PAFlogo_small1.jpg (7865 bytes) Home ] PAF News ] Checklist ] PAF Feedback ] PAF Search Page ]

PanArctic Flora Project ]

Back ] Up ]


During the Panarctic Flora Year at the Centre of Advanced Studies (1998-99), eight workshops combining experts from different countries and taxonomic ‘traditions’ were held in Oslo and St. Petersburg. The plant groups discussed were (1) the Cerastium alpinum/arcticum complex, (2) Draba, (3) Dryas, (4) the Scapiflora/Meconella group of Papaver, (5) the grasses (Poaceae), (6) the Potentilla nivea group, (7) Salix, and (8) Saxifraga.

The Cerastium workshop

The Draba workshop

The Dryas workshop

The Papaver workshop

The grass (Poaceae) workshop

The Potentilla workshop

The Salix workshop

The Saxifraga workshop

Some initial discussions concerning the genus Cerastium were undertaken during a meeting in St. Petersburg in December 1998 by R. Elven and Dr. V. Petrovsky, and a ‘mini’-workshop on the Cerastium alpinum-arcticum complex took place in Oslo 9.-10. March 1999, involving Dr. S. Aiken (Ottawa), Dr. A.K. Brysting (Oslo) and R. Elven (Oslo). The occasion was to discuss the new views as to division of the complex in the North Atlantic area (especially by A. Brysting and A. Hagen, Oslo) and how this perspective might influence the treatment of the complex in arctic Canada. We had a large number of specimens available from The Canadian Museum of Nature (CAN) available. The workshop was followed up by a revision of the specimens from the Canadian Arctic Archipelago by Brysting & Elven, applying to these the criteria previously used for the North Atlantic variation.

In the Arctic, four major species have been accepted: the nearly circumpolar and morphologically unproblematic C. regelii s. l. (including C. jenisejense), the broadly amphi-Beringian C. beeringianum, and the broadly amphi-Atlantic C. alpinum and C. arcticum s. l. The studies in recent years of the North Atlantic area confirm that the probably tetraploid C. alpinum and the probably hexaploid C. arcticum s. l. are closely related and C. alpinum may be responsible for a large part of the genes in C. arcticum s. l. (see Brysting & Hagen in the this volume). Cerastium arcticum s. l. is, however, morphologically and genetically heterogeneous, with a comparatively southern entity in Fennoscandia, the British Isles, Iceland, and possibly Southwest Greenland, and a northern entity in arctic European Russia, Svalbard, most of Greenland, and northeastern North America. They should be recognized as two separate species: the name C. nigrescens applied to the southern entity (typified from Shetland), and C. arcticum to the northern taxon (typified from West Greenland). There still are some unsolved questions as to the SW Greenland materials.

There are two ‘competing’ circumpolar treatments of the taxa. Hultén (1956) emphasized hybridization and introgression, whereas Böcher (1977) saw them as an old, stabilized polyploid complex. Currently, our opinion is closest to Bocher’s. Due to the differing treatments and assumed lack of diagnotic characters, many recent authors treating the Greenlandic and American plants have lumped all into an extremely widely defined C. alpinum. The questions concerning the Canadian materials were, therefore:

  1. The applicability in America of the characters found useful in separating entities in the European part of the North Atlantic area;
  2. The extent eastwards in America of C. beeringianum and the distinctness of this entity as compared with the others;
  3. The presence and extent westwards of C. alpinum in America; (d) The entities of the North Atlantic C. arcticum s. l. to which the American material might belong; and
  4. The usefulness for the proposed subspecific divisions of C. arcticum (Hultén, Böcher) and C. beeringianum (Hultén).

The results were comparatively clear on all these points. The diagnostic characters applied in Europe were equally applicable to American material of C. alpinum and C. arcticum s. l., and a set of characters was found to separate C. beeringianum equally well. Cerastium beeringianum was found distributed eastwards to the Hudson Bay area, Ellesmere Island, and probably also NW Greenland, but it was mainly replaced by C. arcticum in the east. Cerastium alpinum was identified clearly in the materials from the easternmost parts of Canada, especially from Baffin Island, but a comparison with the European subspecific entities remains to be done. The American C. arcticum is of the northern type, corresponding to C. arcticum s. str., conforming very well with the Greenland and Svalbard material. Cerastium arcticum reaches west to the central parts of the Arctic Archipelago, where it becomes completely replaced by C. beeringianum. Although we found a great deal of variation in C. arcticum, there was little support for a further split of C. arcticum (or any of the other species) into clearly definable geographical races.

button-arrow-back.gif (860 bytes) Contents

The Draba workshop

By Reidar Elven, Vladislav V. Petrovsky & Christian Brochmann

The Draba workshop (15.-19. March 1999 in Oslo) was mainly based on the draft checklist of Dr. V.V. Petrovsky (St. Petersburg). It was attended by B. Andersen (Oslo), Dr. C. Brochmann (Oslo), R. Elven (Oslo), H.H. Grundt (Oslo), Dr. V.V. Petrovsky (St. Petersburg) and A.-C. Scheen (Oslo). Some pre-workshop work was done by Grundt, Petrovsky and Elven in the St. Petersburg herbarium in November and December 1998. In addition to the draft checklist, abundant herbarium material from the North Atlantic area and a fairly large representation from Siberia, the Russian Far East, and from Alaska was available. Live plants from Alaska were also studied in the Phytotrone, Biological Institute, University of Oslo. All major arctic groups and most species were represented in the material studied.

