The units of the inflorescence of Poaceae are the spikelets, nearly always numerous in a panicle or spike-like inflorescence. Spikelets are composed of 2 glumes (bracts for the spikelet) and one or more flowers (the term used below) or rather floral units, often named ‘florets’ because we do not know what is the exact flower. A flower or floret is composed of a lemma with one mid vein (probably the floral bract), a palea with 2 mid veins (either fused bracteoles or perhaps 2 fused perianth leaves), 3 small organs called ‘lodiculae’ and essential in the opening of the flower at anthesis (possibly transformed perianth leaves or transformed stamens), 3 stamens (mostly), and a gynoecium of 2 fused carpels with 2 feathery stigmas and one seed.
Inflorescence a dense or interrupted panicle 5—7(8) cm long, narrowly elongate, dark violet, with erect to spreading (but never patent) branches; panicle occupying 1/3—1/4 of total culm length. Base of panicle (lowermost node) with 2 supporting brown scales, withering early during flowering. Panicle with 7—9 nodes, the lowermost one(s) at some distance (often 1.5—2 cm) from the others, with (1)3—4(6) branches at each of the lower nodes. Mid axis and branches of panicle densely scabrous. Branches 10—30 mm long, the longer ones with (2)3—5(6) spikelets more or less spread along their entire length. Spikelets (3)4—5(6) × 0.9—1.5 mm, unusually short-stalked, with one fully developed flower but often with a rudiment of a second flower. Bracts (glumes and lemmas) with rounded backs, i.e., without a keel, but with a marked mid vein. Glumes of approximately the same length, (3)3.5—4.5(5) mm, 3/4 or more of the entire spikelet length, lanceolate, acute or acuminate but apex sometimes lacerate, dark violet with a white hyaline apex, 1-veined, smooth and glabrous (but see var. hirta in Comments below). Lemmas 4–5 mm, oblong, obtuse to subacute or often lacerate, violet with a broad, white hyaline margin in upper 1/3 and apex, 1-veined, scabrous on vein in the upper part, and often slightly hairy besides the vein, hairs short and straight (but see var. hirta below). Paleas as long as lemmas, scabrous on the keels. No stamens observed by us in the field in Svalbard or in the Svalbard herbarium material.
Arctagrostis latifolia resembles Calamagrostis neglecta subsp. groenlandica in its violet panicle but is easily distinguished from the latter by its broad, flat leaves, by the glumes in the spikelet distinctly shorter than the lemma, and by the lemmas often obtuse, with a broad, hyaline apical part, and scabrous without a tuft of hairs at base. Calamagrostis neglecta has leaves narrow and involute, glumes as long as or longer than lemmas, and lemmas acute, with a narrow hyaline margin and with a tuft of basal hairs.
When not flowering, the only native grass in Svalbard with similarly broad, flat leaves is Alopecurus ovatus. However, whereas Alopecurus has long leaves with short (1—2 mm), obtuse ligulas, Arctagrostis has conspicuously short, broad leaves with long (3—6 mm), often lacerate ligulas. Arctagrostis also has papillose leaves with scabrous margins, whereas Alopecurus has non-papillose leaves with smooth margins.
Thermophilous. In the middle arctic tundra zone, mainly in the inner fjord parts, i.e., in the transitional and clearly continental sections.
Found only on Spitsbergen, mainly along the middle and inner parts of Isfjorden and Wijdefjorden, with one stand documented in W Sørkapp Land and one at Stuphallet by Kongsfjorden west of Ny-Ålesund in Oscar II Land. We have excluded three literary records, two of them because they have been made from localities extensively visited by later botanists without any new fines, and one because it is very far from the otherwise documented range. The two former are from Bjørndalen near Longyearbyen (Nordenskiöld Land; Ekstam 1898) and Colesbukta (Nordenskiöld Land; Hadač 1944, based on Flovik 1938); the last one is from NW Edgeøya (Dahl & Hadač 1946, cited by Neilson 1970). In all cases we suspect some confusion with other broad-leaved grasses.
Thermophilous. Confined to the middle arctic tundra zone, mainly in the inner fjord parts, i.e., to the transitional and clearly continental sections. Found only on Spitsbergen, mainly in the middle and inner parts by Isfjorden and Wijdefjorden, with one stand documented in W Sørkapp Land and one at Stuphallet by Kongsfjorden west of Ny-Ålesund in Oscar II Land. The majority of the stands are spatially limited and separated from other stands by long stretches with numerous apparently suitable sites where the species is absent. The entire pattern has a relict appearance.
