862 Report oF THE HorRTICULTURAL DEPARTMENT OF THB 
these seeds were not separated according to specific gravity, but 
trials with other seeds of the same variety showed that all dark 
brown seeds and the larger part of the light brown ones are 
floaters. All pepper seeds at all discolored are low in specific 
gravity. 
Wolffenstein believed that he had found a correlation between 
color of wheat kernels and their specific gravity. The present 
writer has verified this observation. In the case of wheat the 
difference in color is due to differences in gluten content, as 
Snyder* has pointed out. The higher the specific gravity the 
larger the percentage of gluten, which is dark in color. 
RELATION BETWEEN SPECIFIC GRAVITY OF SEEDS 
AND THEIR VIABILITY. 
There is some evidence that a correlation exists between the 
specific gravity of a seed and its viability. For instance, in the 
case of a sample of mustard seed (see Table V), known to be at 
least one and a half years old, the specific gravities ranged from 
1.01 to 1.21; the only germinations which took place, however, 
on the first day of germination were between 1.14 and 1.19. 
Not until the second day after did any germination take place 
among the seeds lying between 1.13 and 1.01, below the optimum 
of distribution, nor in the separates 1.20 and 1.21, above it. In 
all, eighty-nine seeds were set, of which eighty-one lay between 
1.12 and 1.19, and of these forty-three, or 53 per ct., germinated 
the first day of germination, and 66, or 82 per ct. in the entire 
test. Outside of these limits, both above and below, were eight 
seeds, none of which, as stated before, germinated until the 
third day of germination, and in the entire test only two of these, 
or 25 per ct., germinated. 
In the case of some seeds of the Swedish turnip, also old, only 
12 per ct. of those below 1.12 germinated and all of these very 
weakly. Of the seeds in the separates 1.12 and 1.13, 31 per ct. 
germinated and the seedlings were generally vigorous. Above 
1.13 to 1.17 (the maximum) only one germination. occurred. 
Ordinary germination tests showed that the percentage of germ- 
ination in these particular samples of seed was very low. While 
*Minn. Agr. Exp. Sta., Bul. No. 85. 