The genus Draba is one of the larger in the Arctic, with between 50 and 60 accepted species (and some subspecies). The subgeneric structure applied today to the Arctic taxa is mainly that of Tolmachev (1939, 1975) from the Flora USSR and Arctic Flora of USSR, in which the arctic species are distributed among about 14 series. As this system applies only to the Russian species, some additions are needed to encompass a few Cordilleran species that do not cross the Bering Strait. Some of the series are obviously so closely related that it is difficult to find biologically meaningful distinguishing criteria, especially between Lacteae and Pilosae and between Cinereae and Hirtae (and partly also Rupestres).

When these series are merged, the majority of the arctic species falls into six groups: Alpinae with 8 or more species, Cinereae/Hirtae/Rupestres with somewhere between 10 and 15 species, Incanae with 2 to 3 species, Lacteae/Pilosae with ca. 10 species, 'Micropetalae' (Tolmachev's Oblongatae must change its name) with 2 to 3 species, and Nivales with 4 species. In these, except for Incanae, there seem to have been speciation in or close to the Arctic as at least one species in each group is nearly or entirely restricted to the Arctic. The remaining series are represented in the Arctic by 1 to 2 species each, obviously of temperate-alpine or temperate-boreal origin.

Several species are known only from very restricted areas and from few collections, and there is probably an over-description of such local species. Examples are D. allenii from northeasternmost America, D. boecheri from Greenland (one site), and D. sambukii from Taimyr. A very careful study of the specimens, and in some cases also investigation of genetic markers, are needed before such species can be confirmed or rejected.

A second problem is the effect of taxonomic traditions; taxa are often not treated in the same way across national boundaries. Obvious cases are found in: (1) ‘Micropetalae’ where Russian, Norwegian and some Danish (Greenland) authors recognize 2 to 3 species at different ploidy levels, whereas American authors recognize only one in their equally heterogenous material; (2) Alpinae where the name D. alpina often is applied in a wide sense for entities from diploid to 14-ploid levels in America and Russia but in a narrow sense for a decaploid in the North Atlantic area; and (3) Cinereae, where Russian, Norwegian, and Danish (Greenland) authors distinguish more or less clearly between the decaploid D. arctica and the hexaploid D. cinerea whereas North American authors unite them. Several additional examples might have been cited.

There is, however, no essential difference in how scientists in the different botanical traditions recognize and circumscribe species in Draba. We agree in most aspects, but the ‘intensity’ of investigation in various groups has differed in different areas. The main task now lies in comparing materials to reach cross-boundary consensus. The usefulness of such an approach was demonstrated during the workshop(s) recognizing two new entities for Europe - D. oblongata R. Br. for Svalbard and D. parvisiliquosa Tolm. for North Fennoscandia - two for Alaska - D. kamtschatica (Ledeb.) N. Busch and D. pilosa DC. - and finding an unnamed parallel to the North Atlantic D. oxycarpa Sommerf. in the Beringian area.

button-arrow-back.gif (860 bytes) Contents

The Dryas workshop

By Inger Nordal, Boris A. Yurtsev & Marianne Phillip

The Dryas workshop (15.-18. March 1999 in Oslo) was attended by Dr. B.A. Yurtsev, Dr. M. Phillip and Dr. I. Nordal. Discussions were based on a draft checklist by Yurtsev, the key by Yurtsev for Arctic Flora of USSR (1984) and his survey of the genus (Yurtsev 1997), the genetic and morphologic studies of Dryas from Greenland and Svalbard by Phillip and collaborators (see Siegismund & Phillip in this volume), and on herbarium specimens, mainly from the North Atlantic areas, but also some American and Beringian ones. The Russian key was translated into English, and a slightly simplified translation (excluding some hybrids included in the original key) is presented at the end of this summary to facilitate further work with the genus.

Yurtsev (1997) recognizes 15 species globally, of these nine reach the Arctic: D. grandis Juz. and D. drummondii Richardson (in sect. Nothodryas) and D. ajanensis Juz., D. chamissonis Spreng. ex Jurtz., D. crenulata Juz., D. incisa Juz., D. integrifolia Vahl, D. octopetala L., and D. punctata Juz. (all in Sect. Dryas). Two of these - D. ajanensis and D. punctata - are represented by two subspecies each in the Arctic, and D. octopetala is represented by its subsp. subincisa Jurtz. The genus has its center of diversity in northeastern Asia and (partly) northwestern North America, whereas few entities are found in the North Atlantic areas, i.e. northeastern North America, Greenland and northern Europe.

