furtherinore, as already mentioned, many southern species do 
not require high summer temperatures. 

The majority of true Arctic species occurs also on nunataks (Schwarzen- 
bach, 1961) and may be very old members of the Greenland flora which 
arrived during Interglacial times or even during the Late Tertiary and sur- 
vived in unglaciated areas. 

Nothing, however, would in my opinion prevent an early (Interglacial) 
migration of the montane North Atlantic species to Greenland and westwards. 
They have had long periods during which they became established in west 
Greenland (e.g. Angelica archangelica) and reached eastern America (e.g. 
Saxifvaga stellaris). Some of them even seem to have suffered extinction in 
certain areas of Greenland. Thus, north of south Greenland a species like 
Ranunculus acris is now very rare and is found only in natural vegetation at 
places where hardly any Norsemen have lived (Kangerdluarsuk ungatdleq, 
stations at Angmagssalik), and where the topography indicates that during 
the Last Tee Age only local glaciation has taken place. 



It is probable that a large number of south-facing slopes, during a very 
long period of time and also during the last glaciation, have served as bases 
for migration. We know such coastal areas today (e.g. the Blosseville Coast) 
which are composed of half-nunataks separated by big glaciers reaching the 
sea, and we find in such places many southern species of the montane North 

Ranunculus glacialis 
O Pedicularis hirsuta 

Fig. 4. On the left: areas in middle West Greenland with mountains more than 
900 m high and the distribution of Anemone richardsoni (dots). Only the areas 
occupied by this species have sharp peaks or ridges. On the right: eastern limit 
of smoothed mountain form and the southernmost stations in east Greenland for 
Ranunculus glacialis and Pedicularis hirsuta. (From Bocher, 1956, 1963.) 

Atlantic type. But half-nunataks which had suitable localities during the 
Last Glaciation are probably found only where high mountains reach the sea. 
There are large areas of this kind in east Greenland, including the Cape Far- 
well area in the south. In southwest Greenland one small coastal high moun- 
tain area is found at Arsuk and a large one between the two Isortoq fjords 
(66° 30-67° 30'), the latter only interrupted by a rather small area south of 
Holsteinsborg. Anemone richardsonii is confined to the coastal part of the latter 
area in which the mountains are formed by local glaciers only (Fig. 4), and a 
species like Athyrium aJpestre is restricted to the southeast coast (including 
Cape Farwell) and the Arsuk area. These coincidences of plant occurrences 
and alpine topography suggest survival but cannot be regarded as a definite 
evidence for it (see Bocher, 1951, 1956, 1963, and Fig. 4). 

As a whole, Greenland has mostly been considered to have an Arctic flora. 
This is a result of the definition of the Arctic range as the area north of the 
timber-line. But, in the North Atlantic the timber-line depends also on oceanity 
and the species which in other areas have limits at the timber-line do not 
behave in this manner in the Atlantic area. From a plant geographical point 
of view the timber-line is more a physiognomical division than a floristic 



Fig. 5. Above: North Atlantic areas of a number of species forming the fioristic 
boundary in Greenland (Sihbaldia prociimbens (without hatching), Fhleiim 
commutatum (dotted area), Leiiconhis albida (vertical hatching), Alchemilla 
filicaiilis (horizontal hatching), and A. alpina (oblique hatching)). Below: North 
Atlantic area showing the annual isohyet 250 mm (broken line), and lines for mean 
annual temperature range of IS'^C (southernmost full line) and 25°C (northern- 
most full line). Dotted line indicates the 10°C. July isotherm. Continental Arctic 
area indicated by hatching and dotting. (From Bocher, 1954.) 

boundary. In this situation the boundary previously mentioned, between a 
flora which is mainly montane (connected with humid mountains south of the 
true Arctic areas) and a true Arctic flora (connected with the dry Arctic areas), 
seems to be much more significant. The northern limit of sylvicolous Boreal 
plants such as Coptis, Pyrola minor, Platanthera hyperborea, Deschampsia 
flexiiosa, Chamaenerion angusti folium, Cormis suecica, and some of the 
montane Sub-Arctic or Low Arctic species, e.g. Alchemilla alpina, Angelica, 
Gnaphalium norvegicum, etc., gives a much more interesting boundary line, 


It divides the North Atlantic area into a true Arctic (High Arctic) part, 
including Baffin Island, Ellesmere Island, northern Greenland, Spitsbergen, 
Franz Josephs Land, and northern Novaya Zemlya, and a southern Sub-Low- 
Arctic-Montane part ranging from southern Labrador to the southern part of 
Greenland, Iceland, the Faeroes, the Scottish Highlands, and the Scandina- 
vian Mountains. Connected with the true Arctic part are those continental 
areas warm enough for Boreal or Temperate plants which do not require 
oceanic humidity at their northern limits. Typical examples of such regions 
are found in the inland of middle west Greenland, inland at Scoresbysound, 
and undoubtedly in Canada and northern Russia. In northernmost Scandi- 
navia many places will approach this transitional type between Arctic and 
Boreal Continental floristic regimes (see Fig. 5). 

Many plant communities are distributed on one of two sides of this main 
boundary hne. In a previous paper (Bocher, 1954) an attempt was made to 
divide the southwest Greenland vegetation into two complexes, an Oceanic 
and a Continental. Ecologically and floristically the communities of these 
complexes are clearly closely related and merge into one another. A natural 
group of the Oceanic complex is formed by snow-bed vegetation, herb fields, 
snow-protected heaths of the Phyllodoce type, and willow copses with hygro- 
philous herb vegetation. A similar Continental complex consists of dry rock 
vegetation, steppe communities, dry heath vegetation, and willow copses with 
xerophytes. In spite of the interest, however, connected with the study of 
such complexes it is felt that the old dividing system using life-forms as 
primary criteria should not be abandoned. In a recent paper (Bocher, 1963) 
an attempt has been made to classify the middle Greenland vegetation using 
life-forms in the main division and distributional types in the first sub- 
division while dominance and floristic composition are used only in the defini- 
tions of the smallest entities, e.g. the plant sociations. 


Bocher, T. W. (1933). Phytogeographical studies of the Greenland flora. Medd. om 
Gronl. 104 (3), 1-56. 

Bocher, T. W. (1938). Biological distributional types in the flora of Greenland. Medd. om 
Gronl. 106 (2), 1-339. 

Bocher, T. W. (1951). Distributions of plants in the circumpolar area in relation to ecologi- 
cal and historical factors. J. Ecology 39, 376-395. 

Bocher, T. W. (1954). Oceanic and continental vegetational complexes in Southwest 
Greenland. Medd. om Gronl. 148 (1), 1-336. 

Bocher, T. W. (1956). Area-limits and isolations of plants in relation to physiography 
of the southern parts of Greenland. Medd. om Gronl. 124 (8), 1-40. 

Bocher, T. W. (1963). Phytogecgraphy of Middle West Greenland. Medd. om Gronl. 148 
(3) (in press). 

Bocher, T. W., Holmen, Kj., and Jakobsen, K. (1957). Grdnlands Flora. Copenhagen. 

Bocher, T. W., Holmen, Kj., and Jakobsen. K. (1959). A synoptical study of the Green- 
land Flora. Medd. om Gronl. 163 (1), 1-32. 


BocHtR, T. W. and L/tGAARD, S. (1962). Botanical studies along the Arfersiorfik Fjord, 

West Greenland. Bot. Tidskr. 58, 168-190. 
Fredskild B. (1961). Floristic and ecological studies near Jakobshavn, West Greenland. 

Medd. oin Groeiil. 163 (4), 1-82. 
HoLMEN, Kj. (1957). The vascular plants of Peary Land, North Greenland. Medd. oni 

Groenl. 124 (9), 1-149. 
HuLTEN, E. (1950). Atlas over vdxlenuis utbredniug i Norden. Faneiogamer och ormbunks- 

vdxter {AxXas of the distribution of vascular plants in N.W. Europe). Stockholm. 
HuLTEN, E. (1958). The Amphi-Atiantic plants and their geographical connections. Kgl. Sv. 

Vetensk. Akad. H ndl. 4, Ser. 1 (1), 1-340. 
IvERSEN, J. (1953). Origin of the flora of Western Greenland in the light of pollen analysis. 

Oikos 4, 85-103. 
LiNDROTH, C. (1960). Is Davis Strait— between Greenland and Baffin Island— a floristic 

barrier? Bot. Notiser 113, 129-140. 
PoRSiLD, A. E. (1955). The vascular plants of the western Canadian Arctic Archipelago. 

Nat. Miis. Canad. Bull. 135, 1-226. 
PoRSiLD, A. E. (1957). Illustrated flora of the Canadian Arctic Archipelago. Nat. Mus. 

Canad. Bull. 146, 1-209. 
Raup, H. M. (1947). The botany of southwestern Mackenzie. Saigentia 6, 1-275. 
ScHWARZENBACH, F. H. (1961 ). Botanische Beobachtungen in der Nunatakkerzone Ostgron- 

lands zwischen 74" and 75^ N. Br. Medd. oin Gronl. 163 (5), 1-172. 
SoRENSEN, Th. (1953). A revision of the Greenland species of Puccinellia Pari. Medd. om 

Groenl. 136 (3), 1-179. 
ToLMACHEV, A. I. (1960). Flora Arctica USSR. I. Acad. Sci. SSSR. Inst. Bot. Komarova. 

M oskwa-Leningrad. 


Eythor Einarsson 

Museum of Natural History, Reykjavik, Iceland 

The scientific investigation of the Icelandic flora started a little more than two 
centuries ago with the observations by Eggert Olafsson and Bjarni Palsson 
during their journeys through Iceland in the years 1752-57. In their travelogue 
(Olafsson and Palsson, 1 772) considerable information is given on floristics and 
about 130 species of vascular plants are mentioned. 

The first scientific treatise on the Icelandic flora, however, was written by 
the Dane O. Fr. Muller and pubhshed in 1770. It was based entirely on the 
investigations and collections made in Iceland by J. G. Konig in 1764-65, and 
in it 337 species of vascular plants are mentioned. The first Icelandic Excur- 
sion Flora was written by Oddur Hjaltalin and pubhshed in 1830. This flora 
was based only partly on Hjaltalin's own investigations, but mostly on a 
Danish manual by J. W. Hornemann (1821), which also contained descrip- 
tions of all Icelandic spermatophytes known at that itme. In his flora, 
Hjaltahn records 337 species of vascular plants from Iceland. In 1871, 
C. C. Babington pubhshed A Revision of the Flora of Iceland, based on his 
own investigations of the Icelandic flora and all previously published plant 
lists from Iceland, together with information extracted from the herbaria of 
Icelandic plants preserved in Copenhagen by Joh. Lange. In his paper, 
Babington states that the Icelandic flora is essentially European; only 62 
species are found which do not grow in the British Isles, nearly all the 
species inhabit Scandinavia and not more than 3 species are decidedly Arctic. 
Babington finally records 467 species of vascular plants as found in Iceland. 

During the years 1870-85, Chr. Gronlund pubhshed some papers on the 
Icelandic flora, the most important one being his Islands Flora (Gronlund, 
1881), where 357 species of Icelandic vascular plants are described. In a 
later paper Gronlund (1884) adds nine species to this number and states that 
the number of Icelandic vascular plants given by Babington (1871) and some 
other authors of previous Icelandic plant hsts is much too high. In this same 
paper, Gronlund (loc. cit.) compares the flora of Iceland to that of Greenland, 
Scandinavia and the Faeroes. His conclusions are practically the same as 
those of Babington, i.e. that the Icelandic flora is mostly north European 
and afl the vascular plants with the exception of five species are found in 


298 eyth6r einarsson 

Scandinavia. More than half of the Icelandic species are also found in Green- 
land, and in the Faeroes the situation is about the same. Because many of 
the north European species are found throughout the Arctic, a considerable 
part of the Icelandic species are also found in Siberia, Spitsbergen, and 
northern North America. The only endemic species of vascular plants in 
Iceland, according to Gronlund, is Carex lyngbyei, and in addition there are 
three varieties of more widely distributed species. 

Stromfelt (1884) mentions 371 species of vascular plants from Iceland, 
Like Babington and Gronlund, he considered that only five of the Icelandic 
species, or 1.5 per cent are not found in Scandinavia, while 35.7 per cent of 
the species are not found in Greenland, among them some of the most 
common Icelandic lowland plants; 39.5 per cent of the Icelandic species are not 
found in the Faeroes, and that the character of the Icelandic flora, thus, is 
almost entirely Scandinavian. 

In his statistical treatise on "The flora of Greenland, Iceland and the 
Faeroes", Warming (1888) mentions 417 species of vascular plants from 
Iceland and says that the Icelandic flora is a typical European one. He 
divides the flora of these countries into 20 groups. He counts more than one- 
third of the Icelandic plants, or 151 species in his Group 3, Temperate-zone 
species; 70 belong to Group 2, Boreal-zone species; 64 to Group 1, Circum- 
polar species; and 48 belong to his group 4, European-American species. 
Three Icelandic species have a western distribution (not found in Europe), 
whereas 74 have an eastern distribution (not found in America). 

During the last decade of the nineteenth century, the Icelandic flora was 
carefully investigated, mostly by Helgi Jonsson and Stefan Stefansson, and 
many papers were published on the subject. Helgi Jonsson (1896) records 435 
species of vascular plants from Iceland, but in a later paper (Jonsson 1905) 
he sets the number at 360. In his first paper on the Icelandic flora, Stefan 
Stefansson (1890) reports 423 species of vascular plants known there. His 
Flora Islands (Flora of Iceland, 1901), a valuable manual firmly based on his 
own investigations of the flora, however, contains only 359 species of vascular 
plants, 10 of which belong to the genera Taraxacum and Hieracium. 

Thus, Jonsson and Stefansson are even more sceptical than Gronlund 
regarding some of the species mentioned in the old plant lists and especially 
in Babington's paper, but never found again — and perhaps not found at all — 
in Iceland. 

In 1924, the second edition of Stefansson's Flora islands appeared, contain- 
ing 363 species of vascular plants, five "species" of Taraxacum and 43 
"species" of Hieracium excluded. Some few species, new to the Icelandic flora, 
had been added; some of the excluded species from the old plant lists had 
been rediscovered and some of the accidentally introduced species had been 
naturalized since 1901. 

Molholm-Hansen (1930) divided the Icelandic vascular plants (375 species. 


excluding most species of Taraxacum and Hieracium) into two main groups, 
A and E. Group A comprises Arctic and Subarctic species, having their main 
distribution near or north of the forest limit; they are common in Greenland, 
Spitsbergen, and on the Scandinavian mountains, but absent or occurring 
sporadically in more southerly countries. To this group belong 151 species or 
ca. 40 per cent of the Icelandic vascular plants. Group E comprises species of 
common occurrence in central Europe, which have their main distribution 
south of the forest limit. This group consists of 224 species, or ca. 60 per cent 
of the vascular plants of Iceland, about half of which are also found in Green- 
land. According to Molholm-Hansen, there is also some difference in the 
distribution of the groups in various parts of Iceland. In north Iceland 
60.3 per cent of the species belonging to Group A are common but only 
38.8 per cent of the species belonging to Group E, In southwest Iceland, on 
the contrary, 53.6 per cent of the A species and 41.5 per cent of the E species 
are common. 

Ostenfeld and Grontved (1934) mention about 390 species of vascular 
plants from Iceland, Taraxacum and Hieracium excluded. In 1942, Grontved 
published a treatise on the Icelandic flora, in which every species of vascular 
plants ever recorded from Iceland is mentioned, i.e. about 660 species. The 
author considered critically which of these really belong to the Icelandic 
flora. His conclusions are that about 400 species, Taraxacum and Hieracium 
excluded, should be considered as Icelandic. During the last 20 years, however, 
some of the species from the old plant lists, which were excluded by Grontved, 
have been rediscovered in Iceland. 

Askell Love (1945) records 425 species of vascular plants from Iceland, 
Taraxacum and Hieracium excluded, and finally in the third edition of Stefans- 
son's manual, Steindor Steindorsson (1948) mentions 429 Icelandic vascular 
plants, indigenous and naturalized, introduced species, with Taraxacum and 
Hieracium excluded. Besides, both Stefansson (1901, 1924), Ostenfeld and 
Grontved (1934), Grontved (1942), Love (1945), Steindorsson (1948) and 
other authors record a considerable number of accidentally introduced, but 
not naturalized species from Iceland. 

In a paper on the age and immigration of the Icelandic flora, Steindorsson 
(1954) is of the opinion that between 430 and 440 species of vascular plants, 
Taraxacum and Hieracium excluded, are native in Iceland. No sexually 
reproducing species can be considered as endemic, but some endemic varieties 
and subspecies of wider distributed species are found in Iceland. According to 
Steindorsson (1954), about 21 per cent {ca. 90 species) of the 430-440 species 
have been introduced to Iceland by man during the last 1100 years and are 
now more or less naturaHzed in the country. Steindorsson points out, as do 
all the previous authors, that the main part of the Icelandic vascular plants 
is European, mostly Scandinavian species. Only six of the species, i.e. 1.4 
per cent of the flora, are western and not found in Europe, whereas 106 

300 eyth6r einarsson 

species or almost 25 per cent are eastern and not found in America. He 
states further that 74 Icelandic species, or 17 per cent of the flora, are not 
found in the British Isles, whereas 184 species, or ca. 42 per cent are not found 
in Greenland. 

Love and Love (1956) set the total number of spermatophyte species in 
Iceland at 540, of which 387 are regarded as being definitely indigenous, the 
rest alien. They divide the indigenous species into five groups, or elements: To 
the Circumpolar element, comprising all species with a Circumpolar area of 
distribution and some species with an almost Circumpolar area of distribution 
but with a gap in their distribution in the Pacific region, belong 148 Icelandic 
spermatophytes. The majority of these Circumpolar plants do not show any 
variation from the representatives of these species in other countries. Some of 
them, however, show a closer relationship to the same species in Scandinavia 
than elsewhere, whereas others are more closely related to the populations in 
the British Isles. Still others are represented in Iceland by the same races as 
in Greenland and differ somewhat from the European ones. 

The second element is the bis- Atlantic, comprising species met with on both 
sides of the Atlantic, but not reaching very far inland in continental Eurasia 
or North America. It includes 1 13 Icelandic species. Some of these are found 
mainly west of Iceland, while others have their main area east of Iceland. The 
third element is the eastern one, comprising those species which have a 
European or Eurasiatic distribution; some of them however, are also found 
in western North America. There are 95 Icelandic spermatophytes in this 
element. Some of them show strong affinities to Scandinavian specimens 
of these species, but still others bear more hkeness to British specimens. 

The fourth element is the western one, comprising species with their main 
distributional area west of Iceland. The authors, Love and Love (1956) point 
out, that the majority of the 14 species which they classify as western, are 
probably not very old in Iceland, and some of them probably are the most 
recent, prehistoric invaders of the flora. Some of these 14 taxa are varieties 
or subspecies of wider distributed species. Finally, regarding the 17 taxa 
belonging to the endemic element, they are also for the most part races of 
species known from other regions, Circumpolar or European as, for example, 
Papaver, or belonging to genera reproducing by apomixis or autogamy, as 
Alchemilla and Euphrasia. The only exception is AlchemiUa faeroeensis, an 
endemic species growing in eastern Iceland and the Faeroes. 

The great majority of the Icelandic "species" belonging to Hieracium 
are endemic (Oskarsson, 1955), but for the most part they show a close 
relationship to Scandinavian, Faeroese, or British "species". 

According to my own opinion, the Icelandic vascular plant species number 
ca. 440, including naturalized, introduced species, but 116 "species" of 
Taraxacum and ca. 180 "species" o^ Hieracium excluded. 

About 10 of these 440 species are somewhat doubtful; they have only been 


found once in a single locality each and never again. During the last years, 
however, some few of the species mentioned from Iceland in the plant hsts 
from the nineteenth century, but excluded from the Icelandic flora by all later 
authors, have been rediscovered in Iceland. Besides that, about 140 accident- 
ally introduced species have been recorded from Iceland, and most likely 
some of them will later become naturahzed. About 290 of these 440 species, 
or ca. 66 per cent, are found also in Greenland; some of them are without 
doubt introduced in Greenland and they may be so in Iceland, too. Most of 
the Icelandic species, not found in Greenland, are European or Eurasiatic 
Boreal taxa. 

About 250 of the Icelandic plants, ca. 57 per cent are found also in the 
Faeroes. The majority of the Icelandic plants, ca. 85 per cent, grow in the 
British Isles as well, whereas the remaining 15 per cent comprise mostly 
Arctic species and some species with a western distribution. Not less than 426 
of the Icelandic species of vascular plants, or ca. 97 per cent, are found also 
in Scandinavia, a few of them, however, being represented in Iceland by 
other races or subspecies than in Scandinavia. Ten species, or 2.3 per cent 
of the vascular plants of Iceland, are western species, which are not found on 
the European continent nor in the British Isles. When the genera Taraxacum 
and Hieracium are excluded, the endemic element of the Icelandic vascular 
flora is very small, and is entirely composed of "species" of other genera 
reproducing by apomixis, like Alc/iemilla, some varieties and subspecies of 
taxa with a much wider distribution area, as well as some Euphrasia species. 

In his work on the Amphi- Atlantic plants, Hulten (1958) enumerates 278 
species. Of these, 146 belong to the Icelandic flora. EarUer, Hulten (1950) had 
divided all the vascular plants of northwest Europe into 48 distribution 
groups. It is possible to fit into these groups all the 426 Icelandic species also 
occurring in Scandinavia. Using Hulten's classification, the Icelandic flora 
can be divided as follows: 118 species, ca. 27 per cent of the Icelandic vascular 
plants, belong to the Boreal-circumpolar plants (Hulten's Groups 16, 17, 29, 
30, 31, 32, and 33); 90 species, ca. 20 per cent, belong to the European- 
Eurasiatic plants (Groups 12, 13, 14, 15, 25, 26, 27, 28), which are found in 
the Boreal region of Eurasia; 71 species, ca. 16 per cent, belong to the Arctic- 
alpine-circumpolar plants (Groups 6, 7, 8, 9), while 14 species, or ca. 3 per 
cent, belong to the Arctic-circumpolar plants (Groups 1 and 2). The Amphi- 
Atlantic plants (Groups 10 and 11), comprise 32 species, or ca. 1 per cent of 
the vascular plants of Iceland. European coastal, Atlantic, and Sub-Atlantic 
plants (Groups 18, 19, 20) comprise 19 species, 4.3 per cent; and Circumpolar 
coastal and Sub-oceanic plants (Groups 21 and 22) comprise 10 species, or 
2.3 per cent of the Icelandic vascular flora. Group 41, including plants with 
two or more widely separated areas of distribution, comprises 1 3 species, or 
ca. 3 per cent of the flora. Group 46, including species strongly dispersed by 
man, comprises 29 species in Iceland, or 6.6 per cent; and Group 48, including 


taxa with insufficiently known areas of distribution, comprises 12 species, 
or 2.7 per cent of the Icelandic vascular plants. Finally, as mentioned above, 
the 10 western species make up 2.3 per cent of the flora. 

As seen by this classification, more than half of the Icelandic vascular 
plants are species with a Boreal distribution, whereas Arctic-Alpine species 
comprise only ca. 33 per cent of the flora. Most of the Arctic ones have a 
Low-Arctic distribution, but High Arctic species are practically absent from 
Iceland. Those with an Eastern distribution are more than nine times as 
numerous as the Western species in Iceland. Many of the Icelandic vascular 
plants have an Oceanic or Sub-oceanic distribution, whereas typical Continen- 
tal species are practically absent from Iceland. 


Babington, C. C. (1871). A revision of the flora of Iceland. /. Linn. Soc. Botany 11, 1-68. 

Gronlund, C. (1881). Islands Flora. Kjobenhavn. 

Gronlund, C. (1884). Karakteiistik af Plantevfeksten paa Island, sammenlignet med 

Floraen i flere andre Lande. Naturhist. For. Festskr. 1884, 1-39. 
Grontved, J. (1942). The Pteridophyta and Spermatophyta of Iceland. Botany of Iceland 

4, 1-427. 
Hjaltalin, O. J. (1830). Islenzk grasafraedi. Kaupmannahofn. 
HoRNEMANN, I. W. (1821). Forsog til en d.Jisk oekonomisk Plant ehere. Forste Deel. 

Tredie, forogede Oplag. Kjobenhavn. 
HuLTEN, E. (1950). Atlas over vdxternas iitbredning i Norden. Stockholm. 
HuLTEN, E. (1958). The Amphi-Atlantic plants and their phytogeographical connections. 

Kgl. Svenska Vetensk. Akad. Handl. 4, No. 7 (1), 1-340. 
JoNSSON, H. (1896). Bidrag til Ost-lslands flora. Bot. Tidsskr. 20, 327-357. 
JoNSSON, H. (1905). Vegetationen i Syd-lsland. Bot. Tidsskr. 27, 1-82. 
Love, A. (1945). Islenzkar jurtir. Kaupmannahofn. 
Love, A. and Love D. (1956). Cytotaxonomical conspectus of the Icelandic flora. Acta 

Horti Gotoburg. 20, 65-291. 
MxJLLER, O. F. (1770). Enumeratio stirpium in Islandia sponte crescentium. Nova Acta 

Acad. Nat. Curios. IV, 203-216. 
Molholm-Hansen, H. (1930). Studies on the vegetation of Iceland. Botany of Iceland "i, 

Olafsson, E. og Palsson, B. (1772). Reise igiennem Island, etc. Soroe. 
OsKARSSON, I. (1955). Um undafifla. Ndttiirufr. 25, 72-86. 

OsTENFELD, C. H. and Grontved, J. (1934). The Flora of Iceland and the Faeroes. Copen- 
Stefansson, S. (1890). Fra Islands Vtekstrige I. Nogle "nye'' og sjaeldne Karplanter 

samlede i Aarene 1888-89. Vidensk. Medd. Naturhist. Foren. i Kjobenhavn 1890, 1-16. 
StefAnsson, S. (1901). Flora Islands. Kaupmannahofn. 
Stefansson, S. (1924). Flora Islands, 2. litgafa, aukin. Kaupmannahofn. 
Steindorsson, S. (1948). Stefansson, S.: Flora .slands, 3. litgafa, aukin. Akureyri. 
Steindorsson, S. (1954). Um aidur eg innflutning islenzku florunnar. Arsrit R:ektunarfel. 

Nordurlands 51, 3-23, 53-72, 101-115. 
Stromfelt, H. F. G. (1884). Islands karlvaxter, betraktade fran vaxtgeografisk och floristisk 

synpunkt. Ofvers. Kgl. Vetensk. Akad. Forh. 18S4 (8), 79-124. 
Warming, E. (1888). Tabellarisk Oversigt over Gronlands, Islands og Fasroernes Flora i 

1887. Vidensk. Medd. Naturhist. Foren. i Kjobenhavn 1887, 236-292. 



Steindor Steindorsson 

Akureyri College, Akureyri, Iceland 

One of the most striking features of the flora of Iceland is its paucity of 
species. According to Flora Islands (Flora of Iceland; Stefansson, 1948), only 
about 430 species of Pteridophyta and Phanerogams can be considered as 
native to the whole country. Without doubt some species have been added 
since that time, but the entire number does not noticeably exceed 440. To 
this number we can add the genera Hieracium and Taraxacum, but about 
190 species of Hieracium have been recorded from Iceland and a rather 
uncertain number of Taraxacum, since some specialists put the number at 20, 
others at somewhat more than 100. From the distribution of these genera, one 
could undoubtedly find certain features concerning the problem which is 
dealt with in the present paper, but I have omitted these genera here especially 
because we know so little about their real distribution (cf. Steindorsson, 1962). 

The total number of species, 440, however does not tell the whole story. 
After a closer investigation we find that a considerable number of these have 
been introduced by man since the first settlement of the country nearly eleven 
hundred years ago. In another paper I have produced some arguments to 
show that about 100 species have been brought into the country by human 
agency. Although this number after a closer study may prove too high, it is 
certain that up to 20 per cent of the flora of Iceland {Hieracium and Taraxa- 
cum excepted) is anthropochorous. 

Ahhough our knowledge of the origin of the flora of Iceland and its 
relationship to the floras of our neighboring countries is far from complete, it 
is obvious that it is more closely related to the flora of Scandinavia than to 
any other country. A considerable component of the flora is Circumpolar, or 
at least is found on both sides of the North Atlantic Ocean, and the American 
element is probably greater than hitherto generally considered. The cyto- 
logical investigations of Askell Love point in that direction. It is certain that 
the flora of Iceland does not differ remarkably from that of other countries 
around the North Atlantic Ocean in other respects than its poverty of species. 

Then there is the question whether a considerable element of it is com- 
prised of survivors which have overwintered at least the Last Glaciation or 


304 steind6r steind6rsson 

perhaps the whole Ice Age in more or less isolated refugia. If we consider the 
overwintering a reality, it remains for us to find evidence as to the existence of 
such refugia and where they are situated. 

As we know, the prevaiUng opinion in the last century was that both Scan- 
dinavia and Iceland were fully glaciated during the Ice Age, and consequently 
that all biota must have ceased to exist there. According to this opinion, the so- 
called tabula rasa theory, all biota of these countries must have immigrated 
since the end of the Ice Age, or at the same time as the glaciers melted away. 

During the last fifty to sixty years various scientists have expressed their 
doubts about the correctness of the tabula rasa theory, not only in Scandinavia 
but also in Iceland. Arguments have been put forth that there is great likeU- 
hood that some biota have hved there at least during the last glaciated period 
of the Ice Age. 

Before discussing the possibihties of such refugia I will mention briefly the 
hkehhood of natural immigration of plants to Iceland. We can imagine 
three ways by which plants might have immigrated to Iceland, i.e. by ocean 
currents, by air, and by help of migrating birds. 

Regarding immigration by the aid of ocean currents, a large-scale migra- 
tion seems very improbable except in the case of some beach plants which are 
more or less halophilous. The great distances from other countries make it 
improbable that seeds might have kept their germinating power to any extent. 
On the other hand, the beach of the country is very inhospitable, offering 
unfavorable conditions for the seeds to take root. Furthermore, if a great 
number of plants were brought to the country by the ocean currents, the Gulf 
Stream on one side and the Polar Stream on the other, one might expect a 
greater American element and especially a larger Arctic-Asian element than 
is actually found in the flora of Iceland. 

There is a greater possibility that seeds have been carried by birds. I think 
a considerable influx of plants by the help of birds is quite unlikely because of 
the paucity of species with edible fruits. For example, of the many Rubus 
species growing in the neighboring countries only a single one, R. saxatilis, 
occurs in Iceland. As far as we know, the migration of birds to and from Ice- 
land goes by way of the British Isles. If migratory birds were really an active 
factor in the introduction of plants to Iceland, a greater affinity to the flora of 
the British Isles would at least be very probable. 

Last to be considered is immigration by air currents, which in various 
respects is more probable as we know that small particles can be carried for 
long distances in the air. But if the greater part of the flora has been brought 
to the country in this way, there should certainly be signs of it, e.g. in the 
occurrence of an unusually great number of plants with seeds specifically 
adapted to wind dispersal. 

Furthermore, it is probable that species endowed with such great capacity 
for dispersal would spread over the entire country in a relatively short time, 


wherever conditions were favorable for their growth. But the fact remains 
that only a httle more than half of the Icelandic higher plants occur all 
over the country. Even so, there are often considerable gaps in their 
distribution, although no natural obstacles to their continuous distribution 
be found. 

Then the question is posed: how did the plants originally reach Iceland? 
There we must hold to the theory that a land-bridge connected Iceland to the 
neighboring countries during the Tertiary period, or that at least there may 
have been a lesser distance between Iceland and other countries both before 
and during the Ice Age itself than there is now. This might have faciUtated the 
migration of biota, both before the Ice Age and in the Interglacial periods. 

Assuming that biota have survived the Ice Age in Iceland, we must obtain 
answers to two questions. First, are there geological possibilities for the 
existence of ice-free regions in the country during the Ice Age ? Second, has the 
existence of vegetation in the Interglacial periods been proved ? 

The first question has been answered by Thorarinsson (1937), who pointed 
out at least four districts in the country, which in all probability have only 
been partially glaciated during the Last Glaciation. 

The answer to the second question is also in the affirmative, as fossil 
plants from Interglacial formations are found at several places in the 

Although the evidence of some ice-free areas in the country during the Ice 
Age can be proved, we cannot state with certainty that plants have survived in 
such places when the greater part of the country was covered by glaciers, and 
the climate was entirely of an Arctic character. But inferences may be drawn 
from the conditions prevailing at the present time in Arctic countries such as 
Greenland, where a fair number of species are found on nunataks which are 
completely surrounded by the inland glacier, and the climate cannot differ 
very much from that of the Ice Age. 

In this connection it may be pointed out that near the glaciers in the central 
plateau of Iceland are found patches with luxuriant vegetation on the south 
sides of hills and slopes. 

The first hints about survival of biota in refugia in Iceland were expressed 
by scientists some thirty years ago (Lindroth, 1931; Gelting, 1934). I myself 
took up the question for closer investigation in the years 1936-40. 1 especially 
tried to find out whether the distribution of species in any way indicated the 
possibihty of overwintering in probable refugia. I pubUshed the first remarks on 
this problem in a small paper in 1937, and in a somewhat more exact form in 
a second paper in 1949. In the latter I pointed out five districts in the country 
which I considered as probable refugia. This view was based on the strange 
mode of distribution of almost 100 species of plants which seemed to be 
concentrated in these districts and their immediate vicinity. Further investiga- 
tions have confirmed my early statements on this subject. 


steind6r steindorsson 

The first striking feature in the distribution of a great many Icelandic 
plants is the evident discontinuity, i.e. many of the plants seem to be accumu- 
lated in cer:ain dist icts with large gaps in their distribution where no 
natural obstacles can be observed to explain this discontinuity. 

Searching for the causes of this phenomenon one is led sooner or later to 
the opinion that these species might be remnants of an older flora, one which 
has been isolated in these areas. The only time such an isolation could have 
taken place is during the Ice Age. 

Table 1 
The Distribution in Iceland of West Arctic and North Atlantic Groups 
















3 H-. 






ffi .y 




-, tH 

., ^- 

-v fcn 



u -C 

(D Zi 

(U *- 

(L> -k- 

(U *J 

(U *- 

■G .22 

J3 t« 

J3 .;£ 

P .;£ 

-c ."^ 

xi :£. 





H T3 


H t3 


Campanula imiflora 



Carex luacloviana 



Carex nardina 



Cerastiiim arcticiiin 

X X 




Erigero/i imalaschkensis 



Pediciilaris flainmea 






Sagina caespitosa 

X X 


Stellaiia calycantha 

X X 


Alchemilla faeroeensis 




Poa laxa ssp. fiexuosa 

X X 

X X 

X X 

Saxifiaga Aizoon 



Carex bicolor 


X ,< 


X X 

Carex rufina '] 

Draba norvegica 

Epilobiuni lactifioriini 
Euphrasia frigida 

Distributed o\ 

er the whole count 

ry with 

out any 

distinct center 

Festuca vivipara 

Arenaria norvegica 

A fairly thorough study of the distribution of Icelandic plants made it 
obvious that there are six districts which can be considered as centers of 
distribution and occurrence of the various species; apart from them, a few 
smaller areas were found in different parts of the country where the same 
species occur. Comparing these plant centers with these districts, which 
Thorarinsson (1937) has pointed out as possibly ice-free during the Ice Age, 
on the basis of their landscape forms, it appears that they correspond sur- 
prisingly well. The smaller areas are all situated at such places where there 



have been possibilities for ice-free nunataks during the glaciation. These 
plant centers are as follows: (1) The Breidifjordur district, (2) The Vestfirdir 
district, (3) The Eyjafjordur district, (4) The Austfirdir district, (5) The Myrdalur 
district, (6) The Hvalfjordur district (Fig. 1 and Table 1). 

Fig. 1. Sketch map showing possible refugia areas (I-VI) and single nunataks 
(1-13): I, the Vestfirdir district; II, the Breidifjordur district; III, the Eyjafjordur 
district; IV, the Austfirdir district; V, the Mjrdalur district; VI, the Hvalfjordur 


When looking for possible survivors of the Ice Age, attention will first be 
drawn to those species which have been established with certainty as survivors 
in other northern countries. Of these, the plant group which has been charac- 
terized as "West Arctic" species in Scandinavia, is the most remarkable. 

There is hardly any disagreement among Scandinavian plant-geographers 
about the West Arctic plant group being a survivor of the Ice Age in Scandin- 
avia, and that its plants have migrated from America to Europe at least before 
the Last Glaciation. On the whole, 27 species have been classified with various 
degree of certainty as West Arctic, and at the same time considered as over- 
winterers in Scandinavia. Thirteen of these species are definitely found in 
Iceland, and perhaps 2 more. In Scandinavia all the West Arctic species 
are mountain plants. The same applies to the majority of the West Arctic 
plants found in Iceland, as 8 of them have their main distribution above 
300 m, and some of them are found growing up to 1000 m above sea-level. 

These 13 West Arctic species in Iceland are: Campanula uniflora, Carex 
niacloviana, Carex nardina, Carex rufina, Cerastium Edmondstonii{C. arcticum), 
Draba rupestris (D. norvegica), Epilobium lactiflorum, Erigeron unalaschkensis, 
Euphrasia frigida, Festuca vivipara, Pedicularis flammea, Sagina caespitosa, 
and Stellaria calycantha. 

When we observe the distribution of these 13 species, various peculiarities 



will appear. Five of them, Car ex rufina, Draba mpestris, Epilobium lactiflomm, 
Euphrasia frigida, and Festuca vivipara, are common throughout the whole 
country, so we cannot point out any center for their occurrence; but E. 
lactiflorum shows faint tendencies to centricity in three different districts. 

so OOKi 

Fig. 2. Campanula uniflora. 



Fig. 3. Carex Macloviana. 

Five others of these species have a distinct center in the Eyjafjordur district, 
i.e. Campanula uniflora, Carex macloviana, Carex nardina, Erigeron unalasch- 
kensis, and Sagina caespitosa (Figs. 2-5). 

The localities outside the district where these species are found can either 
be explained with a view to direct distribution from the center or as being 



situated on some nunataks. The latter is especially the case with Sagina 
caespitosa and Carex macloviana. 

Fig. 4. Carex nardina. 

Fig. 5. Eiigeron unalaschkensis (dots), Sagina caespitosa (triangles), and Stellaria 

calycantha (squares). 

Cerastium Edmondstonii has two centers, the Eyjaf jordur and the Austfirdir 
districts; Pedicularis flammea seems to be tricentric, and Stellaria calycantha 
(Fig. 5) seems to have a center in the Vestfirdir district. Two of these species, 
Sagina caespitosa and Stellaria calycantha, are among the rarest of plants in 

Closely connected to the West Arctic group are five species which have been 
characterized as North Atlantic, i.e. Alchemillafaeroeensis, Arenaria norvegica, 




Poa laxa ssp. flexuosa, Saxifraga Aizoon, and Carex bicolor. The distribution 
of these species in Iceland is very particular and all these species are considered 
as survivors of the Ice Age. 



