ANDREWS: PROTOPLASMIC STREAMING IN Mucor 461 
the vessel R’, containing R, in a small well-regulated thermostat. 
Unless the generator A is very large it is rarely the case that a 
gas generated by it would be forced through the apparatus shown 
in FIG. 1. In order to make this possible an aspirator, W, is also 
attached and the flow controlled by astopcock, W’. An aspirator 
may be graduated to estimate the outflow and gas drawn through, 
or the stopcock W’ may be adjusted by means of a meter used 
on the outflow stopcock as shown by Detmer.* 
If strong suction is not required a constant flow of any gas 
through the gas chamber U’ may 
be obtained by the use of a float- 
ing siphon shown by Fic. 1a taken 
from the work of N. W. Lord.+ 
The figure shown here is about } 
the size of the apparatus I con- 
structed for this work. It can not 
be used with the apparatus shown 
by FIG. 1. but can be employed 
when the gas has to pass through 
only a shallow liquid, or if gas only 
is to be drawn through. Its value 
lies in the constancy of the flow of 
gas it will cause, while other ap- 
Paratus used for this purpose must 
be continually watched and regu- 
lated to insure accurate results. 
In some cases a suction pump, such 
as are attachable to water pipes, is 
an advantage to use. Wherever 
Possible, and this was generally the 
fase, all glass connections were made by having the tubes fused 
together in one continuous piece. (FIG. 1.) This made leak- 
age impossible, which with hydrogen is very difficult to prevent 
under ordinary circumstances. The other connections, as for 
*xample with the generator, U-tubes, and the gas chamber, were 
made by means of rubber stoppers and sealing wax according to 
Fic. 1a. Floating siphon used to 
draw through a constant stream of 
gas. 
*Detmer, W. Pflanzenphysiologische Praktikum 821. 1905. 
tLord, N. w. Notes on metallurgical analysis 181. 1903. 
