|ARE THERE TOO MANY SPECIES AND SUBSPECIES IN THE PAPAVER RADICATUM
Heidi Solstad, Reidar Elven & Inger Nordal
Material and methods
authors (e.g. Knaben 1959a, 1959b, Elven in Lid & Lid 1994,
Elven & Elvebakk 1996, Solstad 1998, Nilsson in press) currently accept four
species of Papaver sect. Scapiflora Reichenb. (Sect. Maconella
Spach.) within the Flora Nordica area: P. dahlianum Nordh. (Fig. 1a) in Svalbard,
Bear Island and N Norway, P. laestadianum (Nordh.) Nordh. (Fig. 1b) as a local
endemic in N Scandinavia, P. lapponicum (Tolm.) Nordh. (Fig. 1c) with 1-2
subspecies in N Norway, and P. radicatum Rottb. (Fig. 1d) as a North Atlantic
endemic, with about 10 subspecies in S and N Scandinavia and 3-4 subspecies in Iceland,
one of which also reaching the Faeroes. These authors considered P. radicatum as a
geographically restricted species, not reaching westwards to Greenland and not found east
of the Scandinavian mountain range, P. dahlianum as a fairly widely distributed
high-arctic amphi-Atlantic species, and P. lapponicum as a nearly circumpolar
Looking at Russian treatises (e.g.
Tolmachev 1975, see also Petrovsky 1999), three species of arctic-alpine poppies are
reported more or less throughout the Russian and West Siberian Arctic eastwards to the
Beringian area: Papaver lapponicum (Tolm.) Nordh. in three major races, subsp. lapponicum
in Kola peninsula, subsp. jugoricum (Tolm.) Gudshnikov in NE Russia, Novaya Zemlya
and West Siberia, and subsp. orientale Tolm. further east in Siberia. P. polare
(Tolm.) Perfil. in Franz Joseph Land, Novaya Zemlya and N Siberia, and P. radicatum
Rottb. s. lat. here and there throughout Siberia (but not in the European part). In the
Beringian area, several more species are accepted on both sides of the Bering Strait.
On the American side east of the Beringian area, different names are
used in some cases, but probably representing the same taxa (Kiger
& Murray 1997): Papaver lapponicum subsp. occidentale (Lundstr.)
Knaben throughout most of arctic N America and Greenland, P. radicatum subsp. radicatum
widespread in the same area, P. radicatum subsp. polare Tolm. in N and NE
Canada and Greenland. The most recent surveys of parts of Greenland accept only a widely
defined P. radicatum in several 'types' (Feilberg 1984,
Bay 1992, Fredskild 1996).
These treatments are mutually incompatible in several aspects. (1) Papaver
radicatum Rottb. cannot at the same time be an endemic of Iceland and Scandinavia and
widely distributed throughout Greenland, Canada, Alaska and Russia. (2) The majority of
the Greenland-American materials cannot be both a more or less homogeneous species (P.
lapponicum in the sense of Knaben 1959a, 1959b; P.
radicatum in the sense of current Danish authors) and consist of two species (P.
lapponicum and P. radicatum in the sense of Kiger
& Murray 1997). (3) It is improbable that the high arctic NE American and NW
Russian/Siberian plants are P. polare (or P. radicatum subsp. polare),
whereas the high arctic Greenland plants are P. radicatum, and the high arctic
Svalbard plants are P. dahlianum, at least as long as the plants in NE Greenland,
Svalbard and Franz Joseph Land are morphologically inseparable. The focus here will be on
the variation in the N Atlantic area, including Greenland and Svalbard.
These arctic-alpine Papaver taxa have been a taxonomic headache
since the end of the 19th century when it was recognised that P. nudicaule L. (Linnaeus 1753) could not include all variation, and when P.
radicatum Rottb. (Rottbøll 1770) was rediscovered as a
valid name for at least parts of the arctic variation. There are several phases since then
in the development of the species concept in the group. Five important crossroad periods
and the main participants are:
(a) 1894-1923, when Murbeck (1894) and Lundström (1914, 1923) interpreted much of the N Atlantic and N
Russian variation as caused by hybridization between Papaver nudicaule and P.
radicatum. This interpretation was mainly initiated by observations of hybrids in
botanical gardens, but did not take fully into account that the proposed hybrids almost
always occurred in areas where their putative parents were absent (today).