> U-'^yf 

t?/ 2 ap oo>ri» 

Fig. 6. Alchemilla faeroeensis. 




Fig. 7. Poa laxa ssp. flexuosa. 

Alchemilla faeroeensis (Fig. 6) is limited to the Austfirdir district and its 
closest vicinity. Poa laxa ssp. flexuosa (Fig. 7) seems to have two main centers, 
in the Eyjafjordur district and in the southern part of the Austfirdir district; 
it is also found in mountains both in the Breidifjordur and Vestfirdir districts. 
Outside these centers hardly any other Icelandic plant seems so closely 
connected to nunatak areas. 



Saxifraga Aizoon (Fig. 8) has its main center in the Austfirdir district; 
il is found in the Eyjafjordur district and in mountains at Vatnsdalur in 
Hunavalnssysla; a second center seems to be in the Hvalfjordur district or 
more hkely in the nunatak area in the mountain range of Snaefellsnes. 

Arenaria nonegica and Carex hicolor do not have a distinct centricity. 
But several points indicate a center of C. bicolor in the Eyjafjordur district, 
and its distribution is connected to nunatak areas, especially in the highlands 
north of the glacier Vatnajokull. 

* \ \ \ ! 

IJ 50 "00»i» 

Fig. 8. Saxifraga Aizoon. 

When we summarize what has been mentioned about the distribution of 
these 18 species the result will be as follows: eleven species have a distinct 
centric distribution, and 2 more show tendencies of the same kind. Only 
6 species display no centricity at all. The distribution of the centric species 
is mostly limited to four districts, and where the plants are found outside 
these districts, distinct nunataks characterize the landscape. 

Seeing that the majority of the species in this group, which with certainty is 
considered as overwintering in Scandinavia, is almost exclusively found in 
limited areas in Iceland, one niay ask whether the peculiar distribution of these 
species is a mere accident. If that were the case, it would certainly be a strange 
accident. It would be more natural to seek the cause of this peculiar distribu- 
tion in the conclusion that these species have actually survived the Ice Age in 
refugia in the districts mentioned above, and that they have not been able to 
disperse more widely through the country in Post-glacial times. 

Furthermore, the West Arctic plants are not the only ones in Iceland which 
have such a peculiar distribution; about 80 other species have been found to 



have a very similar distribution, chiefly within the same districts and nunatak 
areas. If these facts are taken into consideration, the accidental spread of the 
species in the districts mentioned above is still more improbable. 

It would take too long to present a detailed description of the distribution of 
all the centric species, but an example should be given: 

Papaver radicatum (Fig. 9) is one of the species, which is considered as a 
survivor in Scandinavia. No other Icelandic plant seems to be so closely 
connected to the probable refugia and nunataks as P. radicatum. Although 
the Icelandic material of P. radicatum has been divided into some subspecies 
and varieties, it will be treated here as a single species only. 

Fig. 9. Papaver radicatum. 

The chief center of the species is the Vestfirdir and Breidifjordur districts 
together with their closest vicinities. Over the whole of the Vestfirdir district, 
P. radicatum is very common in the lowlands. On the other hand, the species 
is not found at higher altitudes on the slopes and mountains of that district. 
Outside the Vestfirdir district the species grows almost exclusively in the 
mountains, with the exception of the peninsula Vatnsnes where its habitats 
are similar to those of the Vestfirdir district. 

In the Vestfirdir and Breidifjordur districts the distribution of P. radicatum 
is almost continuous, but outside this center its spread is broken and the 
species is found only in isolated mountains and mountain ranges, as far as 
we know, both in north and west Iceland. In the Austfirdir district, on the 
other hand, the distribution is more continuous, ahhough there seems to be a 
gap in the central part of the district. Hardly any other Icelandic species shows 
such distinct affinity to the possible refugium areas, since it is more or less 



common in all the refugia except one, and outside of them it grows only in 
such mountain ranges where there possibly have been ice-free peaks during the 
glaciation of the Ice Age. In other locahties in the country P. radicatum is not 

As mentioned before, there are six districts in the country where species 
with centric distribution are concentrated. Outside these districts no such 
tendencies are found, except for some anthropochorous plants, which have 
obviously been introduced a comparatively short time ago. Below I will give 
a short survey of the districts in question and the species with a central dis- 
tribution occurring in each of them. 

Fig. 10. Botrychiuin boreale. 

1. The Eyjafjordur district is situated between the fjords Skagafjordur to- 
wards the west and Skjalfandifloi and the valley Bardardalur towards the east. 

As mentioned before, 8 species of the West Arctic and North Atlantic 
groups have distinct centers in the district, either only there or in some other 
centers as well, i.e. being bi- or tricentric. Besides these, 33 species have a 
fairly similar distribution, i.e. they have their central area in the Eyjafjordur 
district, although most of them are also found in other possible refugia. 

I have divided these species into two groups, alpine plants and lowland 
plants, according to whether their main distribution is above or below the 
200 m line above sea level. 

The alpine group contains 18 species. Only 3 of them can be charac- 
terized as unicentric in the Eyjafjordur district, i.e. Carex glacialis, Diapensia 
lapponica, and Saxifraga foliolosa, but all of them occur in isolated localities 
at nunatak areas outside the district. C. glacialis has a fairly wide distribution 
east of the Eyjafjordur district, in almost continuous connection with it. 

Six species of the group are distinctly bicentric, Antewmria alpina, Botry- 
chium boreale (Fig. 10), and Luzula sudetica have their centers in the Eyja- 



fjordur and Vestfirdir districts, but Antennaria may also have a center north of 
Vatnajokull. Draho olpina, Erigeron eriocephalus, and Phyllodoce coerulea 
(Fig. 11 ) are bicentric in the Eyjafjordur and Austfirdir districts. 

Five species are tri-centric in the Eyjafjordur, Austfirdir and Vestfirdir dis- 
tricts: Botrychium lanceolatiinh Draba nivalis, Erigeron uniflorus, Phippsia 
algida, and Ranunculus glacialis, but all of them except Botrychium also 
occur in nunatak areas. Ranunculus glacialis does not have such a distinct 
centric distribution as most of the centric species. 


Fig. 1 1. Phyllodoce coerulea. 

Finally, there are 4 polycentric species: Cardamine bellidifolia, Minu- 
artia biflora, Saxifraga cernua and Ranunculus pygniaeus. The first 2 occur 
in all the possible refugia except Myrdalur and Hvalfjordur, but the last 2 
are missing in Myrdalur and Breidifjordur. All of them occur in nunatak 

All the species of this group are considered either genuine or at least 
probable survivors in Scandinavia, and some of them in North America and 
Greenland as well. 

The lo}\ land group contains 15 species. Three are unicentric, i.e. Agropyron 
trachycaulum {Roegneria borealis var. islandica), Carex flara (Fig. 12). and 
Crepis paludosa (Fig. 13). Four species are bicentric: Agropyron {Roegneria) 
caninum, Carex livida (Fig. 12), Primula stricia (Fig. 14), and \'iola epipsila 
(Fig. 15) have a second center in the Austfirdir district, but Carex livida seems 
to have a third center in the nunatak area of the Sniefellsnes. Potentilla 
Egedii occurs also in the Hvalfjordur district. None of these species except 
Carex livida occurs in nunatak areas. 



Carex bnmnescens and Carex subspathacea are tricentric; Carex rupestris, 
Gentiana detonsa. Oxycoccus microcarpus, Pyrola rotundifo/ici, and Pyroja 
secunda are polycentric. The only species of this group that are found in the 
Myrdalur district are P. secunda and Oxycoccus. 

Fig. 12. Carex fiava (dots), Carex livic/a (triangles). 

Fig. 13. Crepis paludosa. 

Since the latest edition of Flora islands appeared (Stefansson 1948), 3 
species new to the flora of Iceland have been established in the Eyjafjordur 
district, i.e. Asplenium septentrionale. Primula egaliksensis (Fig. 14), formerly 
classified as a variety of P. stricta, and Roegueria douiana var. Stefanssoiiii, 



formerly classified as Agropyron trachycaulum. At least the last 2 are 
probable overwinterers. 


Fig. 14. Primula strict a (dots and squares), Primula egaliksensis (triangle). 

Fig. 15. Viola epipsila. 

2. The Vestfirdir district is the Vestfirdir Peninsula, and in close connection 
with it are the mountains at the head of the fjord Breidifjordur, which I have 
called the Breidifjordur district. 

Of the West Arctic group, Stellaria calycantha is the only species which has 
its chief distribution in the Vestfirdir, and of the centric species of the group 
only PedicuJaris flammea occurs there. 



Besides these plants 22 species are found to have their chief distribution 
in the Vestfirdir district. In contrast to the Eyjafjordur district all the centric 
species are lowland plants. 

so <x »» 

Fig. 16. Cornus suecica. 

Fig. 17. Jimciis castaneus (dots), Juncus squanosiis (triangles). 

Carex adelostoma and Cryptogramma crispa are unicentric. Cornus suecica 
(Fig. 16), Juncus castaneus (Fig. \1), Juncus squarrosus (Fig. \1), Equisetum 
silvaticum, and Melampyrum sUvaticum are bicentric. The second center of 
these species is the Austfirdir district, except in the case of Cornus which has 



its second center in the Eyjafjordur district, but is also found in a nunatak 
area closely connected with the Austfirdir district. All the other species are 

Admittedly, only half of the centric species of the Vestfirdir district are 
considered as overwinterers in other countries. But the similarity between 
their distribution and that of other centric species can hardly be explained if 
overwintering were not the case. 

As an example of such species 1 will take Melampyrum silvaticum 
(Fig. 18). Its main center is in the Vestfirdir district where the locahties are 
concentrated in three separate groups in the fjords: Thorskafjordur, Isaf- 
jordur, and Steingrimsfjordur-Bjarnarfjordur. Although the distance between 

Fig. 18. Melampyrum silvaticum. 

these areas is not very great, it is not likely that the plants have been dis- 
tributed from the same center. It is more probable that they have survived in 
small refugia and have not spread beyond their immediate vicinity, which has 
also been the case with many of the species that have their centers in the 
refugial districts. It is difficult to believe in Post-glacial migration of this 
species into these remote places where the climate is rather rough and cold, 
as long as the plant has not yet established itself in the forests in other parts of 
the country. Outside Vestfirdir the species occurs at only one locality. 
Faskrudsfjordur in the Austfirdir district. As already mentioned, the distribu- 
tion of M. silvaticum indicates distinctly that the species is an overwinterer. 
And if that is so, and this is a species rather sensitive to cold and also a forest 
plant which has been able to survive in refugia throughout the last Glacial 
period, the possibilities are undeniably great for the survival of a great part of 
the Icelandic flora in such refugia. 



3. The Austfirdir district comprises the true Austfirdir and the mountain 
range behind the fjords from the Heradsfloi in the north and southwards to 
Hornafjordur. In close connection to the main district is the Austur-Skafta- 
fellssysla as a whole. 

The great majority of the West Atlantic group is found here, and Cerastium 
Edmondstonii has its chief center here. All the North Atlantic plants occur here 
too, and two of them, Alchemilla faero^nsis and Saxifraga Aizoon, have their 
main centers here. Plant-geographically speaking, there is an obvious relation- 
ship between the Austfirdir and the Eyjafjordur districts. None of the central 
species is a real alpine plant. 

In this district, 6 species are monocentric, i.e. Aspleniuni riride, Lycopodium 
clavatiinu O.xalis acetosella, Saxifraga aizoides (Fig. 19), Saxifraga cotyledon, 
and I'accininni vilis-idaea. and none of them is found outside the district 

Fig. 19. Saxifraga aizoides. 

except Vaccinium litis idaea which occurs in the nunatak area in north- 
eastern Iceland, where Roegneria doniana var. Stefanssonii is also found. 
Last year. Asplenium triciwmanes was found in the Orccfi district, and it may 
possibly also be an overwinterer. 

Four species are bicentric: Ranunculus auricomus, Trientalis europaea, 
Carex pulicaris, and Populus trenmla. and finally there are 2 tricentric, or 
perhaps rather polycentric. species: Campanula rolundifolia and Carex 
pi lu I if era. 

4. The Myrdalur and Hvalfjordur districts have few distinctly centric species, 
and none of them will be dealt with specifically here. On the whole, the 
Myrdalur district may be considered a dubious refugium area. 

320 STEiNDOR steind6rsson 

There still remain 15 species which either occur in these 2 last mentioned 
districts or are found in all the six possible refugial districts, but not outside of 
them except in nunatak areas. 

In the foregoing, six districts have been pointed out as probable refugia for 
biota during the Ice Age. In them some 100 species occur, the modern distribu- 
tion of which is almost exclusively limited to inside the districts mentioned, 
or outside of them only in such places where the landscape indicates nunataks 
during the glaciation. 

Postglacial immigration of these plants does not seem probable, as in that 
case they might have taken root just as easily in other parts of the country 
which they do not. The same would hold true for all other sections of Iceland. 
If we put it the other way around, i.e. that ice-free areas existed at least 
during the last glaciation, the concentration of centric species in special 
hmited areas is quite natural. As far as our present knowledge reaches, no 
more probable explanation of that phenomenon can be put forth, especially 
since a great majority of the centric species in Iceland are exactly the same as 
established survivors of the Ice Age in other northern countries. 


Gelting, p. (1934). Studies on the vascular plants of east Greenland between Franz 
Josephs Fjord and Dove bay. Medd. om Gronl. 101 (2), 1-340. 

LiNDROTH, C. H. (1931). Die Insektenfauna Islands und ihre Probleme. Zool. Bidrag fr. 
Uppsala 13, 105-599. 

Love, A. and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic Flora. Acta 
HortiGotoburg. 20, 65-291. 

NoRDHAGEN, R. (1935). Om Aienaria humifusa og dens betydning for utforskningen av 
Skandinavias eldste floraelement. Berg. Mm. Arbock 1935, Naturv. rekke 1, 1-183. 

StefAnsson, S. (1948). Flora Islands, 3. utg. (Steindor Steindorsson ed.) Akureyri. 

Steindorsson, S. (1937). Jurtagrodurinn og jokultiminn. Ndttitrufr. 7, 93-100. 

Steindorsson, S. (1949). Florunyjungar 1948. Ndtturufr. 19, 110-121. 

Steindorsson, S. (1962). On the age and immigration of the Icelandic flora. Soc. Sclent, hi. 
35, 1-157. 

Thorarinsson, S. (1937). Vatnajokull. Scientific results of the Swedish-Icelandic investiga- 
tions 1936-1937, Chapter II. The main geological and topographical features of Ice- 
land. Geogr. Ann. 1937, 161-175. 


Department of Geography, University of Stockholm, Stockholm, Sweden 

Peculiarities in the distribution of plants and animals in northwestern 
Scandinavia and the occurrence of endemic species in that region have induced 
many biologists to postulate the existence of ice-free areas during the last 
(Wiirm) glaciation. Two main refugia have been localized tentatively, one 
consisting of coastal areas in northern Norway, including the Lofoten and 
Vesteralen Islands, the other situated in western Norway and composed of 
the Stad-Sunmore area and the districts at the mouths of Sognefjord and 
Stavangerfjord. The "unglaciated" areas are said to be of two kinds : small 
nunataks rising above the inland ice and larger foreland areas situated above 
as well as below the present shoreUne. 

In order to prove the existence of refugia, other arguments than biological 
ones have been used. Areas characterized by glacial cirques, such as the 
Lofoten Islands, have obviously been sculptured by glaciers, but the preserva- 
tion of these landforms is interpreted as evidence that there was no continuous 
inland ice overriding them; if so it should have smoothed out the cirque 
topography. Pinnacle-like mountain peaks have been regarded as former 
nunataks. Mountain-top detritus ("Felsenmeer") and the deep weathering of 
rocks, both supposed to require a very long time to develop, have been 
considered as other indications of non-glaciation. Furthermore, the continen- 
tal slope is said to be the definite limit of an inland ice with its fringe of ice 
shelves; as this slope is not very far from the present shoreUne — outside of 
Vesteralen only about 10 km — and the inland ice surface probably did not 
slope steeply, conditions suitable for the existence of nunataks should have 
been present. Finally, the absence of erratics and glacial striae supported the 
idea of unglaciated areas. 

However, among other groups of scientists, especially geologists and physi- 
cal geographers, there was strong opposition to the refugium theory. Some 
of the arguments for this hypothesis, for instance certain deep-weathered 
profiles, were rejected. Observations of glacial striae and erratics made in 
"refugium areas" were said to disprove the hypothesis in those areas. 

The positions for and against the refugium hypothesis were taken up 
several decades ago, and the main arguments were presented at the same time. 



The discussion, however, has continued and to a certain extent has also been 
affected by the progress in different scientific fields connected with the actual 
problem, by the development of new techniques, by parallels with other 
areas, etc. The present paper should be regarded primarily as an attempt 
to apply experience from other fields and areas to the refugium 


During the last 15 years a large number of expeditions visited Antarctica 
and have added immensely to our knowledge of that continent. Many of the 
results are pertinent to the problem under consideration, and reference should 
be made to some of them. Naturally this discussion will not solve anything, 
but it may give some ideas as to the nature of the inland ice of northern Europe, 
and of what is possible and what is impossible. 1 might first mention that the 
size of the ice sheet of northern Europe is open to debate, but it ought to have 
been between 35 and 50 per cent as large as that in Antarctica. 

Seismic, gravity and altimetry measurements have given much information 
about the thickness of the Antarctic ice sheet. In East Antarctica a value of 
more than 3000 m was found in a vast area around the U.S.S.R. bases Vostok 
I and KomsomoFskaya (Shumskiy, 1959). The bedrock topography is totally 
hidden; the thickness of the ice, however, may be as httle as 1000 m in places. 
For the smaller region of West Antarctica a mean ice thickness of 1500 to 
2000 m has been suggested. The greatest thickness, 4270 m, was measured in 
a place where the bedrock surface lies 2500 m below present sea level. This 
depth is in a 400 km wide sub-ice channel, which probably bisects West 
Antarctica. Even if the ice should melt and the rock surface rebound to isostatic 
equiUbrium, the deepest part of the channel would still be as much as 1500 m 
below present sea level. It has been suggested that the West Antarctic ice 
sheet originated as two separate ice sheets in two separate mountainous areas. 
As these sheets expanded they converged over the intervening open water; 
they were probably joined at first by a floating ice shelf, which then grew 
thick enough to fiU the trough completely and produce the present single ice 
sheet which is aground (Bentley and Ostenso, 1961, p. 895). The minimum 
time required to build up such a thickness has been computed by Wexler 
(1961) on the basis of a series of assumptions. Thereby one alternative gives 
about 20,000 years, another about 40,000 years. Quite recently new calcula- 
tions of the mean thickness of the Antarctic and other ice sheets have been 
undertaken (cf. Donn, Farrand and Ewing, 1962). The values obtained for 
Antarctica vary between 2000 m and 2500 m. On the basis of such data and 
the statement by Nye (1959) that the thickness of an ice sheet at equilibrium 
is essentially a function of the areal dimensions of the sheet rather than the 
accumulation, Donn, Farrand and Ewing have calculated a mean thickness 
of 1700 to 1800 m for the Scandinavian ice sheet during the WUrm maximum. 


The lowering of sea level at the same time is computed as being between 105 
and 123 m. 

Another result of the measurements in Antarctica is an increased knowledge 
of the subglacial relief. In several places the ice sheet covers alpine landscapes 
with valleys of the fjord type as well as well-defined peaks (Fig. 1). Such 
observations demonstrate the impossibility of recognizing a real "nunatak 
topography". Furthermore Robin (1958, p. 130) found in the "mountain ice 
sheet area"' of Queen Maud Land that "in the case of larger features, such as 
the valleys at 180, 310 and 410 km (Fig. 1) from Maudheim, the rock relief 
controls the direction of ice flow, so that the greatest velocities of movement 
will be along the line of such valleys. The erosion resulting from the flow of the 
glacier will therefore tend to accentuate the underlying relief." 

The slope of the ice surface varies considerably. For instance, a number of 
outlet glaciers leading to the Ross Ice Shelf have slopes between 1 :90 and 
1 :25 (Kosack, 1955, p. 86), whereas Robin (1958, p. 109) has Hsted inclinations 
between and 1:16. For a part of the "mountain ice sheet area" Robin's 
thorough investigation has confirmed Nye's general hypothesis that the 
thinnest ice cover is indicated by the steepest surface slope. 

In many cases ice shelves exist over considerable depths of water. In front 
of the Ross Ice Shelf systematic soundings have demonstrated that depths of 
500 to 700 m are normal (according to U.S. Navy Hydr. Office, 6636, 1957), 
while the Filchner Ice Shelf seems to extend over depths of more than 1000 m 
(Thiel and Ostenso, 1961, p. 828). The considerable depths under the inland 
ice — referred to above — emphasize that sea depths greater than those of a 
normal shelf do not form a definite limit for expanding ice sheets. "Whether 
or not the ice would extend further out to sea would depend largely on the 
heat supply and circulation of the sea water" (Robin 1958, p. 132). 

According to Bauer (1955) 0.6 x 10" km^ or about 4 per cent of the Antarc- 
tic continent is ice-free. Nunataks are common features, especially in the 
marginal zones. At the edge of the continent ice-free lowlands appear, forming 
re-entrants in the ice sheet. Such so-called "oases" are known from almost 
all parts of the coastal region of Antarctica. The largest one, Bunger's 
"oasis" (Fig. 2), has been described in several Russian reports (for instance: 
Avsyuk, Markov and Shumskiy, 1956a, 1956b). It has an area of about 
600 km^ and is surrounded by ice, which consists in part, however, of shelf 
ice and old pack ice. Roches moutonnees, striae, erratics, and till demonstrate 
that the "oasis" once was glaciated, but it seems to have been ice-free for at 
least 4500 years (loc. cit. p. 15; cf. Rozycki, 1961): it now seems to be quite 
stable. The deglaciation of this "oasis" as well as of others (cf. Bull, McKelvey 
and Webb, 1962) is believed to be the result of starvation, caused by a decrease 
in the surface level of the ice sheet in connection with an increasing influence 
of rock thresholds. A quite unique climate is formed in the "oasis". Precipita- 
tion is rather heavy, probably 600-700 mm a year, only in the solid state. In 








spite of this high value the snow cover, which is very unevenly distributed, 
quickly disappears during the short warm season, when temperatures of 
25 to 30°C were measured on rock surfaces. Evaporation is more important 
than melting because of the dryness of the air. A normal value for the relative 

Fig. 2. Bunger"s Oasis, East Antarctica (after Dolgushin, 1958). 

humidity seems to be only 25 per cent. The flora of Hunger's "oasis" is very 
meagre (Gollerbach and Syrojeckowskij, 1958), and probably the extreme 
dryness of the climate is one of the main reasons for this. However, lichens 
and algae are found everywhere, and mosses appear in wetter places. The 
dark colours of the plants are very characteristic, making possible a strong 
absorption of solar radiation. 




The Wiirm glaciation seems to have begun about 70,000 years ago. As a 
result of investigations of several Globigerina-ooze cores from the Caribbean 
Sea and the Atlantic Ocean, a temperature minimum of the early Wiirm 
(Wisconsin) was dated at about 60,000 years ago, while the temperature 
maximum of the last interglacial occurred 95,000 years ago (Rosholt, 1961). 

The Scandinavian ice sheet began as a glacierization in the highest parts of 
the mountain chain, the Scandes. Cirque and valley glaciers first coalesced to 
form transection glaciers in the mountains and piedmont glaciers in the 
forelands; further enlargement led to the development of glacier caps. 
Finally the ice caps from different mountainous areas coalesced into a 
continuously expanding inland ice sheet. This development was followed by 
continuing movements of the ice divides; in northern Scandinavia, for 
instance, an ice divide was situated for a long, though undetermined, time 
far to the east of the Scandes. 

Penck's classical scheme considered the Wiirm to represent a single climatic 
cycle (1922). This idea has been supported more recently by BUdel (1960), 
Graul (1952), Weidenbach (1953) and Fink (1961), among others. (The 
discussion about the Late Pleistocene climate of Europe has been summarized 
in an excellent way by Wright (1961).) A different scheme of subdivision of 
the Wiirm was developed by Soergel (1919), based on loess stratigraphy; it 
presumes a bi- or tripartition. Interstadials have been suggested, one more 
marked 44,000 to 28,000 years ago ("Gottweig") and another representing a 
minor warm oscillation 25,000 years ago ("Paudorf") by Gross (1958), 
Woldstedt (1958), and others; however, "the recognition of a major early 
Wiirm interstadial should be considered tentative" (Wright, 1961, p. 965). 
Recently the idea of an interstadial has received support from a series of C^* 
datings on a marine clay between beds of gravel and sand in the neighbourhood 
of Gothenburg, southwestern Sweden, i.e. well within the area covered by the 
last Scandinavian ice sheet. The result of the datings is 26,000 to 30,000 years 
(Brotzen, 1961). Radiocarbon datings of shells, which occur on raised beaches 
at 44-47 m but which are believed to originate in the underlying till in 
Nordaustlandet, Spitsbergen, have given an age of 35,000 to 40,000 years. 
Since risks for contamination exist, however, this must be regarded as a 
minimum value. "Thus the ice-free period was in all probability pre-40,000 
years ago" (Blake, 1961) and it cannot yet be placed in the Pleistocene 
timetable. It is obvious that the duration and possible division of the Wiirm 
period must have had a great influence on the extension and thickness of the 
ice sheet in northern Europe. 

Numerous scientists have tried to evaluate the climatological situation in 
Europe during the Wiirm "Pleni-Glacial" on the basis of the depression of the 
snow-line and the distribution of plants, animals, and frost features (Biidel, 


1951; Gross, 1958; Klein, 1953; Klute, 1951; Manley, 1951; Mortensen. 
1952; Penck, 1938; Poser, 1948; Woldstedt, 1958; etc.). For central Europe 
a reduction of 10 to 13°C in the mean annual temperature is commonly 
accepted; for western Europe a somewhat smaller temperature decrease is 
thought to have occurred. From a biological .iewpoint, however, information 
about the summer temperature ought to be of even greater interest. The arctic 
tree line which approximately coincides with the July 10°C isotherm, seems 
to have run south of the Pyrenees in western Europe (Biidel. 1951; Frenzel 
and Troll, 1952); for the coldest part of the WUrm, Gross (1958) has calculated 
a temperature of about 5°C in middle Germany, i.e. about 14°C lower than 
now. Naturally such temperature calculations are very rough; but at least 
they justify the hypothesis that the temperature must have been very low on 
the northwestern side of the Scandinavian ice sheet, much too low to make 
forest vegetation possible on ice-free areas, if such existed. 

The meteorological situation also has been treated in connection with the 
climatological considerations. One such discussion may be mentioned, owing 
to the importance of the problems involved and the special competence of the 
author. On the basis of his meteorological studies in Antarctica, Liljequist 
(1956) deals with the effect of the growing Scandinavian ice sheet. According 
to him an accentuated temperature contrast between snow- and ice-covered 
areas and bare ground should have developed. As the ice sheet grew this 
frontal zone — and the cyclonic tracks corresponding to this zone — moved 
southward. During the maximum of the glacial epoch the main cyclonic 
tracks thus ought to have come from west-northwest or west, passing middle 
Europe on their way to the eastern part of the Mediterranean region and 
southern Russia (Fig. 3). This atmospheric circulation ought also to have 
influenced the sea-currents according to Liljequist. South of the main 
cyclonic tracks westerly and southwesterly winds dominated, whereas to 
the north, easterly winds prevailed. This should have caused the warm 
North Atlantic Drift to bend toward the Bay of Biscay and Spain, while a 
cold current probably went to the west between Iceland and the British Isles, 

The regressive stage of the Wiirm glaciation at first was characterized by 
lability, by fluctuations in the climatic trend. The Younger Dryas was a 
rather cold period, when the ice border according to general opinion made 
long stillstands at the Salpausselka (Finland) — Middle Swedish moraines — 
Raerne (Norway). During that time the July temperature in Denmark went 
down again to about 10 C (Iversen, 1954); in the Netherlands cryoturbation 
in the earlier deposited Allerod layers has been demonstrated to have occurred 
and the existence of permafrost during Younger Dryas seems thereby to have 
been proved (van der Hammen and Maarleveld, 1952). The last ice remnants, 
still active glaciers, receded to the highest mountain regions, such as the 
Sarek area in northern Sweden (Hoppe, 1960), and disappeared 7000 to 8000 
years ago. 



For the following discussion it seems convenient to deal with three different 
cases: (1) nunataks, (2) foreland refugia in western Norway, and (3) foreland 
refugia in northern Norway. 

Fig. 3. Probable circulation pattern in atmosphere and sea during the last glaci- 
ation, ace. to Liljequist (1956). 

1 . Nunatak refugia. It must be admitted that the chance that nunataks rose 
above the Scandinavian ice sheet is rather great. This conclusion is based on 
the topography of the Scandes in their western part, where the ice sheet 
probably was not very thick, and on the occurrence of nunataks near the 
coasts of Greenland and Antarctica. 

On the other hand we do not have yet, as far as I can see, any means by 
which we are able to identify former nunataks. So-called "nunatak topo- 
graphy" can be preserved or even created under an ice-sheet surface (cf. 
above). Nor does a topography characterized by well-developed glacial 
cirques guarantee any ice-free areas between them, resp. lack of inland ice 
glaciation. In Sweden, on the eastern side of the mountain chain, for instance, 
there are many well developed cirques, as low in altitude as 800 to 1000 m 
in areas that unquestionably have been covered by the last inland ice sheet. 

Dahl (1955 and 1961) has argued that the occurrence of mountain top 
detritus ("Felsenmeer") should be a proof of a nunatak. He also gives a map 
of the northwestern part of southern Norway (Dahl, 1961, p. 89), demon- 


strating that the lower hmit of such detritus — a limit not necessarily repre- 
senting the ice surface of Wiirm time but rather that of the maximum Pleisto- 
cene glaciation (loc. cit., p. 92) — is only hundreds of meters above present 
sea level in the coastal areas, whereas it reaches values of 1700 to 1900 m in 
the southwestern part of the investigated region. This hypothesis of Dahl's 
does not agree with the distribution pattern of mountain top detritus in 
Sweden. For instance, in the southern part of the Scandes at about the same 
latitude and at altitudes between 900 to 1200 m, many of the peaks have well 
developed boulder fields of autochtonous character. From the province of 
Dalarna, Lundqvist (1951, p. 81, etc.) mentions Slugufjallet, Nipfjallet, 
Stadjan, Drevfjallet, Fulufjallet and Getsjohoa. An early reference to these 
boulder fields on mountain tops was made by Samuelsson (1914). Glacial and 
glaciofluvial features of diiferent kinds demonstrate also that the peaks must 
have been covered by the last inland ice sheet at a very late stage. 

From more northern parts of the Swedish mountains, Rapp and Rudberg 
(1960) have assembled some data on the lower Hmit of the boulder field zone: 
southern Jamtland, 1 500-1600 m, Yasterbotten, 1 100-1200 m, the amphibolite 
mountains at Abisko, 1000-1200 m. These figures refer to areas where 
covering by the ice sheet is indisputable. In addition, in Finnish Lapland 
low-lying mountain top detritus has been observed, for instance, on Levitun- 
turi (about 20 km. north of Kittila) 531 m above sea level (V. Okko, pers. 
comm. 1961). 

The formation of mountain top detritus certainly is a complicated process 
with many variables, such as the properties of the bedrock, its condition before 
glacierization, the velocity and eroding capacity of the ice sheet in different 
situations, the conditions of deglaciation, postglacial chmate, etc. I think 
that something similar can be said about the more fine-grained weathering 
material, including the clay minerals from mountain tops which have also 
been used by Dahl as a proof of nunatak situations. Here as elsewhere, 
however, Dahl stimulated the debate in an effective way. 

Initially it was stated that nunataks very probably existed during Wiirm 
time. Finally it should be added that plants and animals, if any survived, 
must have lived under extremely harsh conditions. 

2. Foreland refugia in western Norway. The presence of glacial striae and 
erratics in many places in areas postulated as foreland refugia has been taken 
as evidence of glacierization by the opponents of this concept. The proponents 
of the refugium hypothesis reply, however, that such evidence may remain 
from a previous glacial stage, but this is energetically denied by the opponents. 

The successive destruction of striated surfaces can be followed everywhere 
in Sweden and plotted against a quite detailed time scale. In many areas it is 
absolutely impossible to find any striae on bedrock surfaces which have been 
exposed only a few thousand years (e.g. since the emergence of the land 


above the surface of the sea); this seems to be especially true in coarse- 
grained rocks, in other areas, however, quite well preserved striae appear, 
even after 10,000 years. Striae may also be preserved for long periods of 
time on outcrops of quite soft rocks, particularly where these rocks are 
fine-grained and the surfaces are well polished by the ice. Such observations, 
however, do not change my opinion that glacial striae on exposed rock 
surfaces cannot have survived since the Riss Glaciation (possibly with 
extremely rare exceptions). However, the discussion would be easier if the 
weathering of rock surfaces — including the destruction of striae — could be 
illustrated by quantitative data. 

Andersen (1954, 1960) has investigated the end moraines of southwestern 
Norway. Outside the Ra moraines, 10,300 to 10,800 years old, are three sets 
of end moraines. These are partly below sea level, and they have been tenta- 
tively dated as being 12,500 to 14,000 years old by Andersen. His very 
cautiously expressed conclusion is that the front of the inland ice sheet during 
earlier stages of the Last Glaciation was far outside the present coastline. 
This conclusion fits well with the idea expressed at different times (Hoppe 
1959) that the Scandinavian and British ice sheets were connected during 
WUrm time, as well as during earlier glaciations. Recently Valentin (1957) 
has advocated this connection on the basis of studies of the morphology of 
eastern England and the bottom topography of the North Sea (Fig. 4). This 
interpretation is also supported by the pattern of glacial striae in the Shetland 
Islands, presented more than 80 years ago by Peach and Home (1879) which 
clearly demonstrates an ice sheet moving westward, whose front retreated to 
the east (Fig. 5). If the two ice sheets coalesced, it is most unlikely that ice- 
free foreland refugia existed in western Norway ; therefore the Shetland Islands 
may play a decisive role in the refugium discussion, and thorough investiga- 
tions there must be considered very desirable. 

3. Foreland refugia in northern Norway. Thanks to studies of end moraines 
and strandlines in combination with C^^ datings the glacial history of northern 
Norway has become estabhshed on a firmer basis during the last few years, 
Marthinussen (1960, 1961), Andersen and Feyling-Hanssen (both quoted by 
Holtedahl, 1960), and Bergstrom (pers. comm.) have definitely demonstrated 
the position and age of a well-developed "Ra-line" of end moraines as well 
as the existence of older end moraines outside. The conclusion of Marthinus- 
sen (1960, p. 421) concerning Finnmark, based in part also on the degree of 
weathering and the occurrence of glacial striae and erratics, is that "there can 
be no doubt but that the ice sheet of the last glacial period at its maximum 
has covered all the present land". According to the investigations by Berg- 
strom, the Lofoten and Vesteralen Islands have glacial erratics of continental 
origin (conclusion by T. Vogt, according to Bergstrom, pers. comm.) on the 
higher peaks, showing clearly that these islands must have been covered by an 



inland ice sheet at least once. Referring to the fact that the erratics are often 
lying in very unstable positions, Bergstrom suggests (1959, p. 122) that they 
were transported and deposited during Wurrn time. The proximity of the 
continental slope does not justify any definite conclusion about nunataks. 
A number of C^^ datings have been made of samples from this area with the 
intention of throwing light on this particular problem but in no case a higher 
age than 9000 years has been obtained. 

10 n u" 

Fig. 4. The North Sea area during the maximum of the Wiirm glaciation, ace. 

to Valentin (1957). 

At this time we do not know the position of the ice-front north of Scandi- 
navia during the Wlirm maximum. The Barents Sea is quite shallow, with 
depths seldom exceeding 400 m and over vast areas much less. An ice sheet 
could easily have covered this region. It has been proposed in recent papers 
(BiJdel, 1960: Corbel. 1960). that the Scandinavian and Spitsbergen ice 
sheets joined during Wiirm time. However, only very weak arguments favor- 
ing this hypothesis have been presented as yet. In 1934, Krasnov found 
boulders at the Shapkina River in northermost Russia which Likharev and 
Yakovlev suggested were erratics from Spitsbergen or Bear Island on the 



Fig. 5. Glacial striae on the Shetland Islands, ace. to Peach and Home (1879). 


basis of paleontological correlation. They lay in "the middle moraine". (The 
author is indebted to Dr. L. Serebryanny, Moscow, for the information.) 

As long as we are unable to distinguish definitely between real till and 
glacial marine sediments in samples from the sea bottom, the possibility of 
solving this problem is relatively small. 


Until now there is no geological or geomorphological proof of any refugium 
areas in Scandinavia during the Wiirm glaciation; on the contrary strong 
arguments can be raised against the existence of each such "refugium". The 
presence of nunataks, however, cannot be excluded but parts of them must 
have had a more or less permanent snow cover, and the microclimate must 
have been very harsh. 

One of the strongest arguments for the refugium hypothesis is the existence 
of plants and animals with a so-called bicentric distribution. There may, 
however, be other ways of explaining such distributions. Attention may be 
called to the fact that the supposed refugium areas probably were deglaciated 
earlier than the rest of northern and western Scandinavia, and thus both 
flora and fauna have had a longer time to become estabhshed. 


Andersen, B. G. (1954). Randmorener i S0rvest-Norge. Norsk Geogr. Tidsskrift, 14, 273-342. 

Andersen, B. G. (1960). Sorlandet i sen- og postglacial tid. Norges Geol. Unders, 210, 1-142. 

AvsYUK, G. A., Markov, K. K. and Shumskiy, P. A. (1956a). Kholodnaya pustynya v 
Antarktide. Izvestiya Akadeinii Nauk SSSR, Seriya Geograficheskaya, Moscow, 
1956, 4, 16-25. (Also translated: "Cold wasteland in the Antarctic continent". The 
Israel program for scientific translations; "Geographic and oceanographic observations 
in the Antarctic"). 

AvsYUK, G. A., Markov, K. K. and Shumskiy, P. A. (1956b). Geograficheskiye nablyu- 
deniya v antarkticheskom "oazise". Izvestiya Vsesoyuznogo Geograficheskogo Ob- 
shchestva (Moscow), 88 (4), 316-350. (Also translated: "Geographical observations in 
an Antarctic "oasis", The Israel program for scientific translations: "Geographic and 
oceanographic observations in the Antarctic"). 

Bauer, A. (1955). Uber die in der heutigen Vergletscherung der Erde als Eis gebundene 
Wassermasse. Eiszeitalter und Gegenwait 6, 60-70. 

Bentley, C. R. and Ostenso, N. A. (1961). Glacial and subglacial topography of West 
Antarctica. /. Glaciology 3, 29. 822-910. 

Bergstrom, E. (1959). Utgjorde Lofoten och Vesteralen ett refugium under sista istiden? 
Svensk Naturvetenskcip, 1959, 116-22. 

