Mat-forming herb growing in very open and often very large stands (several 100 m² or 1000 m²) due to extensive, branched rhizome. Rhizome branches 1.5–5 mm thick, white, rooting at all nodes, with very variable distance between aerial shoots, from centimetres to several decimetres. The end of each rhizome branch typically has three stages of shoots, apically the budding shoot for next season, behind that a single vegetative shoot, and behind that again a single reproductive shoot. Shoots are dimorphic. Early in season appears pink or reddish, erect, stout stems (up to 8 mm thick), mostly 8–15 cm tall at the flower stage, elongating to 20–30 cm or more in the fruit stage, with 4–7 brownish or reddish, scaly, mainly non-assimilating leaves (see below) and an inflorescence. Later in season appears shoots with 2–4 assimilating leaves at ground level. Petioles, stems and inflorescence branches have a more or less dense pubescence of arachnoid hairs (very thin hairs intermingled like a cobweb). Plants are functionally unisexual. Stems of male plants wither very soon after flowering, those of female plants elongate and last until fruit stage.

Leaves alternate (also scales on reproductive stems), with a petiole of 2–10 cm, about the same length as the blade or longer, and a blade 1.5–5 × 2.5–8 cm, broadly sagittate or triangular, at base truncate or shallowly cordate, with 4–8 coarse, triangular teeth or lobes on each side; upper surface dark green, glabrous or sparsely pubescent; lower surface grey or whitish due to a dense pubescence of arachnoid hairs (see above). Scaly leaves of reproductive stems with very broad, sheath-like petioles clasping the stem, of very variable length and width, without or rarely with a small, green blade of variable shape and size.

The primary inflorescence of Asteraceae is a head (capitulum) surrounded by an involucrum of one or more rows of phyllaries (involucral bracts). The flowers sit on a flat, concave or convex receptacle, sometimes with scales (the bracts of the single flowers). The flowers are epigynous with perianth at top of the gynoecium. The sepals are always transformed into a pappus, mostly by hairs or sometimes by scales. The 5 stamens are inserted in the corolla tube and the anthers form a ring through which the style grows and pushes the pollen outwards. Gynoecium of two fused carpels, 2 stigmas. The fruit is an achene with one seed.

Functionally dioecious with rudiments of the other sex found in the flowers, especially as styles with sexually non-functional stigmas in male flowers (see Comments for other function). Inflorescence a raceme of heads, 2–5 × 2–4 cm in flowering stage. Heads 3–5(9) (often more in more southern regions), on 1–2 cm peduncles, elongating up to 5 cm in fruit. Heads ca. 1 × 1 cm, with 8–14(16) phyllaries in 1–2 rows; phyllaries ca. 10 × 2 mm, narrowly lanceolate or triangular, obtuse or subacute, often with an irregular apex, pubescent along the middle but otherwise usually glabrous, dark purple or dark violet. Heads with two kinds of flowers and corollas. Marginal flowers monosymmetric with a ligula of 5 fused petals with ca. 3 teeth apically. Central flowers radially symmetric with a tubular corolla with 5 teeth apically.

Corollas pale violet or lilac. Male heads with few (1–10) and sterile marginal flowers with ligulas 1.5–3 mm and emerging (non-functional styles), and numerous (more than 10) fertile central flowers with non-functional styles pushing the pollen out of the corolla tubes (see more details in Comments). Female heads with many (20–50 or more) fertile marginal flowers with ligulas 0.8–4.5 mm, and few (1–10), sterile central flowers.

Fruit a cypsela (with 1 seed) where the seed coat and fruit wall is fused. Calyx transformed into a pappus, a circle of thin, obscurely dentate, shiny white hairs up to 15 mm on top of the fruit, developing only on female plants.

Sexual reproduction by seeds, at least potentially; efficient local vegetative reproduction by extensive growth of rhizomes, and the large clones may easily break apart into physiologically separate individuals due to soil movement or just death of old parts of the rhizome system. Seed set is uncertain in Svalbard under prevailing climate, see Comments. Seeds did not germinate in an experiment (Alsos et al. 2013).

Fruits (if they are produced) are adapted to dispersal by wind over some distance (perhaps kilometres as a regular dispersal distance), due to the pappus hairs.

There is nothing similar in the Svalbard flora.

Shallow mires and moss tundras, always in permanently or periodically moist sites and almost always in sites with a dense vegetation cover and fairly stable soil conditions. This plant is taller than most other plants in similar sites and not easily over-grown. It is not very demanding as to mineral nutrients but is absent from the more acidic areas.

