TRANSACTIONS OF THE SECTIONS. 43 



the only two possible methods of explanation, we end by admitting the 

 production and presence of a true haloid salt, the terbromide of gold. 

 It cannot be declared possible that the bromide of gold, as soon as 

 formed, has decomposed water, and become an hydrobromate, since, 

 by reference to the lower diagram, it will be seen that the production 

 of the bromide, according to the Amphide theory, implied the re-solu- 

 tion of hydrobromic acid and oxide of gold into water and bromide of 

 gold. The production of water was declared essential to the change ; 

 its decomposition cannot therefore be assumed as equally possible, since 

 the previous half of the explanation would thus be negatived, and the 

 theorist should only exclude the bromide from the list of haloid bodies 

 by admitting the chloride, or vice versa. The nature and value of this 

 mode of experimenting is summed up in the following proposition : The 

 dissolved terchloride of gold is decomposed by hydrobromic acid, with 

 the production of terbromide of gold, in circumstances which make it 

 certain that one of these salts is a haloid in solution : but if one of 

 these be a haloid the other is also, and so are all the salts of the same 

 class. 



The demonstration is further shown to be afforded in another way : 

 it will be seen, by reference to both diagrams, that each view recog- 

 nises the presence of bromide of gold and hydrochloric acid. But if the 

 bromide became a hydrobromate, it would at once be decomposed by 

 the hydrochloric acid ; it is permanent, however, and must therefore be 

 a bromide while in solution. 



A similar inquiry was stated to have been followed out with the 

 haloid salts of platinum, and with equally decisive results ; and the 

 author observed that the method proposed would aiford an equally un- 

 equivocal demonstration for the haloid salts of all the metals which have 

 a less attraction for hydrogen than the radicals of the hydracids have ; 

 i. e. it would apply to all the electro-negative metals. It was further ob- 

 served, as an incidental conclusion from the experiments recorded, that 

 they afforded a direct proof of the quasi-metallic character of hydrogen, 

 so much insisted on by the advocates of the binary theory of salts ; and 

 that they supplied more direct evidence than any previous trials re- 

 garding this, since they not only demonstrate hydrogen to have the 

 power of displacing many metals, but at the same time assign to it, as 

 its proper place in its metallic character, a position intermediate be- 

 tween the electro-positive and electro-negative metals. 



On the limits within v)hich the Atomic Weights of Elementary Bodies 

 have been ascertained. By Thomas Clark, M.D., Professor of 

 Chemistry in Marischal College and University , Aberdeen. [This 

 paper is given entire at the recommendation of the Committee of the 

 Chemical Section, confirmed by the General Committee.] 



The atomic weights of elementary bodies, as they occur in the re- 

 ceived tables, commonly represent the mean result, or a selected result, 

 of the experiments on which they are founded. It may be now useful, 