Blake, W. Jr., (1961). Radiocarbon dating of raised beaches in Nordaustlandet, Spits- 
bergen. Geology of the Arctic, 133-45. 

Brotzen, F. (1961). An interstadial (radiocarbon dated) and the substages of the last 
glaciation in Sweden. Geol. for. forh. 83, 144-50. 

Bull, C, McKelvey, B. C. and Webb, P. N. (1962). Quaternary glacial ions in southern 
Victoria Land, Antarctica. J. Glaciology 4, 31, 63-78. 

BiJDEL, J. (1951). Die Klimazonen des Eiszeitalters. Eiszeitalter und Gegenwart, 1, 16-21. 

BOdel, J. (1960). Die Gliederung der Wiirmkaltzeit. Wiirzburger Geogr. Arbeiten, 8, 1-45. 

Corbel, J. (1960). Le soulevement des terres autour de la mer de Barentz. Revue de Geogr. 
de Lyon 35, 253-74 


Dahl, E. (1955). Biogeographic and geologic indications of unglaciatedare as in Scandi- 
navia during the glacial ages. Bull. Geol. Soc. Amer. 66, 1499-1519. 
Dahl, E. (1961). Refugieproblemet og de kvartaergeologiske metodene. Svensk Natiir- 

vetenskap, 1961. 81-96. 
DoLGUSHiN, L. D. (1958). Glyaciologicheskie nablyudeniya v Antarktide. (Glaciological 

observations in Antarctica). Izvestiva Akacleiuii Nauk SSSR, Seriya Geogiaficheskava, 

Moskow, 1958, 6, 16-25. 
DoNN, W. L., Farrand, W. R. and Ewing, M. (1962). Pleistocene ice volumes and sea- 
level lowering. /. Geology, 70, 206-214. 
Fink, J. (1961). Die Gliederung der Wiirmeiszeit in Osterrich im Lichte der Terrassen- 

und Loessforschungen. Abstracts of papers, INQUA, Vltli Congress, 152-153. 
Frenzel, B. and Troll, C. (1952). Die Vegetationszonen des nordlichen Eurasiens wahrend 

der letzten Eiszeit. Eiszeitalter and Gegenwart, 2, 159-167. 
GoLLERBACH, M. M. and Syrojeckowskij, E. E. (1958). Biogeograficeskie issledovanija v 

Antarktide 1957 g. (Biogeographical observations in Antarctica, 1957). Izvestiya 

Akademii Nauk SSSR, Seriya Geograficheskaya, Moscow, 1958, 6, 59-68. 
Graul, H. (1952). Bemerkungen zur Wiirmstratigraphie im Alpenvorland. Geol. 

Bavarica 14, 124-131. 
Gross, H. (1958). Die bisherigen Ergebnisse von C^* -Messungen und palaontologischen 

Untersuchungen fiir die Gliederung und Chronologic des Jungpleistozans in Mittel- 

europa und den Nachbargebieten. Eiszeitalter und Gegenwart, 9, 155-187. 
VAN DER Hammen, T. and Maarleveld, G. C. (1952). Genesis and dating of the periglacial 

deposits at the eastern fringe of the Veluwe. Geologie en Mijnbouw, Nw serie, 2, 14e 

Jaargang, 47-54. 
HoLTEDAHL, O. (1960). Gcology of Norway. Norges geol. Unders. 208, 1-540. 
HoppE, G. (1959). Nagra kritiska kommentarer till diskussionen om isfria refugier. Svensk 

Naturvetenskap, 1959, 123-134. 
HoppE, G. (1960). Glacial morphology and inland ice recession in northern Sweden. 

Geogr. Annaler 41, 193-212. 
Iversen, J. (1954). The late-glacial flora of Denmark and its relation to climate and soil. 

Danmarks geol. Unders., II Raeke, 80, 87-119. 
Klein, A. (1953). Die Niederschlage in Europa im Maximum der letzten Eiszeit. Peter- 

manns Geogr. Mitt. 97, 98-104. 
Klute, F. (1951). Das Klima Europas wahrend das Maximums der Weichsel-Wiirm 

Eiszeit und die Anderungen bis zur Jetztzeit. Erdkunde, 5, 273-283. 
KosACK, H.-P. (1955). Die Antarktis. Heidelburg, 310 p. 

Liuequist, G. H. (1956). Meterorologiska synpunkter pa istidsproblemet. Ynierl6, 59-74. 
LuNDQviST, G. (1951). Beskrivning till jordartskarta over Kopparbergs Ian. Sveriges Geol. 

Unders., Ser. Ca, 21, 1-213. 
Manley, G. (1951). The range of variation of the British climate. Geogr. J. 117. 
MARTfflNUSSEN, M. (1960). Coast and fjord area of Finnmark. In O. Holtedahl: Geology of 

Norway. Norges Geol. Unders. 208, 416-431. 
Marthinussen, M. (1961). Brerandstadier og avsmeltningsforhold i Repparfjord-Stab- 

bursdal-omradet, Finnmark. Norges Geol. Unders. 213, 118-169 
Mortensen, H. (1952). Heutiger Firnriickgang und Eiszeitklima. Erdkunde 6, 145-160 
Nye, J. F. (1959). The motion of ice sheets and glaciers. J. Glaciology 3, 26. 493-507 
Peach, B. N. and Horne, J. (1879). The glaciation of the Shetland Isles. Quartern. J. Geol. 

Soc. London 35, nS-811. 
Penck, a. (1922). Ablagerungen und Schichtstorungen der letzten Interglazialzeit in den 

nordlichen Alpen. Preuss. Akad. Wiss., Phys.-Math. KL, Sitzungsber, 214-251. 
Penck, A. (1938). Das Klima der Eiszeit. Verhandl. d. III. Intermit. Quartdr-Konferenz. 

Wien, 1936, 1,83-97 
Poser, H. (1948). Boden- und Klimaverhaltnisse in Mittel- und West- Europa wahrend der 

Wurmeiszeit. Erdkunde 2, 53-68. 
Rapp, a. and Rudberg, S. (1960). Recent periglacial phenomena in Sweden. Biiiletyn 

Peryglacjalny H, 143-154, 


Robin, G. de Q. (1958). Seismic shooting and related investigations. Norwegian- British- 
Swedish Antarctic Expedition, 1949-52. Scientific Results, vol. V, 1-134 
RoSHOLT, J. H. (1961). Pleistocene chronology by the Pa-^VTh"" mcihod^ Abstracts of 

papers, IN QUA, Vlth congress, 180. 
RozYCKi. S. Z. (1961). Changements pleistocenes de Textension de I'inlandis en Antarctide 

oriental d'apres Fetude des anciennes plages elevees de TOasis Hunger. Queen Mary 

Land. — Biidetyn Peryglacjalny 10, 257-283. 
Samuelsson, G. (1914). 6m Dalafjiillen. Ymer 34, 331-345. 
Shumskiy, p. a. (1959). Is Antarctica a continent or an archipelago ? y. Giaciology 3. 26, 

SoERGEL, W. (1919). Losse, Eiszeiten, iind paldolithische Kulturen, Jena, 177 p. 
Thiel, E. and Ostenso, N. A. (1961). The contact of the Ross Ice Shelf with the continental 

ice sheet, Antarctica. J. Giaciology 3, 29, 823-832 
Valentin, H. (1957). Glazialmorphologische Untersuchungen in Ostengland. Ahlandl. d. 

Geogr. Inst. d. fieien Univ. Berlin, 1-86. 
Weidenbach, F. (1953). Zeitliche Einordnung der jungpleistoziinen Ablagerungen 

Mitteleuropas. 4th Int. Quat. Congr, Actes, Rome. 
Wexler, H. (1961). Growth and thermal structure of the deep ice in Byrd Land, Antartica. 

/. Giaciology 3. 30, 1075 1087. 
WoLDSTEDT, P. (1958). Das Eiszeitalter. — Grundlinien einer Geologic des Qiiartdrs, 2: 

Europa, Vorderasien und Nordafiika ini Eiszeitalter, 1-438. 
Wright, H. E. Jr. (1961). Late Pleistocene climate of Europe: a review. Btdl. Geol. Soc. 

Amer. 72, 933-982. 





J. D. IVESf 
Geographical Branch, Department of Mines and Technical Surveys, Ottawa, Canada 

The main obstacles to evaluation of the maximum extent of glaciation in 
eastern Canada are simply a combination of vast area, difficulty of access 
and acute shortage of field investigators. These problems naturally pervade 
the entire discussion which follows and they cannot be overstressed : however, 
some of the gaps can be partially filled by reference to work in other sectors 
of the North Atlantic, particularly in Norway and Iceland. It should be 
apparent, moreover, that even these much smaller areas have not been studied 
in great detail and, as members of the symposium will undoubtedly agree, 
the interpretation of the available field evidence has divided us into two camps 
— the protagonists, and the antagonists of the nunatak hypothesis. 

It is apparent from the hterature that much controversy has arisen con- 
cerning the significance of certain glacial erratic blocks on high coastal 
summits. Bergstrom, Hoppe and others inaintain that blocks on the Lofoten 
Island summits are glacial erratics (Bergstrom, 1959; Hoppe, 1959) and the 
tendency is to proceed from this point to the suggestion that all the Lofoten 
summits must therefore have been inundated at one and the same time, 
namely, during the final glaciation. Dahl, on the other hand, has enunciated 
a plea that evidence for the existence of true glacial erratics must include 
positive proof that the rock in question is foreign geologically to the base- 
ment upon which it rests — in other words, the possibiUty that the rock is an 
erosion residual must be definitely excluded (Dahl, 1961). Also, absence of 
proof of glaciation does not warrant the conclusion that glacier ice has not 
passed across a specific area, nor does positive proof of glaciation determine 
which particular glaciation was involved. One of the more difficult field 
problems is to determine the relative age of an erratic. Usually this is impossible 
and the deduction frequently resorted to is highly subjective. Neither is it 

*Published with the permission of the Director, Geographical Branch, Dept. Mines 
and Technical Surveys, Ottawa. 

t Senior Geographer, Geographical Branch, Ottawa Canada. 


338 J. D. IVES 

possible to ascertain wiiether or not the entire coastal zone was submerged 
by ice at one and the same time. An extension of this discussion can be found 
in the Hterature and it cannot be pursued any further in the present instance. 
An additional problem, probably largely confined to the High Arctic, is that 
certain types of inert ice cover could have existed in the past, no trace of which 
remains today. This is especially important in the areas designated as High 
Arctic refugia in the Queen Elizabeth Islands. 

It is upon this basis of fundamental disagreement in interpretation of the 
available evidence, however, that a review will now be made of the physical 
conditions of the eastern Canadian seaboard. The available, and frequently 
conflicting, evidence from this region will then be presented and its significance 


From Cape Breton Island the eastern Canadian seaboard stretches generally 
northwards for approximately 4500 km (Fig. 1). It may be divided into four 
great sectors: the Maritime-Newfoundland sector which has predominantly 
low coastal areas, excepting Long Range and the Shickshock Mountains; 
Labrador, with moderate to low coasts in the southern half and mountainous 
fjord coasts north of Nain; Baffin Island, with a fjord and high mountain 
coast along its entire 1600 km length; here coastal summits exceed 1500 m 
altitude and at least two groups of high rocky islands lie up to 30 miles off" the 
coast; and finally the Devon-Ellesmere sector which extends for more than 
1000 km in a general north-south direction and is somewhat similar in 
character to the Baffin Island sector. 

One very important consideration is that the present climate of the eastern 
seaboard ranges from moist temperate in the south to High Arctic desert 
in the north, and this great range of climate presumably existed throughout 
the Pleistocene period. One reflection of this chmatic range is that present-day 
glacierization increases in extent northwards from the Kaumajet Mountains 
so that from Pond Inlet to northern Ellesmere Island more than 70 per cent 
of the coastal zone is ice-covered. Another important consideration is the 
extent of the continental shelf: little can be said of the area north of Hudson 
Strait, but southwards the shelf is moderately to extremely broad ranging 
from 100 to 400 km. FinaUy, the character of the bedrock geology is vital, 
although in this respect also little systematic data is available. With a few 
small areas excepted, the coastal zone from the Gulf of St. Lawrence to north- 
ern Ellesmere Island is composed of Archaean rocks, principally a suite of 
acidic granitic gneisses, much of which are probably meta-sediments. Abrupt 
local variations occur within the gneissic group yet there is a general lack of 
distinctive rock types. Furthermore, distinctive rock types do not generally 
occur within several hundred kilometers of the east coast. Thus great difficul- 
ties arise when DahTs (1961) criteria for the confirmation of true glacial 

Fig. 1. General map showing the eastern seaboard of Canada and the major 
localities referred to in the text. 


340 J, D. IVES 

erratics are applied. The coarsely crystalline nature of the rock ensures rapid 
destruction of glacial polish and striations. Their meta-sedimentary character 
militates against the successful apphcation of Dahfs X-ray analysis methods 
for determination of weathering products of the surface mantle. Hence it is 
difficult to establish the relative age of the mountain-top detritus. Thus the 
small areas of distinctive rock type become of critical importance. These 
include: the Kaumajet Mountains which are composed of basic lavas of 
probable Proterozoic age; part of the southern Torngat Mountains, embrac- 
ing the slates and quartzites of the Ramah sedimentary series; the distinctive 
Tertiary lavas of Cape Dyer in southeastern Baffin Island and a small area of 
sandstones and shales on northeastern Bylot Island. A factor which emphasizes 
the importance of the first three of these areas is that the sedimentary and 
volcanic rocks occur at elevations in excess of 1200 m immediately overlooking 
the sea. 

Evidence implying extensive glaciation at high altitudes is so far restricted 
to the Shickshock Mountains and northern Labrador. Odell was the first to 
conclude that ice had passed over the highest summits of the Torngat and 
Kaumajet mountains (Odell, 1933), and the author has located several 
glacial erratics at altitudes in excess of 1200 m and two in excess of 1500 m 
(Ives, 1957 and 1958a). Dahl (pers. comm. 1958) maintains that these "erra- 
tics" are probably erosion residuals and the author would agree that they 
do not satisfy Dahl's criteria. The great contrast in degree of weathering and 
in relative abundance of glacial evidence above and below the general 750 m 
level has been noted by several workers (Daly, 1902; Coleman, 1921; Ives, 
1958b; Loken, 1962a and 1962b) and has been interpreted as dependent upon 
the relative time available for sub-aerial weathering. Daly, Coleman and 
Loken (Loken, pers. comm. 1961) have all failed to find any indications of 
glacial inundation at the higher levels. Thus in view of Dahl's refusal to 
accept the validity of the Torngat "erratics," the available data from the 
Kaumajet Mountains is of critical importance. Following Odell's work, 
Wheeler (1958) found positive proof of total glacial submergence of these 
1200 m coastal summits. He describes erratics of gray gneiss, amphibolite 
and pale garnet-biotite gneiss within 100 m of the summit of Mount Brave, 
the highest point, and these erratics lay on the local bedrock which comprises 
a basic volcanic series (Mugford Series) (Wheeler. loc. cit.). Tomlinson has 
since found numerous gneissic erratics on the Mugford volcanics at altitudes 
exceeding 1000 m (Figs. 2 and 3) and states that in places they are so numerous 
as to be used as stepping stones in walking over the contrasting dark ground 
mass of the volcanics (TomHnson, 1958, 1962, and pers. comm. 1962). 
In this instance, Dahfs criteria appear to be fulfilled and his plea that the 
erratics could be weathered-out inclusions in the lavas (Dahl, pers. comm. 



1959) is discounted due to their abundance, to the lack of metamorphic 
alteration rims and to the absence of such inclusions in the existing lavas 
(Wheeler, pers. comm. 1960). Finally, in the Nain-Kiglapait area farther 
south unequivocal erratics have been found within 20 m of the summit of 

Fig. 2. Gneissic boulder resting on disintegrated bedrock ot basic volcanics of the 
Mugford Series. The altitude is in excess of 1000 m. {Photo by R. F. Tomlinson, 
July 1958. Kaitmajet Mountains.) 

Man-O'War Peak which exceeds 1050 m and lies close to the outer coast. 
Here erratics of garnetiferous gneiss rest upon anorthosite bedrock (J. P. 
Johnson Jr., pers. comm. 1962). By extending this argument, it seems satis- 
factory to interpret the Torngat blocks as true glacial erratics. 
The evidence implies only that at some time during the Pleistocene the 

high summits were inundated by ice from the west. The work of Flint, 


J. D. IVtS 

Demarest and Washburn (1942) in the Shickshock Mountains is faced with 
the same problem of relative age of the erratics and Flint has commented 
that such fresh-looking erratics are no proof that the glacial inundation did 
not pre-date the Wisconsin stage (Flint, pers. comm. 1958). Similarly, it is not 
implied that all summits were necessarily inundated at the same time, so that 

Fig. 3. A large block of granite gneiss resting upon Mugford Volcanics. The figure 

gives an indication of the scale. Altitude 650 m. {Photo by R. F. Toiiilinsoiu July 

1958, Kaiimajet Mountains.) 

the possibility of plant refugia remains. Furthermore, from the evidence 
presented below, it is argued, for northern Labrador at least, that the glacia- 
tion responsible for emplacement of the high-level erratics was certainly 
earlier than the Classical V\ isconsin equivalent, and possibly much earher. 

Next it will be profitable to examine the evidence supporting the existence 
of ice-free areas during the Pleistocene; in particular the importance of the 
mountain-top detritus must be evaluated in the light of the available data. 
Both the author and Loken conclude that a distinctive lower limit to the 
mountain-top detritus can be recognized in the Torngat Mountains. This is 
described as a glacial trim line and can be seen to slope from west to east, and 
from south to north from a high point west of the head of Saglek Fjord. The 
slope is equated with that of the upper surface of the inland-ice and outlet 
glaciers during a recent glaciation, termed the Koroksoak Glaciation (Ives, 
1958a). The northerly slope of the trim line traced roughly along the axis of 
the Torngat peninsula is from 830 m in the Saglek Fjord vicinity to 430 m 
south of Telliaosilk Fjord, 160 km farther north (roughly 2,5 m/km). Leken 
has calculated a minimum isostatic tilt of 1:1000 up towards the south- 
southwest (Token, 1962b) so that the slope of the trim line should be reduced 



by at least this amount. Despite this, the loss of elevation towards the north is 
still significant and it is tempting to relate it to the juxtaposition of the moun- 
tains and the broad and deep Hudson Strait which must have acted as a 
major discharge channel for the inland-ice. 

With the exception of the evidence already presented indications of glacial 
activity above the trim line in the zone of mountain-top detritus is entirely 
lacking (Figs. 4 and 5). Coleman, Daly and Loken all maintain that the trim 

Fig. 4. An "old"' weathering surface: mature mountain-top detritus at 1050 m 
west of Saglek Fjord in the Torngat Mountains. (Photo hv the author, August 


line is the upper Umit of glaciation within the area investigated by them, and 
X-ray analyses by Dahl of fines collected by Loken from within the mature 
detritus have led to the suggestion that their contained abundance of vermi- 
culite, hydrobiotite and montmorillonite imply Tertiary, warm-climate weather- 
ing processes (Loken, pers. comm. 1962). Figure 6 provides a provisional 
picture of the distribution of mountain-top detritus. In Labrador the know- 
ledge of its distribution is adequate for the formulation of a working hypo- 
thesis as the strong correlation between occurrence of mature detritus and 
altitude above present sea level is readily apparent. It is postulated, therefore, 
that the detrital trim line marks the upper limit of a glaciation at one particu- 
lar stage, although it is not yet possible to determine which stage. It is recog- 
nized that this hypothesis is fraught with major difficulties and Hoppe and 
Rudberg (pers. comm. 1962) have both maintained that altitudinai variation 



in the lower limit of mature detritus in Sweden is related to rock type rather 
than vertical extent of glaciation and that mature detritus undoubtedly exists 
in areas proven to have been completely inundated during the last glaciation. 
The controversy of the mode and time sequence of development cannot be 

?«-. . 

Fig. 5. Mature moumain-top detritus above JOOO m in the Kaumajet Mountains. 

Note the small gneissic erratic in the foreground indicated by the arrow. {Photo 

by R. F. Toinlimon, July 1958.) 

examined exhaustively within the limits of the present paper. Suffice it to say 
that in northern Labrador, where rock type is relatively uniform and almost 
universally Archaean, it is altitude, and therefore vertical extent of glaciation 
which is the important dominant, rather than rock type. It is urged, therefore, 
that until evidence is available to the contrary, the surfaces which carry 
mature detritus have been subjected to sub-aerial weathering for a far longer 





, Padloping 
♦ Island 




, .^ Cape Oyei 
^1700 m. 




F F 

1 N 






J050m' ,200 n 

. Mature mountain top detritus 
.Highest summits glaciated 


jj', ■ MTNS 
-^ 1670 m 

il200m KAUMAJET 

^fc MTNS 

--0.. < o<^^ MTNS 

-P ■-.<5<a,;^650n 

■ "^a^o, ^Z 

highest summits in metre 

I q 5 100 150 Miles 

SO 50 100 150 ^00 250 Kilomoln 


e ^ 'v^?5o-- 



Fig. 6. Sketch map of the known distribution of mature mountain-top detritus in 

Labrador-Ungava and southern Baffin Island. Maximum altitudes of individual 

mountain groups are shown, together with the location of critical glacial erratics 

and glacially scoured summits. 

346 J. D. IVES 

period than those areas below the trim hue, and that this period embraced 
the maximum of at least the Classical Wisconsin glacial equivalent. 

For 100 to 125 m below the trim line, both Loken and Ives have described 
the occurrence of incipient mountain-top detritus (Ives. 1958b; Loken, 1962a). 
Andrews, furthermore, has described the same phenomenon in the Kiglapait- 
Nain area farther south (Andrews and Matthew, 1961). This may be equated 
with DahTs incipient detritus in Norway (Dahl, 1961) and it is again argued 
that the time factor is vital to an understanding of its development. Its lower 
limit is marked by a striking series of lateral moraines and kame terraces, 
below which evidence of recent glacial activity is everywhere abundant. The 
present author originally interpreted the upper trim line as marking the maxi- 
mum vertical extent of the last or Koroksoak Glaciation, and the lateral 
moraine system as representative of a recessional phase of the Last Glaciation 
(Ives, 1958a, 1960). Andrews followed Dahl by suggesting that the incipient 
detritus developed, in part at least, during the last glaciation and that the 
lateral moraine system marks the upper limit of that glaciation to which he 
gave the name "Sagiek" (Andrews. 1961; Dahl. 1961). In this scheme the 
Koroksoak becomes the penultimate glaciation. and the Torngat Glaciation, 
during which the high level erratics are believed to have been emplaced. older 
still. ^ 

The Saglek lateral moraine system has been traced both in the field and on 
air photographs over a north south distance of more than 400 km (Ives, 1960; 
Loken, 1962a; Tomlinson. pers. comm. 1961 ; Andrews and Matthew. 1961). 
Below them extensive end moraine systems occur and Loken has related 
individual end moraines to Late-Glacial sea level phases. He has also obtained 
a radiocarbon date of 9000 years B.P.* for a condition when some of the north- 
ern fjords already contained no tide-water glaciers, while farther south outlet 
glaciers still reached the Atlantic Ocean (Loken, 1962b). Andrews has studied 
several massive end moraine systems in the Kiglapait-Nain area which 
appear to overlie extensive deposits of varved clays up to 20 m thick. The 
implication is that subsequent to the Saglek lateral moraine phase, several 
major halt and readvance phases occurred. This evidence supports the 
contention that the Saglek moraines are appreciably older than 9000 years 
and may indeed represent the equivalent of the Classical Wisconsin maximum. 

From the foregoing discussion it will be seen that an attempt is being made 
to tackle the problem of the relative age of the mature mountain-top detritus 
indirectly — namely, by considering the relative and absolute ages of the 
extensive lateral and end moraine systems, thereby obtaining a minimum 
age for the mature detritus. One further step in this reasoning process is 

* Personal communication from Dr. W. S. Broecker, 1961. Marine molkiscs were 
collected at 29 m above sea level by Dr. O. Loken and submitted to Dr. Broecker by the 
Geographical Branch. The Lamont Geological Observatory laboratory number of the 
sample is L-642. 


necessary before turning to Baffin Island where conditions are similar to 
those of northeastern Labrador-Ungava. Matthews has collected marine 
molluscs up to an altitude of 120 m above present sea level in the Deception 
Bay vicinity, south Hudson Strait (Matthews, pers. comm. 1961). Deception 
Bay lies nearly 600 km from the open Atlantic Ocean and the local marine 
limit is approximately 150 m above sea level. Marine molluscs collected from 
the 86 m level by Matthews have yielded a radiocarbon age of 10,450 ±250 
years. (Geographical Branch No. J.D.I.-61-S6; Isotopes inc. No. 1-488.) 
This implies that at the time of the Vaiders Readvance in the south, this sector 
of Hudson Strait was not only open to the Late-Glacial marine incursion, 
but that it had been inundated for a sufficient interval of time to allow a 
maximum of some 60 m of relative vertical isostatic recovery prior to 10,450 
years ago. If this single date can be substantiated, the implications are most 
important. Assuming that the radiocarbon age of 18.000 years B.P. is accurate 
for the maximum of the Classical Wisconsin Glaciation in southern Canada 
and northeastern United States, it seems possible that contemporary conditions 
in Hudson Strait were characterized by partial, or total absence of glacier ice. 
This possibility has a most important bearing on the dating and significance 
of the Sagiek moraine system. At least it can be said that the Saglek moraines 
are more than 10.500 years old and the implication is that large areas along 
the Labrador coast, northwards of Nain. remained ice-free at the Classical 
Wisconsin maximum. Thus it seems probable that the areas occupied by 
mature detritus, and possibly even that occupied by incipient detritus, were 
not ice-covered at the Classical Wisconsin maximum. This would corroborate 
conclusions drawn by Dahl in Norway (Dahl, 1961), and concurs with 
Andrews" interpretation of the significance of the Saglek and Koroksoak 
levels (Andrews, 1961). By comparing the enormous difference in glacial 
conditions between an ice stand at the Saglek moraines or the Koroksoak trim 
line, and one that overtopped the highest summits, as during the Torngat 
Glaciation, it seems probable that the Torngat Glaciation is pre-Sangamon 
in age. If this hypothesis can be substantiated it would provide for a consider- 
able period of time for the development of mature mountain-top detritus. 

Admittedly, the preceding hypothetical discussion leans too heavily upon 
two isolated radiocarbon dates and extensive subjective interpretation and 
extrapolation. The argument, therefore, is tentative and is intended as a 
suggestion for a future line of inquiry rather than as a positive conclusion. 


Little information is available for Baffin Island, but some significant 
statements can be made. According to that acute observor, Dr. A. P. Low, 
the glaciers of northern Baffin and Bylot islands were probably never much 
more extensive than they are today (Low, 1906, pp. 233-236), Falconer has 
substantiated this conclusion with the proviso that it must be strictly applied 



to the mountain glacierization of the east coast (Falconer, 1962) for it can be 
shown that the mountains were at least partially inundated by the outlet 
glaciers of an extensive inland-ice with its centre over western Baffin Island 
and Foxe Basin (Ives, 1962; Andrews and Ives, 1963). The development of 
this inland-ice was the predominant factor in the extent of the glacial inunda- 
tion of the fjords and coastal mountains and in this instance the close parallel 

Fig. 7. Map of the end and lateral moraine systems of Baffin Island reduced from 
field and air photograph interpretation. The Cockburn Moraines are not differenti- 
ated but are represented by the almost continuous black lines running northwards 
from the western side of the Penny Ice Cap. {Compilation by J. T. Andrews 
Geographical Branch.) 

with glacial conditions in Labrador-Ungava is stressed. Thus the Clyde 
lateral moraines, which mark the upper known limit of the outlet glaciers in 
the eastern Baffin fjords, are compared with the Saglek lateral moraines of 
Labrador-Ungava. Secondly, an extensive system of end moraines has been 
traced over a distance of 700 km. They lie inland from the heads of the 
fjords, but have many arcuate extensions (convex eastwards) implying that 
outlet glaciers descended to tide-water. These have been named the Cockburn 
end moraines and their extensive development and degree of continuity 
(Fig. 7) prompt the suggestion that they may have full stadial, or at least 



important sub-stadial significance. Whatever the true stratigraphic position 
of the Clyde and Cockburn moraines, it seems certain that, as in northern 
Labrador, extensive coastal tracts remained ice-free during the Classical 
Wisconsin equivalent, and possibly throughout the entire Wisconsin stage. 

From the foregoing discussion the question arises as to whether or not 
potential ice-free areas can be correlated with the occurrence of mountain-top 
detritus. Few observations have been made in Baffin Island, although Gribbon 

Fig. 8. The outer part o( Sam Ford Fjord in the general vicinity of Clyde, Baffin 
Island. In the middle right prominant lateral glacial terraces can be seen which may 
be related to the Clyde lateral moraine system. In this instance their altitude is 
about 300 m above sea level. Above them mature detritus is abundant as can be 
seen in the foreground. (Photo by G. Falconer, September 1961.) 

has described mature mountain-top detritus on summits above 700 m in the 
vicinity of Cape Dyer (Gribbon, 1960, and pers. comm. 1961). Similarly, 
Falconer has made summit landings by helicopter for the Geographical 
Branch in the Cape Dyer and Clyde vicinities, and has had occasion to examine 
mountain tops from the air on southern Bylot Island. In each case it seems 
likely that mature mountain-top detritus occurs (Fig. 8), ahhough this has 
yet to be firmly substantiated (Falconer, pers. comm. 1961). Finally, lying off 
Hall Peninsula in southeastern Baffin Island are two groups of small islands. 
Lady FrankUn and Monumental islands. They are situated about 45 km off the 
coast and rise precipitously to heights exceeding 200 m. Figure 9 is a photo- 
graph of the summit of Lady Franklin Island taken from a helicopter. Light 
snow limits the value of the photograph and no scale is available, although 
it can be seen that the cliffs fall sheer to the sea. The appearance of mountain- 
top detritus is striking, and the Hydrographic Service officer who had been 

350 J. D. IVES 

requested lo make observations for the Geographical Branch, on landing, 
was unable to detect any evidence of glaciation. Clearly these islands are an 
obvious place to look for evidence of niinataks at the Pleistocene maxima 
because of their altitude and position. 

^'- y 

- > * 


p^l ^ ] ' 

**-* ■* -^ J. 


^' ^ ^ 


. * ' 

*^ » « 

■x' ' ••■' 


l^^/-, ■" 




^^ ti 

\ c ' X * 

Fig. 9. The summit of Lady Franklin Island as seen from a helicopter. The cliff 
descends more than 200 m vertically to the sea (black in the photo). The surface 
is mantled with mature detritus. {Photo by Canadian Hydrographic Service, 

September 1961). 


The presentation of the evidence has been restricted to Labrador-Ungava 
and Baffin Island with a few cursory remarks being devoted to the Shick- 
shock Mountains. Nothing can be said about the Devon-Ellesmere islands 

A few major points emerge from the available evidence: first, it seems cer- 
tain that most of the highest summits in northern Labrador, if not all of them, 
have been inundated at some time prior to the Last Glaciation, although no 
evidence is available for Baffin Island; secondly, either the detrital trim line, 
or the Saglek moraines, may be taken to mark the upper limit of the Last 
Glaciation in Labrador, which would indicate that large areas remained ice- 
free; finally, conditions may have been very similar in Baffin Island, and by 
extsion, along the entire eastern Canadian seaboard. 

Much more work remains to be done before anv of these contentions can 



be substantiated and, but for the problems of vast area and scarcity of workers, 
these contentions would never have been proposed until systematic research 
had been completed. However, controversial and hypothetical discussions 
are healthy aspects of research and a somewhat premature publication can be 
justified provided the facts are separated from the hypotheses and the limita- 
tions are adequately stressed. In this manner some directive for future study 
might be developed which would lead to the careful testing of the hypotheses 
in critical areas. 

Fig. 10. Mature mountain-top detritus on gentle slopes above 700 m inland of Cape 
Dyer, southeast Baffin Island. (Photo by G. Falconer, September 1961.) 

Little more can be said about the origins of the mountain-top detritus 
except that a relatively long time was required for its development. In this 
respect the author cannot go as far as Dr. Dahl in his conclusion that the 
existence of mature detritus indicates the preservation of ice-free areas 
throughout the Pleistocene. The mantle could be largely a relic of Tertiary 
weathering processes, but the erratics in the Torngat and Kaumajet Mountains 
indicate that if this is so, then it has been overridden by glacier ice sometime 
in the past. In certain circumstances thin ice patches may well have developed 
over mountain-top detritus and, because of limited ice movement, the 
detritus may have been preserved underneath without modification. This has 
undoubtedly occurred in Baffin Island, and a study of air photographs has 
revealed instances where inature detritus is today emerging from beneath 
such thin plateau ice caps and ice patches. However, the emplacement of 
erratics on the high summits in northern Labrador demands a vigorous 
movement of ice across them. It is inferred that tiie summits must have been 

352 J.D. IVES 

covered by a minimum thickness of 300 m of ice before the ice sheet would 
have been able to drag erratics up from lower elevations to altitudes exceeding 
1000 m (Ives, 1957). 

Another difficulty arises in High Arctic areas where the temperature of the 
ice mass is below the pressure melting point. Under such conditions it 
appears that the ice caps and glaciers may be frozen to their beds; consequent- 
ly the ice becomes a protective rather than an erosive agent. The former 
existence of a thin ice cover in north-central Baffin Island has been deter- 
mined by the study of rock lichens. It has been concluded that up to 70 per 
cent of the interior upland was ice-covered as recently as 1 50 to 200 years 
ago, whereas the present-day coverage is less than 3 per cent (Ives, 1962). 
Once sufficient time has elapsed for rock lichen growth to reach maturity, no 
sign of the former ice cover will remain. In this instance there is strong evidence 
for glacierization during the Pleistocene, but the evidence for a recent, thin 
ice-cover provides an interesting example of the problem. Thus, if the question 
is transferred to the suggested Queen Elizabeth Island refugia, it must be 
borne in mind that the low islands, which today have so far yielded no proof 
of glacierization, could have been completely inundated in the distant past by 
the type of ice referred to in northern interior Baffin Island. With the elapse of 
1000 to 2000 years (the time required for mature development of rock lichens 
according to Beschel, 1961) following the melting off of such an ice cover, no 
physical evidence for its former existence need remain. Recent botanical 
work in the northwest Queen Elizabeth Islands (Savile, 1961), although far 
from complete, points out an unusual scarcity of both species and individual 
plants which is hardly commensurate with the concept that some of these 
islands provided a major plant refugium throughout the Pleistocene. 


1 . Weight of evidence warrants the conclusion that large areas in northern 
Labrador and Baffin Island remained ice-free at the maximum of the Last 
Glaciation (Classical Wisconsin equivalent) and that the situation may have 
been similar along the northern sectors of the eastern seaboard and also in 
the Shickshock Mountains. This prompts the speculation that appreciable 
areas of the continental shelf may have been dry land and thus have provided 
ample habitats for the survival of a wide range of flora and fauna. Similarly 
sectors of the northern Labrador coast, at or below present sea level, where 
backed by high mountains, are Hkely to have remained ice-free. 

2. The mountain-top detritus required a considerable period of time for 
mature development and this indicates that for a similar period areas occupied 
by mature detritus remained ice-free or covered by only thin, stagnant ice for 
part of that time. However, most, if not all of the high Labrador mountains 
were completely submerged some time during the Pleistocene. 


3. Wide areas of the High Arctic may have been covered by cold, thin, 
stagnant ice, of which no trace remains today. 

From these three points it is apparent that much more systematic research 
is needed. The following problems present themselves for urgent 
consideration : 
Outline of Problems for Future Investigation 

1. How does mountain-top detritus form? Detailed quantitative studies 
of its structure and internal temperature conditions are required under 
existing cHmatic conditions. Studies of the structure should establish whether 
or not individual boulder fields pass downwards into bedrock and are 
therefore the product of weathering in situ. Similarly, additional mineralogical 
studies and intensive work on areal and vertical distribution in areas of 
contrasting rock types are needed. 

2. Further investigations are needed to estabhsh the presence or absence of 
glacial erratics on high coastal summits. Thus the Kaumajet Mountains, 
southern Torngat, and Cape Dyer become critical areas. In the Cape Dyer 
area it is beUeved that the outer coastal mountains, which rise to elevations 
exceeding 1200 m, are composed of a Tertiary volcanic sequence. The 
Tertiary outcrop is not more than ten miles wide and farther inland the 
mountain country is composed of the regular Archaean gneisses. This should 
provide an ideal situation for such a study. In any future attempt to estabhsh 
the occurrence of erratics at high altitudes, Dahl's criteria should be used 
(Dahl, 1961). 

3. Perhaps the most fruitful line of approach would be a more extensive 
study of the history of glaciation in Labrador-Ungava and Baffin Island. In 
this, particular attention should be paid to the problem of dating the lateral 
and end moraine systems and to relating them to late and Post-glacial marine 
phases. Especially important is the need to follow up the hypothesis that 
Hudson Strait may have been largely ice-free during the Classical Wisconsin 

All of these problems are being investigated as part of the northern research 
programme of the Geographical Branch, Department of Mines and Technical 
Surveys, Ottawa, Canada. 

Acknowledgments — The author expresses his indebtedness to numerous 
colleagues who have either directly assisted in the field work or who have 
discussed many aspects of the problems of interpretation of the field data. 
Special thanks are due to Professor R. F. Fhnt, Dr. R. P. Goldthwait, Dr. E. 
Dahl, Drs. Askell and Doris Love, Professors G. Hoppe and S. Rudberg, for 
their many critical discussions and suggestions. In particular, the author was 
fortunate to accompany Dr. Dahl on an excursion through parts of central 
and western Norway in 1960 where the striking field evidence relating to the 

354 J. D. IVES 

mountain-top detritus and the nature of its lower limit (detrital trim line) was 
so ably demonstrated. Finally, special thanks are due to Dr. J. Ross Mackay 
for critically reviewing the manuscript. 


Andrews, J. T. (1 961). The glacial geomorphology of the northern Nain-Okak section of 

Labrador. Unpubl. M.Sc. thesis, McGill Univ., Montreal, 1961. 280 p. 
Andrews, J. T. and Ives, J. D. (1963). Studies in the physical geography of north-central 

Baffin Island, N.W.T. Geogr. Bull. No. 19. In press. Geographical Branch, Ottawa. 
Andrews, J. T. and Matthew, E. M. (1961). Geomorphological studies in northeastern 

Labrador-Ungava. Geogr. Paper, No. 29. 29 p. Geographical Branch, Ottawa. 
Bergstrom, E. (1959). Utgjorde Lofoten och Vesteralen ett refugium under sista istiden? 

Svensk Naturv. 12, 116-122. 
Beschel, R. E. (1961). Dating rock surfaces by lichen growth and its application to glaci- 

ology and physiography (lichenometry). In Geology of the Arctic, Univ. of Toronto 

Press, 1044-1062. 
Coleman, A. P. (1921). Northeastern part of Labrador and New Quebec. Geol. Siirv. 

