380 REPORT OF THE HorTICULTURAL DEPARTMENT OF THE 
Buckwheat seeds of high specific gravity, that is, above about 
1.18, are not large but are very plump and firmly closed at the 
end. The testa is completely filled by the kernel and the two are 
separated with comparative difficulty. Beginning at about 1.18 
and thence downward, the kernel does not fill the seed-coat com- 
pletely and the latter is much more easily removed. From about 
1.07 down the difference in size between the volume of the ker- 
nel and that of the inside of the testa becomes marked, and the 
seed-coat is very easily separated from the kernel. Seeds of a 
specific gravity less than 1.00 have a very shrunken appearance 
and the kernel is very small. Kernels from seeds of high spe- 
cific gravity are very firm and withstand considerable pressure — 
as between the fingers — without breaking; but the firmness of 
the kernel gradually decreases with the specific gravity of the 
seed, and the kernel from a seed of specific gravity less than 
1.00 is so soft that it must be handled very carefully indeed, in 
order to get it out of the seed coat without breaking it. Further, 
the kernels of seeds of high specific gravity are solid all the way 
through but those of low specific gravity are not. In the latter 
there is more or less of unoccupied space surrounding the em- 
bryo which is in the middle of the endosperm and extends from 
point to base. 
From these observations it would appear that differences in 
structure are abundantly sufficient to account for the observed 
differences in specific gravity in buckwheat, granting even that 
the chemical composition of all of the seeds is the same. In this 
case specific gravity should be set down as of no theoretical value 
in seed selection as applied to buckwheat. This conclusion is 
supported by a single culture test which the writer made on 2 
very small scale. In this seed test no conspicuous correlation, 
if any at all, was found to exist between the specific gravity of 
the seed and its germination; and, while no measurements were 
made, the plants from seeds of lowest specific gravity appeared 
to be as good as those from the highest. 
Other seeds also of low specific gravity are in many cases 
found to present some peculiar characteristic. For instance, in 
the case of peas it has already been remarked that some seeds of 
abnormally low specific gravity contain considerably more moist- 
ure than do normal air-dry seeds. In another connection some 
