324 PROFESSOR LETTS ON PHOSPHORUS-BETAINES. 
It was then dissolved in water and the solution warmed. 6 grms. of solid 
caustic potash were added (dissolved in a little water), and the two solutions 
mixed. No separation of triethyl-phosphine occurred. Another 6 grms. of 
potash were then added ; triethyl-phosphine then separated, but so far as could 
be judged it amounted to only 2 or 3 grms. 
The aqueous solution was drawn off from it, and it was found that the 
addition of strong caustic potash solution to this caused the separation of an oily 
liquid which rose to the surface and collected in a layer. 
The mixture was repeatedly extracted with ether (which dissolved the oily 
layer), and the ethereal extract separated by a tap funnel, and fractionally 
distilled. 
As soon as the ether, water, and triethyl-phosphine had passed over, the 
thermometer rose to 259°, and remained stationary at that temperature, whilst 
a colourless liquid passed over, which solidified on cooling. 
The boiling-point of this liquid, as well as its properties, left no doubt as 
to its identity with triethyl-phosphine oxide. 
The potash solution from which it had been extracted with ether, precipi- 
tated, during the extraction, a colourless crystalline salt. To obtain more of 
this, a considerable quantity of alcohol mixed with a little ether was added. 
The insoluble salt was then collected on a filter, and washed repeatedly with 
alcohol. It weighed 9 grms., and consisted entirely of bromide of potassium. 
These experiments indicate that bromacetic acid unites with triethyl- 
phosphine to give both the isomers, which, for the sake of convenience, we may 
eall If. and III. For although neither acetate nor bromacetate of potassium 
were specially sought for in the product of action (owing to the difficulty of 
separating them from the large excess of caustic potash present), the pro- 
duction of both triethyl-phosphine and the phosphine oxide may be considered 
as almost conclusive evidence of the production of both isomers, and from the 
quantities of these it would appear that II. is formed in far larger quantity 
than III. 
But shortly after these experiments were made, it was found that hydro- 
chlorate of triethyl-phosphorus betaine also reacts with potash to give the 
phosphine oxide, and both chloride and acetate of potassium, the reaction 
occurring according to the equation, 
Ol 
(CoH) PK +2KHO = (C,H,),PO+KCl+CH,COOK+H,0. 
CH,—COOH 
(see p. 319. 
The question therefore arose—is no hydrobromate of triethyl-phosphorus- 
betaine formed when bromacetic acid acts on triethyl-phosphine ? 
The hydrobromate was, therefore, prepared from the hydrochlorate (see 
