496 Messrs. Wanklyn and Cooper on the 



The actual figures of the experiment were as follows : — 



cub. centim. 

 Distillate between 170° and 250° ... 688 



250° and 300° ... 205 

 Residue 107 



V 



1000 



That nothing should distil over below 170° is most remark- 

 able, inasmuch as there is actually present in kerosene a liquid 

 with a boiling-point of 78° C. The proportion of this low- 

 boiling liquid is indeed small ; but there are other volatile 

 ingredients, and the sum total of the ingredients boiling below 

 170° C. amounts to more than one fifth of the total kerosene. 



The 688 cub. centim. of distillate (which has just been 

 mentioned) were submitted to redistillation, and they yielded 

 505 cub. centim. below 210° C. And in that manner one litre 

 of kerosene was caused to yield 505 cub. centim. of distillate 

 below 210° C. When it is directly distilled a litre of kerosene 

 does not give anything like half its volume of distillate at 

 temperatures below 210° C. 



Continuing the investigation, the 505 cub. centim. of dis- 

 tillate were redistilled, and they began to distil at 125° C, and 

 up to 170° C. 270 cub. centim. came over. And in that 

 manner a litre of kerosene (which when directly distilled 

 yields only one drop of distillate below 170° C.) may be 

 caused to evolve 270 cub. centim. of distillate below 170° C. 



All this illustrates how admixture with a high-boiling liquid 

 raises the boiling-point of low-boiling liquids. 



We have carried out a most laborious investigation of 

 kerosene, and, operating on a considerable scale, after many 

 months of fractionation have separated 24 different liquids, 

 which, with probably a few others, and together with about 

 13 per cent, of residue boiling at temperatures above 290° C, 

 make up the complex mixture known as kerosene. As to the 

 proportions of the different ingredients of the mixture, it is 

 to be noted that there is no one preponderating ingredistfrt. 

 There appears to be a very small proportion of the two most 

 volatile terms of the series, but for the rest the distribution is 

 not very uneven. We doubt whether the proportion of any 

 one ingredient exceeds 5 per cent, of the whole kerosene. 



In the following table are set forth the main results of our 

 work ; — • 



