94 
W. C. Etheridge 
III 
A. Panicles spreading, oval, irregular, short, with short, upstanding branches; apex of panicle 
short and somewhat inclined; culms usually weak; ripening period seasonable; outer 
grains open, mostly concave on their inner side, short, and blunt-pointed, the points 
weaker than in the pointed-grain forms; single grains entirely, or almost, closed and 
less numerous than among the "barley" types. 
a. Bright grains. 
b. Dark grains. 
Side-panicle varieties with similar forms of grain. 
B. Panicles and culms similar to those of A, but larger; ripening period also as in A; grains 
large, thick, plump, and closed or almost closed, concave on their inner side, and with 
slim, weak points; many double-grains. 
a. Bright grains. 
b. Dark grains. 
Side-panicle varieties with similar forms of grain. 
IV 
A. Panicles long, slim-pyramidal in form, sparsely branched, all branches short, the main 
ones horizontal or loose-hanging; culms generally weak; ripening period seasonable 
to late; 1-3-grained spikelets; outer grain widest at upper end of rhachilla, and with 
sharp, stiff, closed or almost closed, points. 
a. Bright grains. 
b. Dark grains. 
B. Panicles and culms similar to those of A; ripening period very early; 2-1-grained spike- 
lets; grains fine-hulled, short, cylindrical, wide open, with blunt, short points. 
a. Bright grain*. 
b. Dark grains. 
Bohmer's classification cannot be said to distinctly differentiate groups 
of varieties. The panicle types adopted from Nilsson's classification 
are, as pointed out in the discussion of that system, lacking in the dis- 
tinctiveness necessary for accurate identification; and the descriptions 
of grain forms selected from Atterberg's classification serve only to char- 
acterize in the most general terms the appearance of grains without dis- 
tinguishing their morphological differences. Bohmer's system, therefore, 
does nothing more than present somewhat indefinite groups of vaguely 
characterized varieties, and it is in no respect a usable system for 
identification. 
SUMMARY 
From the foregoing discussions the outstanding features and the use- 
fulness of previous systems of classification may be briefly summarized 
as follows: 
1. The system of Kornicke and Werner, based on morphological differ- 
ences of panicles, spikelets, and grains, is competent for the distinction 