Arctagrostis latifolia s. str. is circumpolar, common throughout the arctic parts of Russia (European and Asian), North America and Greenland. It also occurs in N Fennoscandia from Finnmark eastwards.
The genus Arctagrostis Griseb. consists of 1 or 2 species, depending on whether A. arundinacea (Trin.) Beal is accepted as a separate species or as A. latifolia ssp. arundinacea (Trin.) Tzvelev. The additional ca. 15 species reported for the genus seem to be local modifications or ecotypes of A. arundinacea (Aiken et al. 1994). The morphological variation is small in Europe, Greenland, and arctic Canada but much larger in Siberia and the Beringian regions where the two species or subspecies are largely sympatric and have been observed growing at the same localities but in different vegetation types, both in W Alaska and E Chukotka (R. Elven & H. Solstad observ.). Whereas many current North American authors prefer treatment as two subspecies (e.g., Aiken & Lefkovitch 1990; Aiken et al. 2007), others prefer two species (e.g., Malyschev & Peschkova 1990; Elven et al. 2011). Arctagrostis latifolia is mainly octoploid (2n = 56) and circumpolar (see Distribution), whereas A. arundinacea is mainly tetraploid (2n = 28) and restricted to Siberia, Russian Far East, and NW North America, and to more southern latitudes than ssp. latifolia. Transitional forms are known, also with intermediate chromosome numbers.
There is a single report of an even higher chromosome number, ca. 9-ploid (2n = 62; Flovik 1938), and this count was made in a morphologically deviating population in Spitsbergen: at Kapp Wijk in the Isfjorden area (Dickson Land). The plants in this population have glumes scabrous on the mid vein and more or less hairy elsewhere, and lemmas densely pilose and with lacerate apex. This plant was described as var. hirta Hadač (Hadač 1942) and is only known from this locality. It is reported to have aborting anthers. In view of the rather extreme polymorphy in A. latifolia s. lat., we are reluctant to accept local races even if they deviate in chromosome number, and especially if they do not seem to reproduce (viz. the anthers).
The plants at the not too many Svalbard localities often differ rather much between localities, partly in vegetative characters (leaf width, how papillose and/or scabrous the leaves are, culm thickness), but more so in panicle and spikelet characters (how acute and fringed glumes and lemmas are, how hairy, whether panicle branches are short or longer, erect or spreading, and how well the hyaline margins are developed. Plants (or perhaps only clonal shoots) within a stand are usually monomorphic but may differ from those of other stands in the same general area, most evident by Wijdefjorden in Andrée Land and Ny-Friesland and by Isfjorden in Dickson Land (areas where several localities are known). It is possible that each stand has its separate immigration history (probably bird dispersal) and that no, or very little, local recruitment by seed has taken place after immigration to Svalbard. All Svalbard "populations" may be clones.
Aiken, S.G. 1999. Challenges of the species concept in arctic grasses based on North American experience. – Skrifter Norske Videnskaps-Akadademi. I. Matematisk-Naturvitenskapelig Klasse, n. s. 38: 161–171.
Aiken, S.G. (ed.), Dallwitz, M.J., Consaul, L.L., McJannet, C.L., Boles, R.L., Argus, G.W., Gillett, J.M., Scott, P.J., Elven, R., LeBlanc, M.C., Gillespie, L.J., Brysting, A.K., Solstad, H. & Harris, J.G. 2007. Flora of the Canadian Arctic Archipelago: descriptions, illustrations, identification, and information retrieval. – [CD-ROM version] National Research Council Canada, Ottawa.
Aiken, S.G. & Lefkovitch, L.P. 1990. Arctagrostis (Poaceae, tribe Pooideae) in North America and Greenland. – Canadian Journal of Botany 68: 2422–2432.
Aiken, S.G., Lefkovitch, L.P., Gardiner, S.E. & Mitchell, W.W. 1994. Evidence against the existence of varieties in Arctagrostis latifolia ssp. arundinacea (Poaceae). – Canadian Journal of Botany 72: 1039–1050
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.
Elven, R., Murray, D.F., Razzhivin, V. & Yurtsev, B.A. (eds.) 2011. Annotated Checklist of the Panarctic Flora (PAF) Vascular plants. http://panarcticflora.org/
Flovik, K. 1938. Cytological studies of arctic grasses. – Hereditas (Lund) 24: 265–326.
Hadač, E. 1942. Notulae ad floram Svalbardiae spectantes. – Studia Botanica Cechica 5: 1–5.
Malyschev, L.I. & Peschkova, G.A. (eds.) 1990. Flora Sibiri. 2. Poaceae (Gramineae). – Nauka, Novosibirsk.