the Effect of Formaldehyde on Living Plants . 4 1 3 
General Experimental Methods. 
The first essential was to grow the plants under as nearly normal con- 
ditions as possible. The whole apparatus (shown diagrainmatically in 
Fig. 1 ) was set up in a sunny conservatory. The cultures, contained in 
three inverted glass bell -jars, were set side by side at the south end of the 
house. For each experiment two controls were used, one with normal air 
containing carbon dioxide, the other with air free from carbon dioxide. 
A continuous current of air was passed through the whole apparatus, so 
that no ‘ damping off’ of the cultures occurred. The organic vapour to be 
tested was introduced into the air-current in minute traces just before it 
was bubbled into the critical culture. A special method of watering was 
used, so that the plants could grow undisturbed in the apparatus for four or 
five weeks. 
Desiccated seeds were used as the starting-point for each culture, 
because their dry weight could be accurately ascertained, and also so that 
the young plants should become accustomed to the treatment from the 
first. When first set they were moistened with pure distilled water ; their 
further requirements were supplied from an automatic waterer, set above 
each culture, in the form of a nutrient solution (containing 0*2 gramme 
each of KH 2 P0 4 , KN0 3 , MgS0 4 , and CaS0 4 per litre). 
After three or four weeks’ growth, the culture without carbon dioxide 
showed signs of flagging, and this condition necessarily stopped the 
particular experiment ; for a vigorous growth of mould or bacteria on 
any of the cultures would have vitiated the weight relations involved. 
All the cultures were therefore removed from their bell-jars and measured 
or photographed. Then they were quickly carried to an oven, heated for 
some hours to 8o°-ioo°, dried in a calcium chloride desiccator, and weighed. 
After subtracting the weight of nutrient salts, calculated from the weight 
of nutrient solution given to the plants, this gave the final dry weights 
of the cultures. These could be compared directly with the original 
dry weights of the seeds used. 
The important difference between these experiments and others on 
similar lines is that here the plants are weighed and put into the apparatus 
as seeds, and growth can continue for some weeks in a continually changing 
atmosphere containing a given amount of any desired vapour. Also, on 
account of the air-current, the vapours to be experimented upon can be 
used in very minute quantities, comparable to the amount of C0 2 normally 
present in air, and the chances of poisoning effects are reduced to a minimum. 
Details of the Experimental Methods. 
The Treatment of the Seeds. 
The seeds were weighed out — after being kept some time in a calcium- 
chloride desiccator— on to small tared glass ‘ Petri dishes ’, about five or 
