Panarctic Flora

630601 Oxytropis deflexa (Pall.) DC.

• DC., Astragalogia, ed. quarto: 66, ed. folio: 53 (1802). - Astragalus deflexus Pall., Acta Acad. Sci. Imp. Petrop. 2: 268 (1779). Lectotype (BM): Siberia: "ad nivalia Dauriae ... in excelsis montibus circa Balyra rivum, aliosque Ononem influentibus", leg. S.P. Pallas (Welsh 1995: 274).

2n= 16 (2x). - Europe (Norway), Canada, U.S.A. - Several reports, the European ones (Laane 1965; Knaben and Engelskjøn 1967) for the local subsp. norvegica.

Geography: European (N) & Asian (C-NE) - amphi-Beringian - North American.

Notes: Hultén and Fries (1986) mapped Oxytropis deflexa with a disjunct range of part areas in northern Norway, northern and central Asia, northeastern Asia, and at least two separate parts of North America. Some populations connecting the Asian and American ranges have been found later. The plants in the part areas differ morphologically.

Hultén (1968a) accepted four taxa. The two small and non-arctic northern Norwegian populations are isolated by more than 80 longitude in both directions from the nearest populations in Asia and North America, and they differ appreciably in morphology from the others. Hultén accepted them as subsp. norvegica Nordh. All the other populations were by him by implication included in subsp. deflexa with three varieties: var. deflexa in Asia and var. foliolosa and var. sericea in North America. In addition to morphology, there is some chemical support for recognition of the Norwegian, Asian, and North American plants as different (Høiland and Laane 1989). Hultén mapped the two North American varieties to be largely sympatric in the west but largely allopatric in the east. In the northwest, they are rather different even in close to sympatric situations and racial rank may not be the most appropriate. They were accepted as varieties by Porsild and Cody (1980) and Welsh (2001). Hultén and Fries (1986) accepted O. foliolosa as a separate, trans-American species and accepted three subspecies of O. deflexa: subsp. norvegica in Norway, subsp. deflexa in Asia, and subsp. retrorsa (= var. sericea) in western North America. Cody (1996), with much experience from an area where the two putative North American races or species co-occur, accepted them as subsp. foliosa and subsp. sericea. Many authors do not accept races in this species within North America (e.g., Cronquist et al. 1977; Gillett et al. 2007). Yurtsev (1986) proposed an additional subspecies in the Russian Far East, subsp. dezhnevii, in the gap between subsp. deflexa in northeastern Asia and subsp. foliolosa in western Alaska.

Our opinion is that the morphological variation is too large and too well structured geographically to fit within a species without races, and also that the North American material must be divided on two races or perhaps species (as done by Hultén). For the Checklist, we accept the variation found in the northern parts as four subspecies (but subsp. norvegica is non-arctic).