The taxa recognized by Yurtsev are mainly defined and delimited by characters from indumentum, glands (presence/absence and types), and leaf shape, incision and structure. In Siberia, all the presumed primitive features are present in some lineages of Sect. Dryas, that is, all the types of hairs and most types of glands. These are also present together with the assumed primitive tendency of compound leaves in Sect. Nothodryas, purportedly a sister group of Sect. Dryas. The only novelty later evolved in Sect. Dryas, according to Yurtsev, is the sessile, convex lens-shaped glands characteristic of D. punctata (including subsp. alaskensis) and in taxa outside the Arctic in subsp. hookeriana from The Rocky Mountains and the related D. viscosa.

In general, the assumptions of a northeastern Asian center, of a subsequent spread from there into the Arctic and other areas, and of morphologic overlap and hybridization, are not disputed. The discussions concerned the interpretation of the morphological variation and its taxonomic significance, especially the use of single characters and of assumption of hybridogenous origins of taxa with intermediate features.

As a test case, the variation in the North Atlantic areas was discussed. Nordal and Phillip recognize two species - D. integrifolia in northeasternmost America and Greenland and D. octopetala (s. l.) in northern Scandinavia, Iceland, Svalbard and East Greenland - and intermediates between them in parts of Greenland. In the same area, Yurtsev recognize four species - D. chamissonis (Notheast America, Greenland), D. integrifolia (NE America, Greenland), D. octopetala subsp. subincisa (East Greenland, Svalbard, northernmost mainland Europe), and D. punctata subsp. punctata (East Greenland, North Europe). These discrepancies have to be resolved before a consensus checklist is attained. No such consensus was reached during the workshop.

Yurtsev recognizes three taxa in Svalbard and in East Greenland: D. octopetala subsp. subincisa, D. punctata subsp. punctata and their hybrid, D. x vagans. The paramount diagnostic character of D. punctata, the lens-shaped glands, was found by Siegismund & Phillip (this volume) to be more or less randomly distributed among and within populations in Svalbard, and the presence of glands was found not to be correlated with leaf morphology or isoenzyme pattern. In a North Atlantic perspective, thus, the gland character appears to Nordic botanists as an intraspecific polymorphism. Greenland variation is a similar case.

According to Yurtsev, the observed patterns in Svalbard and Greenland represent marginal introgression zones for more or less interfertile taxa originating in other areas, D. punctata in Siberia, D. integrifolia in America, and D. octopetala s. str. probably in North Europe. The question is how to handle the situation taxonomically. If the arctic plants of Sect. Dryas constitute only a partially structured panarctic gene-pool created by extensive hybridization, this might be handled in two ways: (1) By attempting to identify both the original species and the hybrid combinations and to characterize and name them accordingly (Yurtsev's proposal); and (2) By including assumedly interfertile and geographically overlapping entities within more widely defined species (Nordal's and Phillip's proposal).

There is an obvious need for modern analysis of both arctic and non-arctic Dryas, involving both more extensive morphological analysis, investigation of molecular markers, and further application of ecogeographic methods. The Dryas logo of the PAF project was partly chosen because this well-known and important arctic genus still remains a taxonomic riddle.

button-arrow-back.gif (860 bytes) Contents

Key to the genus Dryas in the Russian Arctic [hybrids removed]

From Yurtsev (1984, Arctic Flora of USSR 9 (1): 256-260), with emendations and with hybrids removed.

1. Leaves oblong-oblanceolate, widest part above the middle, large with cuneate base and 1-4 separate teeth on upper petiole, blade flat, thick, subcoriaceous, with 6-13 equally short oblique triangular directed upwards teeth on each side, upper surface dark green glossy, with submerged veins, glabrous, the midvein simple hairs, sometimes (along with petiole and stipules) with sparse minute, colourless subsessile, non-viscid glands and without sessile amber, lens-shaped, viscid glands; not rarely leaf lower surface with snow-white tomentum, midvein glabrescent, except for apex, petiole firm, thick, glabrescent, much shorter than the blade with dark brownish, thinly branched hairs sometimes with stipitate glands. Flowers during anthesis campanulate, suberect to slightly nodding with ovate-lanceolate sepals, sepals and hypanthium densely covered with long dark purple glands, petals narrow obovate, white. Styles in fruit 3-5 (-6) cm long, densely covered with long hairs, strongly spirally twisted, violet, shining (nitide) when immature D. grandis (sect. Nothodryas)

– Leaves of not as above, widest at the middle or below, with cordate, sagitate, rotundate (rounded) or truncate, rarely truncate-cuneate, base without teeth on the petiole, petioles not rarely nearly as long as the blade. Flowers during anthesis erect and open. Styles in fruit usually less than 3 cm long, weakly twisted, weakly glossy, rose to greenish when immature (sect. Dryas) 2

