278 Chittenden and Blake—Distribution of 
Classen also found that antimony was deposited quantitatively, 
when the sulphide was dissolved either in potassium or sodium mono- 
sulphide, or in potassium or sodium hydrosulphide. Polysulphides 
must also be absent in this case, and the potassium and sodium must 
be quite free from both iron and aluminum, as by long continued 
electrolytic action, sulphide of iron and aluminum hydroxide may be 
deposited upon the antimony. Classen also recommends the use of 
ammonium sulphydrate and a weak current; a strong current tending 
to cause the separation of the antimony in a pulverulent form, not 
closely adherent to the platinum. 
As it would be necessary in our work, after oxidation of the or- 
ganic matter, to separate the antimony as sulphide, the above method 
seemed particularly advantageous, and experiments were therefore 
tried to ascertain its value when applied to very small quantities. 
Weemployed a battery of four moderate sized gravity cells, giving a 
weak current, and as a rule, exposed the solution to the action of the 
current for at least 15 hours, as we found better results were ob- 
tained by the long continued action of a weak current, than by the 
quicker action of a more rapid one; particularly in such solutions as 
we usually had to work with, containing considerable excess of sul- 
phur and some organic matter. The negative pole of the battery was 
either a small platinum crucible or a platinum dish, while the posi- 
tive pole was a large piece of platinum foil welded onto a good sized 
platinum wire. The deposition of the antimony was much more 
complete, more tightly adherent to the platinum, and as a rule less 
mixed with sulphur under this arrangement, than when the dish was 
made to serve as the positive pole; due, doubtless, simply to the 
broader surface for the deposition of the metal. 
Using ammonium sulphide as a solvent for the antimony sulphide, 
did not give us very good results by electrolysis, the loss being cou- 
siderable. Much better results were obtained by using a solution of 
sodium monosulphide, made by saturating one-half of a 15 per cent. 
solution of sodium hydroxide with hydrogen sulphide, and then add- 
ing the remaining half of the sodium hydroxide solution. 
The method was tested by precipitating definite volumes of a 
standard antimony (tartar emetic) solution with hydrogen sulphide, 
dissolving the metallic sulphide in the sodium monosulphide and then 
exposing the solutions to electrolytic action. When the separation of 
the metal was complete, it was found best to wash the deposit with 
considerable water, without breaking the current; as sometimes, as in 
the presence of tartaric acid, the separated metal rapidly dissolved 