Canada, Mem. 124. Ottawa. 
Dahl, E. (1961). Refugieproblemet og de kvartaergeologiske metodene. Svensk Natiirv. 

14, 81-96. 
Daly, R. A. (1902). The geology of the northeast coast of Labrador. Harv. Coll. Miis. 

Comp. Zoology Bull. 38, 205-270. 
Falconer, G. (1962). Glaciers of Northern Baffin and Bylot Islands. Geog. Paper No. 33. 

In press. Geographical Branch, Ottawa. 
Flint, R. F., Demarest. M. and Washburn, A. L. (1942). Glaciation of Shickshock Moun- 
tains, Gaspe Peninsula. Geol. Soc. Amer. Bull. 53, 1211-1230. 
Gribbon, p. W. F. (1960). Flora and fauna at Bagnall Fiord, central Baffin Island. Unpubl. 

MMS., Royal Military College, Kingston, Ontario. 
HoppE, G. (1959). Nagra kritiska kommentarer till diskussionen om isfria refugier. Svensk 

Naturv. 12, 123-134. 
IvES, J. D. (1957). Glaciation of the Torngat Mountains, northern Labrador. Arctic 10, 

IvES, J. D. (1958a) Glacial geomorphology of the Torngat Mountains, northern Labrador. 

Geogr. Bull. No. 12: 47-75. Geographical Branch, Ottawa. 
Ives. J .D. (1958b) Mountain-top detritus and the extent of the last glaciation in northeastern 

Labrador-Ungava. Can. Geogr. 12, 25-31. 
Ives, J. D. (1960). The deglaciation of Labrador-Ungava — an outline. Cahiers de Geogr. de 

Quebec IV, 323-343. 
Ives, J. D. (1962). Indications of recent extensive glacierization in north-central Baffin 

Island, N.W.T. J. Glaciology 4 (32), 197-205. 
L0KEN, O. (1962a). Deglaciation and postglacial emergence of northernmost Labrador. 

Unpubl. Ph.D. thesis. McGill Univ. 1962. Montreal. 
L0KEN, O. (1962b). The late-glacial and postglacial emergence and the deglaciation of 

northernmost Labrador. Geogr. Bull. No. 17, 23-56. 
Low, A. P. (1906). Cruise of the Neptune, 1903-1904: Report on the Dominion Govern- 
ment E.xpedition to Hudson Pay ad the Arctic Islands. Ottawa. 355 p. 
Odell, N. (1933). The mountains of Northern Labrador. Geogr. J. 82, 193-211 and 315-326. 
Savile, D. B. O. ( 1961 ). The botany of the northwestern Queen Elizabeth Islands. Canad. J. 

Botany 39, 909-942. 
ToMLiNSON, R. F. (1958). Geomorphological investigations in the Kaumajet Mountains and 

Okak Bay (North River) region of Labrador. Arctic 11, 254-256. 
ToMLiNSON, R. F. (1962). Pleistocene evidence related to glacial theory in northeastern 

Labrador. Unpubl. MMS, 1962. 
Wheeler, E. P. 2nd (1958). Pleistocene glaciation in northern Labrador. Geol. Soc. Amer. 

Bull. 69, 343-344. 




Thorleifur Einarsson 

University Research Institute, Reykjavik, Iceland 

The pollen-analytical work in Late and Post-Glaciai bog- and lake-deposits in 
Iceland has recently been started. Previous pollen analyses have been carried 
out by Thorarinsson (1944, 1955), Okko (1956). and Straka (1956). The 
present author started his studies in 1954 and has since examined 20 bog 
profiles from different localities in Iceland. The older investigations fit well 
into the present zonation scheme (Th. Einarsson, 1961). 

This late start of such studies in Iceland is due mainly to the following 
facts. The Icelandic Flora counts only 440 species of vascular plants, when the 
genera Hieracium and Taraxacum are excluded. The only forest tree in 
Iceland is Betula pubescens-tortuosa. Therefore the pollen production is 
very low and there is a scarcity of pollen grains in the young bog and lake 
deposits (peat and diatomite mud). 

The climate of Iceland is oceanic, i.e. humid with cool summers and 
relatively mild winters. The mean temperature in July is 9°-ll^C, in January 
about C^C in southern Iceland and -2 to -6'C in northern Iceland. The mean 
precipitation is as high as 2200 mm in southern Iceland (Vik Myrdal) but only 
465 mm in Akureyri, northern Iceland. The weather is very changeable due 
to the low-pressure tracks in the North Atlantic. The mild climate is a result 
of the strong influence exerted by the Gulf Stream. 

Approximately 10 per cent of Iceland is covered with bogs which can be 
divided into two groups: 

(1) The "Floi" bogs, i.e. topogenic bogs; and, 

(2) The "Hallamyri" bogs, i.e. bogs on hill and mountain sides, which are 
fed mainly by precipitation and run-off water. 

Most of the Icelandic pollen profiles are from bogs of the latter type. 

The first two profiles show the typical trend in ordinary pollen diagrams 
from Iceland. In pollen diagram 1 from Moldhaugar, Eyjafjordur, northern 
Iceland (Fig. 1), three 5e/w/a-maxima can be seen: 

A. The first 5e/w/fl-maximum is probably of Late Glacial age. but in the 
absence of C^^ datings this cannot be verified. 



>*♦ *i 


Profit X 

MOLDHAUGAR oom o.nn) 

10 X 30 to so to 70 10 to IO(r>b 

Fig. 1 . Pollen diagram from a bog, 80 m above sea level, at Moldhaugar, Eyjafjordur, 

N. Iceland. Pollengraphs from left to right: Sali.x, Gramineae, Betiila, Cyperaceae. 

Hatched area: herbs. (Cf. also Fig. 5.) 

B. The first great 5e/i//a-maximum can be equated with the Boreal and 
the lower part of the Atlantic in continental Europe. In this zone falls the 
Hekla volcanic ash layer H5 (6600 B.P.).* This 5ef///a-niaximum is followed by 
a Betula-mimmmTi which corresponds to the wet Atlantic period of continental 
Europe. The birch forest that grew in the "Hallamyri" bogs was receding. In 

*C^'' values on Fig. 1 have not been corrected for Suess-effect. Corrected values appear in text. 


this interval appears a great maximum of Sphagnum, a plant which today is 
not spore-producing in Iceland. 

C. This Betula-minimnm is thereafter followed by a second great Betiila- 
maximum which corresponds to the Sub-Boreal and the lower part of the 
Sub-Atlantic periods. In this pollen zone, most probably in the interval 4000- 
2500 B.P., falls the Hypsithermal period in Iceland. At this time at least 50 
per cent of Iceland was covered with birch forest as contrasted to 1 per cent in 
modern times. Within this pollen zone occur two rhyohthic tephra (volcanic 
ash) layers from Hekla, H4 (4000 B.P.) and H5 (2700 B.P.). 

The numerous, partly C^^-dated, tephra layers in the Icelandic peat 
deposits greatly aid pollen analytical work. The tephra layers have been 
thoroughly studied by Thorarinsson (1944, 1954, 1958) and facilitate the 
geological work in Holocene deposits. They are also of great importance for 
the study of the volcanic history of Iceland in Post-Glacial times. 

Another feature that can be seen from the tephra layers is that the pre- 
historic ones are straight and even in the bog profiles, in contrast to the 
historic ones which are uneven and lenticular. This deformation of the 
historic tephra layers is probably a result of the beginning of the cryoturbant 
"thufur"-(hillock)-formation in Iceland in the last thousand years. The 
"thufur"-formation could indicate a climatic deterioration in historic times or 
be the result of deforestation. In places where forest and/or shrub cover is 
lacking snow is blown away by wind. Absence of the insulating snow cover 
thus results in a much more effective frost action in deforested areas. 

The Betula-cmwQ thereafter decreases generally with some local deviations 
from 2500 B.P. to the time of settlement (ninth century a.d.) in accordance with 
the climatic deterioration at the beginning of the Sub-Atlantic period. At the 
beginning of pollen zone D, historic times, the Betiila-curwQ decreases very 
rapidly and the Gramineae-curve rises, indicating the influence of man. 

In pollen diagram 2 from Sogamyri, Reykjavik, southwestern Iceland 
(Fig. 2), the pollen zone A is Betula-free. This suggests that the Betula was 
growing for a much longer time in northern Iceland east of the Eyjafjordur 
mountains to the Austfjordur mountains. Perhaps it has survived there on 
ice-free areas in the mountains and /or on a dry coastal shelf during the last 
Glacial, as this interpretation of the pollen diagrams seems to indicate. 

As has been mentioned in other lectures in this symposium, the Early 
Tertiary flora of Iceland was characterized mainly by deciduous-trees 
whereas the Late Tertiary was dominated by conifers (Pflug, 1959). With the 
setting in of the Pleistocene Glacials the heat-demanding trees were eliminated 
and only species of A hms, Betula, and Salix survived the First Glacial. In the 
second to last Glacial Alnus, too, disappeared from the flora. 

Kjartansson (1955, 1961) has shown, through studies on Glacial striae, 
that the ice-divide in late stages of the retreat of the ice-sheet in the Last 
Glacial was near the Torfajokull-area, i.e. more than 50 km south of the 






W 20 30 t 

50 60 70 »0 50 ;00°'^ 

- J)././,I 








fi c 



• L 


^ ^ 






— • L 







L •~' 






1 1 L 1 1 1 


y^ J 




c~ ' 
































-^i ^i^^ 









^•w V 



















— - ^ 










WO - 

^~ v/ 














^^ , 















a ■ — 






'■^^ ^ 




too - 





tf .^^ 





















\ V 

iOO - 

— - 
















12 S 

2> -^ 






\ — 







\ — 



9% r 


Fig. 2. Pollen diagram from Sogamyri, southwestern Iceland. 

present water-divide. Therefore it seems probable that the ice-sheet in northern 
Iceland during the Last Glacial was much thinner than in southern Iceland. 
Perhaps there was only a great valley glaciation in northern Iceland, where 
mountains and mountain ridges protruded as nunataks through the ice, as 
indicated from geomorphological studies by Thorarinsson (1937) and 
Trausti Einarsson (1959) (Fig. 3). 

On the northwestern peninsula there was another ice-center but on the 
outer mountain ridges there may have been ice-free areas. The sea level during 



-2 -o *" 

i2 c a 
W 2 P 

c 2 

^ S ^ 


c« — O 

■^ 5 h-I 







































the high point of the Last Glacial was more than 100 m lower than today. 
Due to these facts there have probably been ice-free and dry areas also on the 
northern coastal shelf of Iceland. 

The present interpretation of the pollen diagrams seems to support the 
opinion of botanists (Steindorsson, 1954; Love and Love, 1956) that a part of 
the Icelandic flora has survived on ice-free areas during the Last Glacial 
(see also Fig. 3). 

The Betula immigrates rapidly into southern Iceland about 9000 B.P. as 
the Sogamyri diagram (Fig. 2) indicates. The diagram then shows the first 
5eri//fl-maximum followed by the Atlantic Betula-mmimma.. The younger 
Betula-msLXivmxm. is thereafter very distinct. The slight rise of the Betula-curve 
in pollen zone D (historic time) may be a result of re-bedded pollen grains 
that were blown into the bogs from the hills in the vicinity. The loess soils 
were eroded very rapidly in historic times and the dust, too, was re-bedded 
partly in the bogs. 

Raised beaches are known in all parts of Iceland (Fig. 7). The evidence of 
higher sea-level are gravel terraces, old sea-cliffs and bedded marine clays with 
sub-fossil molluscs. The height of the marine hmit differs from place to place 
in Iceland. In southern Iceland (Holt, Hreppar, Landssveit) the marine Hmit 
is ca. 1 10 m, at Reykjavik ca. 45 m, in Borgarfjordur 80-100 m, and in western, 
northern, eastern, and southeastern Iceland 40-50 m, i.e. the raised beaches 
are much lower at the coast than farther inland. The highest shorehne seems 
to be synchronous all over the island. The varying height of the raised beaches 
is a result of the different downwarping of Iceland caused by the ice-sheet 
during the last glaciation, and of the subsequent isostatic recovery. 

The age of the highest shoreline is still not definitely known. The first 
investigators believed, that the marine limit was of Post-glacial age according 
to the heat-demanding molluscs in the raised marine deposits (all these sub- 
fossil molluscs exist today along the Icelandic coast), equating the Late 
Glacial Icelandic subfossil mollusc fauna with that of Scandinavia, without 
regard to the great eustatic changes of sea level and the different land-sea 
distribution in the Pleistocene. The Gulf Stream had much more influence on 
the Icelandic coast during the Glacials, and especially in Late Glacial times, 
than it did on the coasts of Scandinavia. According to Kjartansson (1958) the 
sea level had sunk approximately 10-15 m below its maximum height at the 
end of the Budi stage. He concludes, that the Biidi stage may be comparable 
to the Salpausselka-Raerne stage in Scandinavia, i.e. Younger Dryas. Pollen- 
analytical studies by the present author indicate that the highest shorelines 
are probably older, but in the absence of C^^ datings this cannot be definitely 
stated yet. 

The isostatic recovery was relatively rapid, as the sea level had reached the 
present level by 9000 B.P. The lagoon Seltjorn near Reykjavik was isolated 
at this time and normal peat formation started in the basin. Only the deepest 




part of the peat contains marine diatoms whereas the higher part has only 
freshwater diatoms (Jonsson, 1956, 1957). The isostatic recovery continued 
and kept pace with the eustatic sea level rise during the Boreal and Atlantic 

A pollen diagram from Seltjorn (Th. Einarsson, 1956, 1961) shows the 
first great Betula-maximum (pollen zone B), the following Atlantic Betula- 
minimum, and the first part of the second great 5e/»/a-maximum. The 
transgression submerged the Seltjorn bog and stopped the peat formation in 
Sub-Boreal time. Seltjorn is today a bay. The subsidence in this area is more 
than 4 m, and the transgression is still continuing. Submerged peat has been 
found in many localities in Iceland (Bardarsson, 1923; Thorarinsson, 1956; 
Jonsson, 1957). Only at Hri'itafjordur, northern Iceland, is there evidence of a 
transgression in Sub-Boreal time, the Purpura (Nucella) transgression, with a 
following regression (Bardarsson, 1910; Thorarinsson, 1955). Trausti 
Einarsson (1961) has argued against the transgressive trend on the Icelandic 
coasts in the last few thousand years, and denies the submerged peat as proof 
for a subsidence. 

As mentioned before, the birch forest dechned from ca. 2500 B.P. to the 
onset of the settlement, the "landnam". In a short history of Iceland (Islend- 
ingabok), written about 1 120, Ari the Wise Thorgilsson (1067-1 148) says that 
Iceland was covered with woods between "coast and mountain", when the 
"landnam" began. Iceland was colonized mainly from Norway in the years 
A.D. 870-930. After the beginning of the "landnam", the birch forest was 
devastated very rapidly as can be seen from the detailed pollen diagram from 
Skalholt in southern Iceland (Fig. 5). Skalholt was probably settled early in 
the tenth century and in 1056 the farmstead became the residence of the 
Icelandic bishops and was until the end of the eighteenth century the cultural 
center of Iceland. In the pollen diagram from Skalholt (Fig. 5) the first column 
shows the sediment (Carex-moss-peat) with the tephra layers K-ca. 5500 
B.P., H4 - 4000 B.P., H3 -2700 B.P., and G, which according to the pollen 
studies was formed during an eruption in the Torfajokull area, southern 
Iceland, during the "landnam" period, i.e. a.d. 870-930. The rhyolithic 
tephra layer Hi was formed in a Hekla eruption in the year a.d. 1104. 
This ash-fall destroyed the district Thjorsardalur 30 km east of Skalholt. 
The two top-most basaltic tephra layers were deposited during the eruptions 
of Hekla in 1693 and of Katla in 1721 respectively. 

Column A shows the proportion of inorganic-organic constituents of the 
peat. The inorganic constituents are much higher in historic times than 
before, because of the sedimentation of eolian dust from the soil erosion that 
began after the settlement (Bjarnason, 1952; Th. Einarsson, 1961; Thorar- 
insson, 1961). This soil deflation began early in historic times. 

Column B shows the proportion of Cyperaceae pollen (shaded) to all other 
pollen grains. Column C shows the proportion of Salix, Gramineae, and 

■I. b ^P-Cyp 
5% 10% 5% 10% 20% 

fHO Anal. Thorteifur Emorsso. 

Fig. 5. Pnllcn diagram from a bug at Skalholl, South Iceland. Explanation of symbols in the figure. 

Facing page 362 


Betula in contrast to the herb pollen (shaded) after the Cyperaceae pollen 
have been subtracted (i^P-Cyp.). The diagram shows clearly the second great 
5erj//a-maximum, whose optimum lies between H4 and H3, i.e. 4000-2700 
B.P. The Betula dechnes thereafter toward the "landnam" but the herbs 
increase (cf. Rosaceae). The Salix-c\iv\Q goes as high as to 10-20 per 
cent in Skalholt, and the Gramineae make up 5-10 per cent. With the "land- 
nam" the "wild" Gramineae increase very rapidly and are very characteristic 
of the influence of man in Iceland. A similar increase of "wild" grasses was 
found by Iversen (1934, 1953) in the vicinity of the Nordic ruins in western 
Greenland, and by Thorarinsson (1944) in Thjorsardalur, southern Iceland. 

Column D shows the differentiated herb curve, calculated on the total 
pollen minus Cyperaceae (ZT-Cyp.). The number of "anthropophilous" 
plants increases. These are Polygonum (cf. aviculare), Caryophyllaceae 
(cf. Stellaria media, cf. Cerastium caespitosum), Rumex, Compositeae, 
Ranunculaceae, Galium (cf. verum), Plantago (cf. maritima) and new species 
like Valeriana, Linum (one grain from Skalholt), Myrica Gale and Artemisia. 
The last two plants do not grow wild in Iceland today, but have probably 
been raised in older times as medicinal plants, or been used for the brewing of 

Another interesting plant is Horcieum, cultivated in Iceland during the first 
centuries of historic times. After the fifteenth or sixteenth centuries Hordeum 
was not grown any more in Iceland, perhaps because the cHmate became 
worse or for some economical or other unknown reasons. At the end of the 
seventeenth century the Gramineae-curve shows a new rise at Skalholt and 
the small Cerealia-msiXimnm perhaps indicates attempts to raise crops. 
Pollen of Elymus arenarius is a serious source of errors in studies of the 
Cerealia, since this species grows all over the country today, in coastal parts 
as well as in the highlands in loose eohan sand deposits. Hordeum and 
Avena sativa thrive in Iceland today again. 

Plants which decreased after the beginning of the "landnam" besides 
Betula and Salix, are the Rosaceae (Comarum and Filipendula) and the 
Umbelliferae (Archangelica officinalis, Angelica silvestris). After the deflation 
had cleared the soils down to the barren-ground moraines, pollen of Thalictrum 
increases in many profiles. 

Column E indicates a relatively small long-distance pollen transport 
of mainly Pinus and Alnus (ca. l%o). There is apparently much more of long- 
distance transport in prehistoric times than later, perhaps because of the 
higher pollen production in Iceland after the "landnam" or because of the 
deforestation of the nearest countries: Iceland lies 1000 km from Norway and 
800 km from Scotland. 

Column F concerns the evidence of spores. Filicales are decreasing with 
the "landnam", but Selaginella (selaginoides) is increasing. The highest part 
of the Atlantic SphagnutJi-nvdximum is clearly seen (300 -^ 60 per cent). 


From the observations referred to above it is obvious that the same Late 
and Post-glacial climatical changes have taken place in Iceland as in continen- 
tal Europe and that they are reflected in the vegetation pattern, though the 
number of indicative species is low. The dating of the changes, especially in 
Post-glacial and historical times, is greatly aided by the tephra layers. In 
historical times man's devastating influence on his surroundings is amply 
verified in the poflen diagrams. These show four pollen zones: 

A. Late Glacial + Pre-Boreal. In northern Iceland there is a small Betula- 
maximum, whereas southern Iceland is Betula-freQ. This could perhaps 
support the theory that a part of the Icelandic flora has survived the Last 
Glacial in ice-free refugia in northern Iceland. 

B. Boreal + Atlanticum. The first great Betula-maximum is followed by a 
Be tula-minimum with a Sphagnum-ma.ximum caused by the higher humidity 
in Atlantic times. 

C. Sub-Boreal + lower part of Sub-Atlanticum. Second great Betula- 
maximum. After 2700 B.P. the chmatic deterioration sets in, 

D. Historic time, from a.d. 870. The influence of man is reflected; the 
Betula-foTQSt decreases rapidly, and there is a sharp increase of the grasses as 
well as the appearance of cultural indicators. 


BArd ARSON, G. B. (1910). Traces of changes of climate and level at Hiinafloi, Northern 

Iceland. Intern. Geol. Congr. : 347-352. Stockholm. 
Bard ARSON, G. B. (1923). Old sea-deposits in Borgarfjordur and Hvalfjordur. Soc. Sci. 

Island 1, 1-118. 
Bjarnasson, O. (1952). Islenzki morinn. Fjolrit Rannsoi ardds 3, 1-100. 
EiNARSSON, Th. (1956). Frjogreining fjorumos ur Seltjoixi. Ndttiintfr. 26, 194-198. 
EiNARSSON, Th. (1957). Tvo frjoHnurit lir islenzkum myrum. Arsrit Skognektarf. Islands 89- 

EiNARSSON, Th. (1961). PoUen-analytische Untersuchungen zur spat- und postglazialen 

Klimageschichte Islands. Sonderveroff. Geol. Inst. Univ. Koln 6, 1-52. 
EiNARSSON, Tr. (1959). Studies on the Pleistocene in Eyjafjordur. Soc. Sci. Islands 33, 1-62. 

EiNARSSON, Tr. (1961). Das Meeresniveau an den Kiisten Islands in post-glazialer Zeit. 

N. Jahrb. Paldontolog., Mh., 9, 443-473. 
IVERSEN, J. (1934). Moorgeologische Untersuchungen auf Gronland. Medd. Dansk. Geol. 

For. 8, 341-358. 
IVERSEN, J. (1953). Origin of the flora of western Greenland in the light of pollen analysis. 

Oikos 4, 85-103. 
JoNSSON, J. (1956). Kisilthorungar i Seltjarnarmonum. Ndtturufr. 26, 199-205. 
JoNSSON, J. (1957). Notes on the change of sea-level m Iceland. Geogr. Ann. 39, 143-212. 
Kjartansson, G. (1943). Arnesingasaga. Reykjavik. 
Kjartansson, G. (1955). Frodlegar jokulrakir. Ndtturufr. 25, 154-171. 
Kjartansson, G. (1958). Jardmyndanir i Holtum og nagrenni. Univ. Res. Inst. Dept. of 

Agric. Rep. B. 11, 1-23. 
Kjartansson, G. (1961). Um jardfraedi vesturhluta Arnessyslu. Arb. Ferdaf. islands 1961, 



Love, A, and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic flora. Acta 

Horti Gotobiirg. 20, 65-291. 
Okko, V. (1956). Glacial drift in Iceland. Acta Geogr. Fenn. 15, 1-133. 
Pflug, H. D. (1959). Sporenbilder aus Island und ihre stratigraphische Bedeutung. A'^. 

Jahrb. Geol. Palcioiitolog., Abh. 107, 141-172. 
Steindorsson, S. (1954). Um aldur og innflutning islenzku florunnar. Arsrit Rcektiinarfel. 

Norduiiands 51, 3-23, 53-72, 101-115. 
Straka, H. (1956). Pollenanalytische Untersuchungen eines Moorprofiles aus Nord- 

Island. N. Jahrb. Geol. Palcioiitolog., Mh. 6, 262-272. 
Thorarinsson, S. (1937). The main geological and geomorphological features of Iceland. 

Geogr. Ami. 19, 161-171. 
Thorarinsson, S. (1944). Tephrokronologiska studier pa Island. Geogr. Ann. 26, \-l\l. 
Thorarinsson, S. (1954). Tephra-fall of Hekla on March 29th 1947. The eruption of Hekla 

1947-1948, II, 3, 1-68. 
Thorarinsson, S. (1955). Nakudungslogin vid Hunafloa i Ijosi nyrra aldursakvardana. 

Ndttiirufr. 25, 172-186. 
Thorarinsson, S. (1956). Morinn i Seltjorn. Ndtturufr. 26, 179-193. 

Thorarinsson, S. (1958). The Orc^fajokull eruption of 1362. Acta Nat. Island U, 2, 1-99. 
Thorarinsson, S. (1958). Uppblastur a Island i Ijosi oskulagarannsokna. Arsrit Skogrcekt- 

arfel. Islands 1961, 17-54. 
Thoroddsen, Th. (1906). Island. Grundriss der Geographic und Geologic. Peterm. Mitt 

Ergans. Heft, 152-153, 1-358. Gotha. 


G. Erdtman 

Palynological Laboratory, Stockholm-Solna. Sweden 

The broad outline of Post-glacial plant development was emphasized by 
Andersson (1909) more than 50 years ago. His conclusions were based mainly 
on the occurrence of megascopic fossils in Scandinavian and Finnish peat 
deposits. It was, however, von Post (1918) who, in the second decade of the 
present century, introduced pollen statistics as a new botanical and geological 
research method, thereby inaugurating the present epoch of pan-global 
research into vegetational and climatical history. The main stages of the 
development, which can be condensed and vividly portrayed by following the 
story of "cryocrats", "protocrats", etc., have been elucidated by von Post 
{loc. cit.) and those who have followed in his footsteps (e.g. Firbas, 1949, 
1952; Iversen, 1958). 

Statocrats. Statocrats, as here defined (Greek statos = stationary, kratos = 
power), live in stable environments. In Quaternary climatic cycles — from an 
Ice Age to an Interglacial period to a new Ice Age and so forth — the first 
statocratic biota are cryocrats (Greek kryos = ice), i.e. elements that can 
live under the severe conditions of an Ice Age. They are gradually replaced 
by protocratic biota (Greek protos = first) which characterize the climatic 
amelioration (after the Ice Age). These, in turn, are succeeded by the meso- 
crats (Greek mesos = middle) of the Interglacial or Post-glacial Chmatic 
Optimum. Eventually telocrats (Greek telos = end) close the circle, forming 
a connecting link between the mesocrats and the cryocrats of a new Ice Age 
(cf. Iversen, 1958). 

Apocrats. Apocrats, as here defined (Greek apo = back, away), do not 
live in stable environments. Apocrats are opportunists, kinetophilous elements 
to which lack of competition generally means more than climatic and edaphic 
conditions. Thus granted freedom from competition, they can appear at 
practically any time, be it during an Ice Age or a Climatic Optimum. They 
invade virgin soil uncovered by a retreating glacier, the sudden drainage of 
an ice-dammed lake, etc. They hkewise spread into areas emerging from the 
sea as the result of eustatic sinking of the Ocean level or isostatic upheaval of 
land. They also enter areas where a temporary freedom from competition 
accompanies the upsetting of the natural balance by axes, plows, bulldozers, 
etc., in fields and forests, along liighways and railroads, in the outskirts of 
villages and towns, etc. Intermittent changes induced by volcanic activity or 



earthquakes, etc., and non-intermittent changes caused by erosion and 
deposition, etc., also favor certain apocrats. They may invade wide stretches 
of land and remain for a long time as the dominating element if the area in 
question — hke the Alvar steppe on the Isle of Oland, Sweden — offers some 
kind of perpetual instabiUty (in Oland the shallow soil is broken up each 
year by frost action). Usually, however, they are confined to narrow belts, 
as, for example, the Sea Buckthorn {Hippophae rhamnoides) in the tension zone 
between sea and forest along the slowly emerging shores of the Gulf of 

But conditions have changed. In this respect Hippophae provides an 
instructive example. As the continental ice withdrew from Sweden, this 
species followed the receding ice-border not only along the coasts but 
practically all over the country. Fossil Hippophae pollen grains were identified 
by Halden (1917), but it was von Post (1918) who emphasized the importance 
of these pollen grains as indicators of open, treeless areas. This, it seems, is 
the first important contribution to the theme "Palynology and Pleistocene 

Surface samples. The present often provides clues for unlocking the 
secrets of the past. However, attempts to draw historical conclusions from the 
actual distribution of biota have not always been successful. It is impossible, 
for instance, to draw any conclusions from the maps of the present-day 
distribution of Artemisia spp., Centaurea cyanus, chenopodiaceous plants, 
Ephedra sp., Helianthemum cf. oeJandicum, etc., regarding the former occur- 
rence of these species in the wake of the receding ice in northwestern Europe. 

On the other hand, it is easy to study, practically all over the globe, the 
way in which the vegetation is reflected by the "pollen rain" of the present day. 
The amount and composition of the pollen rain can be ascertained by trapping 
pollen grains in different ways, including the collection of those that sink in 
lakes, etc. The composition of pollen rains, the effect of long-distance trans- 
port, etc., can also be studied by subjecting surface samples from Uving bogs 
(hchen thaUi, moss tufts, and similar material) to pollen analysis. 

In connection with the theory of pollen analysis, the effect of long-distance 
transport of pollen grains was first studied by Hesselman (1919). A comparison 
between the forests and the distribution of the different tree species in a certain 
area on one hand, and the pollen spectra in surface samples from the same 
area on the other was made a few years later (Erdtman, 1921). In 1943 a 
study of different types of pollen spectra was pubhshed, including "habita- 
tion spectra" particularly rich in pollen grains of "apocrats" (Erdtman, 

Some apocrats produce pollen grains in great profusion, e.g. Ambrosia, 
Artemisia, Urtica, and certain chenopodiaceous species. In northern Sweden 
many surface samples contain at least a trace of Artemisia pollen although 
botanists roaming about in this part of the country may search for days 


without encountering a single specimen. If, however, the apocrats with 
regard to their pollen grains behave like Trifolium pratense, for instance, 
there will be only a shght chance, if any, of finding their pollen in surface 

From the study of surface samples — unfortunately much neglected hither- 
to — it is evident that pollen statistics affords a means of tracing the history 
of at least some apocrats and thus, at the same time, of the ecological condi- 
tions of which they are characteristic. 

Fossil apocrat pollen grains. Hippophae pollen has already been dealt with. 
Another interesting pollen type, at least in part apocratic, is Artemisia. These 
pollen grains were identified after having been referred to for many years 
as ""Salix 2" by early pollen analysts, even though they realized that the 
pollen grains were not produced by species belonging to that genus. 

Hippophae and at least some of the Late Glacial Artemisia are examples of 
ananthropochorous apocrats whose appearance and dispersal have not been 
influenced by man. The history of anthropochorous apocrats as well as other 
plants, the distribution of which was disturbed by deforestation through 
clearing and cultivation, has been dealt with in great detail by Iversen in his 
important "landnam" paper (Iversen, 1941). Special attention should be 
paid to Centaurea cyanus, generally believed to be a typical anthropochorous 
element that attained its present distribution in Europe in connection with the 
cultivation of rye. After a report — in 1948— on findings of pollen grains of 
Centaurea cyanus in Late Glacial deposits, several stray finds of this charac- 
teristic pollen type have been made in various countries. Of particular interest 
is the report by Schmitz (1957) on pollen grains of Centaurea cyanus in a bog 
in northern Germany, from Late Glacial deposits right up to recent layers. 

The finds were made not far from Kiel. In the neighborhood there is a place 
called Schwedeneck where the steep slopes of the Baltic end moraine are 
continuously being eroded and broken down by the sea. In the loose slopes 
grow the moss Dicranella varia and many other plants of a more or less 
decidedly apocratic habit. A narrow path borders the upper margin of the 
slope a foot or two from the declivity. Between this path and the slope is a 
characteristic assemblage of plants, including, i.a., Artemisia and Centaurea. 
There they are secure: nobody, neither man nor beast, walks along the 
edge of an abyss. Theoretically this assemblage could be a rehct of Late 
Glacial vegetation that has maintained itself through all the vicissitudes of 
cUmate and vegetational change in Post-glacial times. Our scanty knowledge 
of the history of these plants seems to be due to the fact that their pollen 
grains cannot be expected to have been accumulated in appreciable numbers 
in bogs and other polliniferous deposits because such are rare in the vicinity. 

A sort of parallel to the conditions at Schwedeneck is found at Mount 
Omberg in southern Sweden. Here it may be possible, as pointed out, e.g. 
by Hedberg (1949), that some plants now growing on or near the mountain 


are direct descendants from pioneers (partially at least of an apocratic habit) 
which, as described by Erdtman (1949) Uved on a terrain that protruded as a 
small nunatak from the surrounding, rapidly shrinking continental ice. 
Among the pioneers were some plants now confined, in their capacity of more 
or less kinetophilous elements, to Oland, particularly on the Alvar steppe 
{Helianthemum cf. oeJondicum, Artemisia cf. riipestris). Kinetophilous plants 
like these, disUking much competition, probably had a chance of surviving, 
e.g. in the sometimes almost vertical slopes of Mount Omberg facing Lake 
Vattern. No traces of them have been found there, however: a continuous 
existence as botanical "cat-burglars" over 10,000 years is not easily accom- 

Detailed studies of the pollen or spore morphology also in "trivial" 
elements in the flora of Iceland and Scandinavia, etc., can lead to new, 
unexpected vistas. Thus, at the Palynological Laboratory in Solna, about 
10,000 pollen slides of Scandinavian plants have been made and distributed 
to a number of scientific institutions in Sweden and Finland. Among these 
were also slides of Rorippa silvestris. One Rorippa slide, however, was returned 
with the comment that a mistake must have been made since pollen grains of 
the same type as those in the slide "do not occur in the Cruciferae". A check 
revealed that there was no mistake. Pollen grains from 40 different specimens 
of Rorippa silvestris from various parts of Europe were investigated. The 
grains in "statocratic" individuals were, it seemed, all normal, 3-colpate, 
whereas those in more or less sterile individuals of a somewhat "apocratic" 
habit (growing along roads, in grass swards in Stockholm and suburbs, etc.) 
were deviating. In certain examples the pollen grains were devoid of apertures 
or were provided with five to six colpi or colpoid apertures. In some specimens 
the outer part of the exine (the sexine) was reticulate, in others retipilate, i.e. 
provided with drumstick-shaped processes arranged in a reticuloid pattern. 
In some grains, finally, the sexine was considerably thinner than the inner 
part of the exine (the nexine). Other grains were, it seemed, destitute of 
nexine or almost so. A cytotaxonomic analysis of the Rorippa silvestris- 
complex is now being carried out at Uppsala (for further notes, illustrations, 
etc., cf. Erdtman, 1958). 

I have deliberately dealt with Rorippa silvestris at some length, mainly for 
two reasons. The first is that this palynological investigation revealed some- 
thing actually going on before taxonomists and cytotaxonomists had ob- 
served that there was anything "wrong" or peculiar with the Linnean species 
Rorippa silvestris. The second reason is strictly palynological: which factors 
are responsible for the variation of aperture patterns and the details of 
exine stratification? This is a question of great interest. Critical investigations 
into this theme would be much more worthwhile than uncritical speculations 
on the taxonomic and phylogenetic bearing of certain pollen grain characters 
now in vogue. 


In Speaking of the border-lines between palynology, cytology, and cyto- 
taxonomy, a few words may be added concerning pollen grains in some species 
of Sanguisorba (cf. Erdtman and Nordborg, 1961). Here, from size and other 
morphological details, it seems possible to determine the chromosome race 
(2n = 28 or 56, etc.) to which the pollen grains belong. This is true both with 
regard to fresh grains and, in certain cases at least, with regard to fossil, 
e.g. Late Glacial, pollen grains. The example favors a view expressed by 
Stebbins (1959): "In view of the fact that the characteristics of pollen which 
distinguish modern species are becoming much better known while discoveries 
and analyses of fossil pollen are also greatly increasing in number, the com- 
bination of these two types of data for the purpose of tracing out the ancestry 
of polyploid complexes would seem to be a valuable new avenue of approach 
which deserves attention." 

Apocrats and the history of the Icelandic biota. To me, mention of the 
nature of Iceland conjures up the vivid description in the book "Till Hackle- 
fjall" by the Swedish author and artist Albert Engstrom. Therefore I must 
restrain myself and point out only one or two things that may possibly be of 
interest to future palynological investigators concerned with ecological and 
phytogeographical conditions in Iceland. 

The mixture of ice and fire, of volcanism and glaciation, contribute to 
make Iceland a rewarding place for the study of apocratic biota, "obligate" 
as well as "facultative". Furthermore, Iceland is only sparsely inhabited and 
one may thus expect a more distinct difference — or at least one more easily 
disentangled — between apocrats and statocrats than in densely populated, 
heavily trafficked areas. 

"Tephrochronology" is a term introduced by Thorarinsson (1944). Not 
only volcanic ash (Greek tefra = ash) as indicated by this term, but also 
other products of volcanic activity have interfered with the natural vegetation. 
Which are, or were, the main phases of regeneration? Which apocrats 
correspond, ecologically, to, for example, the scattered, very characteristic 
Eriogonum spp. around Crater Lake, Oregon, or to the "pioneers" on the slopes 
of Mt. Etna and Mt. Vesuvius? 

And, with regard to the immigration of plants on virgin soil laid bare at the 
retreat of glaciers, which plants in Iceland correspond to those invading 
similar soils in other parts of the world, in Alaska, at the end of the Rhone 
Glacier in Switzerland (cf. Flora 146, p. 386, 1958), etc.? 

The interrelationship between certain apocratic elements along the sea coast 
and common "weeds" as well as certain "continental" elements on high 
mountains, inland plateaus, etc., far from the coasts or other places where 
their nearest relatives occur, ought to be studied (cf. also Godwin, 1960). 
Ten to twenty years ago some of the leading taxonomists and phytogeo- 
graphers apparently paid only scant attention to some of the apocratic news 
of that time (e.g. the occurrence of pollen grains of Artemisia, Centaurea 


cyanus, and Ephedra in Late Glacial deposits; of Artemisia, Chenopodiaceae 
and Cruciferae side by side with Hippophae in Oland, Sweden, and in the 
interior of southern Lapland just at or after the disappearance of the last 
remnants of dead ice). Compare also the finding in the New Siberian Islands 
of pollen grains of Artemisia and chenopodiaceous plants in layers which, it 
seems, were laid down during a particularly apocratic regime. In our days they 
no longer live in these places (cf. Gorodkov, 1954). 

What is needed are more detailed investigations carried out with refined 
techniques and sound judgement, without exaggeration of the duration and 
importance of open ground conditions and so forth. 

