342 REPORT OF THE HorTicuLTURAL DEPARTMENT OF THE 
that the seeds were often found to have changed materially 
in specific gravity before the determination could be completed. 
Various devices have been tried with a view to preventing this 
absorption of water, but apparently without entire success, since 
none of them have become a recognized part of practice. One 
of these methods, suggested by Wollny, was to oil the hands 
lightly and to roll the seeds between them before making the 
determination. In reviewing the literature of the subject the 
present writer has not noticed that this practice has been con- 
tinued by any other investigator. The objection to it is, that 
more or less air is imprisoned and the results vitiated. Another 
device is to coat the seeds with shellac or varnish. In a series 
of careful determinations it was found that this treatment of the 
seeds changed their specific gravity slightly, though not enough 
presumably to be of importance in practice. Obviously, seeds 
thus treated would be valueless for use in culture tests. 
In order to obviate the practical difficulty involved in the use 
of water, various other liquids have been employed. Among 
these are alcohol, naptha, benzine and petroleum. <A quite 
extended study was made by Wolffenstein of the subject of media 
for use in making pycnometer determinations. He it was who 
first suggested the use of petroleum. The employment of this 
medium obtained much currency among investigators. The super- 
iority of this fluid for use in making specific gravity deter- 
minations lies in the fact that the seeds take up almost none of it. 
A very considerable objection to it is, that it changes much in 
volume and consequently in specific gravity with even small 
changes in temperature, so much so that corrections must be 
made for temperature changes so small as one-half degree. 
One practical difficulty which experimenters have to contend 
with, is the adherence of air to the seed when immersed. In 
order to obivate this difficulty Nowacki, after having put seeds 
in petroleum in a pycnometer, placed the apparatus under the 
receiver of an air pump and exhausted the air uniformly to a 
stated pressure, allowing the apparatus thus to remain for fifteen 
minutes, after which the determinations were made. In thus 
treating the seeds a little petroleum was absorbed, but the error 
therefrom is stated not to be of importance, not amounting to 
more than three or five per ct. according to Nowacki. 
