A Classification of the Varieties of Cultivated Oats 113 
character for the same purpose, but Fruwirth (1907) considers it of no 
value. While under given conditions a distinction in the color of the 
glume may be observed, the difference is slight and is largely dependent 
on the stage of ripening at which the observation is made and on the 
climatic environment; therefore the character is an inconstant one and is 
not worthy of use in classification. 
PHYSICAL PROPERTIES OF GRAINS 
Although the absolute weight of the grains, and their proportion of kernel 
and hull, were used in classification by Atterberg (1891) and by Denaiffe 
and Sirodot (1901), no other investigators have employed these characters 
for the separation of a large number of varieties. Bohmer (1908-09) 
argues that the physical properties are too easily influenced by conditions 
of climate and soil to be reliable characters in classification. His extensive 
data show not only that in the weight of grains there was from year to 
year a greater difference than was covered by the classes of Atterberg 
and the French authors, but also that in a given year the position of the 
grain in the spikelet would determine its group relationship in weight 
and kernel content. Thus, in respect to these characters, outer grains 
would fall into higher classes than inner or middle grains. There is an 
abundance of other data to support Bohmer's conclusions. Fruwirth 
(1907) shows the same wide difference between outer and inner grains in 
weight and in proportion of kernel, and he also finds that both properties 
are influenced by the position of the grains in the panicle; grains borne 
by the upper branches are heavier and richer in kernels than those borne 
by the lower branches, on which there are a greater number of sterile 
flowers. Lippoldes (1904) shows further a difference between the weight 
of grains from various stems of the same stool. 
As to the effect of soil and moisture on the physical properties of the 
grains, Biinger (1906) found that poor soils, low in moisture, produced 
light grains, small kernels, and many sterile spikelets. The effect of an 
extreme range in seasons on the weight and the kernel content of the oats 
grain has been shown by Seton (1903), Edler (1905), and Berry (1912), 
all of whom found that much larger and plumper grains were produced 
in cool, moist seasons than in hot, dry seasons. Berry, who classified 
the grains of oats into several types on the basis of physical properties, 
believed that the distinction between his classes might be greatly modified, 