(b) 1923-1932, when Tolmachev (1923, 1927)
and Nordhagen (1932) refuted the hybrid hypothesis and
described the variation they found as several species, especially Papaver dahlianum
(Nordhagen 1932), P. lapponicum (Tolmachev 1923 as subspecies of P. radicatum, Nordhagen 1932 as species) and P. polare (Tolmachev 1923 as subspecies of P. radicatum, Perfiljev 1936 as species), partly also with subspecies for
local and regional races. Those authors initiated a school accepting comparatively wide
species with regional and local variation described as subspecies, in the sense of
(c) 1938-1959, when Horn (1938), Holmen (1952),
Löve (1955) and Knaben (1958,
1959a, 1959b) discovered several ploidy levels in the group. This 'Nordic' reliance on
ploidy differences as criteria of species resulted in certain taxonomic changes and in
evolutionary interpretations. During this phase, Papaver lapponicum (octoploid,
2n=56) was considered to be clearly specific distinct from P. radicatum and P.
dahlianum (both decaploids, 2n=70), and P. laestadianum was recognized as a
species distinct from P. radicatum because of its octoploidy.
(d) 1945-present, when the ploidy differences found made the
typification of Papaver radicatum crucial for the interpretation of naming and for
understanding the geographical distribution of taxa. The typification debate concerned,
firstly, the question of the nomenclatorial type of P. radicatum, and secondly,
whether this 'type' - which happens to be a drawing (Fig. 2) - was made from material
originating in Greenland or in Iceland. If the type was from Iceland, it most probably is
a decaploid, and if from Greenland an octoploid. A typification based on the decaploid
would, if the difference between the ploidy levels is recognised at the species level,
reserve the name 'P. radicatum' for the geographically restricted North Atlantic
decaploid whereas the name 'P. lapponicum' would be the one to apply to the widely
distributed octoploid. Hultén (1945), followed by Löve (1955, 1962a, 1962b), argued for a Greenlandic
typification, but this was disputed by Knaben (1958) and Knaben & Hylander (1970). Nilsson (in press), based on the
arguments of Knaben (1958) and Knaben
& Hylander (1970), argues for and will propose a formalisation of an Icelandic
(i.e. decaploid) typification of P. radicatum, a view we fully agree with.
(e) 1974-present, a synoptical phase with attempts of world-wide (Rändel 1974) and regional surveys (Tolmachev
1975 - Russia; Kadereit 1993 - Europa; Kiger & Murray 1997 - N America; Nilsson in
press - the Nordic area). In this phase, information from several sources is included.
These attempts differ significantly from each other, partly because of different data sets
utilized, partly (still) because of different interpretations of the types of P.
radicatum and P. dahlianum. Kiger & Murray (1997)
separate P. lapponicum form P. radicatum on other criteria
than done by Nordic authors. Rändel (1974) and Kadereit (1993)
include P. dahlianum as a subspecies of P. radicatum, probably because of
its decaploid status, whereas they keep P. laestadianum as a separate species. This
approach is not followed by any Nordic or Russian authors. Kiger
& Murray (1997) doubt the identity between P. dahlianum and P. polare,
Tolmachev (1975) probably, too, as he keep P. polare
as species, whereas Nilsson (in press) and other Nordic authors regard P. polare as
a synonym of P. dahlianum.
