227 



specific heat of the vapour of all this series of hydrocarbons was 

 ascertained to be very small, a fact which we believe accounts in 

 some measure for the difficulties we experienced in the fractional 

 distillations. For the purpose of analysis, we have contented ourselves 

 with selecting such products as boiled within the same 5 C. (9 F.) 

 of the thermometric scale. All the analyses tend to prove that the 

 ratio of carbon to hydrogen increases slowly with an increase in the 

 boiling-point, and to negative the not improbable assumption of the 

 carbon and hydrogen being combined in equal equivalents. The 

 general formula C n H n+ i agrees best with our results, and indicates a 

 probability of the Burmese naphtha containing several radicals or 

 their hydrides. 



Our endeavours to obtain definite substitution-compounds by 

 means of bromine and chlorine have been attended with only partial 

 success. Chlorine gas acts slowly in the dark, but more quickly 

 with the aid of diffused daylight ; pentachloride of antimony, on the 

 other hand, acts with so much violence that explosions frequently 

 ensue. 



Bromine appears to separate the hydrocarbons into two distinct 

 bodies, a circumstance which throws some doubt upon their simple 

 constitution. Hydrated sulphuric acid saturated with the anhydrous 

 acid likewise produces a separation of the hydrocarbons, absorbing 

 one portion and leaving the other unacted upon ; anhydrous sulphuric 

 acid, on the other hand, in some cases completely absorbs the whole, 

 sometimes with evolution of sulphurous acid. The copulated sulpho- 

 acids which are produced in these cases will probably enable us to 

 clear up the enigma of the composition of the hydrocarbons ; we 

 think it better therefore not to lay much stress upon the Eadical or 

 Hydride hypotheses until further experiment has thrown more light 

 upon the subject. Nevertheless we may state, that by operating 

 upon a fluid boiling between 90 and 100 C. (194-212 F.) with 

 chlorine for some months, we at last obtained a cessation of all 

 action, and a chlorine compound resulted, which, when purified and 

 analysed, gave numbers agreeing perfectly with C 26 H 22 C1 6 , corre- 

 sponding to a hydrocarbon, C 26 H 28 , a formula with which the ana- 

 lysis of the original hydrocarbon was perfectly consistent, although 

 its boiling-point pointed rather to a lower formula. 



