210 PROFESSOR LETTS AND N. COLLIE ON THE 
redistilled. They began to boil a little above 100° C. The distillate was 
divided into two fractions, viz., from 100°-200° C., and from 200°-3820° C. 
The first of these was liquid, and on redistillation passed almost entirely 
between 110°-114° C. (chiefly at 112° C.), and had all the properties of toluol. The 
second was solid, and contained a large quantity of free phosphorus. As 
its fractional distillation did not give satisfactory results it was dissolved in 
boiling alcohol. Free phosphorus in some quantity remained undissolved, 
and on filtering and cooling the solution, colourless crystals separated. They 
were collected and recrystallised until their melting point was constant, viz., 
118° C. | 
This is the melting point given by Laurent for stilbene, and the crystalline 
habit which is so characteristic was exactly the same as that of the substance 
under examination. On combustion we obtained numbers agreeing fairly well 
with those calculated for that hydrocarbon. 
0:3135 gave 1-0815 carbonic anhydride=0:29495 carbon =94'1 per cent. 
03135 ,, 0°1965 water =0°02183 hydrogen=6°9 a 
Obtained. Calculated. 
Carbon, , ; ; 94:1 : : : 93:3 
Hydrogen, . : : 6°9 . : : 66 
The mother liquors from which it had been separated were concentrated, and 
yielded a batch of colourless crystals, which were not examined. The mother 
liquors from them were considerably concentrated, and yielded another crop of 
colourless crystals, which, after repeated recrystallisation, ceased to alter in 
melting point. This was 51° C, which is that of dibenzyl. We have not 
analysed the substance, as we considered its identity with dibenzyl proved by 
its melting point and characteristic odour. 
We had thus proved that the resin split up on heating into free phosphorus, 
stilbene, dibenzyl, and toluol—a result which might occur supposing it to 
have consisted of tribenzyl-phosphine. The equation, 
Stilbene. Dibenzyl. Toluol. 
—s es —— 
2(C,H,);P=2P+C,,Hi, + CyHy + 20H. 
shows this. 
This supposition is strengthened by the fact that sulphide of benzyl yields 
stilbene when heated; and one of us has shown that organic compounds of 
phosphorus and sulphur often behave in a similar manner. 
Moreover, subsequent experiments showed that chloride of tetrabenzyl- 
phosphonium is decomposed by heat into hydrochloric acid, and the same pro- 
ducts as we obtained on heating the resin. We also heated the resin with 
chloride of benzyl in a sealed tube for some time at 180°-190°C. Nothing par- 