Pollen morphology, cytotaxonomy, and advanced microscopical techniques. 
In Scandinavia the number of tree species is quite suitable with regard to 
pollen statistics. In the U.S.A. the situation is often more complicated owing 
to the great number of tree species. In Iceland conditions are not ideal either 
because of the low number of species. Yet it is possible (cf., for example, 
Praglowski, 1962) to distinguish pollen grains of Be tula nana from those of the 
tree birches (B. pubescens and B. tortuosa), and — if the grains are typical and 
well preserved — often also between the grains of the two latter species. 
Hybridism — hitherto not adequately studied from a palynological point of 
view — will no doubt make the matter more complicated. Attention should be 
drawn to the importance of suitable embedding media, adequate optical 
facilities, etc. (Berglund, Erdtman and Praglowski, 1959). Electron micro- 
graphs exhibiting the fine rehef of the exine surface (PI. V, 1) are often very 
instructive and helpful also to those working with an ordinary fight micro- 
scope. They demonstrate sharply and precisely some of the features which 
can be seen only dimly by means of the latter. Thus, the white dots and streaks, 
e.g. in PI. I, 1, 2, 5, 7 and in PI. II, 1, 2, 4, 9 and 10, represent small spinules 
and ridges of more or less the same shape as those shown in PI. V, 1 and in 
some places in the UV micrograph PI. V, 2. 

Plates I-IV have no direct bearing on Icelandic Pleistocene problems. 
They have been inserted, however, as examples of what might be done in the 
way of making better and safer specific determinations based on pollen 
grains. Thus, tetraploid pollen grains of Alnus glutinosa (PI. I) are con- 
siderably larger than those of diploid specimens and usually also slightly 
different morphologically (cf. the ringfike "arcus" the center of 9). "Tetraploid" 
grains have, so far, only been noticed once in Swedish Post-glacial deposits 
(unpubl.). They are of much the same shape as fossil grains described from 
the Early Tertiary volcanic districts in Scotland (Simpson, 1961). Detached 
"arcus" are also found in Alnus sieboldiana (Simpson, loc. cit.). The apertures 
in A. glutinosa are, as a rule, more narrow than those in A. incana (PI. II). 
In Sweden the Quercus robur-Q. petraea problem (cf. Pis. Ill and IV) has 
not yet been tackled with pofien statistics. Investigations are, however, being 
carried out in order to determine whether some of the features shown in the 

PL I. Pollen grains of tetraploid Ainus ghitinosa from Ekebo, Sweden. 1300. 


PI. II. Pollen grains of Alniis incana 1300. 

9 "^ 10 

PI. III. Pollen grains of Quercus petraea. x 1300. 

9 — p- TO 

PI. IV. Pollen grains of Quercus robiir. ; . 1300. 

PL V. 1, Replica of pollen wall with spinules and ridges of Betula verrucosa. 
Electron micrograph 8,000. 2, UV-micrograph of Corylus avellana pollen. 

X 2000. 

PI. VI. Pollen grains of the Chinese conifer Cathaya argyrophylla. 950. Pollen 

grains — as those in PI. I. 9 — embedded in glycerine jelly; the pollen grains in 

PI. 1: 1-8, il-IV are in distilled water. 


plates — small, densely spaced processes in Q. petraea, larger, less densely 
spaced processes in Q. robur — are constant or not. Pollen grains of Tilia 
cordata can be easily distinguished from those of T. platyphylla. 

Furthermore: what is the cytotaxonomic status of Geranium silvaticum 
with comparatively small flowers and pollen grains (Erdtman, unpubl., 
specimens from Jarvso, Sweden, 1961); and what is that of the large luxuriant 
specimens of Sedum acre, Hammerfest, Norway, etc. ? 

If someone asks: "Have these things anything to do with the theme 'Paly- 
nology and Pleistocene ecology" (with special reference to Iceland)"? the 
reply is: "Yes and no"'. My intention has been to stress the importance of 
basic pollen morphological research, preferably in connection with cytological 
and cytotaxonomic studies; also to stress the desirability of investigations 
into the pollen grains and spores of statocrats as well as of apocrats, in 
surface samples of various kinds; finally, the importance of pollen statistical 
studies not only of large polliniferous deposits (as in bogs), where pollen 
grains of statocrats necessarily dominate, but also of podzolized soils and of 
any small, local, in one respect or another "queer" deposits of peats and 
sediments. Here interesting, unexpected finds may be made. The palynologist 
in charge must act as a pathfinder. He will perhaps go astray if he pays too 
much attention to the routine palynological approach: new problems demand 
new methods. This may be illustrated by a few words on the debut of 
palynology in Swedish criminology. Four experts, among them a pollen 
analyst, working independently of one another, arrived at the same conclusion : 
some dirt adhering to the clothes of a murdered person could not have come 
from the place where the corpse was found. This statement was of importance 
to the court. Later another palynologist was asked to undertake a control 
investigation. As a result he was able to testify to the great care and skill 
with which his colleague had accomplished his analyses. His conclusion, 
however, was contrary to that arrived at by the four experts: The dirt, or part 
of it, must have come from near the very place where the corpse was found. 
This conclusion was based on the fact that pollen grains of Trifolium prafeiise 
as well as zygospores of a subterranean phycomycete [Endogone sp.; det. by 
Professor J, A. Nannfeldt, Uppsala) were found in the dirt on the clothes as 
well as in one or two soil samples from near the place where the corpse was 
lying. They were not found in any other samples. Compared to these findings, 
the relative frequencies of tree pollen grains and the locally highly varied non- 
tree pollen grains must, in my opinion, be considered of minor significance, 
(cf. Fries in Nordisk Kriminalteknisk Tidskrift 31, 1961; Erdtman, ibid. 
32, 1962). 

Does Endogone occur in Iceland ? What information can be derived from 
tracing the history of Icelandic apocrats? What is the importance of "apo- 
cracy" in the development of biota? Did early Angiosperms live as apocrats, 
leaving practically no fossil record? Did some apocrats suddenly succeed, 


appearing in enormous quantities after having been more or less hidden for 
ages? Did they invade new territories, suppressing the old vegetation? Did 
they leave the soil, estabhshing themselves as "preloranthaceous" parasites, 
eventually killing the Gymnosperms or other plants upon which they lived? 
What is ''Micropinus'" of which small pinoid pollen grains are encountered in 
old Icelandic strata? Have they any connection with the small pollen grains 
in the recently described coniferous genus Cathaya from China (cf. PI. VI)? 
What about the possibility of picking out fossil birch pollen grains, etc., 
from Icelandic Interglacial and Post-glacial deposits, and having them sub- 
jected to study under the electron microscope? Replicas showing the fine 
details of the exine relief can, with some training, be made without too great 
difficulty, as shown by M. Takeoka (unpubl.). Replicas of a number of 
pre-Quaternary pollen grains and spores have been made at the Palynological 
Laboratory, Solna. This would mean another approach to the study of 
stability and instability, in the course of the ages, of certain characteristics in 
special plants. Which species are static, or which are in statu nascendi? 
What is, in evolution, the real importance of apocracy versus statocracy? 


Andersson, G. (1909). The climate of Sweden in the Late-Quartcrnary period. Facts and 

theories. Iv. Geol. Unders. C 218. 
Berglund, B., Erdtman, G., and Praglowski, J. (1959). Nagra ord om betydelsen av 

inbaddningsmediets brytningsindex vid palynologiska undersokningar. Svensk Bot. 

Tidskr. 53, 462-468. 
Erdtman, G. (1921). Pollenanalytische Untersuchungen von Torfmooren und marinen 

Sedimenten in Siidwest-Schweden. Ark.f. Bot. 17 (10), 1-173. 
Erdtman, G. (1943). Pollen spectra from Swedish plant communities. Addendum: Pollen 

analytical soil studies in southern Lapland. Geol. Foren. Stockh. Forhandl. 65, 37-66. 
Erdtman, G. (1949). Palynological aspects of the pioneer phase in the immigration of the 

Swedish flora II. Identification of the pollen grains in Late Glacial samples from Mt. 

Omberg, Ostrogothia. Svensk Bot. Tidskr. 43, 46-55. 
Erdtman, G. (1958). Uber die Pollenmorphologie von Rorippa silvestris. Flora 146, 

Erdtman, G., and Nordborg, G. (1961). tJber Moglichkeiten die Geschichte verschiedener 

Chromosomenzahlenrassen von Sanguisorba officinalis und S. minor pollenanalytisch 

zu beleuchten. Bot. Notiser 114 19-21. 
FiRBAS, F. (1949, 1952). Spat- und nacheiszeitliche Waldgeschichte Mitteleiiropas nordlich 

der Alpen. I, II. Jena. 
Godwin, H. (1960). The history of weeds in Britain. In Harper, J. (ed.): The Biology of 

weeds. Oxford, 1-10. 
GoRODKov, B. N. (1954). La palynologie en Russie 1. Paysages pleistocenes peri-glacieres 

en Asie du Nord. Bot. Notiser 1954, 90-94. 
Halden, B. (1917). Om torvmossar och marina sediment inom norra Halsinglands litorina- 

omrade. Sv. Geol. Unders. C. 280, 1-227. 
Hedberg, O. (1949). Vegetation och flora inom Ombergs skyddsomrade. Svenska Vetensk. 

Akad. Avli. Naliirskyddsdrt. 5, 1-64. 
Hesselman, H. (1919). lakttagelser over skogstradspollens spridnings-formaga. Medd. 

Stat. Skogsfor.wksanst. 16. 


IvERSEN, J. (1941). Landnam i Danmarks stenalder. Damn. Geol. Unders. II (66), 1-68. 
IvERSEN, J. (1958). The bearing of Glacial and Interglacial epochs on the formation and 

extinction of plant taxa. Uppsala Univ. Arsskr. 1958 (6), 210-215. 
Post, L. von (1918). Skogstradspoilen i sydsvenska torfmosselagerfoljder. Fork. 16. 

Skand. naturforskerm. i Kiistiania 1916. 
Praglowski, J. (1962). Notes on the pollen morphology of Swedish trees and shrubs. 

Grana Palynologia 3, 45—65 
Simpson, H. (1961). The Tertiary pollen-flora of Mull and Ardnamurchan. Tram. R. Soc. 

Edinb. 64 (16). 
Stebbins, G. (1959). Genes, chromosomes and evolution. In Turrill, W. B. (ed.), Vista.s in 

Botany, 258-290. 
Thorarinsson, S. (1944). Tefrokronologiska studier pa Island. Geogr. Ann. 1944, 1-217. 





Museum of Natural History, Reykjavik, Iceland 

In early June 1957 a young farmer, Helgi Bjornsson, found a small fragment 
of petrified wood in the scree of Snidagil, a ravine on the western side of 
Svinafellsfjall in the Or^efi district, southeast Iceland (see Fig. 1 and PI. I). 
H. Bjornsson is the youngest of the "Kvisker brothers" who live at the farm 
Kvisker and are renowned as keen observers of nature and very interested in 
various branches of natural science. One of these brothers, Sigurdur Bjorns- 
son, visited Snidagil shortly after his brother had found the petrified wood and 
discovered at the same place some fragmental leaf impressions. These he 
sent to the present writer, who went to Oriefi a little later, studied for some 
days the geology and stratigraphy of the plant-bearing deposits and organized 
more thoroughly the collecting of leaf impressions; this was carried out mainly 
by Sigurdur Bjornsson who has found all the most important species hitherto 
collected there. In July 1958 the author continued his stratigraphical studies 
of Svinafellsfjall with special regard to the remanent magnetism of the rocks. 
He was accompanied by a young Icelandic geologist, Thorleifur Einarsson, 
who collected samples from the plant-bearing layers mainly for pollen- 
analytical and other microscopical studies which he, however, has not yet 
found time to finish. 

The paper here submitted is a short presentation of what is currently known 
about the Svinafell layers. The author expresses his sincere thanks to his 
collaborators, S. Bjornsson and Th. Einarsson, for their valuable help. He is 
also indebted to Dr. Svend Th. Andersen (Geological Survey of Denmark) 
who has determined many of the leaf impressions collected, and to Mr. 
Tryggvi Sami'ielsson, Reykjavik, who has photographed them. 


The plant-bearing layers here given the name "Svinafell layers" form a 
part of the mountain Svinafellsfjall (older name Svinafell) in the district 
Oraefi, southeast Iceland. Svinafellsfjall (Fig. 2) is a mountain bordering on 
the OraefajokuU massif and forms the forward section of the ridge that juts 







Fig. 1. Location of Pleistocene fossil-bearing deposits in Iceland. 

Fig. 2. Svinafellsfjall and its surroundings. Base map: U.S. Army Map Service, 

Sheet 6018 IV, scale 1:50,000, striated: the Svinafell layers. A, Skjolgil; 

B, Godagil; C, Breidagil; D, Snidagil; E, Snidabrekka; F, Gapar. 


forward between SvinafellsjokuU and Falljokull (PI. II, 1). The gravel plain 
west of Svinafellsfjall is about 120 m above sea level and from this plain the 
steep west face of the mountain rises about 450 m. On this west face the 
Svinafell layers are exposed for a distance of 1.7 km, between the ravine 
Snidagil in the south and the north edge of Skjolgil in the north. For a distance 
of 1.2 km between the Godagil and Snidagil ravines, the layers form the base 
of the mountain and can be clearly seen from the highway to the west where 
their hght yellow-brown color contrasts strongly with the darker surround- 
ings (cf. Pis. II and III). It is really curious that they have not previously 
attracted the attention of geologists. It should be noted, though, that Henson 
(1955, p. 45) mentions "a sandstone formation in the succession of Svinafell" 
without further description. The layers are also exposed in a small spot, 
named Gapar, on the eastern face of Svinafellsfjall. 

S. Bjornsson has informed the author that on Skeidararsandur, in the area 
between the Skeidara river and the tourist hut, are boulders and blocks of 
material very similar to the Svinafell layers, and in the spring of 1959 he 
found distinct plant remnants, bits of grass and unclassified foliage, in one 
such block. These blocks must once have been /// situ either beneath Skeidar- 
arjokuU or in the mountains along the eastern margin of that glacier and 
thus at least 20 km distant from Svinafell. Whether these sediments are of the 
same age as the Svinafell layers is an open question. Furthermore it should 
be mentioned that both Bjornsson and the author have observed a layer of 
bedded sediments, about 30 m in thickness, in Breidamerkurfjall on the eastern 
side of Orsfajokull. On closer examination Bjornsson has not succeeded in 
finding any plant remnants in these layers, and in the author's opinion they 
are probably older than the Svinafell layers, although most likely of Pleisto- 
cene age. Their bedding is graded and they are probably glaciolacustrine. 
Petrified boulders of varied glacial sediments from this area have been des- 
cribed by Tryggvason (1952, pp. 96-98). 


As mentioned above, the Svinafell layers formi the base of Svinafellsfjall 
south of Godagil. North of that ravine the base of the layers rises gradually 
towards the north, but the contact is visible in only a few places because of 
scree and soil cover. Just north of Skjolgil the base of the layers is about 160 m 
above the plain and their thickness is about 15 m. The underlying basalt 
layers there are normally magnetized down to a layer about 100 m below the 
sediments, but farther down all layers are reversed. In the reverse layers some 
of the amygdoils are partly filled with crypto-crystalline quartz, and these 
layers are on the whole more old-looking than the normally magnetized 
overlaying layers. There may be a hiatus between the two groups, although a 
clear erosion contact is not exposed (cf. Fig. 3). 





m K 


PI. II. 1 , Svinafellsfjall and the glacier FalljokuU. The Svinafell layers reach upwards 

to the line indicated by the arrow. The peak in the background is Hvannadals- 

hnukur, Iceland's highest peak. {Photo S. Thoran'nsson.) 

2, Snidagil and Snidabrekka (cf. Fig. 4). {Photo S. Thoiarinsson.) 

PI. III. 1, The Svinafell layers at Breidagil. E, Upper grey layer; B, Lower grey layer. 
{Photo S. Thorarinsson.) 

2, Detail of the Svinafell layers near Breidagil. The arrow points to a basalt sill. 
{Photo S. Thorarinsson.) 


PI. TV. 1, The contact between theSvinatell layers and the tillite just north ofBreida- 
gil. (Plwto S. Thorariiisson.) 

2, The contact between the Svinafell layers and the tillite just north of Skjolgil. 
(Photo S. Thoraiinssoii.) 


How deep below the plain the Svinafell layers reach between Godagil and 
Snidagil cannot be said with certainty; in the author's opinion it is hardly 
more than ten meters or so. A normally magnetized basalt layer at their base 
at Breidagil is doubtless a sill. Two normally magnetized dykes cut through 
the layers and sills emerging from these dykes are intercalated between the 
sedimentary beds. One intrusive sheet connected with the northernmost 
dyke is only 5 cm thick and thoroughly conformable with the beds (PI. Ill, 2). 
The maximum visible thickness of the sedimentary layers at Snidagil is about 
140 m and, as their base a little farther north is at least 10 m lower and the 
layers are practically horizontal, the total thickness there has been at least 
150 m. Towards the south the sediments are limited by Snidabrekka, the 
base of which consists of reversely magnetized basalt-globe breccia with 
individually scattered lava nodules. 

On the west face of Svinafellsfjall as well as at Gapar on its east face, the 
Svinafell layers are covered by a 200-300 m thick series of tillites, basalt 
intrusions, and basalt-globe breccias with irregular layers of globular basalt. 
This series is capped by beds of subaerial lava. For the most part the tillite 
rests directly and uncomformably on the Svinafell layers. For a short distance, 
however, a basalt sill is intercalated between the tillite and the layers. The 
tillites at Svinafell have been described by Noe-Nygaard (1953, pp. 223-227; 
Nielsen and Noe-Nygaard, 1936, Figs. 2 and 3). Of their true morainic nature 
there is no doubt; they certainly contain beautifully striated boulders. In 
the author's opinion the two tillite beds as well as all tuff breccias between the 
Svinafell layers and the subaerial basalts 200-300 m higher up belong to the 
same Glacial, whereas a lower tillite bed in the ravines above the Svinafell 
farms may well belong — though this is not certain — to an older Glacial 
(cf. Noe-Nygaard, 1953. p. 224). 

The subaerial lava beds capping the subglacially formed rock series could 
possibly be interpreted as the top layers of a table mountain and in that case 
could belong to the same Glacial as the underlying tuff breccias. More likely, 
however, these layers belong to the following Interglacial. These basalts are 
partly covered by a bed of grayish tuff breccia containing angular boulders. 
The author, who has had only a hasty glance at this breccia, is inclined to 
interpret it as subaerial eruptive breccia of the kind found around some 
Post-glacial explosion craters such as Hverfjall, 

Over the subaerial layers is deposited a basalt-globe breccia forming the 
uppermost part of Svinafellsfjall. 


The Svinafell layers are for a great part lacustrine. The lower part of the 
formation shows a distinct macroscopic and partly graded bedding (rhytmites) 
which is horizontal except for the upper part of the contact zone towards 



Snidabrekka where the layers bend upwards. On the contact with Snida- 
brekka is found rounded gravel, which in all probability is shore gravel. 

Fig. 4. Section through the Svinafell layers in Snidagil (cf. also PI. IT, 2). 

Figure 4 shows a section through the layers in the Snidagil ravine, where the 
base of the layers is about 130 m above sea level. For the height measurements 
the author used a Locke level and an aeroplane altimeter (Kollsman Sensitive 


Altimeter), but the figures for the altitudes should not be regarded as exact. 
In the section we find the following layers : 

A. Sediment mostly covered by scree. Where visible it is like layer B, 
although more yellowish-brown in color. 

B. "Lower gray layer". Mainly consisting of thin (2-3 mm) silty-clayish 
strata; here and there fine-sandy strata, about 1 mm thick, are inter- 
calated. The color of this layer and layer E is more grayish than that of 
the rest of the sediments, which are more or less yellow-brownish. 

C. Layers, 10-50 cm thick, of fine-banded silt alternating with unhanded 
fine-sandy layers. 

D. Unhanded, fine-sandy layer. 

E. "Upper gray layer". As to color and banding very similar to layer B. 

F. Like layer C. 

G. Alternating fine-banded and unhanded layers. Average thickness of the 
unhanded ones about 0.5 m. 

H. Unhanded except for three fine-banded layers about 0.5 m thick. Grain 
size of unhanded sediments gradually increasing upwards from fine to 
about medium sandy. 

L Unhanded sediment, medium to coarse sandy, with embedded layers of 
gravel, 5-15 cm thick, and scattered rounded stones the size of a clenched 

J. Tillite. 

K. Basalt-globe breccia. 

In Breidagil and adjacent ravines the layers are fine-banded from the lowest 
visible base to the tillite-covered erosion surface. Here the rhytmites are more 
regular than in Snidagil. yet hardly regular enough to be described as varves. 
The gray layers are discernible here also and at practically the same height as 
at Snidagil. In the steep-sided Godagil ravine the layers from the base 
upwards are considerably coarser, mainly sandy, yet here and there one 
finds thin lenses and even angular fragments of finely banded and silty material 
of the same type as farther south. There are also thin layers of gravel and 
scattered rounded stones up to 15 cm in diameter. 

Just north of Skjolgil the base of the strata is about 260 m above sea level 
and the layers about 15 m thick, the contact with the overlying tillite being 
just about 5 m higher than in Snidagil (PI. IV). Here the layers consist 
mainly of sand and rounded gravel with distinct current bedding. On the 
whole we thus find an increasing coarseness of sediment both upwards and 
towards the north and it seems to have been transported from the north to a 
lake that was gradually filling up. 

According to the preliminary microscopic studies carried out by Th. 
Einarsson, the fine grained facies of the Svinafell layers show a microscopic 
banding. The felspar crystals (labradorite-bytownite), on the whole, lie 
horizontally. The sediment contains some olivine and occasional grains of 


augite, but most of the grains are basalt glass. The coating of these glass 
grains consists of a diffusely double-refracting palagonite which gives the 
sediments their characteristic yellowish-brown color. The cementing matrix 
is basalt glass without palagonitization. 

The more coarse-grained sediment is more palagonitized. It contains the 
same minerals: felspar, olivin and augite, as well as magnetite. Even here 
basalt glass is dominant and the cementing matrix is glass. 

The great thickness of the Svinafell layers is rather striking considering that 
to a large extent they are lacustrine. The thickness of the lacustrine facies 
probably exceeds 120 m. The deepest lakes now existing in Iceland are found 
either in tectonic depressions (Oskjuvatn, Thingvallavatn), in valleys over- 
deepened by glacier erosion (Logurinn), or in glacier-eroded valleys dammed 
up by constructive volcanic activity (Hvalvatn). As to the genesis of the basin 
in which the Svinafell layers were deposited, the diastrophism seems not to 
have taken any important part in it, whereas glacier erosion and damming 
up by volcanic activity may have been the main factors. The later exposing 
of the layers indicates both a great vertical erosion by ice and water and a 
lateral erosion extending the strandfiat area during the latter part of the 


Macrofossils. As mentioned above, the interest in the Svinafell layers was 
aroused when plant remnants were found there. 

A closer examination revealed that leaf impressions were rather abundant, 
although far from evenly distributed in the sediments. They are plentiful in 
Snidagil, fairly close to the old shore of the lake, and are especially frequent 
in the gray layers B and E (cf. Fig. 4), but they are found in all layers of this 
section that are accessible for closer study, i.e. all those beneath layer H. At 
Breidagil they have been found in the entire pack of sediments, and the sandy 
sediments of Godagil are rather rich in leaf impressions, especially of AInus. 
Farther north, in Halsatorfugil, one leaf of a Salix sp. was found near the 
tillite contact. 

The following plants have been found: 

Alnus. Leaf impressions of alder are by far the most abundant of all found 
in the Svinafell layers. They are especially frequent in the gray layers B and in 
E in Snidagil, probably because of proximity to the old lake shore. The species 
has been determined by Svend Th. Andersen as AInus viridis (cf. PI. V). 

Betula. Only a few impressions of birch leaves have been found and only in 
fragments. The species has not been determined. 

Salix. Leaf impressions of different species of willows have been found. One 
leaf, found by Helgi Bjornsson in a scree in Snidagil, has been determined by 
Svend Th. Andersen as Salix reticulata (PI. VII): one leaf, found by Th. 
Einarsson, in layer F in Snidagil, is almost certainly Salix lanata (PI. VII). 



Furthermore, there are a lot of willow leaves not yet determined. One type of 
leaf, probably a willow (PI. VII, 3, found in two specimens by S. Bjornsson in 
layer C. Snidagil), does not belong to any plant now growing in Iceland. 

Sorbus. An impression of one leaf fragment (PI. VI) was found by S. 
Bjornsson in the uppermost part of layer B in Snidagil. It is not determined 
with certainty, but looks more like the species now growing in Greenland 
{Sorbus decora (Sarg.) Schneid. var. groenlandica) than the recent Icelandic 
species {S. aucuparia L.) 

Primus padus? The impression of one whole leaf (Pi. VI) was found by 
S. Bjornsson in the uppermost part of layer B in Snidagil. 

Vaccinium, Empetrum. S. Bjornsson has told the author that he has seen 
impressions of leaves which he thinks are Vaccinium sp. Among samples from 
the Svinafell layers, collected by a young boy from Reykjavik. Sigurdur 
Sigurjonsson, was one with impressions which in all probability are of 

Table 1 
Pollen Counted in Samples from Layer B in Snidagil 



























































Gramineae. Some impressions of straw stalks have been found; the longest 
one, found in Snidagil, was 17 cm in length. 

Polypodiaceae. In three samples are impressions of leaves of some Dryo- 
pteris sp. 

As mentioned above, the first plant remnant found in the Svinafell layers 
was a petrified twig. Subsequently many twigs have been found, but none has 
been determined so far. 

Pollen and spores. Pollen-analytical study of the Svinafell layers is still at a 
preliminary stage. In the autumn of 1958, Th. Einarsson counted pollen in 
some samples from the lower grey layer at Snidagil. This was done in Reykja- 
vik under very poor technical conditions, and the samples were not treated 
with HF. The sediments seem on the whole to be poor in pollen and some. 


especially those of Betula, were much corroded. The count of pollen in three 
samples is shown in Table 1 . It may be added that one grass pollen is probably 
Elymus arenarhis and the Filicinae spores are mainly those of Dryopteris 

On the whole the pollen spectrum gives the same picture as the macro- 
scopic plant remnants. The most characteristic feature is the dominance of 
A Inns. Possibly the Betiila pollen is somewhat underrepresented, compared 
with Alnus, as it seems to be more sensitive to corrosion. 

Further palynological studies of the sediments are highly desirable, both 
in order to identify more species and to find out possible changes in the 
flora during the period represented by the Svinafell layers. 


Paleomagnetic dating, based on the repeated reversals of the remanent mag- 
netism of the basaltic rocks, was introduced in Iceland by Hospers (1953) and 
has since been applied in this country mainly by Einarsson (1957 a. b) and 
Sigurgeirsson (1957). It has proved very useful for the establishment of a 
chronology for the Tertiary and Pleistocene rocks in Iceland. This method 
has also been of much help in the determination of the age of the Svinafell 
layers. The author measured the magnetization of the rock of Svinafellsfjall 
at many places in the field, and for further checking, a lot of rock samples 
were measured in the laboratory by Professor Th. Sigurgeirsson. 

That the Svinafell layers are interglacial is, in the author's opinion, already 
evident from the fact that they are definitely younger than the basalt-globe 
breccia of Snidabrekka, which is a very typical subglacial breccia. The 
question is then, to which Interglacial do the layers belong? Hospers (1953, 
pp. 472-473) and Einarsson (1957b, pp. 223-224) agree with Pjeturss (1903), 
Bardarsson (1929) and Askelsson (1938, 1960) in regarding the Biilandshofdi- 
Brimlarhofdi (Stodin) fossiliferous sediments (cf. Fig. 1) as interglacial, and 
state that they are covered by series of basalt layers of reverse polarity. 
According to Einarsson, moraine-like sediments are found between some of 
these basalt beds. The Svinafell layers, on the other hand, rest on normally 
magnetized basalts that have been greatly eroded, presumably mainly by ice, 
before the deposition of the plant-bearing sediments. Thus, they cannot 
belong to the same Interglacial as the Biilandshofdi-Brimlarhofdi layers, 
which in all probability belong to the Giinz-Mindel Interglacial (cf. also 
Askelsson, 1938, 1960). The Svinafell layers therfore cannot be placed lower 
than the Mindel-Riss Interglacial. As stated above, the rocks of Svinafellsfjall 
resting on the Svinafell layers represent two thick series of subglacially formed 
rocks, separated by subaerial lavas. All these rocks are normally magnetized. 
In the author's opinion there is hardly any doubt that these rocks represent two 
Glacials, and the Svinafell layers therefore cannot be placed as high as the 
Riss-Wurm Interglacial. Consequently they must be placed in the Mindel-Riss 

PI. V. 1, Alnus viridis. § nat. size. Snidagil, layer B. 
2, Alnus viridis. | nat. size. Breidagii, layer B. 

PI. VI. 1, Sorhiis sp. n iiat. size. Snidagil, layer B. 
2, Primus /Hic/iis? | nat. size. Snidagil, uppermost part of layer B. 

PI. VTI. 1, Salix reticulata, f nat. size. Snidagil. scree. 

2, Salix lanatal f nat. size. Snidagil. layer F. 

3, Salix sp. ? f nat. size. Snidagil, layer C. 

4, Salix sp. I nat. size. Snidagil. layer B. 

5, Salix sp. nat. size, f Near Breidagil, Layer B. 

6, Fragment of leaf with palmate venation, 'i nat. Snidagil, layer E. 



IntergJacial. As to the intricate question, where to place the first reversal of 
magnetism in relation to the Pleistocene climatic chronology, we have on 
one hand Hosper's and Einarsson's observations on Snaefellsnes, which 
indicate that this reversal did not occur earlier than the end of the Giinz- 
Mindel Interglacial. On the other hand, the reversal took place long before 
the deposition of the Svinafell layers, as the basalts of normal polarity 
underlying them are greatly eroded and that erosion is probably by glaciers. 
From a study of all the evidence it seems most reasonable to conclude that 
the first magnetic reversal occurred towards the beginning of the Mindel 

Table 2 
Pollen Counted in Interglacial Sediments in Iceland 





(layer H) 

(layer K) 















A Inns 

















































Other herbs 



2 ! 2 








116 100 

111 100 



* According to Askelsson, 1938, p. 312. 
t According to Lindal, 1939, p. 269. 


Plant-bearing deposits regarded as interglacial have been described from 
the above-mentioned Brimlarhofdi (Stodin) on Snaefellsnes (Askelsson, 1938, 
1960), from Bakkabrunir in Vididalur, north Iceland (Lindal, 1935, 1939), 
and from Ellidavogur near Reykjavik (Thorkelsson, 1935; Love and Love, 
1956). Common to the sediments in Brimlarhofdi, Bakkabri'mir, and Svina- 
fellsfjall is the dominance of Alnus viridis as shown in Table 2. The occurrence 
of conifer pollen in layer H in Brimlarhofdi is somewhat suspect. Askelsson 
seems to be doubtful whether to interpret it as secondary or not; the present 
writer is inclined to regard it as secondary. As to the age of the Bakkabrunir 
formation, opinions differ. Lindal and Askelsson regard it as interglacial. 


whereas Einarsson (1958) regards it as Pliocene. The present author is 
inclined to place it in the Giinz-Mindel Interglacial. According to Lindal, 
leaf impressions of Dryas octopetala and Betula nana are common in these 
sediments and on the whole they must indicate a somewhat colder climate 
than does the Svinafell flora. But here one must take into account the 
difference in climate between the Svinafell and the Bakkabrunir areas as 

In the plant-bearing sediments at Ellidavogur no remnants of Alnus have 
been found. If we regard these sediments as most likely belonging to the 
Riss-Wijrm Intergalcial, a reasonable although not the only possible con- 
clusion will be that Pinus possibly, but not likely, survived the Giinz Glacial, 
that Alnus survived both Giinz and Mindel, but became thoroughly extinct 
during the Riss Glacial, and that Betula survived all four main Pleistocene 

The investigation of the Svinafell layers is still in its early stages, and what 
has been stated about them here is primarily intended to draw attention to 
the fact that these are layers which require systematic and detailed research, 
both macroscopic and microscopic, and both of the organic remains and of 
the sediments as such. Morphometric measurements are particularly desirable. 
The extent of the layers and their stratigraphy must also be more closely 

What has been stated here about the need for closer investigation of the 
Svinafell layers also applies to all other fossil-bearing interglacial sediments in 
Iceland. There is probably no other place where so much of such sediments 
has been preserved in a country that has been for the most part covered by 
ice during the Pleistocene Glacials, the conditions for their preservation being 
rather unique, as volcanoes were very active during both the Glacials and 
Interglacials and sediments could therefore easily be covered either by lava or 
tephra and thus be protected against glacial or fluvial erosion. Thorough and 
systematic study of the fossil-bearing interglacial sediments in Iceland is one 
of the most urgent tasks in the Quaternary geology of Iceland. 


77?^ "'Svinafell layers"" are plant-bearing sedimentary deposits which form 
the base of the mountain Svinafellsfjall in southeast Iceland. These deposits were 
discovered in 1955 and have hitherto been only provisionally studied. The 
minimum thickness of these sediments is 150 metres. Impressions of leaves are 
abundant. Characteristic for the flora is the dominance o/' Alnus. 77?^ species 
has been determined as A. viridis. Other trees that have grown in the area are 
Betula sp., Sorbus sp. and probably Prunus padus, besides many species of 
Salix. The flora indicates a climate as warm as now or somewhat warmer. 
Stratigraphical and magneto-geological studies place the Svinafell layers in the 


Mindel-Riss ( Yarmouth) Interglacial and indicate that the youngest reversal of 
the Earth's magnetic field occurred near the beginning of the Mindel (Kansan) 
Glacial. Other plant-bearing interglacial formations in Iceland are discussed 


AsKELSSON, J. (1938). Quartargeologische Studien auf Island II. Inter-glaziale Pflanzenabla- 

gerungen. Medd. Dansli Geol. Foren. 9, 300-319. 
AsKELSsoN, J. (1960). Pliocene and Pleistocene fossiliferous deposits. On the geology and 

geoplnsics of Iceland. Guide to excur ion No. A2, Int. Geol. Congr. 1960, 28-32. 
Bardarsson, G. G. (1929). Nogle geologiske profiler fra Snaefellsnes. Rep. of the ISth 

Scand navian Natiralist, Copenhagen. 
EiNARSSON, Tr-. (1957a) Der Palaomagnetismus der islandischen Basalte und seine strati- 

graphische Bedeutung. Neiies Jahrb. Geol. Palciont. Mh. 4, 159-176. 
EiNARSSON, Tr-. (1957b). Magneto-geological mapping in Iceland with the use of a compass. 

Phil. Mag. Siippl. 6 (22), 232-239. 
EiNARSSON, Tr-. (1958). Landslag a Skagafjaligardi. Ndtturufr. 28, 1-25. 
Henson, F. H. (1955). The geology of Iceland. University of Nottingham, Survey 5, 34-46. 
HosPERS, J. (1953). Reversals of the main geomagnetic field. I. Koninkl. Nederl. Akad. van 

Wetensch. Proc. Ser. B, 56 (5), 467^76. 
Li'ndal, J. H. (1935). Mobergsmyndanir i Bakkakotsbrunum og steingervingar theirra. 

Ndttwufr. 5, 97-114. 
Lindal, J. H. (1939). The interglacial formation in Vididal, Northern Iceland. Quart. 

Journ. Geol. Soc. London 95, IbX-llli. 
Love, A and Love D. (1956). Cytotaxonomical conspectus of the Icelandic Flora. Acta 

Horti Gotoburg. 20, 65-291. 
Nielsen, N. and Noe-Nygaard, A. (1936). Om den islandske "Palagonitformations" 

Oprindelse. Geogr. Tidskr. 39, 3-36. 
Noe-Nygaard, A. (1940). Suhglacial volcanic activity in ancient and recent times. 

Geogr. Dan. 1 (2), 1-67. 
Noe-Nygaard, A. (1953). Notes on the nature of some indurated moraines in South 

Iceland. Geogr. Tidsskr. 52, 222-231. 
Pjeturss. H. (1903). Shelly boulder-clay in the Palagonite-formation of Iceland. Quart. 

Journ. Geol. Soc. London 59, 356-361. 
ScHWARZBACH, M. (1955). Allgcmeiner Uberblick der Klimageschichte Islands. Neue.s 

Jahrh. Geol. Palciont. Mh. 3, 97-130. 
SiGURGEiRSSON, Th. (1957). Direction of magnetization in Icelandic basalt. Phil. Mag. 

Suppl. 6(2), 240-246. 
Thorarinsson, S. (1958). The Onefajokull eruption of 1362. Acta Nat. hi. 2 (2), 1-99. 
Tryggvason T. (1952). Steinrunniiin hvarfleir. No/r///////-. 22 96-98. 


AsKELL Love 

Institut Botanique de TUniversite de Montreal, Montreal, Canada 

The science of biogeography, in the true and wide meaning of the term,"may 
be defined as the exploration of the universe of Hving beings and the appHca- 
tion of the laws of nature to the interpretation of the history of the distribution 
and dispersal of biota. In this application the laws themselves are put to a 
test and important explanations are constructed. Such was the case with the 
theory of evolution which arose from the study of distribution, in time and 
space, of plants and animals. 

It has been said that the most important step in getting a work started is 
the recognition of a problem. Regarding the North Atlantic biota, this was 
achieved already by Hooker (1862), or perhaps even by Humboldt (1817) 
who recognized the close relationships between the biota in the far north of 
Scandinavia and America. The second important step in the development of 
this case was the collection of evidence for and against several working hypo- 
theses coined to explain the relationship. Out of this emerged the theory of 
survival of plants within the glaciated areas of Scandinavia (Blytt, 1876, 1 882; 
Fries, 1913) replacing the now merely historical tabula rasa idea. 

The theory of glacial survival has been found to be one of the most fertile 
ideas of biogeography, and much evidence has been sought in support of it 
for almost a century. Naturally, most of this evidence has come from Scandi- 
navia, but some also from North America, Greenland, Iceland, Svalbard, 
and the British Isles. Many Scandinavian biogeographers have agreed that 
this theory is the only available explanation for the distribution patterns of 
certain biota in Scandinavia (cf. Dahl, 1961 ; Gjaerevoll, 1959; Lindroth, 1958; 
Nannfeldt, 1958; Nordhagen, 1936; and in this symposium). 

Some of the reasoning regarding the North Atlantic biota and their history 
may seem to have been somewhat circular in that biologists proposed geo- 
logical possibilites and geologists tended to accept such conclusions as final. 
Recently, this has changed in such a way that most biogeographers base their 
conclusions on biological evidence only, at the same time as the conclusions 
of most geologists rest on geological observations alone. Lately, the latter 
have tended to doubt the possibility of recognizing ice-free refugia even from 
the last glaciation (Holtedahl and Rosenquist, 1958; Hoppe, 1959; and in 
this symposium), though they have not regarded it wise to deny completely 
this possibiUty because of the strong biological evidence. 




It is extremely difficult to reconstruct biological history when there are no 
fossils and if all conclusions must be based on the distribution patterns of the 
surviving populations. This has been the main problem in the investigations of 
glacial survival in Scandinavia and North America. Although palynologica, 
studies have added much information on the occurrence, recolonization and 
extinction of some plants and on certain variations in distribution patterns, 
such studies usually cover only the Late and Post-glacial periods in areas south 
of the ice-border or made available by the retreating ice (cf. Deevey, 1949; 
Godwin, 1956; Iversen, 1958). 

