1887.] Prof. Letts and W. Wheeler on Benzyl Phos'pliines. 79 
Accordingly we prepared a new specimen of the brominated body 
with great care (specimen E). A sample of it was evaporated three 
times to dryness with alcoholic potash, then carefully recrystallised 
from alcohol. In appearance the product thus obtained exactly 
resembled the original dibenzyl phosphine. Its melting point (205°) 
was the same, and when treated with bromine it gave the charac- 
teristic yellow compound. On analysis numbers were obtained 
agreeing with those required for dibenzyl phosphine. 
Calculated for 
Obtained. Bzg H P 
Carbon, . . .78-0 78-5 
Hydrogen, ... 7*1 7'0 
We are of opinion that these experiments prove that the 
brominated body is a product of addition, and that the alkali simply 
removes the bromine ; but as a compound of one atom bromine 
with a molecule of dibenzyl phosphine is not in harmony with the 
modern views of atomicity, the following structural formula may 
be written for the brominated compound ; — 
Bz — P — Br 
h/ 
Bz 
H 
H 
yP— Br 
/ 
— indicating that the formula for dibenzyl phosphine itself should be 
doubled, and this would certainly account for its remarkable inert- 
ness, in which respect it differs from all secondary phosphines 
hitherto obtained. The compounds which it forms with the 
hydracids also favour this view of its constitution, for it will have 
been noticed that both the hydriodate and hydrobromate contain a 
single molecule of the hydracid for the double molecule of the 
phosphine. 
Action of Chlorine on Dihenzyl Phosphine . — Chlorine was passed 
into a solution of the phosphine in glacial acetic acid, when a white 
body was first precipitated, but this dissolved up partly, and even- 
tually a light yeUow crystalline substance separated. 
Bz^HPCl 
14-2 
Chlorine, 
. 12'0 