At present, there is therefore no agreement about taxa or
delimitations, in spite of this being one of the most thoroughly investigated of all
groups of arctic plants. This may be a warning when going into other complicated arctic
polyploid complexes, i.e. Draba, Oxytropis or Poa. The first major
experimental investigations of Papaver were undertaken in the 1950ies (Knaben 1959a, 1959b), utilizing morphology, cytology and
crossing experiments, and with univariate statistics, before the introduction of more
sophisticated statistical analysis and especially before the use of genetic markers. Later
investigations have mainly had as aims to test the findings of Knaben, by multivariate
morphometry (Selin & Prentice 1988, Selin 1999) and by
chemotaxonomy (Folkestad et al. 1988). A study applying
genetic markers was therefore due, especially as Knaben (1959a,
1959b) proposed a hybrid origin of the decaploids P. radicatum and P.
dahlianum from the octoploid P. lapponicum and the presumed diploids P.
alpinum and P. nudicaule, respectively. Later studies have, however, shown that
P. nudicaule in the sense of Knaben probably is P. croceum Ledeb., whereas
the true P. nudicaule is a polyploid.
The aims of the present study are:
- To analyse the genetic variation within and among subspecies of P. radicatum in
the Nordic area by isoenzymes and RAPD,
- To report some results on reproduction and demography for P. radicatum,
- To analyse the genetic variation between P. radicatum and other Scapiflora
taxa - P. alpinum, P. dahlianum, P. laestadianum, P. lapponicum
and P. 'nudicaule' by isoenzymes,
- To compare genetic variation between P. dahlianum in Svalbard and in N Norway by
isoenzymes to test their identity, questioned by Kiger &
- To test the hypothesis of allopolyploid origins of the decaploids P. dahlianum
and P. radicatum from taxa at diploid and octoploid levels,
- To indicate taxonomic and phytogeographical implications of these results.
Other results from this study were presented by Nordal et al. (1997), Solstad et al. (1997), and Solstad (1998).
A more detailed version is under preparation (Solstad et al. in prep.).
Drawings of the four North
Atlantic species of Papaver sect. Scapiflora: a. P. dahlianum Nordh., b. P.
laestadianum (Nordh.) Nordh., c. P. lapponicum (Tolm.) Nordh. and d P.
radicatum Rottb. Drawings by Miranda Bødtker.
The plate representing the
type of Papaver radicatum, from Rottb?ll (1770).
set of investigations (isoenzymes, RAPD, demography, habitat choice) were undertaken on
materials from 20 populations of the six described South Norwegian subspecies of P.
radicatum. Offspring from a single plant of an Icelandic subspecies was also included
in investigations of both types of genetical markers. The reproductive system was
investigated in two South Norwegian subspecies. Isoenzyme variaton was also investigated
in one population each of P. alpinum subsp. sendtnerii (Austria), P.
dahlianum (Svalbard), P. laestadianum (North Norway), P. lapponicum
(North Norway) and P. 'nudicaule' (garden material, probably P. croceum
little variation in the genetic markers was found within and among the South Norwegian and
Icelandic populations/subspecies of Papaver radicatum, both as to isoenzymes (Fig.
3a) and RAPD-markers (Fig. 3b). The small variation found was within populations and
subspecies, not among them: One deviant isoenzyme multilocus phenotype was found in one
population of subsp. gjaerevollii (in the South Norwegian Trollheimen mountains)
and another deviant phenotype in some of the plants of the Icelandic subsp. steindorssonianum.
The plants in the two populations (of subspp. oeksendalense and ovatilobum)
investigated for reproductive system proved to be mainly autogamous with autodeposition,
but with a possibility of outcrossing. Segregation in the offspring from the single
Icelandic plant studied also indicates that some outcrossing takes place. These plants can
therefore be classified as mixed maters, but probably with self-pollination as the rule.
Demographically, there seems to be rapid turnover of generations in
dynamic populations. Populations with many or predominantly old plants occur, but they are
few. In habitat choice, the S Norwegian subspecies do not seem to differ; apparent
differences may be ascribed to presence or absence of suitable habitats as scree slopes,
riverbanks, road verges, alpine slopes or moraine ridges in the areas where they occur.
The two major habitat types are calcareous screes and river banks. Suitable river banks
are lacking in the areas where subspp. gjarevollii, oeksendalense and relictum
are found. Calcareous screes are rare in the area of subsp. intermedium whereas
riverbanks are common. In areas where all suitable habitats are abundant, as in the
Dovrefjell area with subsp. ovatilobum, P. radicatum occurs in all.
