348 ‘EXPERIMENTS ON THE RELATION 
requires no water of composition, that we can ever expect to 
obtain the combined or latent water of sulphuric acid, or of 
muriate of ammonia. By this general principle, all my efforts 
have been directed. 
I shall now endeavour to demonstrate, by particular experi- 
ments, the accuracy of these views. 
1, Sal ammoniac was put down to the sealed end of a glass- 
tube; ‘over it was placed a few inches of river sand, which, af- 
ter having been digested with muriatic acid, had been well 
edulcorated with water, and ignited, in a platina crucible. On 
passing the vapour of the salt, through the ignited sand, liquid 
of a reddish-brown colour was copiously condensed in the pro- 
jecting and cool part of the tube. This liquid was acidulous 
muriate of iron. I naturally ascribed the origin of this to the 
red oxide of iron, still contained in the sand after the above 
operations. 
2. To verify this idea, 1 then took ‘pounded flints, and ob- 
tained from the same quantity of salt, in various repetitions of 
the process, a very small quantity of blue liquid, which seemed 
to be an ammoniacal solution of copper, by its colour and 
smell. 
3. I next had recourse to quartz nearly pure. The quanti- 
ty of liquid obtained in the same circumstances, was now very 
inconsiderable indeed ; and I conceived it might be ascribed to 
some interspersed particles of mica and felspar, whose alu- 
mina might fix a little of the dry acid, and leave water of am- 
monia. 
4, When pure rock-crystal was employed, the aqueous pro- 
duct became almost evanescent. The salt sublimed through 
the ignited quartz powder, without any apparent decomposi- 
tion. 
5. When 
