198 J. M. Ordway on Sesquisalts. ras 
amorphous, which, in reality, admit of very distinct cry 
ti 
Nor is it altogether a matter of indifference what precipitant 18 
used; for under certain influences m the sesquioxyds are 
liable, even in the hydrated condition, to assume an inert allo- 
tropic state. For iron, chrome, and cerium, ammonia does well. 
But to throw down alumina, it is best to use carbonate of ammo- 
nia or an alkaline sulphuret. 
In the second place, it is sometimes possible by direct oxyda- 
tion to make a soluble salt with less than the normal quantity © 
acid. Thus, in cold weather, 2Fe, and 3N.Os may be made toyield 
N Oz and ¥e fz. When crystals of protosulphate of iron are put 
into half their weight of nitric acid of sp. gr. 1, 2, a rapid oxyda- 
tion ensues, with the evolution of nitric oxyd, and the resulting 
dark red liquid must contain three equivalents of perox d 
iron, held in solution by six equivalents of sulphuric acid ant 
one of nitric. Another instance is afforded in the change of aes 
protochlorid of tin into a soluble bibasic perchlorid, by meats — 
of chlorate of potash. ee 
ird plan, applicable only in some rare cases, is to expel a 
portion of the acid from a normal salt by a moderate and prd- 
tracted heating. Nitrate of alumina erystals exposed for several 
hours to a steam heat, were found to have lost nearly half their 
acid, the perfectly soluble residue having about the compost 
tion Al, 8;+Hs3. A solution of chlorid of aluminium by evapo 
ration to dryness, parted with one-third of its acid without losing 
its solubility. : 
___ Fourthly, we may often operate advantageously by precipita 
tion. A portion of the acid can be removed by another 
hole 
way is to neutralize part of the acid of 4 
A og Spi part ox tnt .g left in 