In a comparison between Papaver radicatum and other Scapiflora
species, comparatively little variation of isoenzymes was found. The first axis in a PCO
analysis (Fig. 3a) based on the isoenzyme multilocus phenotypes separated between P.
dahlianum and the remaining taxa. The second axis separated between P. nudicaule
and the majority of P. alpinum at one end and the group of P. radicatum, P.
laestadianum, P. lapponicum and some few P. alpinum plants at the other end.
The isoenzyme multilocus phenotype of Papaver laestadianum was identical to
the majority phenotype in P. radicatum. The deviating multilocus phenotype
in P. radicatum subsp. steindorssonianum was identical with one of
the four phenotypes found in P. alpinum, and closest to the deviating
multilocus phenotype in subsp. gjaerevollii.
Two multilocus phenotypes were found in P. dahlianum from
Svalbard (one population Fig. 3a). The isozyme phenotype found in one of the few
populations in N Norway, close to the type locality of the species, was identical to one
of the Svalbard phenotypes.
No direct evidence of additive patterns in isoenzymes was found between
P. alpinum (2x) / P. lapponicum (8x) and P. radicatum (10x), or
between P. 'nudicaule' (=P. croceum) (probably 2x) / P. lapponicum
(8x) and P. dahlianum (10x).
a. A PCO ordination of the results of analyses
of isoenzymes in Papaver radicatum, P. alpinum P. lapponicum and P. laestadianum. The
calculations are made on the basis of absence (0) or presence (1) of eight homomeric
enzyme bands: Gpi-2a, Gpi-2c, Gpi-2d, Tpi-b, Mdh-1a, Mdh-1d, Mdh-1e, and Pgd-2b. b.
An UPGMA-dendrogram based on the results RAPD analyses of S Norwegian and Icelandic
Papaver radicatum populations. The calculations are made on the basis of absence (0) and
presence (1) of six polymorphic RAPD markers. In both cases are Euclidean distance used as
dissimilarity measure. Abbreviations: Alp=P. alpinum ssp. sendtnerii, Dhl=P.
dahlianum, Lae= P. laestadianum, Lap=P. lapponicum, Nud=P. nudicaule,
Rad=P. radicatum including the 7 analysed subspecies: Rgj= ssp.
gjaerevollii, Rgr= ssp. groevudalense, Rin= ssp. intermedium, Roe=ssp.
oeksendalense, Rov=ssp. ovatilobum, Rre=ssp. relictum, Rst=ssp.
species and subspecies of the Scapiflora Papavers, including P. radicatum, P.
dahlianum, P. laestadianum and P. lapponicum, have been central in the
discussion of survival and speciation of the North European mountain flora during the
Weichselian or throughout all the Pleistocene glaciations see e.g. Nordhagen (1936, 1963), Knaben (1959a, 1959b), Gjærevoll (1963),
Nordal (1985, 1987), and Dahl (1989). Some geographically structured morphological
differences are found within both P. radicatum (see, e.g., Selin
1999) and the other species, but mainly in quantitative characters and mainly in
combinations of them.
The scant variation in genetical markers among populations of P.
radicatum and its subspecies does not support the recognition of subspecies.
It is interesting to note that seeds from one population of the diploid
P. alpinum subsp. sendtneri, are much more genetically (enzymatically)
variable than the entire body of material analysed for P.
radicatum-laestadianum-lapponicum complex. The reason may well be that P. alpinum
is outcrossing whereas the complex of P. radicatum-laestadianum-lapponicum is
Morphologic differences between North Norwegian and Svalbard plants of Papaver
dahlianum were the reason for Kiger & Murray (1997)
to doubt the validity of the name P. dahlianum for the arctic poppies. The
morphologic differences are slight and may easily have been caused by genetic drift in
postglacial time, in these small, mainly selfing, semi-isolated and fluctuating
populations on the mainland. The enzymatic identity between P. dahlianum from N
Norway and one of the multilocus phenotypes from Svalbard supports the view that these
plants should be placed in the same species. By implication, P. dahlianum should
then, due to priority, be the accepted name for this high arctic species both in the
Nordic area, Russia, Greenland, and Canada, replacing the younger name P. polare. Papaver
dahlianum is unambiguously typified (O) whereas Tolmachev
(1923) included sites also of P. lapponicum in his protologue (and
determinations) of P. polare. A possible type of P. polare designed by
Tolmachev (in LE) has been lost. And the name was lectotypified by Egorova (1998) on a material from Svalbard. This typification
further supports the reduction of P. polare to synonymy of P. dahlianum.