One of the most important contributions from this symposium is the 
furnishing of definite paleobotanical evidence in support of the glacial survival 
theory. This evidence comes from Iceland, where volcanic activity throughout 
the Pleistocene has secured the conservation of extensive fossiliferous deposits 
from the Interglacials (Thorarinsson, Th. Einarsson, in this symposium). In 
fact, the Icelandic layers show not only that plants, including trees, survived 
the glaciations within the country, but also that the flora of this island has 
diminished gradually during the Pleistocene and that some species which were 
dominant in the first Interglacials have become extinct in the latter. Logically, 
this must be regarded as a fully satisfactory proof that plants and animals 
have been able to survive the Pleistocene glaciations also in other countries 
where no paleological remains have been found and where the only indications 
left are met with in form of peculiar distribution patterns. 

This carries us back to Scandinavia. As mentioned by Fsegri (this sym- 
posium), it is possible that the bicentric distribution of certain plants in the 
Norwegian mountains may be explained as a consequence of climatical 
variations after the glaciations. This argument is, however, considerably 
weakened by the occurrence of multicentric areas in Iceland for several plants 
whose present distribution within the country must have been affected by the 
glaciation, as shown by Steindorsson (this symposium). There are, however, 
several species in Scandinavia v/hich cannot have invaded their areas from 
southern or eastern localities in Europe following the retreat of the ice for 
the simple reason that they have never grown there. These are the species 
belonging to the element termed "West Arctic" by Blytt (Joe. cit.), i.e. the 
American-Arctic element in the European flora. It is apparent from the 
statements by D. Love and Dahl (this symposium) that long-distance dis- 
persal over oceans or ice is in this case hardly even a remote possibility, because 
the radius of dispersability of these plants is much smaller than the present 
distance to their nearest populations. The hypothesis is also supported by 
the fact that these plants in Scandinavia are very rare and local though 
apparently appropriate habitats are available in many localities close by their 
present areas there. Hence, the "West Arctic" species must have arrived to 
Scandinavia over some kind of a land-connection and hardly only over the 
present lands, as partly suggested by Hulten (this symposium). The same sort 


of dispersal over land may apply also for at least the great majority of the 
species in the Icelandic flora (E. Einarsson, this symposium). It goes almost 
without saying that some of the truly Amphi-Atlantic biota in eastern North 
America must have dispersed westwards in a similar manner. In addition, 
glacial survival in Arctic America seems even more likely than in Scandinavia 
for the simple reason that there a much larger land area may have been free 
from ice north of the glaciation limits (cf. Ewing and Donn. 1956; Ives, this 
symposium; Terasmae, 1961). 

Land-connections have been discussed by several contributors to this 
symposium. Some biogeographers seem to hold the opinion, based on climatic 
requirements of the biota, that a land-bridge must have connected at least 
Iceland and Europe as recently as in late Pliocene (Dahl, this symposium). 
Others have used paleobotanical data to indicate such a hypothetical connec- 
tion as late as in the latter part of the Pleistocene (Sorensen, 1953) or during 
the penultimate glaciation itself (Love and Love, 1956). However, the 
possibility of such relatively recent connections over the Atlantic or parts of it 
remains a matter of dispute and has so far received only limited support from 
geologists (Heezen and Tharp, this symposium), but not from paleoclimatolo- 
gists (Schwarzbach, this symposium). On the contrary, geological (Tr. 
Einarsson, Schwarzbach, this symposium) and certain botanical (Hulten, 
this symposium; Steere, 1937; Love and Love, 1953) as well as zoological 
(Walden, Omodeo, this symposium) evidence seem strongly to uphold the 
opinion that Iceland (or at least "Great Iceland", cf. Barth, 1941 ; Tr. Einars- 
son, this symposium) has been an isolated island ever since the Middle 
Tertiary, and that its present indigenous biota must be regarded as only 
remnants of the flora and fauna which became isolated at that time. It 
follows, if this reasoning is correct, that at least the "West Arctic" plants in 
Scandinavia are survivors not only of the Pleistocene but relics from a still 
more distant past when there was some kind of land-connection at least 
between Greenland and Scandinavia. This is in conformity with the views 
previously expressed by Nordhagen {he. cit.) and Nannfeldt {loc. cit.). 

The question arises what kind of connection over the North Atlantic made 
possible at least a limited exchange of biota until the Middle Tertiary, but 
because of the limited geological knowledge of the sea bottom in these regions, 
the answer remains controversial. Though Tr. Einarsson (this symposium) 
seems to favor the idea of a large Atlantic island (cf. Barth, 1941) somewhat 
Hke the Scandic proposed by De Geer (1912), other possibilities remain open 
and those who prefer, for example, one land-bridge over Iceland and another 
over Svalbard (Cernohorsky, Hadac, Ronning, this symposium), or simply 
some kind of continental displacement (cf. Heezen and Tharp, this sym- 
posium) can still find support for these ideas with equally strong evidence. 
However, the data presented by Lindroth and Walden and even Dahl (this 
symposium) seem to back the opinion by Tr. Einarsson (this symposium) 


that this North Atlantic land-mass had connections northwards and eastwards 
but hardly all the way to the present American continent. Davis Strait may 
thus be older than other parts of the North Atlantic Ocean and have acted as 
an effective barrier to dispersal for a long period of time (cf. Lindroth, 1960). 
It is, however, also evident that the early Tertiary flora of Iceland was 
largely American, or, perhaps more correctly, belonged to the so-called 
"Arcto-Tertiary" or rather "Tertiary-mesophytic" flora that was — and still 
in part is — common to eastern North America and eastern Asia (cf. Li, 1952). 
That Denmark Strait— between Iceland and Greenland — apparently has 
also acted as a strong barrier when this flora was replaced by a more boreal 
one, seems to be indicated by the fact that the Icelandic flora at present is as 
typically European as the Greenland flora is typically American, a puzzle 
not explained at this Symposium. 

Although it is possible to draw conclusions hke these based on our present 
knowledge of the history of the North Atlantic and its biota, one ought not to 
forget that this knowledge still is so hmited as to make many of these con- 
clusions very preliminary, and thus, controversial. 

Even if we feel that the general picture of the history of the North Atlantic 
and its biota is emerging into greater clarity, this symposium also has shown 
that we are in the happy situation of still being confronted by unsolved 
problems of considerable importance for the problem of the geological and 
biological history of this region. The answers to even some of the apparently 
minor questions might aff'ect the major picture considerably. These may be 
geological as well as biological, and it seems futile to try to mention more than 
a few problems in need of consideration. Important geological problems 
are connected with the identification of refugia and unglaciated forelands 
after the ice has left; these questions need to be investigated in the North 
Atlantic region where glaciers are still active, although research in the Antarctic 
may give us some answers. Another geological problem concerns the bottom 
of the North Atlantic and, then, especially the transatlantic submarine ridges; 
bottom cores from carefully selected localities might tell us if thftse ridges ever 
have been raised above the sea, and if so, when and for how long a time. 
Such cores taken close to the existing basaltous islands may perhaps reveal 
in what way the hypothetical land connection has been formed in the Cretace- 
ous. According to a common hypothesis, it should have been formed by 
eustatic changes in the crust or by some kind of continental drift, whereas 
another hypothesis assumes that it has been formed by volcanic eruptions 
only, without drastic changes in sea level (cf. Tr. Einarsson, 1961). If the 
latter hypothesis is correct, the more than 5000 m thick basalt plateau ought 
to stand firmly on palagonite formed during suboceanic volcanic eruptions, 
whereas the lack of such a palagonite formation would at least indicate 
that the other explanations were more likely. Cores from the deeper parts of 
the ocean might solve the mystery of the "Scandic", or perhaps of the 


continental drift, or of the possibility of an expanding earth crust as mentioned 
by Heezen and Tharp (this symposium; cf. also Dicke, 1962). 

A geological question of utmost importance for the ultimate solutions of 
many problems here discussed concerns detailed examination of the Hgnites 
of the North Atlantic basalt formation, which reaches from Scotland and 
Northern Ireland to the Faeroes, Iceland, and Greenland (cf. Tr. Einarsson, 
Rasmussen, this symposium). Only such a study can clarify the particulars of 
the geological and chmatological history of hfe in this part of the world 
during the Tertiary. Still, these hgnites are very sporadically known. As a direct 
continuation of such studies of the lignites ought to follow detailed investiga- 
tions of the Icelandic Pleistocene deposits and of Late Glacial and Post- 
Glacial palynological phenomena in all the countries concerned. 

Turning to the biological field, palynological identifications of all the species 
of the North Atlantic flora is both desirable and essential for the interpreta- 
tion of past dispersal and distribution areas. Such a background may help in 
understanding the formation of the hmits between the Arctic and Subarctic 
discussed by Bocher (this symposium). It could probably also find explanation 
of the pecuhar disappearance and reappearance of the vegetation zonation 
discussed by Sjors (this symposium), and support or contradict the hypothesis 
of dispersal of plant communities rather than of individual diaspores as 
suggested by Hadac (this symposium). Palynology can also be expected to in- 
crease our understanding of the importance of the conservative statocrats and 
the aggressive apocrats (cf. Erdtman, this symposium) in the history of the 
floras concerned. A combined palynological and cytotaxonomical study based 
on this concept is also likely to provide new ideas for an explanation of the fact 
that the frequency of polyploids increases with an increased latitude and is 
highest in the areas where we expect the highest frequency of glacial survivors 
(Love, 1953, 1959; Love and Love, 1949, 1957; Reese, 1958, 1961a, b; 
Favarger, 1961). 

Though the general knowledge of the botanical and zoological conditions 
in the North Atlantic countries is better than almost anywhere else, studies on 
distribution and sociological behavior still remain insufficient in parts of the 
area, and the taxonomy of many species is often incomplete and inexact. 
Detailed studies on minor races need to be intensified, as demonstrated by the 
excellent results achieved by Nordhagen (1931, 1935) and Nannfeldt (1935, 
1940). In addition, the problem of endemism needs special attention by 
cytogeneticists (cf. Love and Love, 1956, 1961). The universally accepted 
idea that the frequency of endemics should reflect the age of a flora or fauna 
is still not completely confirmed, at least not in northern lands where popula- 
tions are medium-sized and competition is limited. Some of the more widely 
distributed North Atlantic endemics, e.g. Armeria maritima ssp. planifolia, 
Geum rivale ssp. islandicum, and Papaver Nordhagenianum ssp. faeroeense. 


may be old and indicate some late land-connection between Shetland, the 
Faeroes, and Iceland, but several other local endemics, such as Sesleria 
albicans ssp. islandica, G/yceria fluitans var. islandica, Roegneria borealis 
ssp. islandica, Roegneria Doniana ssp. Stefanssonii, Dactylorchis maculata 
ssp. islandica, some species of Euphrasia, and the varieties of Papaver Nord- 
hagenianum and P. radicatum (cf. Love, 1962a, b) are possibly neoendemics 
formed in Post-glacial times by strong natural selection in small populations, 
or simply by genetic drift. 

Venturing to summarize the total outcome of this symposium, it seems safe 
to conclude that although our views on the history of the North Atlantic and 
its biota have become greatly elucidated by the papers presented, we have 
also become well aware of the many geological and biological problems which 
remain unsettled in this part of the world. Many of these problems require 
thoroughly organized investigations and combined geological and biological 
efforts on basis of refined chemical, physical, geological, oceanographical, 
cytogenetical, taxonomical, palynological, biogeographical, and other methods 
of approach. Most desirable of all, however, are greatly increased facilities 
for such cooperative studies which could lead to a solution of the complicated 
but important problems of the history of biota in the North Atlantic area. 
It seems to us that Iceland, both because of its position and its unique geology, 
holds the key to many of these problems. 


Barth, T. F. W. (1941). IsIamL Oslo. 

Blytt, a. (1876). Essay on the imnugration of the Norwegian flora during alternating dry 

and rainy periods. Christiania. 
Blytt, A. (1882). Die Theorie der wechselnden kontinentalen und insularen Klimate. 

Englers Bot. Jahrb. 2, 1-50. 
Dahl, E. (1961). Refugieproblemet og de kvartcergeologiske metodene. Svensk Natin-vet- 

ens/iap 14, 81-96. 
Deevey, E. S. (1949). Biogeography of the Pleistocene. Part I. Europe and North America. 

Bull. Geol. Soc. Anier. 60, 1315-1416. 
De Geer, G. (1912). Kontinentale Niveauveranderungen im Norden Europas. Compt. 

Rend. Xlme Congr. Geol. Intern. Stocklwlm (1910), 848-860. 
DiCKE, R. H. (1962). The earth and cosmology. Science 138, 653-664. 
EiNARSSON, Tr. (1961). Upphaf Islands og blagrytismyndunin. Nditura Islands, 11-29. 
EwiNG, M. and Donn, W. L. (1956). A theory of Ice Ages. Science 123, 1061-1066. 
Favarger, C. (1961). Sur Temploi des nombres de chromosomes en geographie botanique 

historique. Ber. Geobot. Inst. Riibel 32, 119-146. 
Fries, T. C. E. (1913). Botanische Untersuchungen im nordlichsten Schweden. Uppsala. 
GJ/EREVoll, O. (1959). Overvintringsteoriens stilling i dag. Kgl. Norske Vidensk. Selsk. 

Forh. 32, 1-36. 
Godwin, H. (1956). The History of the British flora. Cambridge. 
HoLTEDAHL, O. and RosENQUiST, I. T. (1958). "Refugie-problemet" pa den skandinaviske 

halvoy fra geologisk synspunkt. Svensk Naturvetenskap 11, 108-1 18. 
Hooker, J. D. (186'?^ Outline of the distribution of Arctic plants. Trans. Linn. Soc. London 

23, 251-348. 


HOPPE, G. (1959). Nagra kritiska kommentarer till diskussionen om isfria refugier. Svensk 

Natiirvetenskap 12, 123-134. 
Humboldt, F. A. von (1817). De distribi/tione geographica plaiitantin seciindiim coeli 

tempeiiem et altitudinem moiitiitin, prolegomena. Lutetia Parisiorum. 
IvERSEN, J. (1958). The bearing of glacial and interglacial epochs on the formation and 

extinction of plant taxa. Uppsala Univ. Arsskr. 1958, (6), 210-215. 
Li. H. L. (1952). Floristic relationships between eastern Asia and eastern North America. 

Trans. Amer. Philos. Soc. N.S. 42. 371-429. 
LiNDROTH. C. H. (1958). Istidsovervintrare bland djuren. Svensk Naturvetenskap 11, 

LiNDROTH, C. H. (1960). Is Davis Strait — between Greenland and Baffin Island — a floristic 

barrier? Bot. Notiser 113, 130-140. 
Love, A. (1953). Subarctic polyploidy. Hereditas. 39, 113-124. 
Love, A. (1959). Origin of the Arctic flora. Piibl. McGUl Univ. Museum 1, 92-95. 
Love, A. (1962a). Typification of Papaver radicatum — a nomenclatural detective story. 

Bot. Notiser 115, 61-84. 
Love, A. (1962b). Nomenclature of North kiXanUc Papaver . Ta.xon 11, 132-138. 
Love. A. and Love, D. (1949). The geobotanical significance of polyploidy. I. Polyploidy 

and latitude. Portug. Acta Biol. (A) R. B. Goldschmidt Vol.. 273-352. 
Love, A. and Love, D. (1953). Studies on Bryoxiphium. Bryologist 56, 73-94, 183-203. 
Love, A. and Love, D. (1956). Cytotaxonomical conspectus of the Icelandic flora. Acta 

Horti Gotob. 20, 1-291. 
Love, A. and Love, D. (1957). Arctic polyploidy. Proc. Genet. Soc. Canada 2, 23-27. 
Love, A. and Love, D. (1961). Chromosome numbers of Central and Northwest European 

plant species. Opera Botanica 5, I-VIII, 1-581. 
Nannfeldt, J. A. (1935). Taxonomical and plant-geographical studies in the Poa laxa 

group. A contribution to the history of the North European mountain floras. Symb. 

Bot. Upsal. I (5), 1-113. 
Nannfeldt, J. A. (1940). On the polymorphy of Poa arctica R. Br. with special reference to 

its Scandinavian forms. Symb. Bot. Upsal. IV (4), 1-86. 
Nannfeldt, J. A. (1958). Denskandinaviskafjallfloranochnedisningarna. Svensk Naturvet- 
enskap 11, 119-133. 
Nordhagen, R. (1931). Studen iiber die skandinavischen Rassen des Papaver radicatum 

Rottb. sowie einige mit denselben verwechselte neue Arten. Bergens Mus. Arbok 1931. 

Naturv. rekke 2. 1-50. 
Nordhagen, R. (1935). Om Arenaria humijiisa Wg. og dens betydning for utforskningen 

av Skandinavias eldste floraelement. Bergens Mus. Arbok 1935. Naturv. rekke 1, 1-183 
Nordhagen, R. (1936). Skandinavias fjellflora og dens relasjoner til siste istid. Nord. (19. 

skand.) naturforskarmotet i Helsingfors 1936, 93-124. 
Reese, G. (1958). Polyploidie und Verbreitung. Zeitschr. Bot. 46, 339-354. 
Reese, G. (1961a) Geobotanische Bedeutung der Chromosomenzahl und Chromosomen- 

struktur. Naturw. Rundschau 14, 140-145. 
Reese, G. (1961b). Karyotype and plant geography. Recent Advances in Botany, 895-900. 
Sorensen, T. (1953). A revision of the Greenland species of Puccinellia Pari. Medd. om 

Grdnl. 136(3), 1-179. 
Steere, W. C. (1937). Bryoxiphium norvegicum, the sword moss, as a preglacial and inter- 
glacial relic. Ecology 18, 346-359. 
Terasmae, J. (1961). Notes on Late-Quaternary climatic changes in Canada. Ann. N.Y. 

Acad. Sci. 95, 658-675. 


List of those participating or attending the Symposium. 

Hugo Andersson, Lund, Sweden 
Pall Bergthorsson, Reykjavik, Iceland 
Tyge W. Bocher, Copenhagen, Denmark 
HoGNi BoDVARSSON, Luud, Sweden 
Eggert V. Briem, Pennsylvania, U.S.A. 
SiGURDUR J. Briem, Reykjavik, Iceland 
Max E. Brixton, Washington, D.C., U.S.A. 
Zd. Cernohorsky, Praha, Czechoslovakia 
R. Charpentier, Lund, Sweden 
EiLiF Dahl, Vollebekk, Norway 
Jean Dahl, Vollebekk, Norway 
Truls Dahl, Vollebekk, Norway 
Ingolfur Davidsson, Reykjavik, Iceland 
Margaret Davis, Ann Arbor, Mich., U.S.A. 
Per Douwes, Lund, Sweden 
Heather Drummond, Sydney, Australia 
Kristbjorg Duadottir, Akureyri, Iceland 
Eythor Einarsson, Reykjavik, Iceland 
Thorleifur Einarsson, Reykjavik, Iceland 
Trausti Einarsson, Reykjavik, Iceland 
Gunnar Erdtman, Stockholm, Sweden 
GuNNi Erdtman, Stockholm, Sweden 
Jon Eythorsson, Reykjavik, Iceland 
Knut FiCGRi, Bergen, Norway 
Sturla Fridriksson, Reykjavik, Iceland 
Walter Friedrich, Cologne, Germany 
Arnthor Gardarsson, Reykjavik, Iceland 
Camille Gervais, Montreal, Canada 
Geir Gigja, Reykjavik, Iceland 
Olav Gji^REVOLL, Trondheim, Norway 
Finnur Gudmundsson, Reykjavik, Iceland 
Teresia Gudmundsson, Reykjavik, Iceland 
Petur Gunnarsson, Reykjavik, Iceland 
Emil Hadac, Plzen, Czechoslovakia 
Helgi Hallgrimsson, Reykjavik, Iceland 
Haye Walter Hansen, Hamburg, Germany 



A. M. Harvill, Jr., Murray, Kentucky, U.S.A. 

Bruce C. Heezen, Palisades, N.Y., U.S.A. 

H. J. Helms, NATO, Paris, France 

Steingrimur Hermannsson, Reykjavik, Iceland 

GuNNAR HoppE, Stockholm, Sweden 

Eric Hulten, Stockholm, Sweden 

J. D. Ives, Ottawa, Canada 

Sveinn Jakobsson, Reykjavik, Iceland 

Bergthor Johannsson, Reykjavik, Iceland 

Baldur Johnsen, Reykjavik, Iceland 

Bengt Jonsell, Uppsala, Sweden 

Margareta Jonsell, Uppsala, Sweden 

Jon Jonsson, Hafnarfjordur, Iceland 

GuDMUNDUR Kjartansson, Hafnarfjordur, Iceland 

Britt Kjellqvist, Lund, Sweden 

Ebbe Kjellqvist, Lund, Sweden 

Carl H. Lindroth, Lund, Sweden 

Gun Lindroth, Lund, Sweden 

AsKELL Love, Montreal, Canada 

Doris Love, Montreal, Canada 

GuNNLAUG Love, Montreal, Canada 

LoA Love, Montreal, Canada 

Thrainn Love, Reykjavik, Iceland 

J. R. Mackay, Vancouver, B.C., Canada 

Pierre Morisset, Montreal, Canada 

J. A. Nannfeldt, Uppsala, Sweden 

Rolf Nordhagen, Oslo, Norway * 

Knut Norstog, Springfield, Ohio, U.S.A. 

Steinunn Olafsdottir, Reykjavik, Iceland 

Svandis Olafsdottir, Reykjavik, Iceland 

P. Omodeo, Sienna, Italy, 

U. Omodeo, Siena, Italy 

Ake Persson, Lund, Sweden 

SiGURDUR Petursson, Reykjavik, Iceland 

Frank A. Pitelka, Berkeley, California, U.S.A. 

Joannes Rasmussen, Torshavn, The Faeroes 

Olof I. Ronning, Trondheim, Norway 

Martin Schwarzbach, Cologne, Germany 

Bjorn Sigurbjornsson, Reykjavik, Iceland 

Flosi Hrafn Sigurdsson, Reykjavik, Iceland 

Sven Th. Sigurdsson, Reykjavik, Iceland 

* Did not attend in person, but gave a lecture from a tape. 


Thorbjorn Sigurgeirsson, Reykjavik, Iceland 
GuDMUNDUR E. SiGVALDASON, Reykjavik, Iceland 
Hugo Sjors, Uppsala, Sweden 
Armann Sn.€varr, Reykjavik, Iceland 
Steindor Steindorsson, Akureyri, Iceland 
Brita Stenar-Nilsson, Uppsala, Sweden 
Marie Tharp, Palisades, N.Y., U.S.A. 
Sigurdur Thorarinsson, Reykjavik, Iceland 
GiJNTER TiMMERMANN, Hamburg, Germany 
Haukur Tomasson, Reykjavik, Iceland 
Eysteinn Tryggvason, Reykjavik, Iceland 
ToMAS Tryggvason, Reykjavik, Iceland 
Kari Valsson, Strond, Iceland 
Henrik W. Walden, Gothenburg, Sweden 
Virginia Weadock, New York, N.Y., U.S.A. 


Agnew, a. D. Q. 212 

Ahmadjian, v. 236, 239 

Ahorner, L. 14 

Ahti, T. 110, 118, 120, 123 

Albertson, N. 228, 231 

Aleksandrova, V. D. 200, 202, 203 

Allan, T. 29, 30, 43 

Andersen, B. G. 330, 333 

Andersen, S. T. 377, 384 

Andersson, G. 367, 364 

Andrews, J. T. 346, 347, 348, 354 

Andreyev, V. N. 119,123 

Areschoug, F. W. C. 241, 259 

Arldt, T. 73, 84 

Arndt, W. 74, 84 

Arwidsson, T. 282 

Askelsson, J. 9, 15, 18, 82, 84, 387, 388, 

AvsYUK, G. A. 323, 333 

Babington, C. C. 297, 298, 302 

Baker, F. C. 170 

Bardarsson, G. G. 16, 18, 362, 386, 389 

Barth, T. F. W. 187, 188, 393, 396 

Bauer, A. 323, 333 

Beijerink, W. 196, 199, 203 

Bentley, C. R. 322, 333 

Berg, R. 253 

Berglund, B. 372, 374 

Bergstrom, E. 330, 331, 333, 337, 354 

Berry, E. W. 15, 18 

Beschel, R. E. 252, 254 

Bjarnason, O. 362, 364 

Bjornsson, H. 377, 384 

Bjornsson, S. 377, 379, 385 

Blake, W. 326, 333 

Bliss, L. C. 193, 203 

Blytt, a. 173, 176, 188, 221, 241, 243, 

245, 252, 253, 259, 261, 269, 282, 391, 

392, 396 
BocHER, T. W. 74, 84, 173, 174. 180, 181, 

188, 195, 202, 203, 285-296, 395 
Br/€ndegaard, J. 75 
Briem, S. J. X 
Broecker, W. S. 346 
Brooks, B. W. 166, 170 
Brooks, S. T. 166, 170 
Brotzen, F. 326, 333 
Brown, R. J. E. 122, 123 
Bryan, M. S. 17, 18 
Bucher, W. H. 21, 26 

BiJDEL, J. 326, 327, 331, 333 
Bull, C. 323, 333 
Buss, I. O. 191, 203 

Cernohorsky, Z. 233, 235, 239, 393 
Cernosvitov, I. 147, 150 
Chaney, R. W. 8,9,15,18 
Clarke, A. H. 169, 170 
Coleman, A. P. 340, 343, 354 
Conolly, a. 229, 231 
Corbel, J. 331,333 
Craig, B. C. 17, 18 
Crary, a. p. 26 

Dahl, E. 73, 84, 183, 186, 188, 194, 200, 
204, 207, 211, 219, 233, 235-239, 279, 
280, 282, 328, 329, 334, 337, 338, 340, 
343, 346, 347, 351, 353, 354, 391, 392, 
393 396 

Dahlstedt, H. 248, 250, 259, 271, 282 

Daly, R. A. 340, 343, 354 

Darlington, P. J. 141, 150, 170 

Darwin, C. 173 

Davidson, C. F. 30, 44 

Davies, a. M. 170 

Deevey, E. S. 166, 170, 392, 396 

De Geer, G. 393, 396 

Degelius, G. 238, 239 

Demarest, M. 342, 354 

Dicke, R. H. 395, 396 

DiETZ, R. S. 23, 26 

DoLGUSHiN, L. D. 325, 334 

DoNN, W. L. 25, 26, 201, 204, 322, 334, 
393, 396 

Durango, S. 191, 204 

Du RiETZ, G. E. 110, 119, 120, 122, 123 

Ebeling, F. 120, 123 

Ehrmann, P. 120, 123 

EiNARSSON, E. X, 297, 393 

Einarsson, Th. 71, 81, 83, 84, 355, 362, 

364, 377, 383, 385, 392, 396 
Einarsson, Tr. 1, 9, 358, 362, 364, 

386-389, 393-395 
Ekman, E. 251, 259 
Elfstrand, M. 251, 259 
Ellis, A. E. 166, 168, 171 
Elson, J. 200 
Emiliani, C. 17, 18 




Engstrom, a. 371 

Erdtman, G. 87, 367, 368, 370-374, 395 

Ericson, D. B. 22, 26 

Erskine, J. S. 169, 170 

EwiNG, J. I. 22, 26 

EwiNG, M. 21, 22, 25, 26, 27, 201, 204, 

322, 334, 393, 396 
EwiNG, W. M. 22, 26 

F^egri, K. 221, 229, 231, 392 

Fairbridge, R. W. 25, 27 

Falconer, G. 347, 348, 349, 351, 354 

Farrand, W. R. 25, 26, 27, 322, 334 

Favarger, C. 231, 395, 396 

Fernald, M. L. 173, 174, 188, 201, 204 

Feyling-Hansen 330 

Fink, J. 326, 334 

FiRBAS, F. 367, 374 

Fischer, A. G. 163, 171 

Flint, R. F. 201, 204, 341, 342, 353, 354 

forcart, l. 166, 171 

forchhammer, j. g. 29, 30, 43 

Fredskild, B. 285, 296 

Frenzel, B. 327, 334 

Fries, M. 373 

Fries, T. C. E. 246, 248, 259, 260, 261, 

282, 391, 396 
Fries T. M. 105, 107 
Fristrup, B. 202, 204 
Fyles, J. G. 17,18 

Gates, G. E. 129, 147, 149, 150 

Geiger, R. 183, 188, 197, 200, 204 

Geikie, J. 29,30,41,42,43 

Gelting, p. 305, 320 

GiTZ 75 

Gj/erevoll, O. 243, 254, 260, 261, 391, 

Godwin, H. 168, 171, 371, 374, 392, 396 
goldthwait, r. p. 353 
Gollerbach, M. M. 325, 334 
GoRODKOv, B. N. 372, 374 
Graul, H. 326, 334 
Gregory, J. W. 42, 44 
Gribbon, p. W. F. 349, 354 
Gronlie, a. 277, 282 
Gronlie, O. T. 269, 282 
Gronlund, C. 297, 298, 302 
Grontved, J. 299, 302 
Gross, H. 327, 334 
Grossman, K. 42, 44 

Hadac, E. 102, 107, 194, 195, 200, 204, 
207,211,214,219, 393, 395 

Hafsten, U. 227, 231 

Hagen, E. 194, 204 

Halden, B. 368, 374 

Halliday, W. E. D. 114, 123 

Hamilton, E. L. 22, 27 

Hammen, T. van der, 327, 334 

Hammer, M. 75 

Hansen, A. M. 245, 260 

Hanssen, O. 194, 204 

Hare, F. K. 110, 113, 114, 115, 118, 

Hedberg, O. 369, 374 
Heer, O. 8, 9, 15, 18 
Heezen, B. C. 21, 23, 27, 393, 395 
Helland, A. 29,30,41,44 
Heltzen, a. M. 252, 260 
Henderson, J. 153, 156, 171 
Henriksen, K. L. 75 
Henson, F. H. 379, 389 
Heslop-Harrison, J. 194, 204 
Hesselman, B. 228, 231 
Hesselman, H. 368, 374 
Hibbard, C. W. 154, 171 
Hill, M. N. 21, 27 
Hjaltalin, O. J. 297, 302 
Hochbaum, H. a. 191, 193, 204 
Hojer, E. W. 120, 123 
Holm, A. 75 
HOLMBOE, J. 222, 229, 231 
HoLMEN, K. 74, 84, 174, 180, 181, 188, 

195, 203, 280, 282, 285, 286, 290, 295, 

HOLMSEN, G. 268 
HoLTEDAHL, O. 245, 251, 260, 330, 334, 

391, 396 
Hooker, J. D. 173, 391, 396 
Hope, E. R. 25, 27 
Hoppe, G. 171, 321, 327, 330, 334, 337, 

343, 353, 354, 391, 397 
Horn, G. 105, 107, 219 
Horne, J. 330, 332, 334 
Hornemann, I. W. 297, 302 
HosPERS, J. 386, 387, 389 

HUBENDICK, B. 154, 171 

HULTEN, E. 45, 73, 74, 84, 89, 91, 97, 173, 
183, 188, 195, 201, 204, 243, 250, 254, 
257, 260, 263, 264, 265, 266, 269, 282, 
286, 296, 301, 302, 392, 393 
Humboldt, F. A. von xi, 391, 397 
HusTiCH, I. 114, 118, 119, 121, 124 

IVERSEN, J. 87, 174, 188, 223, 231, 285, 
291, 296, 327, 334, 363, 364, 367, 369, 
375, 392, 397 346, 348, 

Ives, J. D. 190, 337, 340, 342' 
352, 354, 393 



Jakobsen, K. 74, 84, 174, 180, 181, U 

195, 203, 285, 286, 290, 295 
Jalas, J. 113, 124 
Jeffre, E. p. 202, 204 
Jessen, K. 43, 44 
johansen, e. 229, 231 
Johnson, J. P. 341 
JONSSON, H. 298, 302 
JONSSON, J. 16, 18, 81, 84, 362, 364 
JOLEAUD, L. 132, 150 
JORGENSEN, C. A. 195. 204 
JOURDAIN, F. C. R. 205 
Jux, U. 8, 9, 15, 19 

Kalela, a. 110, 119, 120, 124 

Kennard, a. J. 168, 171 

Kew, H. W. 151 

Kjartansson, G. 207, 219, 357, 360, 

Klein, A. 327, 334 
Klimes, L. 219 
Klute, F. 327, 334 
Knaben, G. 91, 92, 93, 94, 97, 184, 185, 

188, 219, 280, 281, 282 
Koch, L. 280, 282 
KoLBE, R. W. 23, 27 
Konig, J. G. 297 
KosACK, H. P. 323, 334 
Krasnov, 331 
KujALA, V. 110,124 
Kupriyanova, L. a. 215, 219 

L/EGAARD, S. 285, 288, 295 

Lamb, I. M. 236, 239 

Lange, J. 297 

Larson, L. M. ix 

Larsson, S. G. 81, 84 

Lavrenko, E. M. 113, 118. 119, 121, 122 

Li, H. L. 394, 397 

Licharev, L M. 154, 169, 171 

Lid, J. 194,204,222,228,231,259 

Likharev 331 

Liuequist, G. H. 327, 328, 334 

LiNDAL, J. H. 387, 388, 389 

Lindquist, B. 207,219,229,231 

LiNDROTH, C. H. 73-79, 82, 83. 85, 147, 
150, 154, 169, 171, 173, 174, 176, 179, 
181, 188, 229, 231, 290, 296, 305, 320, 
391, 393, 394, 397 

Llano, G. A. 238, 239 

lohmander, h. 166, 168, 171 

Loken, O. 340, 342, 343, 346, 354 

LoKKEN, S. 251, 256, 260 

LoMAS, J. 42, 44 

LoTZE, F. 13, 19 

Love, A. x, xi, 25, 27, 74, 81, 85, 91-94, 
97, 173, 174, 179. 188, 194, 195, 200. 
203, 204, 217, 219, 299, 300, 302, 303, 
320, 353, 360, 365, 387, 391, 393. 395, 
396, 397 

Love, D. x, 25, 27. 74, 81, 85, 97, 173, 174, 
179, 188, 189, 194, 195, 200, 203, 204, 
214, 217, 219, 300, 302, 320. 353, 360, 
363, 387, 389, 392, 393, 395, 397 

Low, A. P. 347, 354 

LowENSKjoLD, H. L. 191, 204 

LuNDQViST, G. 329, 334 

Lynge, B. 105, 107, 173, 233, 234, 236, 
237, 239, 240 

Maarleveld. G. C. 327, 334 
Mackay, J. R. 354 
Macpherson, J. 252 
McKelvey, B. C. 323. 333 
Mackenzie, G. 29. 30. 42. 44 
McVean, D. N. 121. 124 
Mandahl-Barth, G. 166.171 
Manley, G. 327, 334 
Manum, S. 15, 19 
Marie-Victorin, Fr. 201,204 
Markov, K. K. 219, 323, 333 
Marthinussen, M . 246, 252, 260, 330, 334 
Matthew, E. M. 346, 347, 354 
Matthew, W. D. 140, 141, 142, 150 
Meriel, Y. 226, 232 
Meyer, B. L. 8, 9, 15, 19 
Michaelsen, W. 132, 144, 147, 151 
Miller, D. J. 16, 19 
Molholm-Hansen, H. 298, 299, 302 
MoNTEiTH, J. L. 202, 204 
Moreau, R. E. 161,171 
Mortensen, H. 326, 327, 334 
MuLLER, O. F. 297, 302 

Nannfeldt, J. A. 87, 89, 90, 94, 95, 97, 
173, 263, 265, 275, 276, 280, 282, 373, 
391, 393, 395, 397 

Nansen, F. 105, 107 

Nathorst, a. G. 13, 19, 87, 107, 173 

Nielsen, N. 381, 389 

Noe-Nygaard, a. 30. 44, 381, 389 

NoRDBORG, G. 371, 374 

Nordhagen, R. 91. 97. 105, 107, 122, 
124, 173, 229, 232, 241, 243, 247, 248, 
251, 252, 258, 260, 263, 265, 266, 267, 
268, 276, 280, 282, 391, 393, 395, 397 

NoviKOV, V. 26, 27 

Nydal, R. 228, 232 

Nye, J. F. 322, 334 

Nygren, a. 89, 94, 96, 97 



Odell, N. 340, 354 

Oekland, F. 166, 171 

Okko, V. 329, 355, 365 

Olafsson, E. 14, 297, 302 

Omodeo, p. 127, 140, 149, 150, 151, 393 

Orvig, S. 200 

Orvin, a. K. 105, 107, 219 

Osborne, S. 195, 204 

oskarsson, i. 300, 302 

OsTENFELD, C. H. 173, 174, 188, 299, 302 

OSTENSO, N. A. 322, 323, 333, 335 

OuREN, T. 221, 232 

Palsson, B. 297, 302 

Peach, B. N. 330, 332, 334 

Peacock, M. A. 30, 42, 44 

Penck, a. 327, 334 

Pettersson, B. 96, 97 

Pflug, H. D. 5, 7, 8, 9, 15, 16, 19, 79, 80, 

81, 365 
Pickford, G. E. 132,151 
PiLSBRY, H. A. 153, 154, 156, 164, 166, 

169, 171 
PiRRiT, J. 8, 9, 15, 19 
Pjeturss, H. 386, 389 
PoLUNiN, N. 121,124,267,286 
Porsild, a. E. 118, 124, 196, 204, 276, 

286, 296 
Porsild, M. P. 78, 85, 174, 188, 199 
Poser, H. 327, 334 
Post, L. von 367, 368, 375 

Quick, H.E. 154,171 

Raasch, G. O. 19 

Rammelmeyer, E. S. 154, 169, 171 

Rapp, a. 329, 334 

Rasmussen, J. 29, 44, 395 

Rasmussen, R. 43, 44, 194, 195, 204 

Ratcliffe, D. a. 121, 124 

Raup, H. M. 286, 296 

Rebristaya, O. 232 

Reese, G. 395, 397 

Regel, C. 110,124 

Reid, E. M. 171 

Rensch, B. 145, 151 

Reusch, H. 280, 283 

Ridley, H. N. 191, 193, 195, 197, 200, 205 

RiEDLOVA, J. 219 

Ritchie, J. C. 114,124 

Robin, G. de Q. 323, 324, 335 

RoNNiNG, O. I. 99, 103, 107, 194, 195, 205, 

RosENQUiST, I. T. 391, 396 

RosHEViZH, R. Y. 215, 219 
RosHOLT, J. H. 326, 335 
Rouleau, E. 180, 181, 188 
Rousseau. J. 110, 114, 118, 124 
RowE, J. S. 112,113,114,118,124 
RozYCKi, S. Z. 323, 335 
Rudberg, S. 329, 334, 343, 353 
Runcorn, S. K. 22, 27 
Runemark, H. 238, 239, 240 