2. Leaves with entire margin or with few teeth at the base of some leaves or at least with uppermost part of blade without teeth, if blades crenulate along full length, then most teeth are very minute, shorter than they are wide, rotundate and involute; blade coriaceous, thick with slightly submerged veins (in D. crenulata); leaves lack branched hairs and stipitate glands (appearance of glands or branched hairs on some leaves is the result of hybridization with species of other sections) (subsect. Tenella) 3

– Leaves more or less dentate along full length, usually with blade rugulose, owing to deeply submerged veins 5

3. All leaves with entire margin, rarely with teeth in lower part or at the base of blade, with wide sagitate base, upper surface not rugose, glabrous or glabrescent except for median vein, or dense pubsecent (var. canescens), lower surface white, tomentose or rarely glabrate D. integrifolia

– Leaf blade (at least in many leaves) dentate or crenulate to apex or nearly so 4

4. Leaves narrow lanceolate-oblongate or almost linear, strongly involute, margins not rarely touching, coriaceous, thick, the upper surface glossy, not dark or bright green glabrous or glabrescent without strongly submerged lateral veins, gradually tapering towards the base, base rotundate (rounded), truncate, seldom subcordate, crenulate almost along full length, lower part rarely with oblique acute teeth, lower surface white-tomentose, with glabrous petioles and ciliate stipules. Scape slender, long, somewhat twisted, not rarely with a subulate bracteole D. crenulata

– Leaves normally wider, often ovate with broad sagitate, cordate, rarely truncate base, thinner, at least some or almost all dentate to apex or nearly so, teeth in lower half not rarely large, oblique triangular, the upper surface glabrous or glabrescent, except for midvein, or densely pubsescent (in var. incana), the lower surface white-tomentose. Scape thick or slender, not twisted, without bracteole D. chamissonis

5. Leaves lacking dotted sessile viscid or minute colorless, clavate glands on upper surface, lower surface lacking branched and glanduliferous, hairs on midvein and petiole, with slightly involute margins, dentate along full length with elongate, oblique triangular teeth, the sinus between teeth, in lower half of leaf, sometimes reaches midvein, not rarely gradually tapering to the base, usually coriaceous, glossy rugulose, in living state often pale green, upper surface except for veins glabrous or with tomentum (var. cana), lower surface with snowy-white tomentum or vein and petiole sometimes softly spreadingly white pilose. Sepals narrowly lanceolate or linear lanceolate, densely covered with dark purple glanduliferous hairs, without a mixture of branched hairs .D. incisa s.str. (subsect. Incisa)

– Leaves with branched hairs and/or sessile and stipitate viscid or minute colourless glands on midvein of lower surface and on petiole, upper surface without glands or with dotted glands or minute colourless subsessile glands 6

6. Leaves covered with sessile, convex lens shaped, amber yellow, yellow green or reddish, dotted viscid glands on upper surface, resin on older leaves due to hardened white secretions, lower surface with sessile or stipitate yellow, reddish brown or almost purple, viscid glands on veins and petioles, often with a mixture of brown branched and sometimes simple white hairs; blade elongate elliptical to elongate-oblong or elliptical, widest at middle with wide, obtuse or sometimes acute, long or short teeth, lateral veins on lower surface not concealed by tomentum. Stipules with viscid glands (D. punctata, subsect. Punctatae) 7

– Leaves lacking dotted viscid glands on upper surface, also lack sessile and stipitate large viscid glands on veins and petioles of lower surface; midvein of lower surface and petiole with branched, brown hairs, with a mixture of simple white hairs; upper surface, veins on lower surface, and petiole and stipules some times with minute, colourless subsessile glands not producing secrections that harden to form white spots. Sepals, hypanthium, and upper part of scape often with mixture of long purple glanduliferous and brown branched hairs (subsect. Dryas) 8

7. Leaves, covered on lower surface of median and lateral veins with a mixture of brown branched hairs and viscid sessile and stipitate, yellowish, brown or purple glands; upper surface with glabrous or glabrescent submerged veins, with abundant viscid, dotted glands, more rarely with dense white pubescence concealing sparse dotted glands (var. cinerea Jurtz.; with veins on lower surface which are some time covered with a white tomentum, in addition to abundant branched hairs, and inconspicuous glands) D. punctata subsp. punctata

– Median and lateral veins of lower leaf surface and petioles lacking branched brown hairs, but densely covered with purple brownish or yellowish stipitate glands, often with an abundant mixture of sessile, viscid verruciform glands D. punctata subsp. alaskensis

8. Leaves usually ovate-oblongate with cordate or nearly saggitate base, the widest part below the middle, sub-coriaceous, with involut margins, often glossy with submerged veins, rugulose, margins obliquely acute dentate, sinus between lower teeth can almost reach the midvein, upper surface glabrescent, pubescent mostly only on veins, or with densely adpressed pubescence (var. argentea), white tomentose on lower surface, between and on the lateral veins, sometimes a part of leaves lacking branched hairs. Sepals usually acute-lanceolate to linear-lanceolate, glabrous on the upper surface. Scapes with sparse dark stipulate glands and/or branched hairs in upper part. Petioles not rarely without branched hairs D. octopetala subsp. subincisa