The relationship of Papaver dahlianum to other species within
the genus is, however, still unsolved. Knaben's hybrid origin hypothesis is not supported
by the presented results. Papaver dahlianum seems to represent a distinct species,
at least in the N Atlantic area, much less related to the others (P. laestadianum, P.
lapponicum, P. radicatum) than those are among themselves. It probably has an
evolutionary history different from the other North European taxa, and we should perhaps
look towards the Beringian area where, P. gorodkovii Tolm. & Petrovsky appears
to be a closer relative.
The C European diploid Papaver alpinum may have furnished
genomes to the North European polyploid complex, and an allopolyploid origin of P.
radicatum from P. alpinum and P. lapponicum can not be excluded by our
results. It is, however, not the simplest hypothesis. A simpler alternative might be
auto-polyploidization from P. alpinum stock (or a similar stock) to reach both the
octoploid (P. lapponicum and P. laestadianum) and the decaploid (P.
radicatum) levels. The very low genetic diversity in the group of P.
radicatum-laestadianum-lapponicum argues against an allopolyploid origin. It was
striking that most isozymes displayed fewer bands when compared to other arctic polyploids
of allopolyploid origin (see Solstad et al. in prep.).
Papaver laestadianum is so similar to P. radicatum
isozymatically that it may differ from P. radicatum only in having one
genome in single dose that P. radicatum has in double dose. Maintenance of P.
laestadianum as a separate species, even if octoploid, is therefore problematic. The
documented morphological differences are also small, not larger than the morphological
differences between the recognized subspecies of P. radicatum. One of the
very few Scandinavian mountain range endemics might therefore belong to the P. radicatum
species which is not an endemic.
The implications are clear: there are much smaller differences
genetically among many of the N Atlantic Scapiflora poppies than previously
assumed. The differences between P. radicatum and P. lapponicum might be so
small that they might be included within one single species.
The analyses of the N Atlantic Papaver Sect. Scapiflora
may also throw some light on the problems of the 'Species Concept in the Arctic'. It is a
classical case of different results reached in different 'political' areas, e.g. the
Nordic countries, Russia, America and Greenland. It has often been assumed that the
Russians, following the tradition of Komarov (1934 and
later), have applied a more narrow species concept than most Europeans and Americans. In
the present case, the major splitters seem to have been most active in the N Atlantic
area. If we consider both species and subspecies, about as many taxa of Papaver
Sect. Scapiflora have been described within the very small variation found around
the N Atlantic as within the much more impressive variation found in the Beringian area.
The investigations should now be extended to includ the taxa recognized
as P. lapponicum and P. radicatum in Greenland and North America, the
Russian taxa described within P. lapponicum, and to possible progenitors of the
high-ploid arctic complex(es), if extant in the Beringian, Cordilleran and C and E Asian
had inspiring discussions with Dr. D.F. Murray, Dr. O. Nilsson and Dr. V.V. Petrovsky on
taxonomical questions. We thank for all the comments, and a special thank to Dr. D.F.
Murray for refereeing the manuscript.
1992. A phytogeographical study of the vascular plants of northern Greenland - north of 74
º northern latitude. Meddel. Gronland, Bioscience 36.
Dahl, E. 1989.
Nunatakk-teorien II. - Endemismeproblemet. Blyttia 47: 163-172.
Egorova T.V. 1998. Genus Papaver
L. (Papaveraceae) in the flora
Eastern Europe. [Rod Papaver L. (Papaveraceae) vo flore Vostochnoy Evropy]. Nov.
sist. vysshykh rast. 31: 90-118 (in Russian).
Elven, R. & Elvebakk,
A. 1996. Part 1. Vascular plants. In: Elvebakk, A. & Prestrud, P. (Eds) A
catalogue of Svalbard plants, fungi, algae, and cyanobacteria. - Norsk Polarinst. Skr.
