the Effect of Formaldehyde on Living Plants. 429 
through this bulb after leaving the cultures and the drying-tube. To 
prevent a possible backrush of air containing carbon dioxide, another 
potash bubbler was connected on to the exit-tube of the weighed Liebig 
bulb. It was found that the potash solution in the Liebig bulb had to be 
absolutely saturated, at room temperature, and the calcium chloride fresh 
and perfectly dry, or water escaped. If these precautions were observed, 
and the weighed apparatus was kept covered from dust, results could be 
obtained quite as accurate as the weighings of the seeds with which they 
were to be compared. 
The Absorption of Water. 
The quantitative determination of the water relations of the plants 
was a very fascinating problem, but the practical difficulties involved were 
so great that no really satisfactory method has so far been found. The 
observations are, however, sufficiently accurate to indicate the general 
nature of the relations. 
The first important point is to keep the same water-absorbing medium 
throughout the apparatus, so that the air should not be more efficiently 
dried at one point than another, and so take up water from a reagent which 
is supposed to dry it, as is the case, for example, when air dried in sulphuric 
acid passes over calcium chloride. Calcium chloride was chosen as this 
drying reagent, not because it is the most efficient, but because it is clean 
and convenient, and gives no possibility of acid gases, even when acted 
upon by organic vapours. 
Before commencing each experiment every precaution must be taken 
to exclude water from the whole apparatus. The desiccators, tap grease, 
tubing, &c., must all be dried with calcium chloride before use. 
The first absorbent for water is a calcium chloride U-tube next to the 
desiccator containing the culture. This was kept open towards the desic- 
cator, even when no air-current was passing, in order to keep the exit- 
tube dry. 
In a few preliminary experiments it was found that while the U-tube 
absorbed about 1*5 grm. of water, about 2 grm. were lost in some way. 
The only possibility of loss was by condensation on the walls of the 
desiccator. To obviate this a second absorbent for water was made. This 
was a small ‘ house ’ of asbestos board, made to fit into the desiccator, which 
held the waterer and culture and had a hole in the roof for the air-tubes. 
The ‘ house 5 was varnished over the outside with shellac varnish, so that it 
could be weighed before and after each experiment. The rough inner 
surface of the ‘house’ absorbed much of the water of transpiration. 
A third measure of water was by the original and final wet weighings 
of the seeds. For these weighings, in which a free evaporating surface of 
water was present, special methods had to be employed. 