– Leaves elliptic, more rarely short- or elongate-elliptical, or ovate-elliptical to oblongate-oblanceolate, the widest part near or somewhat above the middle, with cordate, rotundate (rounded) or truncate base, with short teeth, upper surface dull dark green, leaf thin, upper surface with densely appressed pubescence sometime only between the veins, more rarely glabrescent, pubescent on the deeply submerged veins, lower surface often with glabrate median and lateral veins along with petiole densely covered with more or less short branched hairs, usually also with numerous minute colourless, subsessile, non-viscid glands abundant on upper surface of leaf, petiole, and especially stipules. Sepals ovate-lanceolate to oblongate-ovate, acute or obtuse, some times pubescent on the upper surface (D. ajanensis) 9

9. Leaf upper surface canescent from appressed hairs, sometimes with glabrate veins or green and glabrous between submerged veins (var. viridis), leaves short-elliptical to elliptical with somewhat remote, often obtuse, or double unequal teeth (can be similar in general appearance to leaves of D. punctata, but shorter and less incised); lateral veins at divering up to right angles from the midvein, as a rule not concealed by tomentum; minute colourless, subsessile, non viscid, glands often very abundant, especially on petiole and stipules. A plant of eastern part of Chukotka Peninsula, Alaska, and Yukon D. ajanensis subsp. beringensis

– Leaf upper surface always canescent from dense adpressed hairs not rarely with glabrate veins, oblong-oblanceolate or oblong-obovate with short oblique acute, equal teeth; lateral veins almost parallel, directed oblique upwards, forming a smaller angle with the midvein; minute colourless glands on petiole and stipules inconspicuous (the plants of northern coast of Sea of Okhotsk) D. ajanensis subsp. ochotensis

button-arrow-back.gif (860 bytes) Contents

The Papaver workshop

By Inger Nordal, David F. Murray, Orjan Nilsson, Vladislav V. Petrovsky & Reidar Elven

The Papaver workshop (October 1998 in Oslo) followed immediately on the Symposium, with many of the same participants: Dr. D.F. Murray (Fairbanks), Dr. V. Petrovsky (St. Petersburg), Dr. Ö. Nilsson (Uppsala), and Dr. I. Nordal, H. Solstad and R. Elven (Oslo). The workshop based its discussions and other work on the discrepancies in treatments of the Scapiflora Section of Papaver by Petrovsky and Solstad et al. (in this volume), Nilsson in Jonsell (in press, Flora Nordica Vol. 2), and Kiger & Murray (1997, in Flora of North America Vol. 3). Materials were available from the North Atlantic area and some from arctic Canada, Alaska and Siberia. The workshop, in an extended sense, was continued in some analysis of genetic markers and a morphometric attempt in autumn 1998 and study of type specimens and other materials in the Komarov Botanical Institute in February 1999.

Two problematic themes were discussed during the workshop and subsequent works: The lack of common species characterization and naming across the North Atlantic (see Solstad et al. in this volume), and the level of species variation in the Beringian area (see Petrovsky in this volume).

Superficial studies of herbarium specimens alone did not clarify the application of names in the North Atlantic where the names P. radicatum, P. lapponicum, P. dahlianum and P. polare are used with partly different meanings in different areas. We made a morphometric program for registering variation in most available characters and measured a selection of specimens from most major species, including several Beringian ones. A numerical analysis failed to find clear limits, and also failed to classify together specimens belonging to the same of to closely related species. The reasons for this are probably that the large number of plastic characters included obscured the pattern formed by presumed taxonomically important characters, and that the total variation among the species is too large for a 'collective' analysis. The analysis should rather focus on smaller groups of clearly related species within the genus.

Some investigations of molecular markers (isoenzymes) were undertaken on a selection of North Atlantic and Beringian taxa currently in culture in Oslo. Some very initial results are: (1) that also morphologically very different Beringian species (P. macounii in two subspecies, P. walpolei) showed much general molecular similarity with the North Atlantic plants; (2) that the multilocus isoenzyme phenotype of mainland Norwegian P. dahlianum (from where it is described and typified) was identical with one of the phenotypes found in Svalbard (from where P. polare is typified), more or less proving identity between P. dahlianum and P. polare at species level; and (3) that one of the markers differentiating P. dahlianum/polare from P. radicatum/lapponicum was present in one of the Beringian species (P. walpolei). Investigations of molecular markers are obviously a powerful tool in solving the problems in this complicated group. A fairly large material of live plants and seeds was collected in the summer of 1999 through Greenland and arctic America, as a start of this work.