Feilberg, J. 1984. A
phytogeographical study of South Greenland. Meddel. Grønland, Bioscience 15.
Folkestad, K., Høiland,
K., Paulsen, B.C. & Malterud, K.E. 1988. Alkaloid chemotaxonomy of Nordic Papaver
sect. Scapiflora (Papaveraceae). Nord. J. Bot. 8: 139-146.
Fredskild, B. 1996. A
phytogeographical study of the vascular plants of West Greenland (62 º20' - 74º 00').
Meddel. Grønland, Bioscience 45.
Gjærevoll, O. 1963.
Survival of plants on nunataks in Norway during the Pleistocene glaciation. In:
Löve, Á. & Löve, D. (Eds) North Atlantic biota and their history. Pergamon Press,
Oxford. pp. 261-283
Holmen, K. 1952.
Cytological studies in the flora of Peary Land, North Greenland. - Medd. Grønl. 128:
Horn, K. 1938. Chromosome
numbers in Scandinavian Papaver species. - Avh. Norske Vidensk. Akad. Oslo., I.
Mat.-Naturv. Kl. 1938, 5.
Hultén, E. 1945. Flora of
Alaska and Yukon. V. - Lunds Univ. Arsskr., N.F., Avd. 2, 41, 1: 799.978.
Kadereit, J.W. 1993. Papaver
L. In: Tutin, T.G. et al. (Eds) Flora Europaea 1, 2. ed. Cambridge Univ. Press,
Cambridge. pp. 297-301
Kiger, R.W. & Murray,
D.F. 1997. Papaver L., In: Morin, N.R. et al.(Eds) Flora of North America
north of Mexico. 3. Magnoliophyta: Magnoliidae and Hamamelidae. Oxford Univ. Press, New
York. pp. 323-333
Knaben, G. 1958. Papaver-studier,
med et forsvar for P. radicatum Rottb. som en islandsk-skandinavisk art.
Blyttia 16: 61-79.
Knaben, G. 1959a. On the
evolution of the Radicatum-group of the Scapiflora papavers as studied in 70
and 56 chromosome species. A. Cytotaxonomical aspects. Opera Bot. 2, 3: 1-74.
Knaben, G. 1959b. On the
evolution of the Radicatum-group of the Scapiflora papavers as studied in 70
and 56 chromosome species. B. Experimental studies. Opera Bot. 3, 3: 1-96.
Knaben, G. & Hylander,
N. 1970. On the typification of Papaver radicatum Rottb. and its nomenclatural
consequences. - Bot. Not. 123: 338-345.
Komarov, V.L. (Ed)
1934-1960. Flora of USSR [Flora SSSR]. - Izdatelstvo AN SSSR, Leningrad (in Russian).
Lid, J. & Lid, D.T.
1994. Norsk Flora, 6. ed. by R. Elven. Det Norske Samlaget, Oslo.
Linnaeus, C. 1753. Species
plantarum. - Holmiae.
Löve, Á. 1955.
Cytotaxonomical remarks on the Icelandic Papaver. - Nytt Mag. Bot. 4: 5-18.
Löve, Á. 1962a.
Typification of Papaver radicatum - a nomenclatural detective story. - Bot. Not.
Löve, Á. 1962b.
Nomenclature of North Atlantic papavers. - Taxon 11: 132-138.
Lundström, E. 1914.
Beobachtungen und Studien bei den in den Jahren 1912-13 ausgeführten Pflanzenbestimmungen
im Botanischen Garten Bergielund (Hortus Botanicus Bergianus). - Acta Hort. Berg. 5, 3:
Lundström, E. 1923. Über
Papaver nudicaule L. und P. radicatum Rottb. in Fennoskandia und Arktis
sowie uber einige mit P. nudicaule verwandte Arten. - Acta Hort. Berg. 7: 403-430.
Murbeck, S. 1894. Neue
oder wenig bekannte Hybriden in dem Botanischen Garten Bergielund (Hortus Botanicus
Bergianus). - Acta Hort. Berg. 2, 5: 1-24.
