192 KSSAY XVI. 



Having already learned, from the experiments of others 

 as well as my own, that pure air never constitutes more than 

 one-third of the whole atmosphere, I pasted on the outside 

 of the glass cylinder a piece of paper, which was equal in 

 length to one-third of the contents of the glass, or 11 oz.- 

 measures of water. I divided this paper into equal parts, 

 which I marked with black lines and small ciphers, so 

 that each line showed one - thirty - third part of the space 

 in the glass. The whole piece of paper I varnished over 

 with oil-varnish, in order to prevent the effect of water 

 upon it. 



On the 1st of January 1778, I began my observations. 

 After having filled the vessel (c) with the mixture of iron 

 and sulphur, I inverted the cylinder over it, observing the 

 degree of the barometer and thermometer. The water 

 began slowly to ascend into the cylindrical glass. After 

 eight hours it stopped at No. 9. Though I left the glass 

 for six hours longer in the same situation, the water did 

 not rise any higher. In the meantime the barometer and 

 thermometer had undergone no change. The next day I 

 repeated the experiment, by admitting fresh air into l the 

 cylinder; but the water remained at the same mark. The 

 3rd of January the air was the same. The 4th, upon 

 making a new experiment, the water rose more slowly, and 

 arrived only at its former height in fourteen hours' time; 

 whence I concluded that the mixture in the vessel (c) had 

 lost most of its loosely adhering phlogiston ; I therefore 

 afterwards filled, on occasion of every fourth experiment, 



1 In order to be sure that there may not be any corrupt air remaining 

 from the first experiment, I always fill the vessel with water, Avhereby 

 the air is entirely expelled. This water I afterwards pour out in the air. 

 When I make experiments upon air enclosed in a chamber, I proceed in 

 the same manner, and pour the water out of the glass into a vessel in the 

 same room. 



