122 Mr Johnston on the atomic cofistitution 



Carbonic acid. Azote. Cyanogen. Atomic proportion. 



No. 1, as inches 1.78 1.9 6.6 



2, 3.75 1.88 1.875 6.5 



3, 3.62 1.78 1.81 6.21 



4, 3.67 1.9 1.835 6.32 



Mean atomic proportion, 6.407 



These results agree as much among themselves, and come 

 about as near the truth as by either of the former methods. 

 The azote also differs a little from half the volume of carbonic 

 acid. 



4. Let us now take the mean result of the whole three me- 

 .thods, and we shall probably not be far from the truth. 



Mean result by peroxide of copper — 6.54 

 _ mercury =z 6.68 



chlorate of potash = 6.407 



of the whole = 6.54 



That is to say, 25 grains of mercury when converted into 

 cyanide are combined with 6.54 grains of cyanogen by ex- 

 periment, it is obvious, therefore, that the true composition of 

 the salt is 



Mercury one atom = 25 

 Cyanogen two atoms = 6.5 



And the atom of bi-cyanide weighs 31.5 



This result leads us to another analogy between chlorine 

 and cyanogen. The fti-chloride like the 6i-cyanide of mercury 

 is a soluble salt, while the proto-chloride (calomel) is nearly 

 insoluble. It is probable, therefore, that there is also an in- 

 soluble ^roto-cyanide not hitherto met with. In a note to a 

 paper on the carburets of azote ^ published in this Journal, I 

 have mentioned a series of insoluble compounds, which may 

 possibly prove to be pro^o-cyanides. 



5. This constitution of the salt may be verified by estimat- 



