Panarctic Flora

672106-09 The Draba alpina aggregate D. alpina, D. glacialis, D. oxycarpa, D. "pseudo-oxycarpa"

Geography: North American (NE) - amphi-Atlantic - European (N) - Asian (N) - amphi-Beringian (W).

Notes: The Linnaean name Draba alpina has been in use for almost every perennial, scapose, yellow-flowered Draba in northern Europe, Greenland, northern North America, and northern Asia. The availability of D. alpina as a collective name has prevented an analysis of the variation among the yellow-flowered Drabas in many regions. A type for the name D. alpina was designated late, by Elven in Jonsell and Jarvis (2002), and the name is now connected to a fairly narrowly amphi-Atlantic decaploid (2n = 80). After this, it is obvious that the major parts of the arctic material named D. alpina does not belong to (or even close to) that species. All applications of this name outside the North Atlantic regions are either suspect or obviously erroneous as contrary to the type now designated. As an example, Rollins (1993) reported for North American D. alpina the chromosome numbers 2n = 64, 80, 120, and ca. 160 !, indicating inclusion of several yellow-flowered species, among them D. oxycarpa (2n = 64) and D. corymbosa (2n = 120), both genetically far from D. alpina s. str. (Brochmann et al. 1992, 1993).

A survey in 2003 of the material from the Canadian Arctic Archipelago (CAN, DAO) proved D. alpina from the easternmost parts but not from elsewhere in these islands. A re-investigation in 2009 (CAN) threw doubts on the identifications of this species from the Hudson Bay area. The majority of the specimens that had been assigned to D. alpina from non-Beringian arctic Canada were reassigned to D. corymbosa, D. pilosa, D. simmonsii, and the Hudson Bay plants perhaps to D. glacialis (see below).

A more superficial survey of the material annotated as D. alpina in Alaska and the Yukon Territory (ALA, 2003) showed that no part of it corresponds to D. alpina as typified. About half the material fits into other named species (namely D. corymbosa, D. macounii, and D. pilosa), whereas the other half divides on at least three morphologically distinct plants that either are nameless or may have names on the Russian side: an Alaska Range plant, a widespread plant, and a Seward Peninsula plant provisionally included below as D. "pseudo-oxycarpa".

Field investigations in northern Siberia in 2004, where D. alpina also has been reported as frequent, gave better results. The majority of material fitted into the D. micropetala aggregate, D. corymbosa, D. glacialis, and D. pilosa, but still with 1-2 plants we could not fit into the accepted species. No plant fitted D. alpina as typified.

There is an obvious need for a critical, circumpolar study of this group. However, no meaningful molecular work of the D. alpina group is possible unless it is based on a morphological analysis revealing (and naming) taxa. Its closest non-arctic relatives are probably found in southern Siberia and central Asia, e.g., the Altai, Tuva, and Burjat-Mongolia, from where several similar species are described. The mountains of Pacific Asia and North America (including the Cordilleran mountains) may have contributed some basic genomes, whereas there are no known close relatives in European mountains or in the eastern parts of North America.