36 
There appears to be no difference between A. Gmelini var. Pringlei 
Scribn. & Sm. and A. Scribneri Vas., except that in the former the leaves 
are strigose above, a trivial character which at the very most may be 
employed as a varietal one. The two constitute an apparently good 
species, totally different from Ledebour’s Trilicum caninum var. Gmelini 
and also different from the so-called A. caninum of North America. From 
the latter they differ by having the anthers much longer, about 2*4 mm. 
Pease and Moore further include var, iatiglume, var. andinum, var. 
Hornemanni and its f. pilosifolium. 
Concerning var. Iatiglume, originally described as a variety of violaceum 
Vasey by Scribner and Smith (16, page 30), it is a plant having “empty 
glumes oblanceolate, acute, with broad, scarious margins, short-awned or 
awnless, becoming flat with age; flowering glumes rounded on the back, 
densely pubescent/’ The oblanceolate, or even obovate, broadly scarious- 
margined, 3-5 nerved empty glumes, which are abruptly contracted toward 
the apex, are very characteristic features, although it must be admitted 
that, particularly in Gaspe peninsula, Quebec, forms occur in which 
they are less pronounced. These forms, as Pease and Moore point out 
(13, page 66), appear to connect Iatiglume with A. tenerum Vas. They 
have the empty glumes narrower, less abruptly contracted toward the 
apex, and with a narrower scaiious margin. Nevertheless, the writer is 
inclined to consider Iatiglume a species of its own, distinct from other 
members of the caninum group. It is chiefly arctic-alpine, occurring, 
outside of North America, in Greenland, northern Scandinavia, northern 
Russia, and Siberia. As amended by Vestergren apud Holmberg (3, 
page 272), A. Iatiglume (Scribii. & Sm.) Rydb. includes forms having the 
lemma either glabrous or pubescent. It also includes long-awmed forms, 
one of the latter with glabrous lemma being var. andinum / The forms 
appearing to connect Iatiglume w'ith tenerum in Gaspe peninsula are appar- 
ently analogous to A. Iatiglume subsp. subalpinum (Neum.) Vestergren 
apud Holmberg, a plant less alpine in character than Iatiglume itself. 
What the variety Hornemanni is, is a somewhat involved question. 
When Koch described it, as a variety of Triticum bijlorum (6, page 
953), he cited, as a synonym, T. violaceum Hornem. fl. dan. t. 2044, saying 
also that it is a variety found “in regionibus maxime borealibus,” There 
is, therefore, no doubt but that Koch’s Hornemanni is a Greenland plant. 
Ascherson and Graebner (1, page 654), giving a fuller description of 
Koch’s Hornemanni , state that it is not identical with Hornemann’s T. 
violaceum in Flora Danica. On the other hand, however, they quote as a 
synonym Agropyrum violaceum Lange, Consp. Fl. Groenl. (1880). Pease 
and Moore quote, among other synonyms of var. Hornemanni, Triticum 
violaceum Hornem. and Agropyrum violaceum (Hornem.) Lange. Vester- 
gren in Holmberg (3, page 272) cites Triticum biflorum var. Horne- 
manni of Koch and Ascherson and Graebner as a synonym of A . Iatiglume, 
asserting at the same time that the latter is A. violaceum Lange apud 
Rink (1857), ex parte. 
‘A. Iatiglume var. andinum (Scribn. & Sm.) n. comb. 
A, violaceum (Horn.) Lge. var. andinum Scribn. & Sm., U.S. Dept. Agr. Div. Agrost. Bull, 4, 30 (1897). 
A. biflorum (Brign.) R. & S. subsp. andinum (Scribn. & Sm.) Piper, Bull. Torr. Bot. Club XXXII, 547 
(1905). 
A. andinum (Scribn. & Sm.) Rydb., Fl. Colo., 54 (1906). 
