290 Kreider— Quantitative Determination of Perchlorates. 
In (19) a mixture of cadmium iodide and potassium iodide 
taken in the proportion of their molecular weigats was substi- 
tuted for the zine chloride. The known salt corresponding to 
the formula CdlI,.2KI+2H,O was not so convenient because 
of its high melting point —230° C; but when the two iodides 
are taken in the proportion of their molecular weights, the 
mixture fuses at about 200° C. Although this mass after 
‘fusion was more easily soluble than the zine residue, the blank 
determination revealed a source of error equally disparaging. 
Gaseous hydriodic acid passed into a mixture of the per- 
chlorate and meta-phosphoriec acid at a temperature between 
200° and 300° C, was markedly less effective than the genera- 
tion of the acid on the spot; and the distilling of the perchloric 
acid by meta-phosphoric acid into a receiver of potassium 
iodide yielded only a trifling amount of iodine, while the pas- 
sage of hydriodic acid over the fusing perchlorate in a short. 
combustion tube was precluded by the high melting point of 
the perchlorate endangering the dissociation of the halogen. 
The invariably high results obtained by the use of meta- 
phosphoric acid in all those experiments in which there was 
a reasonable assurance that the hydriodic acid was held till the 
perchlorate was completely broken up—experiments (13), (14), 
(15), (17), (18) and (19)—indiecated either a dissociation of 
hydriodic acid or a partial reduction of the meta-phosphoric 
acid. Of the latter cause there were some grounds for sus- 
picion, but as its determination led too far from the object of | 
the investigation, the use of phosphoric acid was abandoned. So 
far as our experience extended there remained then nothing 
among the reagents of the wet methods which was sufficiently 
active and stable enough to warrant its application. Fusion 
alone seemed capable of extracting the oxygen from the per- 
chlorate. A number of preliminary tests were therefore made 
on certain salts of the halogens, in the hope of finding one 
which would be acted upon by the oxygen of the perchlorate 
with the liberation of the halogen, which latter could be col- 
lected in a receiver of potassium iodide and titrated with thio- 
sulfate. 
The double chloride of aluminum and sodium, A1,Cl,.,NaCl, 
melts at about 200° C, and was in other respects desirable. 
When fused with potassium perchlorate, chlorine was copiously 
evolved. The action of air on the fusion also liberated chlo- 
rine; but blank determinations in an atmosphere of carbon 
dioxide proved that under these conditions not a trace of chlo- 
rine was evolved. The apparatus employed for a quantitative 
test of this reaction on perchlorates consisted of a small distil- 
ling flask of about 20 cm* capacity, into the tubulature of 
which was ground a piece of glass tubing reaching well into 
