io8 
William Seifriz 
A io per cent, solution of alcohol will kill nearly all of the cells of 
an Elodea leaf within two hours (80 per cent, are killed within one 
hour). The time is too brief in which to arouse the cells to activity 
before death ensues. In 8 per cent, alcohol several hours’ treatment 
is usually enough to arouse the cells to activity. Eighteen to twenty- 
four hours is sufficient in a 5 per cent, solution, while from two to 
three days is necessary in 3 per cent, alcohol. The lower grades, viz. 
3 per cent, to 5 per cent., with long treatment are more satisfactory. 
In a leaf which has undergone such treatment about one-half of the 
cells will exhibit active streaming. In some instances I have observed 
nearly every cell in the leaf, with the exception of the few which have 
been killed, in an active state of streaming. In many of them the 
chloroplasts are racing around in the cell at a remarkably high 
speed. 
While the amount and rate of streaming is very pronounced in 
alcohol treated cells, the most interesting feature of this stimulation 
to protoplasmic activity is the great variety of abnormal types of 
streaming. The usual type of streaming in a normal cell, as seen in 
optical section, is a movement of cell inclusions around the inner 
surface of the cell wall in a plane parallel to the surface of the leaf, 
and ordinarily all chloroplasts take part. In an alcohol treated cell, 
however, the chloroplasts may be grouped in a variety of ways. 
One very common type of streaming in these alcohol treated cells 
is that in which the chloroplasts are arranged in the form of a belt 
encircling the protoplast at its centre in a plane perpendicular to the 
surface of the leaf. When the belt is broad the protoplast resembles 
a revolving cylinder. At times the belt consists of two apparently 
independent streams moving at different rates. Again, there may 
be two such belts of chloroplasts, one at each end of the cell situated 
on the very tip of the plasmolysed protoplast, or intermediate to the 
ends in a plane oblique to the cell axis. What is especially interesting 
is that two such belts in a single cell revolve, in some instances, in 
opposite directions. 
The second reagent in which I observed a stimulation of proto¬ 
plasmic streaming in Elodea cells was saponin, including the similar 
glucosides, smilacin and senegin. The results are very like those just 
described for alcohol; namely, a great amount of streaming, much of 
it at a remarkably high rate, and many abnormal types. The saponins 
used in my experimentation were all chemical preparations; but for 
obtaining a saponin suitable for treating Elodea leaves in order to 
arouse the cells to activity it is convenient to extract the saponin 
