378 Dr. F. F. Blackman and Mr. A. M. Smith. [Dec. 19, 
wholesome for the plants than distilled water. A stock of this was 
saturated with CO, which had been carefully scrubbed free from HCl, and 
diluted with more boiled tap-water to obtain the various strengths needed. 
Water cannot be saturated with COz by merely bubbling the gas through it, 
even in a very long time; prolonged violent shaking of the water with 
a large volume of pure COz will, however, achieve this. 
The properly diluted CO--solution is filled into the bottle A whence it can 
flow to the chamber B by the tube 0. As the solution flows out of the 
bottle air is drawn in down the central tube V, and on the principle of 
“ Mariotte’s bottle,” the rate of outflow is constant and independent of the 
level of the liquid in the bottle. 
If laboratory air were to be drawn in this way into the bottle it would 
carry some of the CO: out of solution with it to the top of the bottle and 
the liquid would get progressively weaker. To prevent this the incoming 
air is charged with just as much COz as should be in equilibrium with the 
solution of CO, in the bottle, and alteration of the strength is thereby 
prevented. This addition of CO, to the air-current is accomplished by 
making the air traverse a CO.-generating tower S, on its way to the bottle. 
Hydrochloric acid of determined strength drops regularly from R on to 
the marble in this tower, trickles down and flows away below as a neutral 
liquid at T. The air entering above at (a) carries away the CO» generated 
and is thus enriched to the desired amount before it leaves by the tube U. 
Minor adjustment of the amount of CO2 generated may be made during the 
course of an experiment by altering the rate of flow of the HCl by raising 
or lowering the dropper of the acid bottle by means of a screw attachment 
at the top of 8S. | 
The strength of acid required for any experiment was determined 
empirically and it was found that this whole procedure was quite satisfactory 
for maintaining constancy of CQO:-content in the water flowing into the 
assimilation chamber. The constancy attained in most experiments is 
shown in the successive analyses of samples withdrawn at m at regular 
intervals of time. 
(f) Lstimation of the COz in the Solutions.—The fundamental proceeding of 
the whole method is the determination of the amount of dissolved COs, in 
(1) a sample of the liquid flowing from A to the assimilating plant in its 
chamber B and in (2) a sample of the effluent in D or E after it has passed 
over the plant. The samples analysed were always 200 c.c. of the liquid. 
For (1) a sainple was drawn from the supply bottle by the tube m into 
a 200 c.c. pipette W at any given time and for (2) the 200 ce. of liquid 
accumulated in pipette D or pipette E through which the effluent had been 
