Mr. J. A. Wanklyn on Fractional Distillation. 129 



as possible about 200° F., — this also acting as a regulator for 

 the amount of steam thrown into the chamber. 



3. That in " starting " a chamber, sulphuric acid should be 

 run en the bottom in preference to water, which is at present 

 generally employed. 



XVII. On Fractional Distillation. By J. Alfred Wanklyn, 

 Corresponding Member of the Royal Bavarian Academy of 

 Sciences*. 



IT is an extraordinary fact that there is, up to the present 

 time, no theory of fractional distillation. Chemists are per- 

 fectly aware that, on distilling a mixture of two unequally vola- 

 tile liquids, they will find some of the less volatile liquid in 

 an early portion of the distillate, and some of the more volatile 

 liquid in a later portion of the distillate. And, on the other 

 hand, chemists are quite as well aware that, notwithstanding this 

 clinging together of different liquids, almost perfect separations 

 are attainable when fractional distillation is laboriously and 

 systematically practised. Every chemist's own experience sup- 

 plies instances of the kind ; and most of us have our own peculiar 

 methods of performing the operation. My object, on the pre- 

 sent occasion, is to lay down a general theory of this process, and 

 to explain how it comes to pass that it is at once so laborious 

 and yet so successful when duly performed. 



In order to gain a clear idea of the effect of admixture, let the 

 hypothetical case of two liquids of precisely the same volatility 

 be supposed. In such a case each fraction of the distillate will 

 have the same composition as every other fraction and as the 

 original mixture. If the two liquids constituting the mixture 

 be named A and B respectively, and if the relative quantities of 

 them existing in the original mixture be expressed by a and b, 

 then a : b will express the ratio in which A and B are present in 

 the original liquid and in every fraction of the distillate from 

 first to last. If we picture to ourselves the course of events on 

 heating such a mixture, we perceive that, on applying heat to it, 

 the quantity of heat which is received by each constituent is 

 exactly proportional to the quantity of each constituent present 

 in the mixture. If a and b be the relative quantities of A and 

 B, then a and b will also be the relative quantities of heat re- 

 ceived by A and B in a given period of time. In the case under 

 consideration the volatilities of A and B are supposed to be equal ; 

 therefore a and b, which express the original quantities of xl and 

 B, and which (as just shown) express the relative quantities of 



* Communicated by the Author. 

 Phil. May. S. 4. Vol. 45. No. 298. Feb. 1873. K 



