326 J9r. Thomas Thomson on the Atomic Weight 



between 7*8 and 8 grains. Now, 7 atoms of water weigh 

 7-875, which is within one per cent, of 7*8, the quantity 

 actually got. I concluded from these facts, that sulphate of 

 nickel contains 7 atoms water ; nor do I think any doubts 

 can be entertained on the subject. It is clear, then, that 

 17*125 grains of crystals of sulphate of nickel contain 

 Sulphuric acid, . . 5* 



Water, 7-875 



What is wanting to complete the sum, must be the weight 

 of an atom of protoxide of nickel. Now, this quantity is 

 4-25. Therefore, 4-25 is the weight of an atom of pro- 

 toxide of nickel, and, consequently, 3*25 is the atomic 

 weight of metallic nickel. 



Such are the results which I still continue to obtain, when 

 the experiment is made in the way above stated, however 

 pure the sulphate of nickel and chloride of barium which are 

 employed. But, when we subject crystallized sulphate of 

 nickel to analysis, the results which we obtain are different. 

 The result, however, of this analysis is quite incompatible 

 with the atomic weight of nickel as stated by Berzelius, 

 which he makes 3-69675. I do not know on what data 

 this number is founded ; for I am not aware of any experi- 

 ments which he has made on the subject. He quotes 

 Rothoff^s experiments, the result of which only, so far as 

 I know, has been published by Berzelius himself. Accord- 

 ing to RothofF's experiment, protoxide of nickel has an 

 ash-gray colour, and is composed of 100 nickel and 26-909 

 oxygen. This is equivalent to 



Nickel, . . 3-716 

 Oxygen, . . 1- 

 So that it does not appear upon what authority Berzelius 

 has made the atomic weight of nickel 3*69675. 



We have an analysis of sulphate of nickel by Mr. Richard 

 Phillips, well known for the accuracy of his experiment&. 

 According to him, it is composed of 



Oxide of nickel, . . 26*30 or 4-669 

 Sulphuric acid, . . . 28*16 or 5* 

 Water, 45*54 or 8*08 



10000=^ 



Annals of Philosophy (2iul Series), vi. 440. 