The investigation of types and other specimens in St. Petersburg convinced two of us (Elven, Solstad) that at least the majority of species described from the NE Russian areas represent real taxa, and that the problems in the arctic parts of the genus cannot be solved without a thorough and modern investigation of these.

button-arrow-back.gif (860 bytes) Contents

The grass (Poaceae) workshop

By Reidar Elven, Nicolai N. Tzvelev & Susan G. Aiken

The workshop on grasses (Poaceae) took place in St. Petersburg (22.-26. February 1999), with attendance of Dr. S. Aiken (Ottawa), R. Elven (Oslo), Dr. N.N. Tzvelev (St. Petersburg), and with translation and active participation by Dr. B. Yurtsev (St. Petersburg). The discussions were mainly based on a draft checklist by Tzvelev and the abundant specimens, including many types, present in the herbarium of the Komarov Botanical Institute, but also with reference to the Poaceae part of the Flora of Canadian Arctic Archipelago and a North Atlantic checklist.

Poaceae is the largest family of vascular plants in the Arctic. Tzvelev's draft includes 282 species and subspecies, that is 10-15% or the estimated arctic vascular flora. Most of the problems found throughout the arctic flora are represented here, e.g. how to treat morphologically similar entities at different ploidy levels, agamospermy, entities reproducing mainly or entirely vegetatively, polyploid complexes with possible polyphyletic and hybridogenous origins, and different treatments of similar (or identical) taxa across national boundaries.

The discussions revealed several discrepancies between the treatments, discrepancies that will need much more discussion before a consensus can be reached. Some of the major points were:

(1) Species are conceptualized more narrowly in the Russian draft than commonly done now in North American and North European treatments. The draft accepts some species in cases where ‘westerners’ now accept only subspecies as, e.g., in the Poa arctica and P. pratensis complexes and in Arctagrostis arundinacea/latifolia, and accepts species also where the criteria used only defines varieties or ecotypes in ‘western’ opinion or pure synonymy as, e.g., in Hierochloe orthantha, several Poas, and in one species of Arctagrostis. Even if both species and subspecies are applied in all treatments, they are applied at slightly different levels.

(2) The importance of hybridization and agamospermy/-vivipary is viewed differently. In Tzvelev's opinion, the very numerous cases of polyploid series in grasses represent past hybridization events, in some cases also across current generic boundaries and often across sectorial boundaries. This view influences the acceptance and interpretation of both species and genera. It is, however, not uniformly accepted and should not, in the opinion of Aiken and Elven, be used as taxonomic criteria without proof. The opposite is the case with agamospermy and vivipary. This is often interpreted as results of past hybridization in ‘western’ tradition, and a sharp limit, as species, is often drawn between presumed diploid sexual parents and their polyploid agamic offspring, e.g. in Calamagrostis and Hierochloe. The same is the case with separating between diploids and presumed or documented allopolyploid derivatives, e.g., in Anthoxanthum. Russian tradition places more emphasis on morphological differences than on assumed evolutionary pathways.

(3) The general acceptance of subspecies for zonal variation and cases of intermediacy. In the draft, there are numerous cases of 2 to 3 vicariant subspecies or species along the south-north gradient in cases where ‘western’ traditions only accept one species - e.g. in Alopecurus borealis and Calamagrostis lapponica and C. stricta - the two or three entities often being separated by quantitative characters only. The ‘western’ interpretation is that these may be examples of clinal variation, which is normal in most widely distributed species. The taxonomic decision to accept separate taxa must depend on the presence of discontinuities in the variation, similarly the decision to reject is based on an absence of breaks in the pattern.

We also looked at some specific cases and in the Festuca brachyphylla complex we reached a unified opinion on how to define the species in a circumpolar context and were then able to revise important parts of the Russian materials available. In the intricate Deschampsia cespitosa complex, we agreed that both ‘western’ and ‘eastern’ traditional treatments were unsatisfactory. A future treatment must be a Panarctic effort based on a large-scale biosystematic study.

button-arrow-back.gif (860 bytes) Contents

The Potentilla workshop

By Bente Eriksen, Inger Nordal, Jorun Nylehn & Boris A. Yurtsev

The Potentilla workshop (November 1998 and a few days in March 1999 in Oslo) was attended by B. Eriksen (Gothenburg), J. Nyléhn (Oslo), B. Yurtsev (St. Petersburg), and partly also by I. Nordal (Oslo). In the focus of the discussion were the variation and classification of hair types found in different species aggregates within Potentilla sect. Niveae. A scanning electron microscope study of hair types and hair ultra structure was initiated, which during the year resulted in a joint paper on terminology and systematic distribution presented in this paper (Eriksen & Yurtsev 1999). It is clear from this study that P. sect. Niveae consists of three main species aggregates (see below), confirming the preliminary results obtained of Eriksen (1997). It was also seen that a mixture of parental hair types is not always expressed in the putative hybrids of the section. It was concluded that the hair types are not inherited in a co-dominant way, the hybrids in question are hybrids of other parents, or the plants in question are not hybrids at all.

