ACTION OF PHOSPHIDE OF SODIUM ON HALOID ETHERS. 197 
We naturally supposed that the hydrate of tetrabenzyl-phosphonium would 
be produced, Thus— 
(C,H,),PHSO, + Ba(OH), =(C,H,),P(OH) + BaSO,+ H,0. 
In our first attempts to prepare the base by this method 3 grms. of the 
acid sulphate were dissolved in about half a litre of water, and a solution of 
caustic baryta (in boiling water) was added until no more sulphate of barium 
was precipitated. The solution was then filtered and a stream of carbonic acid 
passed through it until all excess of baryta was converted into carbonate. It 
was then boiled and filtered, and the filtered solution concentrated until it 
amounted to about 50 cc. in volume. Tufts of needles separated out, though 
in small quantity ; but on allowing the solution to stand for a couple of days, 
large rhombohedral plates appeared, which were colourless and of striking re- 
fractive power—resembling, in fact, iodide of phosphonium. They were washed 
with a little water and their mother liquor evaporated down, when a second 
crop of tabular crystals separated. The two crops were mixed and recrystal- 
lised from water. The resulting crystals were analysed. 
Heated in a drying oven to 110° C. they became opaque, and lost in weight, 
thus showing that they contained water of crystallisation. As, however, dur- 
ing drying they developed a smell, recalling that of bitter almonds, we cannot 
place much reliance on the water determination, as the compound no doubt 
volatilised slightly. It was, however, necessary to dry it for organic analysis, 
and we therefore heated it quickly in a small beaker placed in an oil bath until 
it had partially fused. As this only occurred at a temperature above 200° C., 
there can be no doubt that we obtained a dry product. 
Its combustion thus dried gave numbers agreeing with those required for 
the base. 
0:2065 grm. gave 0620 CO, = 81°8 per cent. carbon. 
O2065 95 3, 01256 HO = 6:5 ‘ hydrogen. 
Obtained. Calculated for (C;H;),P(OH). 
Carbon, f : 2 81'8 : 81:5 
Hydrogen, . ; 6'8 : 5 70 
Wishing to prepare another quantity of the base, we proceeded in a similar 
manner, using however more concentrated solutions; but on filtering and 
evaporating (after getting rid of excess of baryta), no base could be obtained. 
This curious result induced us to examine the precipitated sulphate of barium, 
- which we found to contain a substance readily soluble in alcohol, though quite 
insoluble in water. It separated from its alcoholic solution in white needles. 
These were recrystallised until of constant melting point, viz., 210-212° C. 
VOL. XXX. PART I. 2H 
