the Effect of Formaldehyde on Living Plants. 417 
was then divided equally between the two cultures by allowing it to fall 
into the opening of a ‘divider’, i. e. a symmetrically bent glass tube open 
at both ends, and with another opening blown out at its apex (Text-fig. 1, 
Divider). The amount of nutrient solution given to the cultures could be 
accurately determined by weighing the waterers before and after each 
experiment. 
The Air Supply . 
A continual stream of air was kept passing through the apparatus, 
which prevented the ‘ damping off ’ of the cultures, so that healthy growth 
was maintained throughout. The importance of this in physiological 
experiments has been emphasized by F. F. Blackman. 1 In order to prevent 
any chance of the entrance of C 0 2 through a possible leakage, and also to 
facilitate smooth running, pressure was used to drive the air through the 
apparatus, instead of suction. This is easily obtained from an ordinary 
water-suction pump by allowing the water and air drawn through with it to 
run through a trebly bored rubber cork into a bottle. The air is then 
forced out through a glass tube at the top of the bottle, while the water 
runs out through a siphon-tube reaching to the bottom. The use of 
a siphon-tube, instead of an outlet-tube worked with a tap, is to allow for 
variation in the water-pressure. The siphon adjusts the level of the water 
in the bottle, automatically, through a considerable range of pressures 
(Text-fig. 1, Pump). 
The Purification of the Air-current. (See Text-fig. 1.) 
From the pump the air was led first through an empty ‘safety’ bottle, 
then through calcium chloride to dry it, then through soda lime, and finally 
through two bubblers, filled with strong potash solution, to free it from 
carbon dioxide. As a final precaution, the bell-jars containing the cultures 
without carbon dioxide were inverted in troughs of dilute potash, and the 
entering air was bubbled through this. By adjustment of the screw clips 
next to the two glass T-pieces, the air could be divided equally between 
the three cultures, and was passed simultaneously through all the bell-jars. 
At one time, two experiments were running together, using five bell-jars, all 
worked from the same pump. The pressure required is small, provided the 
whole apparatus is air-tight. 
The Introduction of Small Traces of Formaldehyde Vapour into the 
Air-current. 
This was effected in three ways : 
(a) The air was bubbled through ordinary commercial ‘formalin’, which 
is an aqueous solution of formaldehyde of about 40 per cent, concentration, 
1 Blackman : Phil. Trans. Roy. Soc., clxxxvi, 1895, B, pp. 485-502. 