The different hair types found in the above-mentioned study are included in a list of characters and character states to be used for a morphometric study of Potentilla sect. Niveae. This list, including also diagnostic characters of leaves, flowers and fruits, might be found on the homepage of the "Potentilla Society":

The list, a joint venture from the Potentilla workshop, is presented in the hope that more botanists will eventually become interested and join the project. Above all we need help to collect data in various herbaria. Inclusion of type specimens in the study is of particular importance since misidentification is a large problem within the group.

A draft check-list on the genus Potentilla for the Panarctic Flora has been compiled by Yurtsev. It includes 70 taxa, of which 33 are proposed to be of hybrid origin. In its present form this list reflects the taxonomy presented by Yurtsev (1984) in the Arctic Flora of USSR, and this will represent the starting point for further work and discussions. It was agreed that the hypotheses on hybrid taxa must be examined more thoroughly, and preferably also with molecular methods.

Under the section Niveae, three species aggregates identified in the study on hair characters have been incorporated in the draft check list: The Potentilla nivea aggregate with floccose tomentum and smooth straight hairs (including taxa like P. nivea and P. crebidens in the Arctic); the P. arenosa aggregate (= the P. hookeriana aggregate of Eriksen and Yurtsev 1999) with floccose tomentum and verrucose, straight hairs (including taxa like P. arenosa, P. tomentulosa, and P. chamissonis in the Arctic); and the P. uniflora aggregate with crispate tomentum and smooth straight hairs (including taxa like P. uniflora and P. villosa in the Arctic).

Phenotypic plasticity may cause large variation in leaf morphology (Eriksen & Nyléhn 1999). A shift in leaflet number and position from trifoliate to digitate or digitate to pinnate induced by favorable growth conditions indicates that supernumerary leaflets may be the result of mechanisms other than hybridization.

Further, during the workshop, the genus Potentilla has been formatted for the Delta database by Kjell-Tore Hansen in collaboration with Susan Aiken, using Svalbard material as input.


Eriksen, B. 1997. Morphometric analysis of Alaskan members of the genus Potentilla sect. Niveae (Rosaceae) – Nord. J. Bot. 17: 621–630.

Eriksen, B. & Nyléhn, J. (1999). Cases of Phenotypic plasticity in leaves of Potentilla L. (Rosaceae). – Det Norske Videnskaps-Akademi. I. Mat. Naturv. Klasse, Skrifter, Ny serie 32: 191-199.

Eriksen, B. & Yurtsev, B.A. 1999. Hair types in Potentilla sect. Niveae (Rosaceae) and related taxa – terminology and systematic distribution. - Det Norske Videnskaps - Akademi. I. Mat. Naturv. Klasse, Skrifter, Ny serie 32: 201-221.

Yurtsev, B. A. 1984. Genus Potentilla L. [Rod Potentilla L.]. – In: Jurtsev, B. A. (Ed) Arctic Flora of USSR 9 (1): 137–234 (in Russian).

button-arrow-back.gif (860 bytes) Contents

The Salix workshop

By Reidar Elven & George Argus

The Salix workshop (19.-27. April 1999 in Oslo) was based on drafts of a Panarctic checklist from Dr. A. Skvortsov (Moscow) and Dr. G. Argus (Ottawa), In addition, treatments for Alaska/Yukon (Argus 1973), the Russian Arctic (Skvortsov in Tolmachev 1966) and for the whole of Russia (Skvortsov 1968), the Nordic area (R. Elven et al. in Jonsell in press), a list for the North Atlantic areas (Elven unpubl.), and the infrageneric classification in North America (Argus 1997) were reviewed. The workshop was attended by G. Argus (Ottawa) and R. Elven (Oslo), but the comments of A. Skvortsov were solicited and have been included. A fairly large selection of specimens from Siberia and the Russian Far East, some specimens from Alaska and Canada, and an abundant material from the North Atlantic areas, including NW Russia and Greenland, were at hand. Subsequent work has resulted in a revised draft, which is presented in this volume.

Salix is one of the larger genera in the Arctic, with 50-60 species and subspecies. There were surprisingly few and small discrepancies among the American, Russian and Nordic treatments. Skvortsov tended to accept broader species than usually practiced by northwestern European and American authors in some groups, e.g., the S. glauca, S. lanata and S. myrsinifolia groups, whereas Argus and Elven had a broader species view than Skvortsov in, e.g., the S. arctica and S. rotundifolia complexes. The draft proposal presented in the Salix checklist (in this volume) is a consensus proposal rather than a compromise, even if points of contention remain.

The species of Salix extending to the Arctic are derived from numerous ‘branches’ of the genus and belong to three subgenera and numerous sections. The temperate-boreal character of Salix is also reflected in the large number of species touching or only slightly extending across our arctic boundary. There are probably no exclusively arctic species, and even species like S. arctica, S. nummularia, S. polaris and S. rotundifolia have extensive alpine extensions, especially in East Asia and western North America.

