60 Existence of Radicals in the Amphide Salts disproved. 
that if, as the author alleges, certain bodies have been classed as 
salts, on account of their similarity in this respect, when dissimi- 
lar they ought not to be so classed. Under this view of the 
question, I propose to examine how far any similitude in proper- 
ties exists between the bodies designated as salts by the author, 
or any other chemist. 
20. The salts, hitherto considered as compounds of acids and 
bases, are by Berzelius called amphide salts, being produced sev- 
erally by the union with one or other of his amphigen class, com- 
prising oxygen, sulphur, selenium, and tellurium, with two radi- 
cals, with one of which an acid is formed, with the other a base. 
The binary compounds of his halogen class, comprising chlorine, 
bromine, iodine, fluorine, and cyanogen, are called by him haloid 
salts. I shall use the names thus suggested. 
21. Among the haloid salts we have common salt and Derby- 
shire spar; the gaseous fluorides and chlorides of hydrogen, sili- 
con or boron ; the fuming liquor of Libavius; the acrid butyra- 
ceous chlorides of zinc, bismuth, and antimony; the volatile 
chlorides of magnesium, iron, chromium, and mercury, and the 
fixed chlorides of calcium, barium, strontium, silver, and lead; 
the volatile poison prussic acid, and solid poisonous bicyanide of 
mercury, with various inert cyanides like those of Prussian blue ; 
likewise a great number of ethereal compounds. 
22. Among the amphide salts are the very soluble sulphates of 
zinc, iron, copper, soda, magnesia, &c., and the insoluble stony 
sulphates of baryta and strontia; also ceruse and sugar of lead; 
alabaster, marble, soaps, ethers, and innumerable stony silicates, 
and aluminates. Last, but not among the least discordant, are 
the hydrated acids, and alkaline and earthy hydrates. 
23. When the various sets of bodies, above enumerated, as 
comprised in the two classes under consideration, are contempla- 
ted, is it not evident that, not only between several sets of haloid 
and amphide salts, but also between several sets in either class, 
there is an extreme discordancy in properties; so that making 
properties the test, would involve not only that various sets in 
one class could not be coupled with certain sets in the other, but, 
also, that in neither class could any one set be selected as exem- 
plifying the characteristics of a salt, without depriving a majority 
of those similarly constituted, of all pretensions to the saline char- 
acter ? 
