1914-15.] Interaction of Methylene Iodide and Silver Nitrate. 229 
alcohol gave the purest product, but it also was contaminated with 
silver iodide. The substance is best kept in an alcoholic solution of 
silver nitrate. It is insoluble in benzene and paraffins and is almost 
insoluble in ether. 
Effect of Water . — On addition of water the compound moderately 
rapidly decomposes; the crystals lose their crystalline form and become 
yellowish, and the supernatant fluid becomes acid. When 8’562 gms. was 
heated with 150 c.c. water under a reflux condenser for one hour, 3*800 
gms. silver iodide was recovered. The supernatant liquid was 0075 
normal acid, and contained 0*618 gm. silver nitrate. The acidity was 
mainly due to nitric acid. Formic acid was present in much smaller 
amount, and there was a questionable trace of oxalic acid. On distillation 
the liquid yielded a small quantity of a colourless oily substance which 
was proved to be methylene iodide. This substance was also isolated from 
the mixture in the distilling flask, but owing to difficulties in separation 
the amount could not be quantitatively estimated. The decomposition 
appears to vary somewhat with the conditions of the experiment. For- 
maldehyde is at first formed, and the earlier part of the reaction may 
probably be expressed as — 
SAgNOg . CH 2 I 2 + H 2 O = 2AgI + AgNOg + 2HNOg + 2 CH 2 I 2 + HCOH. 
Effect of Caustic Alkali . — The action of an aqueous solution of caustic 
alkali was similar to that of water except that a large proportion of the 
precipitate formed consisted of silver hydroxide. When boiled for an hour 
with 10 per cent, aqueous potash 14*8 per cent, of the silver appeared as 
silver hydroxide and 9*5 per cent, as silver iodide. 
Effect of Potassium Cyanide. — When the substance was treated 
with twice the molecular quantity of potassium cyanide in aqueous 
solution, the mixture separated into two layers, an upper aqueous 
layer of solution of silver potassium cyanide and a lower layer of 
methylene iodide. The quantity of methylene iodide recovered was 
58*3 per cent, (theoretical yield = 61*1 per cent.). The silver in the 
aqueous layer was estimated, after previous boiling with nitric acid, 
by Volhard’s method, and was found to be 22*4 per cent, (theoretical 
yield = 24*7 per cent.). A small quantity of insoluble residue, too small 
for further examination, accounted for the slight difference between the 
actual and theoretical yields. 
Discussion . — There seems little doubt that the substance is an unstable 
additive compound of molecular proportions of silver nitrate and methylene 
iodide. The compound is interesting because the two neighbouring iodo- 
