Solitary herb with basal caudex sparsely or moderately densely covered by marcescent leaf remains, branching into one to several, small or moderately large rosettes, forming dense or loose tussocks. Each rosette potentially with one flowering stem without or with 1–2 leaves, up to 10 or more stems per tussock. Flowering stems ascending to erect, more or less elongating during and after flowering, up to 10–15 cm in some populations but mostly shorter, sparsely to densely pubescent with simple, forked, and/or multibranched or stellate hairs, rarely glabrous.

Leaf rosettes up to 3–5 cm. Leaves alternate, up to 15 × 5 mm, elliptic, obovate or oblanceolate, acute, entire or with 1–2 teeth per side, mid vein not or moderately prominent, dull green. Upper leaf surface sparsely to densely pubescent with mostly forked, multibranched, and/or stalked cruciform or stellate hairs (up to 0.3–0.4 mm), rarely also with short, simple hairs; lower leaf surface moderately or densely pubescent with the same hair types; margin with the same hair types and often with long, simple hairs up to 1 mm, especially in the proximal parts. Some populations have glabrous or nearly glabrous leaves.

Inflorescence a raceme, mostly with 5–10 flowers, very short in the flower stage, moderately or strongly elongating in the fruit stage up to 2–4 cm. Pedicel short (2–4 mm), much shorter than fruit and often less than 1/2 of fruit length, stout, attached with culm at an angle of 20–30°, with predominance of multibranched hairs.

Flowers radially symmetric with 4 free sepals and petals. Sepals up to 3 × 2 mm, ovate or elliptic, pale greyish green and with narrow to moderately broad white margins. Petals 2.5–3.2 × 1.0–1.4 mm, not overlapping or contiguous, erectopatent (making the flower half open), narrowly obovate or spathulate, notched, white (‘skimmed milk’).

Fruit a silicule 5–8 × 2–4 mm, erectopatent, ovate or elliptic, more rarely lanceolate, glabrous or pubescent with simple, forked, and/or multibranched, stalked cruciform or stellate hairs, greyish green,. Style short, up to 0.3 mm. Seeds 6–8(10) in each room, medium brown, up to 1.2 × 0.9 mm.

Sexual reproduction by seeds; no vegetative reproduction. Flowers are moderately large and probably occasionally pollinated by insects (i.e., a mixed mater). Flowering and seed-set is regular and abundant in most years; mature seeds are regularly observed. Seeds germinate to about 37 % in an experiment (Alsos et al. 2013).

There is no special adaptation to seed dispersal.

Draba rupestris is, perhaps, the most polymorphic and problematic of the species of Draba in Svalbard. It can be mistaken for several other species. Among the white-flowered ones, it differs consistently from D. arctica and D. nivalis in having a mixed pubescence on leaves and flowering stems of simple, forked, multibranched, and sometimes stalked cruciform or stellate hairs with few branches, whereas the two others have a dense pubescence of uniform, many-branched stellate hairs, sometimes with long, simple hairs in the leaf margins. When D. rupestris has pubescent fruits, the hairs are simple, forked or three-branched (but plants with fruits with stellate hairs are known), whereas the fruits of D. arctica are densely covered by small, many-branched stellate hairs. The fairly uniform stellate hairs on leaves and flowering stems distinguish also D. hirta from D. rupestris, as do the larger flowers with cream white petals (vs. ‘skimmed milk’ white). The perhaps most similar among the white-flowered species is D. lactea, differing from D. rupestris in leaves with a combination of long, simple, marginal hairs and very small, multibranched hairs on the lower surface (but never true stellate hairs). This combination is never found in D. rupestris. The flowering stems of D. lactea are usually glabrous, whereas those of D. rupestris usually are hairy. However, populations of entirely glabrous plants can be found in both species, and such plants can be very difficult to assign to species on morphological features. Generally, D. lactea has larger and more patent petals than D. rupestris (petal size 3.5–4.0 vs. 2.5–3.2 mm), a shorter infrutescence, and ovate–lanceolate fruits (vs. ovate–elliptic). The fruits of D. lactea are always glabrous, whereas a majority of Svalbard plants of D. rupestris have hairy fruits.

A main difference from the yellow-flowered species, when not in flower, is that plants with one or two stem leaves are found in nearly all populations of D. rupestris. No yellow-flowered species of Draba (in Svalbard) have stem leaves. Plants with rosettes only can often not be assigned with certainty to any species.

Occupies sites with little vegetation cover and high insolation rates. Common growth sites are gravelly slopes, dry ridges, river bars and fans, polar desert, and also often found on road verges and in other disturbed sites, occasionally on cliff ledges. On fine textured or mixed substrates with acidic or circumneutral soil reaction (pH), rarely on basic soils, usually moderately drained, but also in more moist sites. Probably little grazed by reindeer and geese.

Present in all zones and sections but very rare in the polar desert and in the clearly continental section. The species is frequent on Bjørnøya and in W and C Spitsbergen, rare on Nordaustlandet and Prins Karls Forland, and not yet recorded from Edgeøya, Barentsøya or the smaller off-lying islands.

Draba rupestris is amphi-Atlantic, confirmed from E North America (mainly or only in Canada), Greenland, all the North Atlantic islands from Britain north to Svalbard, Fennoscandia, and N European Russia. There are reports from other areas, e.g., Siberia and NW Canada, but these need confirmation.

