Solitary to mat-forming herb growing in tussocks or dense to loose stands due to rhizome with short creeping branches (1−3 cm). Bases of tussocks covered by broad (1−3 mm), withered leaves and sheaths from previous years, with distinct veins, shiny, purple or reddish brown. Culms several from each tussock, rarely single, 3−10(15) cm, erect or nodding at the top, terete, glabrous (except for sheaths of leaves and bracts), green or reddish tinged. Leaves mainly basal; culms with 2−3 leaves, the uppermost usually above the middle of the culm.

Leaves canaliculate (especially in apex) or sometimes flat. Basal leaves 3−10 cm long, 1−2 mm broad, gradually tapering in upper half, with more or less distinct mid vein and several indistinct lateral veins, margins smooth or with very sparse, low teeth, with a few, long hairs, green, yellowish green or reddish.

Inflorescence an open, 1−2 times branched cyme with several loose, small heads or clusters. Heads or clusters few-flowered (rarely more than 10 flowers per head or cluster), of various size, rarely more than 4−5 × 4−5 mm, dark reddish brown. The central cluster usually not larger than the other ones. Inflorescence branches slender (0.1−0.2 mm broad), arcuate. Inflorescence subtended by several bracts 3−6 mm long, with sheaths 1−3 mm and very short, ovate blades, reddish brown and more or less hyaline, fringed with numerous white hairs ca. 0.5 mm long. Bracteoles 0.5−1 mm, orbicular to ovate, rounded to subacute, dark reddish brown.

Flowers radially symmetric with 6 (3 + 3) tepals. Tepals subequal, 1.4−2.1 mm, lanceolate, acute, longer than fruit, reddish brown, with a very narrow hyaline margin and apex. Stamens 6. Gynoecioum of 3 carpels with 3 stigmas.

Fruit a one-roomed capsule with 3 seeds. Capsule broadly ovoid, subacute, with a very short style, shiny reddish brown. Seeds 1.0−1.2 mm, without a distinct elaiosome (a fatty appendage which is an adaptation to ant dispersal, uniformly present in boreal and temperate Luzula). The elaiosomes are stunted and non-functional in this and almost all other arctic and high alpine species of the genus.

Sexual reproduction by seeds; local vegetative reproduction by rhizomes and fragmentation. Wind pollinated. Seed set is probably regular (both in Bjørnøya and Spitsbergen) as opened and emptied fruits and well-formed seed regularly are observed.

Seed dispersal is probably mainly passive.

Luzula arcuata is similar to L. confusa, and these two are not clearly distinguished in the North Atlantic regions (see Comments). Luzula arcuata normally has several arcuate branches in the inflorescence, often branched anew, with loose clusters of few flowers; L. confusa normally has none or very few, stiff and erect branches in the inflorescence, never branched anew, with compact, head-shaped clusters of many flowers (more than 10). Except for these differences, these two species are difficult to keep apart.

Moist heaths and early snowbeds, mostly on substrates with circumneutral to acidic soil reaction (pH). The substrates may be fine grained or coarse but always with good drainage. The species tolerates inundation only for short periods during snow melt.

In the middle and northern arctic tundra zones and the weakly oceanic and transitional sections. Luzula arcuata is confirmed from numerous sites on Bjørnøya and a few sites along the west coast of Spitsbergen from Sørkapp Land north to Krossfjorden (Albert I Land), perhaps to Magdalenafjorden. It belongs to a west coast element including, e.g., Cerastium alpinum, Harrimanella hypnoides, Ranunculus glacialis, Rhodiola rosea and Salix herbacea.

Luzula arcuata is a disjunct Atlantic and Beringian species. One part of its range is around the North Atlantic in Greenland (long known from SE Greenland, confirmed 2013 also from W Greenland by R. Elven & H. Solstad), Jan Mayen, Bjørnøya, and Fennoscandia, another part is in W Alaska and NE Asia (Chukchi Peninsula). These two parts of the range are separated by ca. 100° latitude in North America and ca. 120° latitude in Eurasia. See Elven et al. (2011).

The main reason why we treat L. arcuata and L. confusa as two separate species is how they behave outside the North Atlantic regions. In Beringia, they keep apart without any transition whatsoever (see Elven et al. 2011). In the North Atlantic regions, however, intermediate forms are common, both in Fennoscandia and in the arctic parts (see Elven et al. 2011). Whether this is a hybridization situation or something else, we do not know. The following is part of a comment by Egorova et al. in Elven et al. (2011): "The main division runs between the circumpolar Luzula confusa and the more oceanically distributed L. arcuata s. lat. Luzula confusa has long been known as sympatric with L. arcuata in the Atlantic regions but seems to be so also in the Beringian regions. The two species differ disjunctly morphologically in Beringia, whereas transitional forms occur around the North Atlantic in eastern Greenland, Iceland, Fennoscandia, Jan Mayen, and Svalbard. The two taxa should nevertheless be treated as different species."

This statement may have to be modified a little; we (R. Elven & H. Solstad) have not been able to confirm L. confusa from Iceland, whereas we in 2013 found it also in W Greenland. Both species seem to have the same two chromosome numbers, 2n = 36 and 48, and as in other species of Luzula, this does not necessarily mean a ploidy difference; it can be different degree of fragmentation of chromosomes (see Kirschner et al. 2002).

The situation in Svalbard is, at present, not very clear. Some populations from Bjørnøya and from some sites on the west coast of Spitsbergen, at least north to Krossfjorden but perhaps also farther north to Magdalenafjorden (Albert I Land), are not separable from Fennoscandian L. arcuata (from the type region) but many other populations in these coastal areas are problematic to assign. We have some tens of collections from the western parts that we are unable to assign to either L. arcuata or L. confusa. This is not unexpected; in the Scandinavian mountains, one of the species may be alone on a mountain, both present but distinct on another mountain in the vicinity, and all transistions found on a third mountain, without any explainable pattern (E. Hultén in identifications in the herbaria; R. Elven field experiences etc.). In the other parts of Svalbard, the material belongs to L. confusa and is inseparable from plants in N Greenland, arctic Canada, and arctic Russia.

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

Kirschner, J. et al. 2002. Species plantarum: flora of the world. Part 6. Juncaceae: Rostkovia to Luzula. – Australian Biological Resources Study, Canberra.