182 Dr. Thomas Thomsons Observations 



in contact with water. The air was kept in a bottle furnished 

 with a ground stopper, and it stood inverted over water ; 

 the stopper being taken out only when a portion of the air 

 was wanted for use. 



The air was measured in a glass tube shut at one end, and 

 capable of containing rather more than a cubic inch. This 

 tube I had divided into hundredths of a cubic inch, by 

 means of mercury, which I added by r^oth of a cubic inch 

 at a time, and marked the height, of the mercury after 

 every addition, by a fine three-cornered file.* 



I always took, for every experiment, one cubic inch or 

 100 volumes of air. The tube was quite clean, and after 

 being filled, it was allowed to remain in a perpendicular 

 position for three minutes, to give the water time to run 

 down the sides. The air was then transferred to a Volta's 

 eudiometer. The hydrogen gas was measured in the tube 

 precisely in the same manner, and then transferred to the 

 eudiometer and mixed with the air. The mixture was 

 agitated by moving the eudiometer gently backwards and 

 forwards for some minutes, in order to ensure the equal 

 mixture of the two gases. The result of the experiments 

 was as follows. 



When 100 volumes of air were mixed with 40 volumes of 

 hydrogen gas, and an electric spark passed through the 

 the mixture, a detonation took place, and the diminution of 

 bulk, determined by transferring the residue after com- 

 bustion into the tube in which it had been originally mea- 

 sured, amounted to 57 measures. Now, the third part of 

 this diminution amounts to the volume of oxygen gas which 

 has disappeared, — water being a compound of one volume 

 of oxygen and two volumes of hydrogen gases. But the 

 third part of 57 is 19, which denotes the bulk of oxygen 

 gas separated from 100 air. This makes the proportion of 

 oxygen gas in air much less than Davy, Humboldt and 

 Gay Lussac found it to be. 



When the amount of hydrogen gas was made as high 



* The method of proceeding was this : A narrow slip of gummed paper was 

 pasted longitudinally on the tube. After every addition of mercury a tangent line 

 to the upper surface of the mercury was drawn with a pencil across this paper. 

 And these lines were afterwards cut through, and a mark made on the tube by 

 a file. 