Salisbury, E. J. 194, 197, 200, 205 

Salomonsen, F. 75 

Samuelsson, G. 191, 193, 194, 205, 329, 

Samuelsson, T. 377 
Savile, D. B. O. 201, 202, 205, 352, 354 
Schaaning, H. L. T. 191, 205 
Scharff, R. F. 73, 79, 85 
Schloemer- Jaeger, A. 15, 19 
Schmidt, W. 183 
Schmitz 369 

Scholander, p. F. 194, 205, 233, 239 
Schonfeld, E. 15, 19 
Schwarzbach, M. 7, 9, 11, 15, 19, 79, 80, 

81, 82, 85, 389, 393 
Schwarzenbach, F. H. 285, 291, 296 
ScoGGAN, H. J. 180, 181, 188 
Selander, S. 267, 283 
Serebryanny, L. 333 
Sernander, R. 224, 232, 245, 260 
Shamurin, V. F. 202, 205 
Shantz, H. L. 112,124 
Shtepa, V. S. 202, 205 
Shumskiy, p. a. 322, 323, 333, 335 
Sigurgeirsson, T. 8, 9, 386, 389 
Sigurjonsson, S. 385 
Simmons, H. G. 82, 85 
Simpson, G. G. 73, 85, 140, 151, 164, 171 
Simpson, H. 372, 375 
Sjors, H. 109, 110, 113, 118, 119, 120, 

124, 395 
Sladen, W. J. 191,205 
Sli'pka, J. 212 
Sn.-evarr, a. X 
Sochava, V. B. (Sotchava) 113, 118, 119, 

121, 122, 124 
Soergel, W. 326, 335 
Sorensen, N. a. 277 
SoRENSEN, T. 103, 107, 195, 202, 204, 205, 

216, 217, 219, 296, 393, 397 
Sparks, B. W. 159, 171 
Spence, D. H. N. 121, 125 
Srodon, a. 214, 216, 219 
Stebbins, G. L. 375, 371 
Steere, W. C. 393, 397 
Stefansson, S. 298, 299, 302, 315, 320 



Steindorsson, S. 83, 85, 299, 302, 320, 

360, 365, 392 
Sterner, R. 228, 232 
Stille, H. 12, 19 
Strachov, N. M. 207, 208, 219 
Straka, H. 355, 365 
Stromfelt, H. F. G. 298, 302 
SvERDRUP, H. 200, 201, 205 
Swan, L. W. 202, 205 
Syrojeckowsku, E. E. 325, 334 
SZAFER, W. 160, 171 

Tateoka, M. 374 

Taylor, D. W. 154, 171 

Terasmae, J. 17,19,393,397 

Tharp, M. 21, 27, 393, 397 

Thiel, E. 323, 335 

Thorarinsson, S. X, 17, 18, 19, 82, 85, 

305, 306, 320, 355, 357, 358, 362, 365, 

371, 375, 377, 389, 392 
Thorgilsson, a. 362 
Thorkelsson, T. 82, 85, 387 
Thoroddsen, T. 365 
TiCEHURST, N. F. 205 
Ti^HOMiROV, B. A. 118. 119, 122, 125, 

200, 202, 205, 215, 219 
ToLMACHEV, A. I. 219, 232, 286, 296 
TOMLINSON, R. F. 340, 341, 344, 346, 354 
Trevelyan, W. C. 29, 44 
Troll, C. 327, 334 
Tryggvason, E. 21, 27 
Tryggvason, T. 379, 389 
Tucker, B. W. 205 

Undas, I. 245, 251, 252, 260, 277 

Vassil'ev, V. N. 214,216,219 
Vogt, T. 246, 260, 330 

Walden, H. W. 153, 154, 171, 393 
Walker, F. 30, 44 
Walker, G. P. L. 7, 8, 9 
Wallace, A. R. 127 
Warming, E. 173,298,302 
Washburn, A. L. 342, 354 
Weadock, V. X 
Webb, P. N. 323, 333 
Wegener, A. 142, 144, 186 
Weidenbach, F. 326, 335 
Wendelbo, p. 225, 232 
Wenz, W. 153, 154, 171 
Wesenberg-Lund, C. 75, 85 
Westergaard, M. 195, 204 
Wexler, H. 322, 335 
Wheeler, E. P. 340,341,354 
WiLCKE, D. E. 145, 146, 151 
Wille, N. 245, 246, 260 
Willis, E. H. 168, 171 
Wilson, C. 109, 125 
Wilson, J. T. 23, 27 

WiSCHMANN, F. 258 

Witherby, H. F. 191, 205 
WoLDSTEDT, P. 326, 327, 335 
Wollin, G. 26 
Woodward, B. B. 168, 171 
Wright, H. E. 326, 335 

Yakovlev 331 
YocoM, C. F. 191,205 

Valentin, H. 330, 331, 335 

Zilch, A. 153, 154, 171 
ZoN, R. 112, 124 



Abies 15, 113 

halsamea 112, 115, 118 
lasiocarpa 1 1 8 
pectinata 1 97 
Abisko 329 

Acadian forest region 1 12 
Acanlhiniila 1 67 
Acanthodrilidae 127, 129, 131, 132, 138, 

139, 142 
Acanthodrilinae 127, 133, 134. 135, 138, 

141, 143 
Acanthodiiliis 133 
Acer 15, 111 

Negimdo 113, 115, 117 
plataiwides 112, 119 
pseudoplatanus 112 
sacchaium 112, 115 
spicafiiDi 1 1 5 
Aciciila 167 

adaptation 141, 142, 178, 202 
Aegopinella 1 67 
affinities of Icelandic flora 297flF 
Afghanistan, earthworms 149 

diatoms carried by wind from 23 
earthworms 127, 129, 138 
gastropods 161 
relationships of fauna 138 
submerged shoreline 25 
agamospecies 88 
age of arctic plants 207 
cauiiuim 3 1 4 
trachvcaiiliim 314, 316 
Agyioplioia n'gidci 235, 238 
air plankton 200 
air turbulence and dispersal 200 
Akureyri 355 
Alaska 216, 268, 371 
plants 89 

unglaciated areas 21 1 
Alberta 115 
Alchemilla 300, 301 
alpimi 287, 293 

faeroeensis 300, 306, 309, 310, 319 
filicaiilis 287, 293 
A lector id nitidiila 235 
Alesund 252 

Alftanes end-moraines 359 
algae in Antarctica 325 
Algonquin Park 1 1 1 
Alleghany earthworms 1 27 

Allerod layers 327 

caliginosa 146, 149 
chlorotica 1 46 
liiuicola 146, 147 

longa 146, 147 

rosea 1 46 
alluvial deposits in the Faeroes 43 
Alma 129, 145 
Ainus 7, 1 5, 82, 1 1 3, 226, 357, 363, 387, 388 

glitlinosa 226, 372 

incana 1 14, 372 

riigosa 1 1 4 

Sieholdiaua ill 

viridis 384-388 
Alopeciiriis alpiniis 88, 99, 215 
Alpine-Atlantic plants 216 
alpine barrens 122 
alpine fir 118 

alpine flora 104, 209, 261, 262, 313 
alpine forest-tundra 1 18 
alpine zones 121, 122 
Alps 1 22, 1 76, 209, 2 1 6, 234, 244, 259, 290 

endemic insects 81 

Poa sect. Oreinos 89 

subalpine forests 113 
Altai 215,217,243 

subalpine forests 1 13 
Alvar steppe 370 

apocrats 368 
Amazon Basin earthworms 129 
Amazon River 131 
amber 81 

icefloe transport of 229 
Ambrosia 368 
America 45 

earthworms 138 

plants 54, 71 
American element 

in British flora 66, 194 

in European Interglaciai flora 237 

gradual transition eastward 173 

in Iceland 74 

in Scandinavian mountain flora 241 
American-Arctic 216 

element in European flora 292 
American-Eastern Asiatic plants 54 
American-Greenland element in Scandina- 
vian flora 245 
American gastropods 154, 156 
American plants 

in the British Isles 66, 194 




American plants, in Greenland 54, 66 

in Iceland 66, 298, 300 

reaching Atlantic islands 54 
amphi-atlantic biota 74, 185 

of North America 393 

Pleistocene survival 188 
amphi-atlantic earthworms 140 
amphi-atlantic element 173 

its anthropochorous nature 169 

long-distance dispersal 178, 181, 182, 

eastern, in North America 174, 177, 188 

western, in Europe 174, 175, 181, 301 
amphi-atlantic plants 46, 54, 73, 174, 178 

arctic 216 

arctic-subarctic 216 

in the birch-belt 186 

and land-connection 72, 176 

and long-distance dispersal 176, 177, 

migration across Bering Sea 183 

vicariant taxa 183, 184 
amphi-atlantic zonation 109ff 
amphi-pacific gastropods 164 
amphibians 129, 140 
amphibolite 329, 340 
amplitude, ecological 235 
Amur 1 1 1 
Anademiliis 157 
anamesite 30 

acuta 1 9 1 

platyrhynclui 1 9 1 
Andes, earthworms 129 
anemochorous plants 181, 182 

long-distance dispersal 177, 183 

increase towards south 181 
Anemone Richardsonii 292 
Angelica 293 

Archangelica 174,291 

silvestris 363 
Angola 138 
Angmagsalik 289, 291 
Anguispiia 167 

animals as dispersal agents 190 
anorthosite 341 

albifrons 192 

anser 1 92 

fabalis 192 

hyperboreus 1 92 
Antarctica 26 

ice-free areas 323 

nunataks 323 

oases 323 

plants 325 

refugia 322, 325 

Antarctica, subglacial relief 323 

thickness of ice 322 

affinis 288 

alpina 3 1 3 

carpatica 265 

glabrata 288 

intermedia 288 

apophytes 221 

Cepaea 1 69 

dispersal 78, 168, 222 

plants in Iceland 313 

species 174 
anthropogenous origin 

of Digitalis purpurea 224 

of Plantago lanceolata 224 
anthropophilous plants 363 
Antillean earthworms 129, 132 

of earthworms 144 

of Placopsis gelida 236 
Antofagasta 131 
Apion cruentatum 83 
Apium nodiflorum 192 
apocracy 373, 374 
apocrats 367-371,373,395 
apomicts 385 
apomixis 300 
apophytes 221 
aposporous embryosacs 94 
apothesia 237 

Appalachian earthworms 154 
Arabis alpina 214, 216 
Aranea, in Greenland 75 
archaean rocks 338, 344 
A rchaeo.xesta 1 5 8 
Archangelica officinalis 225, 363 
archeophytes 224 
Arctic 113 

age of flora 207,241 

climate 208 

definition 292 

Ocean 66, 99, 208 

plants 46, 54, 89, 100, 105, 173, 180, 
182, 207, 211, 214, 215, 216, 228, 229, 
235, 238, 241. 269. 287. 301, 395 

survival of flora 210,261 

tree-line 327 

zonation 109. 121 
Arctophila fulva 2 1 7 

alpina 88 
Uvo-ursi 289 
Arcto-Tertiary flora 163, 394 

humifusa 243, 244, 245. 267, 268, 279 



Arenaria, norvegica 257, 306, 309, 311 

pseudofrigida 104, 218 
Arfersiorfikfjord 288 
Arianta arbusforum 167, 168 
Arion 167 

arer 1 68 

intermedins 1 68 
Aimeiia moritiitia 395 

alpina 2 1 5 

angiistifolia 2 1 5 
Arsuk 292 
Artemisia 363, 368, 371 

orctica 265 

norvegica 96, 229, 248, 264, 265, 269, 
270, 276, 279 

rupestris 370 
arthropods, distribution 144 
Artocarpus, in Greenland Cretaceous 13 

septentrionale 3 1 5 

Trichomanes 3 1 9 

viride 3 1 5 
Astragalus kurdiciis 212 
Athyrium al pest re 292 

area 72 

biota 208 

connections 23 

earthworms 127 

effect on zonation 122 

floor 21 ff 

plants 46, 54, 66, 69, 189, 301, 357 
Atlantic Islands 

plants 66, 72 

relationships of floras 45 
Atlantic Ocean 109, 184 

Early Tertiary 22, 23 

Globigerina-ooze 326 

Mesozoic 22, 23 
Atlantic Seaboard I76ff 
Auckland earthworms 131 
augite 383 
Aurland Mt. 245 

Austfirdir 306, 307, 3 1 0, 3 1 1 , 3 1 7, 3 1 9 

crayfishes 1 32 

earthworms 132 

earthworms 129, 131 

gastropods 154 
Austur-Skaftafellssysla 319 
autochtonous gastropods 154 
autogamy 300 
Auvergne 252 
A vena sativa 363 
Aves, in Greenland 75 

Azeka 167 

plants 66 


Baffin Island 181, 190, 294, 338, 339, 350 

distance from Greenland 76 

fauna 76, 181 

glaciation 347 

glacierization 347 

immigration 77, 78 

Tertiary lavas 340 
Bahia Blanca 129 
Baikal 253 
Balea 167 
Balkans 174 

ballast, and plant introductions 174 
Baltic amber 81,229 
Baltic end-moraine 369 
Baltic zonation 109, 122 
Balvatnet 267 
"Banki" 33 
Banks Island 184 
Barbados earthworms 132 
Bardardalur 313 
Barents Sea 105, 106, 331 
basalt 11,33,42,80 
basalt glass 384 
basalt-globe breccia 381 
basalt plateau 187,239 
bathymetric chart of North Atlantic 24 
Batrachium confervoides 192 
beach plant dispersal 304 
Bear Island 100, 103. 190, 195, 236, 253 

coals 13 

endemic plants 102 

flora 45 

glaciation 105, 211 

refugia 211,235 

relic plants 21 1 
Beckwit/iia glacialis 216 
beech 1 1 2 
Belemnites 13 
Bellis perennis 1 92 
Ben/uimia 137 

Benhaminae 131, 137, 138, 143 
Bering Sea 183 

Bering Strait 73, 162, 164, 170, 176 
Beringia 215, 217 
Bessastadaa 8 

Betula 1 5, 82, 1 1 3, 226, 355, 360, 362-364, 

callosa 207 

lutea 111,112,115 

nana 216, 372, 388 

papyrifera 1 1 2 



Betiila, piibescem 119, 120, 355, 356, 372 

loititosa 119, 120, 355. 356, 372 

verrucosa 197 
bicentric plants 93, 94, 100, 238, 246, 251, 
252, 253, 261, 269, 270, 279, 313, 333 
Bimasriis nnddali 146 
biocoenoses and migration 212 
biogeography 391 
biologic time-scale 231 
biota 82, 84, 185, 187, 208, 227, 368, 396 
bioticbelt 110, 111 
biotic zonation 116, 117 
biotite 279 
biotypes 224, 259 
birch \\9, ci. a\%o Betiila 

in Icelandic Interglacials 5, 6, 238 

in Icelandic Tertiary 238 
birch forest 110, 120, 121, 259, 362 
birds and dispersal 74, 140, 190, 191, 193, 

194, 304 
Bistorta vivipara 203, 214 
Bjarnarfjordur 318 
black spruce muskegs 1 15 
Blomvaag 228 
Blosseville coast 285, 289 
Blytt-Sernander concept 226 
Bockfjorden, hot springs 102, 104 
bogs 43, 355 
Bohemia 12 
Bohmerwald 234 
Boothia Peninsula 236 
boreal 46, 54, 71, 110, 111, 112, 113, 114, 

122, 123, 357, 364 
Boreal-Atlantic transition 226 
boreo-nemoral zone 110, 111, 112, 113, 

114, 122 
Borgarfjordur 7, 360 
Boston squantum tillite 13 

boreale 3 1 3 

lanceohiliini 314 
bottom configuration of North Atlantic 

24, 79, 80 
Brachyspira 1 54 
Bradybaena 167 

heniicki 1 92 

leiicopsis 192 

nificoUis 192 

linearis 116, 288, 290 

rwvae-angliae 288 

purpiirascens 104, 215 
Brazilian earthworms 129, 138 
Breidagil 378 
Breidamerkurfjall 379 
Breidifjordur 306, 307, 310, 312, 314, 316 

Brenner Pass 248 
Brimlarhofdi 386 
British Isles I, 75, 76, 176, 276, 300 

American plants 66, 194 

arctic plants 229 

pseudo-tillites 12 

tillites 12 
Brito-Arctic region 30 
Brjansliekur, lignite 2, 3, 14 
Biicepluila islaiulica lA 
Bi'idi end-moraines 259, 260 
Bulandshofdi 386 
Bunger's oasis 323, 325 
Biinias orientalis 222 
Bylot Island 340. 347, 349 

C» datings 169, 326, 330, 331, 357 

cacti, in Manitoba 201 


edentiila 66, 194, 196, 218 

moritinm 196, 218 
calcareous rocks 269, 276 
Calidivitrina 1 54 
California earthworms 127. 141 
Calhina vulgaris 43 
Caltlia paluslris 192 
Camaenidae 1 54 
Cameroon 1 3 1 

bar ha I a 88 

diihia 199, 203 

rot iindi folia 3 1 9 

uni flora 100, 101, 216, 270, 306, 
307, 308 
Campbell Island, earthworms 131, 132 
Canada 109, 115, 173, 201, 294 

boreal subdivision 114, 115, 122 

earthworms 127, 129, 148 

Huronian Glaciation 12 

tundra 119 

type of arctic refugia 235 
Canadian Arctic Archipelago 184 
Canary Islands, gastropods 160 
Cape Breton Island 338 
Cape Dyer 340,349,351,353 
Cape Farewell 292 
Cape of Good Hope, earthworms 132 
Cardanu'ne hellidi folia 215. 314 
Care.x 192,362 

adelosloina 3 1 7 

aiuhlyorhyiului 218 

arctogena 27 1 . 276 

bicolor 276,306,310,311 

Boecheriana 288 

briiniieseens 3 1 5 

capillar is 1 02 



Caiex, deflexa 291 

distans 2 1 2 

finiHi 209 

flava 314,315 

gkicialis 3 1 3 

livida 314,315 

Lyngbyei 74, 298 

maclovicma 306, 307, 308 

misandni 215, 271, 273, 279 

nardimi 216, 306, 307, 308, 309 

panicea 212 

pamllela 100,218,271,279 

piliilifera 3 1 9 

pseiidolagopiiia 218 

piilicaris 3 1 9 

iiifina 306, 307, 308 

nipestiis 214, 228, 255, 315 

saxotilis 2 1 5 

scirpoidea 230, 242, 244, 268 

stellulala 43 

stylaris 291 

subspatluueo 215, 315 

siipina 288 

wraV/fl 99, 216 
Caribbean Sea, Globigen'na-ooze 326 
Carpathians 174, 209, 216, 244, 252 
Carpinus Betiiliis 1 1 2 
CarvY/ 117 
Carychium 163, 167 
Caryophyllaceae pollen 363 

hypnoides 1 1 

tetragona 100,215 
Cassiopetum tetragoiiae 211, 213 
Castanea 1 1 7 
Catabrosa aqiiatica 192 
Cat hay a 374 

Catinella 157, 159, 160, 166, 167 
Caucasus earthworms 149 
Cecilioides 157, 160, 165 

Cramw 87,368,369,371,372 

///^ra 225 

pseudoplvygia 225 
Central American earthworms 138 
centric gastropods 163 
centric plants 100, 104, 229, 261, 270, 271, 

307, 311 
Cepaea 160, 161, 164, 167 

hortensis 73, 168, 169 

subfossil 168, 169 

arcticum 306, 307 

caespito.siim 363 

ceiastoides 100, 102, 212 

Edmondslonii 192, 306, 307, 309, 319 

Regelii 99, 216 

Cercidip/iy/liiiii arciiciiiii 1 5 
C£'/'<?«//w-maximum 363 
Cetraria chrysantlui llid 
Ceylon, earthworms 132 

angiistifoliiiin 293 

latifoliiim 88 
Chenopodiaceae pollen 372 
Chilean earthworms 129, 131, 132, 138 
Chi lor o 131, 135, 143 
chromosomes 280, 281, 282, 371 
chronology of animals 145 
ChrysospleniiiDi tenandniiii 104, 217 
circumpolar animals 75 
circumpolar plants 45, 46, 48, 49, 50, 51, 
53, 54, 70, 72, 207, 214, 215, 216, 301 
cirque glaciers 326 
Cladium 226, 227 
Cladonia 115,234 
Classical Wisconsin Glacial 346 
Clausilia 1 67 
climate 1 1, 12, 13, 227, 234, 326, 327, 355, 

364, 367 
climatic deterioration 16, 225, 227, 357 
climatic figures 109 
climax vegetation 224, 227 
clubmosses 200 
Clyde lateral moraines 348, 349 
coastal refugia 84, 236, 252, 261 
Cochlearia 192, 195 

groenlandica 195, 196 
Cochlicelki 1 67 
Cochlicopa 163, 167 
Cochlodina 1 67 
Cockburn end-moraines 348 
Coeloglossum viride 198 
Coilostele 160, 166 
Coleanthiis siibtilis 222 
Coleoptera 75, 76, 77, 81, 83 
Collembola 75 

vacillans 103 

Vahlianum 103 
Colymhetes dolabraliis 74 
Coiiimello 163, 167 
Coiiiaruiii 363 
Coniastoiua teneUa 217 
Compositae pollen 363 
conifers 1 10, 1 1 1 , 1 1 2, 1 1 3, 223, 357 
conglomerates 4, 33 
continental displacement 22, 23, 393 
continental drift 16, 142, 144, 147, 186, 

continental ice sheet 201 
continental refugia 208 
continental shelf 338 
continental slop3 321 



continentality 1 1 1 
continents, stability of 140 
convection currents 22 

lapponicuni 217 
spitsbergense 211, 218 
Coptis 293 

trifolia 290 
Corallorhiza trifida 198 
Cor nils 1 5 

suecica 293, 317 
Coronilla Emerus 228 
corresponding species 54-57, 66 
Corvlus 15, 225, 226 
Ave liana 112, 119 
Costa Rica earthworms 129, 132 
Crataegus 1 80 
Crater Lake 371 
crayfish, austral 132 

Gnwlinii 2?i2 

multicautis 230, 232, 242, 245 
paludosa 314, 315 
Cretaceous 1, 13 
Crino somnieri 74 
Criodrilus 127, 131 
Crozet, earthworms 131, 132 
Cruciferae pollen 372 
Crynwdes e.xidis 74 
cryocrats 367 
cryoturbation 327 
Cryptogranuna crispa 317 
Cryptomeria 1 5 
Czechoslovakia 235, 236 
cyclonic belts 226 
cyclonic tracks, during Wiirm 327 
Cygnus cygniis 192 
Cyperaceae pollen 356, 362, 387 
Cy si op ten's Dickieana 214 
cytologica! investigations 184, 185, 303, 

371, 395 
cytotaxonomy 214, 371, 372, 395 

Dactylina 234 

arctica 234, 236 

niadreporifornus 234, 236 

raimihsa 234, 236 

maadata 198, 396 

purpiirella 1 98 
Dalarna 329 
Darien 131 
Davis Strait, biological barrier 77, 80, 

168, 181, 394 
Deception Bay 347 
deciduous forest zone 110 

deglaciation 245, 252, 253, 256, 259 
Deha 193 
liortensis 1 46 
nianinialis 146 
octaedra 146, 150 
riihida 141, 146 
Denmark 226, 228 

Denmark Strait, biological barrier 25, 394 
Dermatina major 238 
Deroceras 163, 167 
alpina 103, 216 
brevifolia 2 1 5 
caespitosa 2 1 2 
flexiiosa 293 
detritus, mountain-top 277, 280, 328, 329, 

340, 343, 344, 345, 347, 349, 351 
Devon Island 338, 339 
Diapensia lapponka 229, 313 
Diaptonuis minimus 74 
diaspores 222, 227, 237 
diastrophism 384 
diatomite mud 355 
diatoms, on Mid-Ocean Ridge 23 
Dichogaster 132, 137, 143 
Dicranella varia 369 
Digitalis 227 

purpurea 224, 252 
Dinodriloides 133 

dioecious plants, and long-distance dis- 
persal 230 
Diplocardiacea 138 
Diplocardia 1 27 
Discus 163, 167 

disjunctive areas 140, 210, 243, 306 
Disko 80,141,287 
Disko Bay 287, 288, 289 
dispersability, radius of 392 
dispersal 72, 76, 77, 78, 87, 140, 156, 157, 
161, 162, 163, 164, 165, 168, 169, 174, 
189, 190, 191, 194, 196, 198, 200, 211, 
222, 228, 237, 243, 290, 304, 391, 393, 
394, 395 
dispersal ability 75, 230, 263, 304 
dispersal ecology 222 
dispersal limits 197 
dispersal means 189flF, 230 
dispersal rate 161 

distribution areas 45ff, 104, 127, 132, 139, 
140, 144, I46ff, 153, 163, 166,201,202, 
214, 221, 289, 294, 303, 368 
dolerite 30 
dolomite 243 
Domlieyopsis 1 5 
dominance 294 
Dovrc Mts. 245, 257, 269, 270, 276 



Draha 103 

alpina 215, 314 

crassifolia 242, 244, 267, 268, 288 

dovrensis 27 1 

fladnizensis 279 

Gredinii 100 

lacrea 215,279 

lane eclat a 288 

nivalis 100, 199, 276, 314 

noivegiea 306 

rupestris 216, 307, 308 

sibiriea 66, 72, 290 

siibeapitata 99, 215 
Dreikanters 4 
Drevfjellet 329 
Driloerius 129, 145 

o//an 129 
Drosera 200 

rotimdifolia 1 99 
Z)r>'a5 228 

flora 216, 218 

Babingtoniana 276 

octopetala 87, 91, 227, 257, 276, 388 
Dryopteris 385 

fragrans 88 

Linneana 386 

Oreopteris 252 
Diipontia Fisheri 2 1 5 
durite 35 
durmast oak 112 
dykes, in Faeroes 29, 32 

earthworms 127ff 

East Antarctica 322 

Eastern Canadian Seaboard 337ff 

ecologic amplitude 238, 239 

ecologic niche 145 

ecotone 121 

ecotype 239 

edible fruits, dispersal 304 

Eikisdalen 252 

Eisenia foetida 146 

Eiseniella tetraedra 146 

elements of Icelandic flora 297ff" 

Eleocharis qiiinqueflora 2 1 2 

Elisma natans 222 

Ellesmere Island 81, 96, 181, 190, 294, 338, 

Ellidavogur 82 
Ely m us arenariiis 363, 386 
Empetrum 192, 193, 385 

Eamesii 214 

hermapliroditiim 101, 214, 216 
Ena 167 
endemics 81, 103, 129, 131, 132, 138, 143, 

154, 156, 164, 395, 396 
end moraines 245, 246, 330, 348, 359 

Endogone 373 
Engadin 250 
eoarctic 209, 214, 215 
Eocambrian Ice Age 12 
Eocambrian dolomites 243 
Eocambrian tillites 12 
Eocene 1, 9, 156, 207 
Eocene-Oligocene 30 
Eodriliis 131, 132, 133, 134, 143 
eolian dust 362 
Ephedra 227, 368, 372 

di St achy a 87 

adenocaulon 222 

lactiflonim 306, 307, 308 

latifoliiim 74 
epiphytic lichens 238 
equilibrium, isostatic 1 
Equisetum 15, 192, 200 

peat 43 

palustre 1 92 

scirpoides 2 1 5 

silvaticiiin 317 

variegatiiin 103, 192, 214 
Ericaceae 385, 387 

canadense 223 

eriocephahis 228, 314 

Iwmilis 288 

imalaschkensis 100, 306, 307, 308, 309 

iinifiorus 3 1 4 

Parkeri 194 

pellucidiiin 1 94 

septangulare 66, 194 
Eriogoniim 371 

angustifoliiini 43 

Scheiichzeii 192, 214 

triste 101, 216 
erosion 323, 337 340 
erratics 321, 323, 329, 331, 337, 340, 351 
eruptions, submarine 29, 394 
Etna 371 

evolution 81, 96, 144, 188, 391, 396 
Euconuhis 163, 167 
Eudrilidae 138 
Eudrilus 133 

eiigeniae 1 42 
Eiiglandina 1 57 
Eiikenia 1 29 
Eiiomphalia 1 67 
Euphrasia 300, 301, 396 

arctica 1 02 

frigida 306, 307, 308 

lapponica 248, 250, 253, 257, 276, 279 

salisburgensis 248 




European plants 45, 54, 67, 68, 173, 185, 

216, 392 
eustatic changes in sea level 18, 21, 79, 80, 

360, 362 
Eutrema Edwardsii 2 1 5 
expanding earth crust 395 
Eyjafjallajokull 83 
Eysturoy 36, 37, 38, 41 

Faeroes-Iceland-Greenland Ridge 16,24, 

Faeroes 25, 76, 79, 173, 190, 195, 243, 282, 
290, 294, 395, 396 
fauna 75, 149 

flora 45, 46, 54, 66, 74, 194, 195, 298 
geology 29ff", 80 
Fag us 15, 111 

silvatica 1 1 2 
Falklands, earthworms 131,132 
Falljokull 379 
Far East, nemoral zone 1 1 1 
Faskrudsfjordur 318 
faunal barrier 77 
faunal exchange 73 
Felsenmeer 280,321,328 
felspar 29, 383 

Fennoscandia 1 1 0, 1 1 9, 1 20, 1 2 1 , 246, 250 
ferns 200 

baffinensis 99 
brachyphylla 99, 215 
hyperborea 2 1 8 
Richaidsonii 2 1 5 
rubra 192 

vivipara 216, 306, 307, 308 
Filchner Ice Shelf 323 
Filica'es spores 363 
Filipendula 363 
Finland 88, 109, 112, 119, 120 
Finnmark 88, 243, 244, 245, 246, 247, 25 1 , 

261, 330 
Finnish Lappland 329 
firn line 235 
fish 132, 140 
fissure eruptions 31 
fjallbarrskogsregionen 120 
flightless insects 75, 78, 83 
flint, icefloe transport of 229 
"floi'' bogs 355 
foreland refugia 329 
forest tundra 1 18 

fossil flora 5, 6, 7, 15, 81, 228, 237, 238 
fossil soil 144 
foundering 1 
Foxe Basin 348 

foxes 190 

Franz Joseph's Land 190, 195, 201, 294 

Fra.xinus 15, 111 

excelsior 112, 197 

nigra 111,112,115 

pennsylvanica 1 1 5 
frequency of polyploids 231,395 
freshwater diatoms 

on Mid-Ocean Ridge 23 

in Seltjorn peat 362 
freshwater fish 132, 140 
freshwater teleostei 129 
frogs, in Iraqi Kurdistan 212 
Fugloy 36,37,40,41,43 
Fugloyarfjordur 41 
Fulufjallet 329 

Galaxidae 1 32 

Brandegei 74 
brevipes 74 
pumilum 192 
verum 363 
Gapar 378 
Garmo 251 

garnet-biotite gneiss 340 
garnetiferous gneiss 341 
Gaspe Peninsula 181 
Gastrocopta 157, 160, 165, 167 
Gastrodonta 157, 165 
Gastropoda 153ff 
Gavia immer 14 
gemmulae 74 

genetic diff"erentiation 184, 185 
genetic drift 396 
aurea 66 

deto/isa 66, 288, 315 
purpurea 88, 225 
Geranium silvaticum 66, 373 
Gerpir 7, 8 
Getsjohoa 329 
Geuni rivale 395 
Gingko 1 5 

Gjevilvasskammene 277, 278, 279 
Gjevilvatnet 277 
glacial cirques 321, 328 
glacial survival 100, 166, 233, 245, 268, 

269, 320, 391, 392, 393, 395 
glacial striae 29, 321, 323, 329, 330, 333, 

342, 357 
glacierization 326, 329, 338 
Glaux 195 

maritima 196 
Globigerina-ooze 326 
Glossoscolecidae 129, 131 




aqualica 1 92 

fluitans 131, 192 
Glvp/nalinia 167 
Godagil 378,379,381,384 
Gondwana, gastropods 158 
Gondwana Ice Ages 1 3 
Gondwanaedrihis 1 32 
Gordiodrilus 1 38 
Gothenburg, marine clay 326 
Gotland 96 
"Gottweig" 326 
Gnap/ialium norvegiciiin 199, 293 
gneiss 338, 340, 344 
gneissic soil 252 
gradient of the ice 277 
gradual transition of biota 173,185 
Grafardalur 7 
Great Britain 

American plants 66, 194 

interglacial fossils 81 
"Great Iceland" 393 
Gramineae 356, 362, 385, 387 
granitic gneisses 338 
gray bottom clay 33 
gray gneiss 340 
Great Lakes area 109 
Great Lakes — St. Lawrence area 1 1 1 
Great Lyakovsky Island 202 
Great Plains 176 

Greenland 11, 12, 13, 15, 16, 18,46,75, 
76, 79, 80, 89, 99, 173, 176, 181, 184, 
185, 190, 195, 201, 217, 218, 230, 243, 
253, 261. 267, 286, 294, 395 

fauna 75, 141, 168, 181 

flora 25, 45, 54, 66, 72, 80, 1 74, 1 76, 2 1 8, 
233, 241, 243, 285ff 

refugia 81,235,292 

unglaciated areas 211,280 
Greenland-Scotland Ridge 79 
Grimsey 193 
Gronfjorden 99 
Gudbrandsdalen 253 
Guinea 131, 138 
Gulf of Bothnia 368 
Gulf of St. Lawrence 1 76, 338 
Gulf Stream 16, 99, 194, 208, 290, 304, 

355, 360 
Gunz-Mindel Interglacial 82, 386, 387 
guyots 22 
gyttja 144 

Habenaria hyperborea 
half-nunataks 292 
Hall Peninsula 349 
"Hallamyri" bogs 355 


halophilous plants, dispersal 177, 224, 304 

Halsatorfugil 384 

Hammerfest 373 

Haphochitonidae 132 

Haplodriliis 129 

Haplotaxidae 142 

Haplotrema 1 67 

Hardanger 257 

Hardangervidda 245, 253, 257, 259, 276 

Harjedalen 245, 257 

Hanimanella 200 

hypnoides 199, 216, 288 
Harteigen Mt. 257 
Harz 234 
Hawaiia 1 67 

minuscula 1 69 
heath, snow-protected 294 
Hebrides 1 2 
Hedera 1 1 7 

Hekla eruptions 356, 362 
Helianthemiini oelandicum 368, 370 
Helicella 167 
Helicigona 1 67 
Helicodiscus 1 67 
Helminthoglyptidae 154 
helo- and hydrophytes 289 
Holt 360 

hemi-alpine ecotone 121 
hemi-arctic 118 
Hendersonia 157 
Hengifoss 9 
Heradsfloi 319 
herb fields 294 
herbs 356 
Hespeiodiihis 1 32 
Hestoy 43 

Heteronieyenici ryderi 74 
Hicoria 1 5 
Hieraciiim 178, 180, 248, 287, 298, 299, 

300, 301, 303, 355 

alpina 100, 101, 215 

orthantha 203, 291 
High Alpine zone 122 
High Arctic 99, 104, 121, 208, 215, 352 
hillocks 357 
Himalaya 141,202,217 
Hippophae 369, 372 

rhamnoides 368 
Hippiiris 1 95 

lanceolata 195, 196 

tetraphvlla 195 

vulgaris 100, 102, 195, 196 
Histn'oniciis histiionicus 14 
Holarctic 127, 146, 156, 157, 163, 167, 181 
Holmatindur 7, 9 
Holocene deposits, dating 357 



Holsteinsborg 290, 292 
Honckenya 195 

diffusa 196 
Hops Island 99 
Hordeiim 363 
Hornafjordur 16,83,319 
hornbeam 112 
Hornsound 112 
horsetails 200 

hot desert belt, in Paleozoic 13 
hot springs, in Svalbard 101, 102, 104 
Howascolex 1 38 
Hredavatn 5 
Hreppar 360 
Hriitafjordur 362 
Hvalbiarfjordur 37 
Hvalfjordur 306, 307, 311, 314, 319 
Hvalvatn 384 
Hvannhagi 37 
Hverfjall 381 
Hudson Bay 122 
Hudson Bay Lowlands 118 
Hudson Strait 338, 343, 347 
human traffic and dispersal 173 
humic soil 227 
humid climate 227, 229 
humid and dry periods 226 
humus and postglacial survival 229 
humus-free soil 227 
Hunavatnssysia 311 
Huperzia Selago 214 
Huronian Glaciation 12 
hydrobiotite 279, 343 
hydrochlorous dispersal 76, 162, 177,224, 

Hydrocotyle vulgaris 66 
hydrophytes 224 
Hydroporus nigiita 82 
Hyla arboiea 212 
Hypnuin cuspidal uiu 192 
Hypshhermal Period 168, 169, 181, 203, 

223, 226, 230, 251, 270 

Ice Ages 12, 13, 201, 209, 215, 216, 228, 

229, 231, 303, 305, 367 
ice divide 326, 357 
icefloe transport 229 
icefree refugia 17, 18, 105, 234, 251, 263, 

269, 305, 320, 321, 323, 342, 347, 349, 

ice islands 72 
ice regression 209 
icesheets, thickness 322, 358 
ice surface slope 323 
Iceland Iff, 16, 29, 30, 46, 173, 179, 185, 

190, 193, 195, 217, 243, 282, 290, 294, 

337, 395, 396 

Iceland, climate 14 

dispersal of flora 211,304 

extension of glaciation 303, 304 

fauna 74, 81, 174, 181. 194, 195, 304 

flora 45, 46, 54, 66, 72, 74, 79, 8 1 , 89, 96, 
174, 181, 194, 195, 233, 236, 298, 304, 
306; cf. Icelandic flora 

geographical position 75, 189, 190, 363 

geology Iff", 80, 187, 358, 359 

impact of Norse colonists on flora 1 74 

Interglacials 17, 188, 305, 377fl" 

marine transgression 360, 361 

Mid-Ocean Ridge in 21,22 

Pleistocene deposits 5, 188, 377fT 

pollen species 9 

pollen zones 364 

raised beaches 360, 361 

refugia 168, 304, 306, 357 

soil erosion 362 

Tertiary history Iff", 14, 357, 394 
Iceland-Faeroe Ridge 26, 243 
Iceland-Greenland Palearctic elements 81 
Iceland-Greenland Ridge 25 
Icelandic flora 

apocrats 371 

cytological investigations 303 

its distribution 306,313,314 

its elements 1 79, 1 86, 297, 298, 299, 300, 
301, 302, 303, 304, 306 

its history 299 

its species number 297, 303, 355 
Ilex 15, 117, 127 
Ilyogenia 138, 144 
immigrant weeds 222, 224 

of Icelandic flora 304 

of Scandinavian flora 221 

speed 226 
interglacial deposits in Iceland 17, 81, 

188, 305, 377, 386 
interglacial flora 245, 261 
introduced plants 

in Iceland 301 

in Newfoundland 174 

in North Atlantic area 45, 66 

occasional 72 
introgressive hybridization 225 
intrusions, in Faeroes 31, 36 
Iphigera 1 67 

Irano-Turanian distribution 212 
Iraqi Kurdistan 211,212 
Ireland, basalt 80 
Iron Age agriculture 225 
irregular meiosis 280 
Isachsen Island 202 
Isafjordur 318 
Isfjord 101, 102 



Isle of Mull 74 

Isortoq Fjords 292 

isohyet in North Atlantic 293 

isolation processes 158, 163 

isostatic equilibrium 1,322 

isostatic movements 79, 80 

isostatic recovery 347, 360, 362 

isostatic sinking 2 

isostatic tilt 342 

isotherm, 10° for July 293, 327 

isthmian links 23 

Italy, earthworms 148 

Jasren 228 

Jakobshavn 287 

James Bay 122 

Jameson Land 13 

Jamtland 228, 245, 257, 329 

Jan Mayen 45, 46, 54, 189, 190, 195 

Januliis 1 65 

Japan 9,110 

Jarvso 373 

Jokuldalur 9 

Jotunheimen 25 1 , 253, 256, 269, 270, 275. 

