New York AGRICULTURAL EXPERIMENT STATION. 157 
moves the coarser deposits but not the bodies of the bacteria. 
To the filtrate was then added enough 95% alcohol to render 
it alcoholic to the desired degree, usually 80%, the precipitate 
allowed to settle, the supernatant alcohol siphoned off, the 
precipitate, collected on filter paper, washed with either 957% 
or absolute alcohol and quickly dried, partially in a current 
of warm air, then in a desiccator over sulphuric acid. The 
dried precipitate, which is gray and somewhat brittle, was 
then powdered before redissolving in water. It is of course 
important to secure quick drying to avoid the possibility of 
alteration as a result of bacterial growth or of chemical 
changes in the precipitate. The drying must also be done at 
so low a temperature as to, preclude danger of injury from 
heat to the sensitive enzym. In our earlier work we washed 
out the 95% alcohol with absolute alcohol in order to hasten 
the drying. Later it was found this made scarcely any dif- 
ference in the time and no difference in the result, providing 
the moist precipitate was properly broken up so as to dry out 
quickly. Spieckermann (1902: 165) used absolute alcohol fol- 
lowed by ether, presumably to secure quick drying. Most of 
our work had been completed before his paper reached us, 
but we thereupon tested this method in comparison with that 
followed by us and have found it unsatisfactory. The pre- 
cipitate when the ether was used showed a diminution in its 
enzymic activity providing it stood in the ether long enough 
to displace the alcohol. Thus, holding the precipitate in ether 
one hour, while it did not injure it, made no appreciable dif- 
ference in its rate of drying; where in ether fifteen hours, it 
was only two-thirds as active and required almost as long 
for drying; where in ether for twenty-four hours it dried 
quickly, but possessed only one-fourth the activity of that 
dried directly from 95% alcohol. While a solution of the latter 
softened radish and turnip tissues in fifteen minutes, the 
former required one hour to accomplish the same result. 
In further trials chloroform was used with part of the 
precipitate and ether with another part to remove the alcohol 
with a view to hastening the drying. Thus one-half of the 