Nilsson, Ö. (in press). Papaver
sect. Scapiflora, In: Jonsell, B. (Ed) Flora Nordica 2.
Nordal, I. 1985.
Overvintringsteori og evolusjonshastighet. Blyttia 43: 33-41.
Nordal, I. 1987. Tabula
rasa after all? Botanical evidence for ice-free refugia in Scandinavia reviewed. J.
Biogeogr. 14: 377-388.
Nordal, I., Hestmark, G.
& Solstad, H. 1997. Reproductive biology and demography of Papaver radicatum -
A key species in Nordic plant geography. Opera Bot. 132: 77-87.
Nordhagen, R. 1932.
Studien über die skandinavischen Rassen des Papaver radicatum Rottb. sowie einigie
mit denselben verwechselte neue Arten. Bergens Mus. Årb. 1931, Naturv. R. 2.
Nordhagen, R. 1936.
Skandinavias fjellflora og dens relasjoner til den siste istid. - Nord. Naturf.-mötet
Helsingfors 1936: 93-124.
Nordhagen, R. 1963. Recent
discoveries in the South Norwegian flora and their significance for the understanding of
the history of the Scandinavian mountain flora during and after the last glaciation.
In: Löve, Á. & Löve, D. (Eds) North Atlantic biota and their history. -
Pergamon Press, Oxford. pp. 242-260.
Perfiljev, I.A. 1934-1936. Flora of the Northern Land [Flora
Severnogo Kraya]. - Sevkraigiz,Arkhangelsk,1: 159 pp; 2-3: 393 pp. (in Russian).
Petrovsky, V.V. 1999. The
problem of the species width in poppies of Russian Arctic. Det Norske Videnskaps -
Akademi. I. Mat. Naturv. Klasse, Skrifter, Ny serie 38: 243-259.
Rändel, U. 1974.
Beiträge zur Kenntnis der Sippenstruktur der Gattung Papaver L. Sectio Scapiflora
Reichenb. (Papaveraceae). Feddes Repert. 84: 655-732.
Rottbøll, C.F. 1770.
Afhandling om en Deel enten ganske nye eller vel forhen bekiendte, men dog for os rare
Planter, som i Island og Grønland ere fundne, tilligemed en kort Indledning om
Urtelærens Tilstand i Dannemark. Skr. Kiöbenhavnske Selsk. L?rd. Elsk. 10:
Selin, E. 1999.
Morphometric differentiation in Papaver radicatum (Papaveraceae). Geographic
pattern and significance for refugial survival theory. Acta Univ. Upsal., Compr.
Summ. Uppsala Diss. Fac. Sci. Techn. 444.
Selin, E. & Prentice,
H.C. 1988. Morphometric analysis of disjunct Papaver radicatum (Papaveraceae)
populations in southern Norway. Pl. Syst. Evol. 159: 237-247.
Solstad, H. 1998.
Genetical and ecological variation in S Norwegian Papaver radicatum Rottb.
Comparisons with related species, and implications for taxonomy and phytogeography.
Cand. scient. thesis, Univ. of Oslo.
Solstad, H., Elven, R.
& Stabbetorp, O. 1997. Habitatvalg og demografi hos de sørnorske fjellvalmuene (Papaver
radicatum). NTNU Vitensk. Mus. Rapp. Bot. Ser. 1997-1: 67-87.
Tolmachev, A. 1923. On the
european races of Papaver radicatum Rottb. [O evropeyskikh rasakh Papaver
radicatum Rottb.] Not. Syst. Herb. Horti Bot. Petrop. 4: 81-90.
Tolmachev, A. 1927. Über
die Formen von Papaver radicatum Rottb. und ihre Verbreitung in Skandinavien.
Sv. Bot. Tidskr. 21: 73-83.
Tolmatchev, A.I. 1975. Genus Papaver L. [ Rod Papaver
L.]. In: Tolmatchev, A.I. (Ed) Arctic Flora USSR. VII. Papaveraceae-Cruciferae.
Nauka, Moskva-Leningrad. pp. 7-32 (in Russian).