Brochmann et al. (1992, 1993) concluded from crossing experiments and molecular data that Draba rupestris (by him as D. norvegica), with samples from mainland Norway and Svalbard, was a polyphyletic polyploid species, i.e., with several origins from possibly different low-ploid parents. The alternative conclusion, supported by the morphological variation pattern, is that it is more than one species. At least two main morphological groups (I and II) have been identified in the North Atlantic material, by Elven in Lid & Lid (2005) assigned under the names D. "norvegica" and D. arctogena (E.Ekman) E.Ekman, the latter assignment erroneous (see Elven et al. 2022). Both groups are present in Svalbard and also in mainland Scandinavia and Iceland. The identity of the plant assigned by Elven in Lid & Lid (2005) to D. arctogena (i.e., group II) is not known and no other name is available for it. We therefore here apply D. rupestris as a collective name for both groups.

Group I plants are usually fairly tall-grown with stems elongating much in the fruit stage. Stem leaves are commonly present and both stem and basal leaves often dentate. The pubescence on leaves and stems usually consists of a mixture of simple, forked and multibranched hairs (but no stellate hairs). The infrutescence elongates strongly and may occupy more than 50 % of the stem length. Fruits are often lanceolate or narrowly ovate or elliptic. They are either glabrous or with a sparse or moderately dense pubescence of simple and/or forked hairs. This is the main plant of this group from the north boreal zone to the middle alpine belt in Scandinavia and also common in the lowlands of Iceland. In Svalbard, it has been recorded mainly from the inner fjord districts of Spitsbergen, i.e., from the middle arctic tundra zone.

Group II plants are low-grown. The stems only rarely have a leaf and both stem and basal leaves are usually entire. The pubescence on leaves and stems is dominated by multibranched, stalked cruciform and sometimes stellate hairs. The infrutescence elongates only slightly and rarely occupies more than 1/3 of the stem length. Fruits are mostly elliptic. They are regularly pubescent with forked, multibranched, stalked cruciform and/or stellate hairs. This is a plant of the higher mountains of Scandinavia and higher and interior parts of Iceland, possibly also of the Faeroes, and it is the single plant of the group on Jan Mayen and Bjørnøya. On Spitsbergen and surrounding islands, it is the predominant form except for the lower altitudes in the inner fjord districts where it is replaced by group I plants.

The correct scientific names of the two species currently (and here) going by the names D. hirta and D. rupestris are undecided. Even if both species may be polyphyletic, they are usually distinctly different morphologically and also differ in ploidy levels (D. hirta is octoploid with 2n = 64, D. rupestris is hexaploid with 2n = 48). The earliest name in the group is D. hirta L. from 1759, a name not yet connected to a designated type. The original material (in the Linnaean Herbarium in London) consists of several parts, some belonging to D. hirta and D. rupestris, some perhaps not identifiable to species. Russian authors have generally accepted D. hirta as the correct and priority name, whereas European authors until the last decades have applied the name D. daurica DC. from 1821, based on a plant from Dahuria (the districts around Lake Baikal in S Siberia) and American and recent European authors have applied the name D. glabella Pursh from 1813 based on a North American plant. All agree now that the names D. daurica DC. and D. glabella Pursh refer to the same plant. If the name D. hirta is typified on one of the Linnaean parts corresponding to this plant, it will have priority, and we have now chosen this solution. However, also the name D. norvegica is included in this problem.

Gunnerus’ name Draba norvegica from 1776 was based on a Norwegian plant described by Oeder in Flora Danica in 1764 as “Draba pyrenaica (?)”. There is no herbarium material, and the holotype (i.e., the only available type specimen) is therefore the illustration in Flora Danica. This shows a plant much different from the current concept of D. norvegica but fitting well with the current concept of D. hirta. This means that the name D. norvegica may become another and early link in the chain D. hirta L. 1759 – D. norvegica Gunnerus 1776 – D. glabella Pursh 1813 – D. daurica DC. 1821. The name D. norvegica is probably not available for the species currently going by that name. The alternative name, which we have chosen here, is D. rupestris W.T.Aiton from 1812, based on plants from the British Isles.

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.

Brochmann, C., Borgen, L. & Stedje, B. 1993. Crossing relationships and chromosome numbers of Nordic populations of Draba (Brassicaceae), with emphasis on the D. alpina complex. – Nordic Journal of Botany 13: 121–147.

Brochmann, C. & Elven, R. 1992. Ecological and genetic consequences of polyploidy in arctic Draba (Brassicaceae). – Evolutionary Trends in Plants 6: 111–124.

Brochmann, C., Soltis, D.E. & Soltis, P.A. 1992. Electrophoretic relationships and phylogeny of Nordic polyploids in Draba (Brassicaceae). – Plant Systematics & Evolution 182: 35–70.

Elven, R., Bjorå, C.S., Fremstad, E., Hegre, H. & Solstad, H. 2022. Norsk flora. Ed. 8. – Det Norske Samlaget, Oslo.

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

Lid, J. & Lid, D.T. 2005. Norsk Flora. Ed. 7 by R. Elven. – Det Norske Samlaget, Oslo.