1 821 .] of the Proportions of Water, 6;c. 49 



was used in the experiments described by MM. Berzelius and 

 Dulong. The gas was then passed in a convenient apparatus 

 over heated oxide of copper. The water thus formed was, in 

 some cases, received in a small vessel in a liquid state, in 

 order that its purity might be examined. In other instances, it 

 was passed with the excess of gas over muriate of lime. The 

 water was found to be quite pure. The results of three experi- 

 ments gave the following proportions of the gases as entering 

 into the composition of water : 



Oxygen. Hydrogen. 



Exper. 1 88-942 11-058 



Exper. 2 88-809 11-191 



Exper. 3 88-954 11-046 



Giving a mean of 



Oxygen 88-9 



Hydrogen 11-1 



100-0 



The mean of these experiments gives 12*488 of hydrogen to 100 

 of oxygen, instead of 13-27, the number adopted. It is, how- 

 ever, to be observed, that this agrees very nearly with the pre- 

 vious determination of Dr. Thomson, according to which, water 

 consists of 



Oxygen 100-0 or 88-88 



Hydrogen 12'5 11-12 



100-00 



1 : I . 



Having thus discovered an error in the usually stated compo- 

 sition of water, the authors proceeded to take the specific gravity 

 of oxygen and hydrogen gases. 



The authors then notice the fact observed by Mr. Dalton, that 

 when any gas which is insoluble in water is kept over it, the 

 gaseous contents of the water rise and contaminate the gas ; 

 and to this cause they think it probable that the error of MM. 

 Biot and Arago may be ascribed. In order to avoid this incon- 

 venience, MM. Berzelius and Dulong covered the surface of the 

 water with a stratum of oil. 



In taking the specific gravity of oxygen, hydrogen, azote, and 

 carbonic acid, the method adopted was to take the weight of the 

 exhausted flask immediately after each weighing of the full flask. 



The oxygen gas used was prepared from chlorate of potash ; 

 and to separate any carbonic acid which it might contain, it 

 was passed over a strong solution of caustic potash. The car- 

 bonic acid was obtained by nitric acid from white marble, and 

 passed over powdered crystals of carbonate of soda before it 

 entered the receiver. The azote was procured by decomposing 



New Series, vol. ii. e 



