on the A tomic Weigh ts of Bodies . 191 



was not sufficiently understood, and, of course, was not 

 guarded against. My results with pure dry hydrogen gas 

 were as follow : 



1st Trial . . . 0-06954 



2nd Trial . . . 0-06933 



3rd Trial . . . 0-06933 



Mean .... 0-0694* 

 Berzelius and Dulong made experiments on this subject 

 during the same year, and they state the specific gravity of 

 pure dry hydrogen gas, at 0*0687. But, from other con- 

 siderations, they pitched upon 0-0688 as the true specific 

 gravity. 



Now, it is admitted on all hands, that water is a com- 

 pound of one volume of oxygen gas, and two volumes of 

 hydrogen gas. If we take my specific gravity, this gives 

 us for the constituents of water by weight, 



Oxygen, . . 1-1111 or 1 

 Hydrogen, . 0-0694 + 2 or 0-1249. 

 This number is within less than xoVo^ P art of 0-125. 



Berzelius and Dulong's specific gravity of hydrogen gas, 

 gives us the composition of water, 



Oxygen . . . 1* 

 Hydrogen . . 0-1238 

 But, Berzelius from other considerations, (thus tacitly 

 giving up the accuracy of his specific gravities,) has adopted 

 0-124796 (or the half of that number, which comes to the 

 same thing) as the weight of an atom of hydrogen, while 

 Dr. Turner adopts my number, or 0-125, as representing 

 the true atomic weight of that body. 



Let us now see what conclusions can be formed from the 

 experiments made, to determine the constituents of water 

 by direct combination. By far the most accurate experi- 

 ments on this subject, are those of Berzelius and Dulong. 

 They passed a current of hydrogen gas through a glass tube 

 filled with black oxide of copper, and heated by a spirit 

 lamp. The hydrogen gas was dried before reaching the 

 oxide of copper, by passing it through a tube filled with 

 fragments of fused chloride of calcium, and the water 

 formed was collected in another glass tube, also filled with 



* Annals of Philosophy, xvi, 168. 



