132 Mr. J. A. Wanklyn on Fractional Distillation, 



ture is — r undistin°;uishable from-: viz. 

 a + b ° a' 



Exp. I. 

 Exp. II. 



1000 1000 



Exp. III. . fT^™^ or 



Ex P- IV - • tjvwwvo ov 



1001000 



1 



1000001 



0'50 

 1000000-5 ( 



0-20 



1000000 



1 



1000000 

 0-50 



1000000 

 0-20 



1000000-2 1000000 



In like manner in these instances the quantity of B in a unit 



bv' bv' 



of distillate is expressed indifferently either by —j-, or by — . 



Now, if it were possible to distil off and work with an infinitesi- 



bv ! 

 mally small quantity of distillate, we should get — for the value 



of B in a unit of distillate ; whence the relation of the value of 

 B in a unit of original liquid to the value of B in a unit of dis- 

 tillate is very simple, viz. - : - v f . Although we cannot distil 



off and work with an infinitely small fraction of distillate, yet 

 we can, in a series of experiments, distil off and work with equal 

 fractions (which will be equal numbers of the infinitesimal frac- 

 tions). In each of the four experiments I took a kilogramme 

 of A and B, and I distilled off 50 grammes (or one twentieth of 

 the whole). The results were as follows : — 



A = water. B = ammonia. 



Q, •, c Quantity of ammonia 



uantnr of ammonia , . J , . ,, ~ * 



«««+«;„«j ;™ „ v* « f contained in the first 



contained in a litre oi cn ,- .. n 



original U<i„id. 6 °"tri" SOf 



milligrms. milligrms. 



Exp. I. . . 1000 480 



Exp. II. . . 1-00 0-50 



Exp. III. . . 0-50 0-235 



Exp. IV. . . 020 0-095 



Now on inspection it will be manifest that, within the limit of 

 experimental error, the strength of the distillate is proportional 

 to the strength of the original liquid. In each of these four ex- 

 periments 1 litre of liquid was taken and 50 cubic centims. dis- 

 tilled off; that is to say, one twentieth of the whole was distilled. 

 If instead of 50 cubic centims. I distil off 100 cubic centims., I 

 also find that the strength of the distillate is proportional to the 



