THE SOUECES OF THE NITROGEN OF VEGETATION, ETC. 487 



of the vegetable cells, dependent on the character and conditions of the carbon com- 

 pounds they contain, was sufficient, under the circumstances specified, to consume all 

 the oxygen (or ozone) that might be present. But the high temperature at which the 

 experiment was conducted must have tended very much to increase this action. In 

 subsequent experiments a different plan of operation was adopted, not open to the same 

 objection. 



Experiments, Series 2. 



In these experiments, as in all those subsequently referred to, the plants were put into 

 a tall glass vessel (fig. 7, Plate XII.) 1-75 inch in diameter, and 14 inches in height. 

 The mouth of this vessel is fitted with a long cork, previously well boiled in bees'-wax. 

 Through the cork, two glass tubes, a and b, are inserted. The vessel being filled with 

 water well boiled and then cooled without access of air, the plant is put in and well 

 shaken to remove adherent air-bubbles. The cork, with its two tubes, is then forced 

 in, taking care that both the tubes become filled with water and that no air remains 

 in the vessel. As a further security for tightness, a piece of wide and thick caoutchouc 

 tubing may be drawn over the neck of the vessel, projecting upwards a little above the 

 cork, and then the cup thus formed partly filled with melted wax, forming a layer over 

 the cork and its joints. A funnel is then attached to the tube b, by means of a caout- 

 chouc tube which can be closed by a strong pinch-cock. Water being admitted through 

 the funnel into the tube b, the tube a becomes filled, and it is then brought into con- 

 nexion, by means of a glass tube and caoutchouc joint fitted with a pinch-cock, with a 

 vessel filled with quicksilver. The connexion being opened, the quicksilver is allowed 

 to flow from the vessel by means of a long tube of more than barometric length fitted 

 into the lower part of it, thus forming a Torricellian vacuum in the mercury vessel. The 

 gas from the plant passes over into this vacuum, and by a simple arrangement is collected 

 in a eudiometer tube for examination. 



The following Table shows the amount and composition of the gas obtained from 

 different plants, in the shade, in the manner above described. 



mdccclxi. 3 x 