Jitglans 15, 117 
Julian 256 
Juncus 200 
alpinus 289 
arcticus 101 
articulatus 1 98 
biglumis 198, 215 

101, 317 






inftexus 212 

ranarius 289 

sqiiarrosus 287, 317 

subnodulosiis 66 

trifidus 192, 203, 288 

triglumis 198, 214 
Juniperus communis 43 
Jutland 88 

Kallsoy 41 

Kangerdluarsuk ungatdleq 
Kangerdlugssuaq 289 
Kamchatka 89, 1 10 
kame terraces 346 
Kanin Peninsula 268 
Kansan Glacial 389 
Kara Sea 200 
Kara Strait 200 


Karavanken 209 

Katla, eruptions 262 

Kaumajet Mts. 338, 339, 340, 341, 342, 

344, 353 
Kautokeino 88 
Kerguelen, earthworms 1 3 1 
Kerriona 1 29 
Kiglapait-Nain area 346 

caricina 2 1 4 

simpliciiisciila 100, 101 
kinetophilous plants 370 
King Charles Land 13,99 
King's Mirror ix 
Kinnekulle 228 
Kittila 329 
Koenigia 229 

islandica 102, 215 
Kola Peninsula 184 

birchwoods 1 20 

plants 104 

type of refugia 235 
"Kolband" 33 
Komsomol'skaya 322 
Konungsskuggsja ix 
Koroksoak Glaciation 342, 346, 347 
Koroksoak trim line 347 
Krakatoa, flora 183 
Krokkleiva 228 
"Kubbaberg" 33 
Kunoy 41 

Kvccnangen Fjord 243 
Kvasrhusho 260 
Kvandalstind 377 
Kvisker 377 

Labrador 118, 122, 180, 181, 182, 184, 

235, 294, 338, 339, 340, 346 
Labrador Sea 22 
Labrador-Ungava 348, 350 
labradorite-bytownite 383 
Lady Franklin Island 349, 350 
Lake Superior 111, 115 
lamellae zones, in Faeroes 36 
Lamellaxis 157 
landbridge, Beringian 73 
landbridge. North Atlantic xii, 23, 25, 54, 
66, 79, 80, 100, 176, 183, 186, 201, 392, 
393, 396 
botanical arguments 72, 78, 79, 176, 

185, 186,211,243 
geological evidence 25, 186 
its age 17, 18, 21, 25, 72, 73, 80, 81, 82, 

84, 186,216, 243, 305, 393 
paleogeographic evidence 73, 80, 186 
rates of evolution 188 



landbridge, submarine relief 24, 25, 207 

zoological arguments 16, 79, 150, 165, 
168, 186 
landnam 362, 363. 369 
Landssveit 360 
Lappland 120, 372 
Loiix 110, 113 

laricina 115 

sibirica 1 1 9 

Sukatchewii 1 1 9 
Last Glaciation 82, 84, 87, 100. 189, 215, 

235, 236, 241, 246, 346. 350 
Last Interglacial 227, 235, 326 
Late Glacial 227, 228, 252, 355, 364 
Late Eocene fossils, in amber 81 
lateral moraines 346, 348 
Lauparen 252 
Laurentia Minor 12 
Lamia 161, 167 
lavas. Tertiary 2 
Lea Valley 168 
leaf impressions 7, 377 
Ledum groeiilaiuUcum 290 
Leiostyla 1 67 
Lehnumnia 1 54 
Lemna 192, 193 
Lena River 174, 253 
Lepidoptera, in Greenland 75 
Leptodiihis 135 
Lesja 253, 255, 260 
Lesjaverk 253 
leso-tundra 118 

albida 74, 287, 293 

St rami Ilea lA 
Levitunturi 329 
Lewisian gneisses 12 
Liard Valley 115 
lichens 105, 200, 233ff, 325. 352 
life-forms 294 

lignites 1, 2, 7, 9, 15, 31flf, 395 
Ligusticum scoticum 1 84 
Umax 154, 160, 161, 164, 167 

marginatus 168, 169 
limestone 244 
limicolous earthworms 132 
Liniiaea borealis 290 
Liiium 363 
Liodriliis 129 
Liquidambar 1 5 
Liriodendion 15, 117 
Listera ova fa 198 
Litlisandur 5, 6, 7 
loess 200 

Lofoten 246,251,321,330,337 
Loiseleiiria 200 

prociimbens 46, 199 

Logurinn 384 

Lom 248, 253, 255, 256 
Lomatogonium lotatum 88 
long-distance dispersal 77, 80, 82, 87, 176, 
177, 178, 179, 185, 189, 194, 195, 197, 
200, 201, 210, 211, 222, 224, 226, 230, 
235, 237, 363, 368, 392 
Long Range Mts. 338 
Lophocolea bidentata 192 
Low Alpine zone 122 
Low Arctic element 181, 182 
Low Arctic zone 121 
Lower California, earthworms 127 
Lower Carboniferous coals 13 
lowering of sea level 25, 323 
Lumbricideae 127, 142, 143, 147 

castaneus 146, 148 

festimis 146, 149 

rubelhis 146, 148 

teirestris 146, 149 
Liiionitim iiatans 222 

arctica 100.271,272,279 

arcuata 1 00 

maxima 252 

nivalis 2 1 5 

spicafa 192, 288 

sudelica 3 1 3 

Wahlenbergii 101,218 
Lycoperdon 197, 200 
Lycopodiiim 197, 200 

clavatiim 319 

Selago 214 

Macaronesian gastropods 160, 165 

Mackenzie River 176 

Macquarie, earthworms 131 

macrofossils 227, 367, 384 

macrolichens 233, 234, 236, 237 

Madagascar, earthworms 131,138 

Madeira, gastropods 160 

magnetic polarity 3, 4, 386 

magnetite 384 

magnetized basalt 3, 4, 379, 381, 386 

Main-Boreal zone 115,120,122 

Malayan earthworms 142 

mammals 145, 153 

mammoth, Siberian 215 

Man, as dispersal agent 190 

Manchuria 111, 149 

Manitoba 111, 113, 115, 193, 201 

Maniwake 111 

Maoridriliis 135 

marine limit, in Labrador 347 

marine transgression, in Iceland 360 



Maritime Provinces 109, 112, 338, 339 

Masi Mt. 243 

"massif de refuge" 81 

Matricaria matricarioides 222 

mature species 210, 216 

Maudheim 323 

mesocrats 367 

Mesodon 1 67 

Mesomphix 1 67 

Mesozoic 12, 13, 143, 144, 155, 158 

meta-sediments 338 

Metaseqiioia occidentalis 1 5 

Mexico 127, 129, 132, 166 

mean temperature 99, 122, 327, 355 

Megascolecidae 127, 141 

Melampyrum silvaticiim 317, 318 


angustifolium 2 1 5 

apetahtm 215, 288 

fiircatitm 2 1 5 
menotrophication 227 
Mertensia 1 95 

maritima 1 96 
mica schist 244, 277 
Micmac Indians, and Cepaea 169 
Micralymna brevilingue 76 
Microchaetidae 142 
niicrolichens 233, 238 
" ' Micropiniis' ' 3 74 
Microscolex 131, 133, 143 

dubiits 141 

phosphoreits 141 
micro-thermocouples 202 
Mid- Atlantic Ridge 22 
Middle Alpine zone 121 
Mid-Oceanic Ridge 21, 22, 23 
Middle Swedish moraines 327 
Middle West of the U.S.A. 1 1 1 
Millsonia 137 
Milium Fjarda 42 
migration 73, 87, 93, 102' 184, 193, 209, 

211, 236,291 
Mindel Ice Age 209,211,235 
Mindel-Riss Interglacial 82, 216, 386, 389 
Minnesota, boreo-nemoral zone HI 

bi flora 314 

Rossii 99, 216 

rubella 271, 272, 279 

verna 1 99 
Miocene 5, 158, 207 
Moberg formation 3 
Mohorovicic discontinuity 21 
Molde 254 
Moldhaugar 355, 356 
moles 148 
molluscs, cf. Gastropoda 

Monacha 1 67 

Mongolia 243 

Monumental Island 349 

moor soil, and earthworms 145 

moraines 4 

More 246, 251, 252, 253, 255, 257, 259, 

261, 276 
mosses 200, 325 
montmorillonite 343 
Mt. Brave 340 
Mt. Washington 203 
mountain-top detritus 277, 280, 321, 328, 

329, 340, 342, 343, 344, 345, 347, 349 
Mucuna 222 

Mugford series 340, 341, 342 
mull 144 
muskegs 115,290 
mutations 239 
mycorhiza 227 
Mykines 31,41,43 
Mykinesfjordur 35 
Myrdalur 306,307,314,315,319 
Myrica Gale 120, 363 

NaCl and gastropods 162 
Nain 338, 347 
Nain-Kiglapait area 341 
Nansen sill 25 
Nansen Straits 25 

Nearctic 74, 75, 77, 80, 127, 128, 129, 181 
Nebria gvllen/iali 82 

nemoral'zone 109, 110, 111, 112, 113, 119 
neo-endemics 66, 396 
Neinatogenia 132, 135, 143 
Neogaster 131, 137, 143 
Neogene 207, 208, 219 
Neolithic man, and ruderal flora 223 
Neotropical earthworms 129, 130 
Nesovitrea 163, 167 
New Caledonia, earthworms 129, 138 
Newfoundland 46, 1 1 5, 1 1 8, 1 22, 1 76, 1 8 1 , 
338, 339 

European plants 45, 174 

relict plants 174 
New Siberian Islands 372 
New South Wales, earthworms 132 
New Zealand, earthworms 131, 132, 138 
Niagara Peninsula 109 
Nigritella nigra 279 
Nipfjallet 329 
Nipur 2 

non-climax vegetation 224 
Nordaustlandet 326 
Nordland 242, 245, 246, 247, 251, 267 
Nordre Stromfjord 288, 290 
Nordreisa 243 



Normandy, tillites and pseudotillites 12 

Norsemen and flora 78, 174, 291 

North America 25, 110, 111, 147, 211 

northern coniferous forest region 119 

Northeast Land 99, 233, 234 

Northern Hemisphere nemoral 1 10 

Northern Ireland 395 

northern tree-line 119 

Northern Pacific gastropods 156, 164 

North Sea continent 229, 331 

Notiophilus aqitaticus 84 

Nova Scotia 169, 174 

Novaya Zemlya 99, 103, 105, 176, 200, 

216, 218, 233, 235, 236, 268, 294 
Novosibirsk Archipelago 202 
Norway 88, 112, 119, 121, 184, 190, 236, 

242, 243, 244, 245, 250, 251, 253, 259, 

261, 282, 290, 337 
Corvlus-non-Pinus cHmate 226 
gastropods 1 68 
survival of plants 24 Iff, 26 Iff 
type of refugia 228, 235, 269, 280 
Nucella transgression 362 
Nugssuak Peninsula 181 
nunataks 107, 236, 255, 256, 261, 277, 

279, 280, 282, 291, 305, 306, 307, 311. 

323, 328, 329, 333, 337, 350, 358, 370 

Oraefi 319, 377 
Orchidaceae 200 

maciilata 66 

mascula 1 98 

Spitzelii 96 
Ordovician 13 
Oregon 371 

oreophytic plants 209, 214 
Oriental region, earthworms 138 
Orkney 190 

Orthilia seciinda 199, 290 
Oskjuvatn 384 
Oslo district, arctic plants 228 
Ostrobottnia 120 
Ot hilts welanocephahis 83 
Ottawa River 1 1 1 
Ottawa — St. Lawrence Valley 1 1 1 
overwintering, cf. Pleistocene survival 
Oxalis Acetosella 319 
OxychiUis 1 67 

alliarius 168 
Oxycocciis microcarpus 3 1 5 
Oxyloma 163, 167 
OxYiia digyna 103,211,214 
Oxvtropis deflexo 88, 96, 230, 242, 245 

O'VOi* isotopes 13 

oases in Antarctica 323 

occasional introductions 72 

ocean basins 2 

ocean floor II, 17, 21 ff 

Ocnerodrilinae 127, 129, 131, 135, 138, 

143, 145 
Ocnerodriliis 138 
Octochaetinae 136, 138, 142 
Octochaetoides 137 

cyaneiiin 1 46 

lacteuin 1 46 
Odadahraun 3 
Oenothera 180 

biennis ll'i 
Oland 112,228,368,370,372 
"Old Iceland" 187 
Old Red Sandstone 13 
Olenek River 243 
Oligochaeta. cf. earthworms 
olivine 383, 384 
Olstappen Lake 251 
Omberg 228, 369, 370 
Omphalodiscus krascheninnikovii 236 
Opdal 248, 253 
Opeas 1 57 
Oia;fajokull 377, 379 

Papaver 91, 92, 93, 94, 197, 248, 249, 300 
bicentric in Scandinavia 91, 92, 93, 94 
cytogenetic analysis 184, 185, 280, 281 
migratory history 93 
sect. Scapiflora 229, 247, 249 
Argemone 197 

Dahlianum 92, 100, 104, 216, 218, 249 
duhiuin 1 97 
hybridum 1 97 
Laestadiamini 92, 249 
lapponicum 92, 184, 249 
Nordhagenianiini 93, 395, 396 
radicatum 92, 93, 1 84, 249, 265, 266, 268, 

279, 280, 281 
relict urn 93, 249, 267 

palagonite 3, 384 

Palearctic 75, 78, 80, 81, 129, 181 

paleographic maps 142ff 

paleogeography of North America 173ff 

Paleocene gastropods 158 

paleoclimatic map 14 

Paleogene 208 

paleomagnetism 14 

Paleotropical earthworms 131 

Paleozoic climates 12, 13, 111 


crocata 2 1 2 
pratensis 2 1 2 



PaWfera 167 

palynology xi, 2, 174, 355ff, 367ff 

"Paudorf 326 

Pangnirtung Peninsula 181 

Parachilota 131, 134, 143 

Parana/Uruguay basin, earthworms 129 

Paravitrea 1 67 

Paris qiiadrifolia 66 

Parmelia 234 

aspera 238 

centrifuga 234, 235 

intestinifonnis 234 

subobsciira 236 
parvifoiiate forest 113 
Panya iiudicaulis 218 
passive dispersal 75, 140, 161, 163, 165, 

Paulistus 129 
Peary Land 18, 202, 280 
peat 43, 355, 360 
peatland 113, 114 
peat soil and earthworms 145 

clasvantha 103,211,218 

flanimea 243, 244, 263, 264, 306, 307, 
309, 316 

hirsuta 215,292 

lahradorica 288 

Oedeii 252, 256, 257, 258, 259, 279 

silvatica 66 
Penny Ice Cap 348 
Penultimate Glaciation 24, 25, 26, 346 
Penultimate Interglacial 218 
"Perapohjola" 120 
periglacial flora 228 
periglacial molluscs 168 
Periodiihis 135 
permafrost 122, 327 
persistence of plants 228; cf. survival 
Petasites frigidus 2 1 8 
Petawawa 1 1 1 
phenology 202 
Phleuni comnnitatiim 287, 293 
Pliilomyciis 157, 167 
Philophorum robust uni 238 

algida 215,277,314 

conciima 100, 218, 265 
Phreodrilidae 132, 144 
Plireodriloides 132 
Phreodriliis 132 
Phreorictidae 142, 145 
Phvllodoce 200, 294 

coerulea 199,288,314 
phytogeographical connections 45ft', 100, 

173, 174 
phytogeography of Greenland 2850" 

Picea 15, 82, 113, 119, 226, 227, 387 

Abies 112, 119, 229 

excelsa 1 97 

glauca 112,115,118 

mariana 17,112,115,118 

obovata 1 1 9 

rubens 207 
Piceetea in Miocene and Pliocene 207 
Pickfordia 137 
piedmont glaciers 326 
pinewoods 113 
pink-footed geese 194 
pinnacle-like mountain peaks 321 
Pimis 15, 82, 113, 226, 363, 387, 388 

Banksiana 1 1 5 

Cembra 1 1 9 

contort a 1 1 5 

nigra 1 1 2 

Pence 1 1 2 

resinosa 111, 112, 115 

sibirica 1 1 9 

silvestris 112,119,197,251,256,259 

Strobns 111,112,115 
Pinns-heh, in Norway 251 
Pinns-non-Corylus climate 226 
Placopsis gelida 236 
Plagiochaeta 135 
Planera 1 5 
Planogyra 1 67 
plant communities 211, 294 
plant sociations 294 
plant succession, laws of 226 

lanceolata 224 

major 224 

maritima 192, 363 
plastic species 210 

hyperborea 293 

oligantha 265 

parvnia 248 
Plat anus 1 5 
plateau basalts Iff", 29ff 
Pleistocene xii, 4, 24, 25, 73, 79, 81, 105, 
158, 166, 168, 170, 214, 215, 216, 219, 
229, 231, 235, 236, 241, 261, 326, 329, 
337, 341, 342, 378, 379, 395 
Pleuropogon Sabinei 99, 215 
Pliocene 4, 5, 8, 16, 79, 158, 160, 186, 216 
Plutellus 127, 129, 141 
Poa 103, 192 

sect. Oreinos 89 

abbreviata 2 1 5 

alpina 89, 216 

arctica 94, 95, 96, 215, 265, 275, 279 

Fernaldiana 89 

flexuosa 89, 90, 93, 279 



Poa, Hartzii 218 
jemtlandica 89, 91 
laxa 89, 306, 310 
stricta 275 
Pohjanmaa 120 
Poiretia 165 

Poland 113, 160, 216, 218 
polar bears 190 
polar coast 120, 121 
"polar flora" 14 
Polar Stream 304 
Pole 9, 13, 46 

Polenwniuin horeale 101, 218 
pollen analysis 87, 224, 225. 227, 355, 367, 

pollen diagrams 226, 355ff 
pollen floras, of Iceland 15 
pollen morphology 372 
pollen species 9 
polster-growth 202 
polycentric species 314 
Polygiridae 154 
Polygonum civiciilare 363 
polygon-tundra 202 
polyploids 203,231,395 
Polypodiaceae 385 
Polystichiim Lonchitis 287 
Polvtiichiim 1 97 
Pond Inlet 338, 339 
Pontodrilus 141 
Popiilus 15, 113 
halsainifera 1 1 5 
giandidenlata 111, 112, 115 
tiemida 113,319 
treimdoides 112, 113 
Popidus-Benda-Piniis/Corvliis forest 226 
porphyritic rocks 29, 30 
Porsanger Fjord 243 
Portugal 174 

Possession, earthworms 131 
Post-glacial climatic optimum, cf. Hypsi- 

Potaniogeton 192, 193 
epihydnis 66 
grainineus 289 
Potent ilia 103 
Grant zii 101,216 
Egedii 314 
hyparctica 268, 288 
multifida 217 
nivea 279 

stipularis 66, 72, 290 
Prairie Provinces 109, 113 
Prairie Region 1 1 1 
Prangetalia 212 
Prangus ferulacea 2 1 2 
Pre-Cambrian 11, 12, 111 


auriculata 2 1 2 

egaliksensis 315, 316 

stricta 288, 314, 315, 316 
Primuleto-Blysmion compressi 212 
Priochaeta 137 
Priodoscolex 137 
Prophysaon 157 
Proterozoic basic lavas 340 
protocrats 367 
Protornatellina 157 
Primus Padus 113,385,388 
Pseudocolumna 1 57 
pseudotillites 12, 22 
Pterostichus diligens 82 

angustata 99, 103, 215 

maritima 192 

phryganodes 103, 217 

svalbardensis 103 

vacillans 2 1 8 

vilfoidea 211,217 
Punctum 163, 167 
Pupilla 163, 167 

muscorum 168 
Pupoides 157, 165, 167 
Pupisoma 157 
Purpura transgression 362 
Pvramidula 167 
Pyrenees 209, 216, 327 
Pyrola 200 

grandi flora 1 99 

)uinor 199, 290, 293 

rotundifolia 3 1 5 

secunda 3 1 5 

quadrupeds xi 
quartzites 340 
Quaternary 16, 17,41, 111, 163,245,367; 

cf. Pleistocene 
Quebec 109, 111, 122 
Quechuona 129 

Queen Charlotte Islands, earthworms 129 
Queen Elizabeth Islands, refugia 202, 

338, 352 
Queen Maude Land 323, 324 
Quercetum mixtum 226 
Quercus 15, 111, 117 

horealis 1 1 2 

macrocarpa 113, 115, 117 

petraea 112,372,373 

puhescens 1 1 7 

Roinu- 112,113,117,372,373 

radiocarbon datings 326, 346 
radius of dispersability 392 



Ra end-moraines 228, 245, 246, 327, 

raised beaches 360 
Ramah sedimentary series 340 
Ramie I la 138 
Rana lidibimda 212 
Ranaskollen 254, 260 
"Rann" 33 

Ranunculaceae pollen 363 

acris 29 1 

auricoinus 88, 319 

Flammula 43 

glarialis 100,292,314 

hederaceus 66 

hvperboreus 215 

nivalis 271,273,276,288 

Pallasii 46 

pedatifidus 288 

platanifolius 88 

pygmaeus 104, 199, 215, 314 

repens 1 92 

Sabinei 202 

spitsbergensis 103, 218 

sulphureu 202 

m lander i 218 
Rauma River 254 
rate of evolution 81,187 
red spruce 1 1 2 
refugia, unglaciated 123. 246, 328, 333 

continental 208 

hypothesis 229, 25 1 , 32 1 , 329 

identification of, 394 

indications of 321 

in Baffin Island 352 

in Canadian Arctic Archipelago 280, 
338, 352 

in Iceland 83, 168, 303, 305, 312, 

in Labrador 342, 352 

in Scandinavia 224, 235, 251, 255, 261, 
266, 280, 321 

in Shickshock Mts. 352 

in Svalbard 106, 107, 238 
regio alpina superior 84 
relationships 138, 140 
relic species 174, 210, 234 
remanent magnetism 377 
reptiles xi, 140, 212 
Retinella 1 67 

Reykjavik ix, 17, 46, 82, 357, 360 
Rhaeto-Liassic floras 13 
Rhamnus 1 5 
Rhiostoma 157 
Rhizocarpon 233 

inarense 239 

super ficiale 238 

Rhododendron 253 

lapponicum 46,25 1 , 253-257, 260, 263, 276 

parvifolium 263 
Rhododrilus 133 
Rhone Glacier 371 
Rhus 1 5 

Rhyocia quadrangula 74 
Rhyt mites 381 
Riesengebirge 234, 235 
Riss Glacial 84, 209, 21 1, 214. 330 
Riss-Wiirm Interglacial 82, 386 
River Plata estuary, earthworms 131 
roches moutonnees 323 
Rocky Mts. 113, 127, 176 

borealis 314, 396 

canina 314, 315 

Doniana 319,396 
Romsdal 253, 254, 255, 276 
Romsdalsfjord 252, 253, 257 
roof clay 33 
Rorippa silveslris 370 
Rosaceae pollen 363, 385 
Ross Ice Shelf 323 
Rost 246 

subg. Eubatus 180 

Chamaenwrus 192, 217 

saxatilis 304 
Rumex 363 

Ruppia spiralis 195, 196 
Russia 109, 113. 118, 119, 122, 294 
Ryfylke 267 

Sagina 200 

caespitosa 100. 270, 279. 306. 307. 308. 

procumbens 1 99 

saginoides 2 1 2 

subulata 1 92 
Sagiek Fjord 342. 343 
Sagiek Glaciation 346. 347 
Sagiek moraines 347. 348, 350 
"5«//.v 2" 369 

Salix 15, 1 13, 356. 357. 362, 384, 385, 387, 

maximum in Iceland 362 

glauca-callicarpaea 74 

herbacea 87, 100, 209, 214, 216, 227, 288 

lanata 66, 384 

polar is 103, 218 

reticulata 192, 216, 384 
Salpaussalka moraines 245, 327 
Saltdal 251, 268 
Salten Fjord 266 
salt water dispersal 141, 194, 195 
Sam Ford Fjord 349 



Sandsdalur 42 
Sangamon Glaciation 347 
Sangiiisoiba 371 
saprophagous earthworms 142 
Sardinia, earthworms 148 
Sarek Mts. 267, 327 
Sargag 287 
Saskatchewan 1 1 5 
Sassafras 1 5 
Sassen area 102 
Saxifraga 192 

aizoides 199, 214, 216, 319 

Aizoon 267,268,306,310,311,319 

cerniia 104, 314 

Cotyledon 66, 319 

flagellaris 99 

foliolosa 3 1 3 

groenlandica 104, 199 

hieraciifolia 248, 251-254, 276 

Hiicidus 2 1 6 

hypeiborea 203 

hvpnoides 66 

nivalis 199, 215 

oppositifolia 104, 214, 227, 255 

rividaris 1 99 

setigera 215 

stellar is 291 
Sayan Mts. 215 
Scandes 87, 88, 91, 326, 329 
Scandic 393, 394 

Scandinavia 76, 99, 103, 122, 173, 176, 
184, 185, 215, 216, 217, 218, 230, 235, 
245, 247, 248, 250, 253, 294 

endemic plants 238, 265 

flora elements 87, 100, 105, 221, 229, 
230, 238, 241, 246, 247, 263, 265 

history of flora 87, 241 

ice divides 326 

mountain flora 89, 91, 92, 93, 95, 100, 
104, 105, 121, 221, 229, 230, 233, 238, 
schist 244, 257, 269, 276, 277 
Schwedeneck 369 

alpimis 243 

aniericaniis 1 92 

emergens 243 

maritinnis 1 92 

pauciftorus 289 

pumilus 243, 244, 245, 265, 266 
Scoresbysound 294 
Scotland 80, 89, 113, 168, 173, 176, 185, 

190, 215, 290, 395 
Scoto-Irish region 30 
Scottish Highlands 294 
sea currents, and dispersal 66, 177 
sea level changes 21, 79, 80, 358, 360 

seamounts 22 

acre 373 

villosum 1 99 
seed weight and dispersal 196 
seismic reflection technique 22 
seismic wave velocities 21 
Selaginella selaginoides 363 
selection and tolerance 203 
self-fertilization and dispersal 162 
Seltjorn 360 
Semilimax 1 54 
Sequoia 1 5 

Langsdorffii 15, 30, 35 
Sesleria albicans 396 
Shapkina River 331 
Shetland 190, 330, 332, 396 
Shickshock Mts. 338, 340, 342, 350 
Sibbaldia procwnbens 102, 288, 293 
Siberia 73,77,113,118,202,211,215,217, 

235, 243 
Sieversia glacialis 202 
Silene acaulis 46, 203 
sills 29ff- 

Sinnilium vittatiini 74 
Sisyrinchiiini 66 

angustifoliuni 1 94 

Bermudianiim 194 

hiberniciini 1 94 

nionlannni 1 94 
Skagafjordur 313 
Skalholt 362 
Skardsstrond 2, 5 
Skeidara 379 
Skeidararsandur 379 
Skjalfandafloi 313 
Skjolgil 378, 379 
skogsodlingsgransen 120 
Sleggjuhekur 5 
slope of ice surface 323 
Slugufjallet 329 

small populations and evolution 96, 396 
Smith Sound 9, 18 
Snccfellsnes 17,311,387 
Snidabrekka 378 
Snidagil 377fi" 

snowbed vegetation 276, 288, 294 
Sogamyri 357, 358, 360 
Sogn 248 
Sogne Mts. 253 
Sognefjord 321 
soil 144, 145, 227, 362 
solar radiation 325 
Solbjergfjell 228 
Solvagtind 230, 242, 268 
Sondre Stromfjord 288, 289 
Sorbus 385, 388 



Sorbus, aiicupaiia 113, 114 

decora 114,385 
soredia 236, 237 
Sorfjord 257 

South Africa, earthworms 131, 132 
South America 25, 131, 138 
Southern Boreal subzone 115 
Southern Hemisphere 

biotic zones 1 10 

continental drift theory 144 
southfacing slopes 292 
South Georgia, earthworms 131, 132 
South Norwegian flora, recent discoveries 

Spain, gastropods 166 
Spaiganimn 193 
Sporganophihis 146, 149, 150 

eiseni 1 27 
speciation during Glaciation 81, 211, 215 
speed of evolution 231 
Spermatophyta, evolution 144 
Spennodea 1 67 
Sphagnum 43, 357, 363 
Spiranthes Ronianzoffiana 66 
Spitsbergen 16, 99, 102, 103, 176, 190, 
195, 201, 216, 217, 218, 236, 243, 267, 
268, 294 

coals 13 

connection with Greenland 25 

flora elements 46, 100, 101, 104, 107, 

fossil floras 15 

icefree refugias 100, 102, 105, 106, 107, 
211, 235 

hot springs 101 

plants 45, 46, 54, 99, 100, 101, 102, 103, 
104, 107, 218, 233, 234 

pollen species 9 

postglacial dispersal 195 

pseudotiilites 12 

tillites 12 

type of refugia 235 
squantum tillite 13 
"Stabbi" 33 

stability of continents 140 
Stadjan 329 
Stad-Sunnmore area 321 
Stafholt 5 
stagnant ice 352 
St. Donat 1 1 1 
St. Lawrence River 1 1 1 
Statice maritima 1 92 
statocracy 374 

statocrats 367, 370, 371, 373, 395 
Stavangerfjord 321 
Steingrimsfjordur 318 


calvcantha 100, 102, 287, 306-309, 316 

crassipes 96, 230, 248, 271, 274, 276 

humifiisa 104,215 

media 363 
Stetwtrema 1 67 
steppe 113,227 
steppe flora 87 
steppe region 1 1 1 
stepping stones ix, 22, 23, 165 
Steyermark 252 
Stockholm 370 
Stodin 386 
Storfjord 252 
Storgravbotn 255 
strandflat, in Iceland 4, 384 
Streymoy 36, 37, 38, 41 
striae 29, 321, 323, 329, 330, 333, 342, 

Striatiira 1 67 

Strobdops 157, 160, 164, 167 
subaerial weathering 340 
subalpine 1 13, 1 14, 1 19, 120 
subalpine-subarctic birch woods 121 
subarctic 1 15, 1 18, 120, 121, 122 
Sub-Atlanticum 225, 301, 357, 364 
Sub-Boreal 43, 357, 364 
Sub-Boreal-Sub-Atlantic zones 226 
subglacial relief 323 

Cepaea in North America 169 

molluscs in Iceland 43 
submarine bogs 43, 360, 362 
submarine eruptions 29, 394 
submarine relief of North Atlantic 24, 25. 

submerged peat 43, 360, 362 
submerged shoreline 25, 106 
Sub-oceanic plants 301 
succession 226 
Succinea 163, 167, 168 

pfeijferi 1 69 

strigata 1 69 
Suduroy 31,35,37,41 
Suduroyarfjordur 35 
Suess elTect 356 
sugar maple 112 
sunken continents 23 
Sunndalen 269, 270, 276 
Sunnmore 321 
surtarbrandur 9, 15 
survival 81, 82, 84, 229 

of animals 142,224 

in the Arctic 210 

in Bear Island 210 

chance 224 

coastal 276, 280 



survival, in Greenland 291, 292 

in Iceland 303, 305, 307, 311, 357, 360, 

and loss of species 230 
on North Sea continent 229 
nunatak 276, 280 
of plants 102, 189, 201, 210, 224, 233, 

235, 236, 237, 238, 241, 261 
in Scandinavia 238, 241, 259, 261, 276, 

in Svaibard 102 
survival theory 246, 251, 252, 259, 261, 

279, 280, 391 
Svaibard, cf. Spitsbergen 
Svinafell 377ff 
Svinafellsfjall 377ff 
Svinafellsjokull 379 
Svinoy 36, 37, 40, 41 
Sweden 12, 112, 119, 120, 122, 228, 242, 

243, 245, 257, 268 
Switzerland 250 
sylvo-tundra 118 

table mountain 381 

"tabula rasa" idea 24 1 , 244, 248. 252, 304, 

Tachhms corticiniis 82 
Tafjord 255 

taiga 118, 119, 120, 123, 207 
Taimyr 216 
"Tak" 33 

Talstadhesten Mt. 276 
Taraxacum 178, 180, 248, 298, 299, 300, 
301, 303, 355 

Arctica group 271 

Ceratophora group 250 

aleuroides 250 

aleurophoriiiu 250 

arctiatm 99, 218 

brachyceras 2 1 8 

cornutum 250 

dovrense 248,271,274,279 

officinale 1 97 

Reichenbachii 248, 271 
Tasmania, earthworms 129 
Tatra Mts. 209 
Taxodiitin 1 5 

distichum 30, 35 
taxonomic differentiation 87, 91 
Taxus 117 

teleostei, freshwater 129 
Telliaosilk Fjord 342 
telocrats 367 

temperature 13,201,202,326,327 
tephra layers 357, 362 
tephrochronology 371 

Tertiary xii, 2, 5, 14, 15, 25, 41, 73 
80, 82, 156, 163, 209, 238, 340, 357, 
Tertiary-mesophytic flora 394 
Thalictrum 363 
Thjorsardalur 362 
Thingvallavatn 384 
Thorisjokull 236 
Thor's Hammer 256 
Thorskafjordur 318 
"thufur"' 357 
Thuja occidentalis 111, 115 
Thulean region 30 
thunderstorms, and dispersal 200 
Thymus Drucei 287 
Thyrea radiata 235 
Tjornes, Pliocene 5, 8, 16 
Tierra del Fuego, earthworms 131, 132 
Tilia 111,120 
americana 1 1 2 
cordala 112, 117, 119, 373 
platvphvUa 373 
till 323 

tillites 12, 16, 381 
timberline 251,292,293 
time-scale, biologic 231 
Timor, earthworms 166 
Tindar 2 

Tindholmur 35, 37 

lateralis 2 1 2 
lunata 2 1 2 
maxima 212 
Tirolian Alps 248 
Titicaca, earthworms 131 
palustris 1 99 
pusilla 2 1 5 
tolerance 141,201,202,203 
Tomentohypnetum involuti 211, 213 
Torfajokull 357, 362 
Torngat Glaciation 346, 347 
Torngat Mts. 340, 342, 343 
Torngat Peninsula 342 
transatlantic submarine ridges 24, 25, 394 
Transcaucasia, earthworms 160 
trap 29, 30 
Trapa 226 

natans 1 92 
Trechus obtusus 83 
trembling aspen 112 
tricentric species 313 
Trichia 1 67 
Trientalis europaea 3 1 9 
pratense 369, 372 
repens 1 92 




tnaritimum 1 92 

pal list re 212 
Triodopsis 1 67 
Trisetetum spicati 211, 213 
Trisetiim spicatum 89, 214 
Trollheimen 269. 270, 275, 276 
Trolitindene 255 
Troms 243, 244, 261 
Trondelag 251,256,259 
Trongisvagur 35, 37 
Tropical reef-belt, in Paleozoic 13 
Trimcatellina 161, 167 
Tsuga 1 5 

canadensis 111, 112 
tuff layers, in Faeroes 33 
tuff-agglomerate zone, in Faeroes 31, 32, 

tundra 1 18, 1 19, 207, 209, 236, 280, 290 
Tungufel! 7, 9 

ubiquitous plants 100,101,104,246 
Ulmus 15,111 

amen'cana 111, 112, 113, 115 

glabra 1 1 2 

scabra 225 
Ulva latissima 192 
Umbelliferae 363, 385 
Unibilicaria Havaasii 235, 238 
unglaciated areas 211, 236, 237, 321, 

unicentric plants 247, 261, 262, 263, 265 
uplift 2, 4 

Upper Devonian coals 13 
Upper Forrestian period 43 
Urals 113, 149, 176, 230 
Vrtica 368 
U.S.A. Ill, 112,223 
Utsjoki 230 

VacciniuDi 15, 385 

Myrtillus 66 

Vitis-idaea 319 
Vasro 246 
Vagar 31,35,37,43 
Valders readvance 347 
Valeriana 363 
Vallonia 163, 167 

cost at a 1 69 

pulchella 1 69 
Van Mijenfjord 101 
Varanger Peninsula 243, 247, 248 
varved clays 347 
Vasterbotten 329 
Vatnajokull 16,311,314 

Vatnsdalur 311 
Vatnsnes 312 
Vattern 370 
Vaygach 104, 218, 268 
vermiculite 279, 343 

fruticans 66 

Wormskjoldii 199, 291 
Vertigo 163, 167 

alpestris 1 69 

pygmaea 1 69 
Vesteralen Islands 321, 330 
Vestfirdir 306, 307, 3 1 0, 3 1 2, 3 1 6, 3 1 7, 3 1 8 
Vesuvius 371 
Viburnum 1 5 
vicariants 112, 183, 184 
Vik i Myrdal 355 

vikings, and Cepaea in North America 169 
Viola epipsila 314, 316 
Vitis 15 
Vilrea 167, 168 
Vitrina 154, 163, 167 
vitrite 35 
vivipary 89, 94 
Vopnafjordur 9 
Vostok I 322 

Wallace's subregions 127, 128 

water dispersal 194 

weathering 321, 343 

weed flora in America 45, 223, 224 

Wegener's hypothesis 16, 142, 144, 147, 

186, 395 
Wegeneriella 131, 137, 143 
West Africa, earthworms 138 
West Arctic element 87. 88, 96, 105, 176, 

178, 221, 229, 230, 237, 245, 270, 306, 

307, 311, 313, 316, 319, 392 
West Spitsbergen 100. 101 
White Sea 122 
Wijdefjord 101 

wind dispersal 197, 198, 200, 201, 304 
wind erosion 277 
Wisconsin Glaciation 326, 342 
woodland-steppe 111,113 
woodland-tundra 114, 118, 120, 121, 122 
Woodsia glabella 2 1 4 
world oceans, surface area 26 
Wurm Glaciation 82, 83, 84, 88, 168, 209, 

322, 326, 327, 330, 331 
Wyville-Thompson Ridge 18, 79 

Xanthoria lobulata 238 
X-ray analysis of weathering 
Xerothermic Period 228 



Yagansia 131 zonation, boreal to arctic 109fF 

Yarmouth Interglacial 389 Zonitoides 163, 167 

Yeniseysk area 265 arboreus 169 

Younger Dryas 225,226,327 zoochorous dispersal 177,181 

Yukon Territory 115,118 Zoogenetes 163,167 

zoogeographical classification 127 
Zostera 192, 195 
ZannicheUia 192, 193 marina 192, 195, 196 

Zizania 192 stenophylla 195, 196