Thermophilous. Mainly in the middle arctic tundra zone and the weakly continental section, but transgressing into the northern arctic tundra zone and the transitional section. Restricted to Spitsbergen where it has a range spanning from N Wedel Jarlsberg Land to the northern parts of the Isfjorden districts, with a single locality on one of the Lovén Islands in Kongsfjorden (Haakon VII Land).

Petasites frigidus is mainly Eurasian and NW North American but with some closely related species or races throughout North America and in the Beringian regions. All European and the majority of Asian plants belong to ssp. frigidus. As the species is missing from Greenland, its affinity in Svalbard is to the east (Russia) or south (Fennoscandia). It is, however, missing from Bjørnøya where hydrologically suitable sites abound. Bjørnøya may be to cool and foggy for this rather thermophilous plant, or too isolated to be reached by wind dispersal.

The variation within the species or species group of Petasites frigidus s. lat. is appreciable but mainly confined to the Beringian regions and North America (Cody 1994, 1996; Cherniawsky & Bayer 1998a, 1998b, 1998c; Bayer et al. 2006). This matter is commented on by Elven et al. (2011). The variation in Europe and in the major parts of Siberia is insignificant.

Petasites frigidus is one of the more impressive plants in the Svalbard tundra, tall-grown and often 4–5 times as tall as any other plant in the vegetation types where it occurs. The crucial question is whether it reproduces in Svalbard under present climatic conditions. This is a plant than can survive for centuries or millennia without setting seed. Our consideration is that it is unlikely to reproduce today. It is dioecious and forms extensive clones, which means that normally only one sex is present at each locality (we have not observed male plants at all in any of the localities we have visited). The pollen presentation mechanism in Asteraceae ensures insect pollination as the only strategy. The anthers are located inside the corolla tube and open inwards. There are hairs on the outside of the style and stigmatic branches. The pollen is brushed out and attached to the outside of the stigmatic branches when they grow through the corolla tube and exposed to insects when they emerge, whereas the stigmatic surface is inside the branches and protected from self pollination (in addition, the genus Petasites has non-functional styles and stigmas in male plants, even if the brush function is retained). The chances that an insect collects pollen in a male population in Svalbard and deposits it on the stigmas in a female population are infinitesimal. Svalbard insects do not fly far, and they are not true to any plant species where concerns pollination (i.e, not like bees and bumblebees). The conclusion is therefore that the present range of Petasites frigidus most probably is a remnant from warmer postglacial times (the Hypsithermal, 7000–4000 years BP). A hypothesis that most or all occurrences in Svalbard are due to single, accidental long distance dispersal events, is far-fetched; the rather heavy, wind-borne fruits do not travel that far very easily, and there are numerous (more than 60) separate populations. Note, for instance, that this species has not reached Greenland even if local conditions there are much more favourable than in Svalbard, and even if the species occurs on both sides (in arctic Canada as another subspecies, in Fennoscandia and Svalbard as ssp. frigidus).

Alsos, I.G., Müller, E. & Eidesen, P.B. 2013. Germinating seeds or bulbils in 87 of 113 tested Arctic species indicate potential for ex situ seed bank storage. – Polar Biology 36: 819–830. Doi 10.1007/s00300-013-1307-7.

Bayer, R.J., Bogle, A.L. & Cherniawsky, D.M. 2006. Petasites Miller. – In: Flora of North America Editorial Committee (eds.), Flora of North America north of Mexico. 20. Magnoliophyta: Asteridae (in part): Asteraceae, part 2: 635–640.

Cherniawsky, D.M. & Bayer, R.J. 1998a. Systematics of North American Petasites (Asteraceae: Senecioneae). I. Morphometric analyses. – Canadian Journal of Botany 76: 23–36.

Cherniawsky, D.M. & Bayer, R.J. 1998b. Systematics of North American Petasites (Asteraceae: Senecioneae). II. Isozyme analysis and population genetic structure. – Canadian Journal of Botany 76: 1476–1487.

Cherniawsky, D.M. & Bayer, R.J. 1998c [1999]. Systematics of North American Petasites (Asteraceae: Senecioneae). III. A taxonomic revision. – Canadian Journal of Botany 76: 2061–2075.

Cody, W.J. 1994. Nomenclatural changes and new taxa for the Yukon flora. – Canadian Field-Naturalist 108: 93–95.

Cody, W.J. 1996. Flora of the Yukon Territory. – NCR Research Press, Ottawa.

Elven, R., Murray, D.F., Razzhivin, V. & Yurtsev, B.A. (eds.) 2011. Annotated Checklist of the Panarctic Flora (PAF) Vascular plants. http://panarcticflora.org/