Panarctic Flora

6709 Braya Sternb. & Hoppe

• Sternb. & Hoppe, Denkschr. Königl.-Baier. Bot. Ges. Regensburg 1, 1: 65 (1815).
• Torularia auct., non (Coss.) O.E. Schulz in Engl., Pflanzenreich IV-105, 86: 213 (1924), nec Bonnemaison (1828, algae).
• Neotorularia auct., non Hedge & J. Léonard, Bull. Jard. Bot. Belg. 56: 393 (1986).

Notes: There are several regional studies of Braya (Böcher 1950, 1956, 1966a; Petrovsky 1975a; Harris 1985, 2006) but no circumpolar monographic treatment. Böcher accepted 6-7 species for Greenland: B. humilis, B. intermedia, B. linearis, B. novae-angliae, B. purpurascens, B. thorild-wulffii, and (with doubts) B. glabella. Harris, also reflected by Rollins (1993), accepted four species from the arctic parts of North America: B. glabella (including B. bartlettiana, B. henryae, and B. purpurascens), B. humilis (with varieties, including B. intermedia and B. novae-angliae), B. pilosa, and B. thorild-wulffii. Petrovsky accepted five species, two with subspecies, from the arctic parts of Russia: B. aenea with subsp. aenea and subsp. pseudoaenea, B. humilis, B. pilosa with subsp. pilosa and subsp. thorild-wulffii, B. purpurascens, and B. siliquosa.

Warwick et al. (2003b) studied variation in ITS sequences of ribosomal DNA and trnL intron of chloroplast DNA in Braya worldwide and in related genera. Except for B. pilosa, all relevant arctic and near arctic species were included. This study is a long step towards understanding the genus. It largely supports Harris' comparatively broad species circumscription. Several regional and local species - especially in North America and Greenland - seem to be either fortuitous parts of a more continuous variation pattern or low-level taxa of recent origin, not reflected in the structure of a larger part of the genome. However, before a fully revised worldwide taxonomy is accepted, this study should be supplemented by a more extensive morphological and cytological analysis and by an investigation applying markers more resolving for possible infraspecific taxa. The results of Warwick et al. (2003b) are reflected in the species we fully accept, whereas some of the other taxa proposed are reflected in comments or entered provisionally. The fully accepted species group into the three molecular subclades of Warwick et al. (2003b): one with B. linearis, B. siliquosa, and the B. humilis aggregate, one with B. glabella, B. thorild-wulffii, and probably also including the non-analysed B. pilosa, and one with B. rosea.

Many authors have accepted Neotorularia (previously as Torularia) apart from Braya, with N. humilis as its arctic representative (Böcher 1950; Léonard 1986; Al-Shehbaz et al. 1999). This species has, however, a very different distribution from the other Neotorularias which are African to southwestern and south-central Asian. Warwick et al. (2003b) presented molecular evidence supporting the genus Neotorularia but without N. humilis which nested within Braya s. str. together with two other previous Neotorularias.