Echinochloa frumentacea Link, Hort. Berol. 1: 204. 1827.

Hoste, Ivan & Verloove, Filip, 2022, Taxonomy of the weed species of the genus Echinochloa (Poaceae, Paniceae) in Southwestern Europe: Exploring the confused current state of affairs, PhytoKeys 197, pp. 1-31 : 1

Echinochloa frumentacea Link, Hort. Berol. 1: 204. 1827.

Type, lectotype designated here.

India, Roxburgh s.n. (K000215131, the specimen on the extreme right on the sheet). Image available at http://specimens.kew.org/herbarium/K000215131.

Note.

The protologue refers to a Roxburgh collection from India ("Roxb. ind. 1. 307. R. S. m. 2. 250. Hab. in India orientali ubi colitur"). The Kew herbarium houses two original but undated Roxburgh collections (sheets K000215131 and K000215132) that can serve for a proper typification. None exactly matches the information provided in the protologue, but since Link described the species in 1827, i.e. well after Roxburgh’s (1751-1815) death, these collections are supposed to have been at his disposal when describing the species. In the apparent absence of other original material, one of the two above-mentioned Kew collections could be chosen as the lectotype for that name. Digital images of both are easily accessible via online resources such as the Kew Herbarium Catalogue, JSTOR or POWO. Sheet K000215131 comprises five stems, four of which have an inflorescence. The extreme left specimen is atypical and might as well represent a different species. The other flowering specimens are representative for the species, and the specimen on the extreme right is here designated as the lectotype for the name E. frumentacea . According to Stafleu and Cowan (1983) considerable sets of duplicates of Roxburgh specimens are stored at BM, BR, E, G and LIV. In some of these herbaria isolectotypes could thus be found although a quick online search did not yield further specimens.

Remarks on E. esculenta and E. frumentacea .

Echinochloa esculenta (syn.: E. utilis Ohwi & Yabuno) and E. frumentacea are cultivated species. Neither is considered a persistent weed in Southwestern Europe. Still, they are included in the key since they are the most frequently occurring non-weedy representatives of the genus in Southwestern Europe, frequently recorded as bird-seed aliens in and along the border of crop fields ( Hanson and Mason 1985). They look similar, and young specimens can be difficult to identify, yet mature ones are easily distinguished by the colour of the spikelets. In both species, the inflorescence varies. In the more typical specimens, the branches are tightly clustered and appressed against the axis, creating a compact inflorescence. Often, however, the inflorescence is rather lax, with the distal part of the branches somewhat curved towards the axis; such specimens are easily mistaken for an awnless form of E. crus-galli . Yabuno (1966) describes the distinct characteristics of the two species, and recent genetic studies have confirmed that they are quite distinct, E. esculenta being derived from E. crus-galli and E. frumentacea from E. colona ( Yamaguchi et al. 2005; Ye et al. 2014). There are arguments for reducing these two taxa to variety rank or, following Banfi and Galasso (2021), subspecies rank under E. crus-galli and E. colona . Yabuno (1966) insinuated that E. esculenta shows more variation, and Michael (1983) added that in this species, the spikelets may be awned (although awned spikelets seem to be rare); this reflects the highly polymorphic nature of the parent species.