Salix is known for frequent polyploid taxa, probably of alloploid origin and often results of intersectional crosses, as alloploid origins in Salix often involve more distantly related diploids rather than close relatives. A subgeneric classification would be better based on diploids rather than on the entire assemblage but this is impossible as some series only include high polyploids.

Hybridization has been deemed of great importance in Salix, especially in the Nordic tradition following Enander and Floderus, whereas it has been given little attention in other traditions or assumed to be taxonomically unimportant. It has been proved experimentally, several times, that very many species of Salix can hybridize. That they also do so in nature is evident from variation patterns. The question is whether this hybridization influences the species to any significant extent, and should be recognized taxonomically. Our opinion is that hybridization (in the modern sense, not of Floderus) seems to have little lasting influence on the species, especially at diploid level, and can be disregarded in the PAF checklist.

button-arrow-back.gif (860 bytes) Contents

The Saxifraga workshop

By Pavel Yu. Zhmylev, Christian Brochmann, Reidar Elven, Inger Nordal & Vladimir Yu. Razzhivin

The Saxifraga workshop was held in Oslo in the spring of 1999 and fully or partly attended by Dr. P. Zhmylev (Moscow), Dr. V. Razzhivin (St. Petersburg), and Drs C. Brochmann, R. Elven, and I. Nordal (all Oslo). Translation was carried out Katya Razzhivina and V. Razzhivin. The discussions were particularly based on a new draft checklist of arctic Saxifraga by Zhmylev, which in turn was based on his recent systematic reviews of the entire genus in Russia and adjacent territories (Zhmylev 1997a, b), on a translation of parts of the Saxifraga treatment in Arctic Flora of USSR (to Norwegian by Jan Wesenberg), and on the fairly recent, comprehensive treatment of the genus by Webb & Gornall (1989), which focuses on European species and represented the ‘western’ tradition in the discussions. Abundant herbarium material was available from the North Atlantic region, but only very limited material was available from Alaska and Russia (in particular western areas). Living material from the N Atlantic region was flowering in the phytotrone in Oslo at the time of the workshop, and this material was important for clarification at some points.

This large, and probably polyphyletic (cf. several recent works by the Soltis group), genus contains some 480 species traditionally classified into several subgenera, many sections, and subsections. A main distinction between Zhmylev’s (1997) and earlier taxonomic treatments is that the new treatment splits the sections into more subsections (the subsection level will probably be excluded from the Panarctic checklist). The arctic species of Saxifraga represent many of the infrageneric taxa, reflecting independent colonization of arctic areas by a number of different evolutionary lineages in the genus. Most of the diversity is found in Siberia and North America, and the lack of material from these areas (as well as the comparatively superficial treatment of Russian and N American species in Webb & Gornall) limited our discussions on the workshop. The North Atlantic area per se presents relatively few taxonomic problems, but the North Atlantic variation represents, in some cases, only part of a more complex circumpolar variation.

During the workshop we tentatively accepted arctic species and subspecies belonging to four subgenera (Micranthes, Diptera, Hirculus, and Saxifraga) and 13 sections. The workshop was, in the main part, characterized by consensus. Examples of some problems as to width of species concepts and points that were clarified are:

1) Zhmylevs divides the arctic Russian material of subsection Rotundifoliatae (sect. Chionophila, subgen. Micranthes) into a several species, which in the American tradition have been recognized as subspecies of S. nelsoniana. A possible compromise is to accept two species.

2) The extreme polymorphism within Saxifraga cespitosa (as well as its distinction towards S. rosacea, in particular in S Greenland and the British Isles) is poorly understood. New and partly unpublished data from Svalbard indicate, for example, that S. cespitosa in this area comprises two genetically fairly divergent lineages, one with large, white or rarely yellowish petals growing primarily in gravelly pioneer sites, and one with smaller, deep yellow petals growing in dense moss vegetation (cf. also Ronning’s last edition of the Svalbard flora).

3) We examined living and pressed material of the postulated Svalbard endemic S. svalbardensis to determine if this material corresponds to one of the forms of S. cernua described in Flora Arctica USSR, which also is stoloniferous and grows in humid places. According to Zhmylev, this Russian form of S. cernua is distinctly different from S. svalbardensis, which thus probably should be maintained as a Svalbard endemic.

4) Examination of a large material of the S. hyperborea/rivularis complex revealed that the western European and Russian interpretations of this complex have emphasized different morphological characters, resulting in different subdivisions of the material. The Russian tradition has emphasized calyx and hair characters. More work must be done in this complex to determine the number of species (e.g., what about S. arctolittoralis?) and their circumscription. A project is currently being carried out in Oslo, including chromosomal, morphological, and molecular analyses.

button-arrow-back.gif (860 bytes) Contents


Send mail to with questions or comments about PAF web site.
Copyright 2000-2005 Panarctic Flora Project
Last modified: 04.02.2005