Panarctic Flora

342006 Calamagrostis neglecta (Ehrh.) P. Gaertn., B. Mey. & Scherb.

• P. Gaertn., B. Mey. & Scherb., Oekon. Fl. Wetterau 1: 94 (1799). - Arundo neglecta Ehrh., Beitr. Naturk. 6: 137 (1791). Type: Sweden: Uppland, "Upsaliae", leg. Ehrhart 118 (according to Tzvelev 1976: 310). Probable isotype in LE.

Geography: Circumboreal-polar.

Notes: Tzvelev and Elven: There has been and still is dispute about the valid name of this species, either Calamagrostis neglecta based on Arundo neglecta Ehrh. 1791 or C. stricta based on Arundo stricta Timm 1795. The previous Arundo stricta Gilib. 1792 does not invalidate Arundo stricta Timm as the former name is inadmissible as published in a work not consistently applying binomial nomenclature (ICBN). The argument for C. stricta is that Arundo neglecta could have been published as a synonym of the previously described A. halleri Willd., which in turn is a synonym of the submediterranean Agrostis calamagrostis L., now Achnatherum calamagrostis (L.) P. Beauv., see Löve (1970b). This was the view of Soreng and Greene in Soreng et al. (2003: 220): "A proposal is in preparation to conserve C. stricta over C. neglecta". Such a proposal has not yet been accepted as part of ICBN (McNeill et al. 2006) and is perhaps unlikely to be so because the name C. neglecta has been in common use for this species before Löve (1970b) and often also afterwards.

Here follows a translation from Tzvelev (1973), justifying the re-instatement of the name C. neglecta and our acceptance of it for the Checklist: "The priority name, recently indicated by me, of the widespread species was replaced by a number of authors based on Löve 1970b with the later name C. stricta (Timm) Koel. The reason is that in the initial description, Arundo neglecta is cited as a synonym of the previously described A. halleri Willd., and therefore A. neglecta becomes a nomen illegitimum. However, when reviewing the primary sources, I had the opportunity to convince myself that there is no reason for the substitution. Arundo halleri Willd. is not a newly described species but a new name for Agrostis calamagrostis L. in the genus Arundo L. where an Arundo calamagrostis L. already existed. Willdenow was mistaken when he indicated A. halleri for the surroundings of Berlin ("In der Jungfernheide rarissime") where this species is absent, and the Berlin plant which was incorrectly identified by Willdenow was correctly defined as a new species by Ehrhart and described by him as Arundo neglecta. That Ehrhart, when he quoted A. halleri as a synonym of A. neglecta, only meant the Berlin specimen erroneously identified by Willdenow, is evident from the distribution indicated by Ehrhart for A. neglecta: "Patria: Germania, Suecia". This does not at all correspond to that of the mainly Mediterranean species Achnatherum calamagrostis (L.) P. Beauv. (= Agrostis calamagrostis L.; = Arundo halleri Willd.)."

Two subspecies have been accepted by Tzvelev (PAF proposal and elsewhere), and also by Soreng and Greene in Soreng et al. (2003) but with a change in circumscription. Subspecies neglecta (stricta) and subsp. groenlandica co-occur throughout the circumpolar range of the species, with subsp. groenlandica the more northern (and including C. holmii in the circumscription of Soreng and Greene in Soreng et al. 2003). The extremes differ morphologically in several characters. Tzvelev (in comment) stated that there is no clinal transition between subsp. neglecta and subsp. groenlandica. This view was supported by a study of a northwestern North American material (ALA) in 2009 and partly by a similar study of a northern European material (BG, O, TRH, TROM) in 2009-2010.

In northwestern North America, the two proposed subspecies of C. neglecta showed morphological differences and only slightly overlapping (parapatric) ranges. Transitional forms were too rare in these regions for the question of anther and pollen fertility to be answered. In Europe, however, the two subspecies were more difficult to separate when applying the diagnostic characters proposed by the Russians. Some specimens were pollen sterile (aborting anthers) or with both aborting and functional anthers, without indications of hybridization with other species (close relatives are absent from the relevant regions). These plants often combined features of the two proposed subspecies and occurred where these meet. There may be two morphologically similar but reproductively largely isolated taxa within a C. neglecta s. lat. We find support for two races but not (yet) for two species. They are too overlapping in many of their characters.