A Classification of the Varieties of Cultivated Oats 105 
while employing certain of the classes suggested by Atterberg, referred 
also to the points of the grain and to its ventral groove. 
In the present studies the form of the grain, considered with reference 
to any or all of its characteristics mentioned above, has been found to 
exhibit a marked transitional tendency; and therefore, in attempting to 
organize into groups a large number of varieties on the basis of differences 
in their grain forms, one soon meets with difficulty in determining the 
group relationship of particular varieties. The characteristics of form 
are also very difficult to describe. Thus, certain forms are not accurately 
defined by the terms plump and long-pointed; only the relative condi- 
tions are stated, and the distinction is left to the judgment of the person 
using the classification. 
Being, then, a transitional character and a relative one, the form of 
the grain can have only a very limited use in classification. It has been 
used occasionally in the present work to divide small groups reduced to 
as few members as possible by previous separations on the basis of more 
sharply defined taxonomic differences. 
COLOR OF THE GRAIN 
The color of the grain, or, more definitely, the color of the lemma when 
ripe, has been accorded various degrees of importance in classification 
by other investigators. Kornicke and Werner (1885) used color alone 
to distinguish the main groups of varieties in A. sativa and A. sativa ori- 
entalis. Nilsson (1901), in arranging the Svalof system, made the color 
of the grains and the coordinating form of panicles the main distinction 
for principal types. Denaiffe and Sirodot (1901) characterized main 
groups of- varieties by stating the range in their color; and to describe 
single varieties they made numerous fine subdivisions of color within 
each group. Dufour and Dassonville (1903) believe that color is one of 
the most important characters for the differentiation of groups, but that 
it must be considered en masse rather than in individual grains. Bohmer 
(1908-09) used color only as a final means of distinguishing varieties 
within groups characterized by the form of panicles, spikelets, and grains. 
Fruwirth (1907), also, believes color to be 1 of little importance in classifi- 
cation. Atterberg (1891) mentions it only as a descriptive character. 
In the present classification color is in some cases made the basis for 
the separation of principal groups. It is the most conspicuous character 
