Vi. 
On a Process of Fractional Condensation ; applicable to the Separation of Bodies having 
small Differences between their Boiling-Points. , 
By C. M. WARREN. 
Communicated May 10th, 1864. 
Ir is well known that the process in general use for the proximate analysis of mix- 
tures of volatile liquids, — viz, that of simple fractional distillation, either from a 
tubulated retort or from a flask with bulbs, as proposed by Wurtz,*— affords but 
very imperfect and unsatisfactory results, and not unfrequently leads to gross errors 
and misconceptions, except in those cases in which the boiling-points of the constitu- 
ents are widely different, or in which some auxiliary method can be advantageously 
employed. 
The want of a more efficient process for effecting such separations has long been 
recognized. There are numerous natural and artificial products, of the highest 
scientific interest, — such as petroleums, essential oils, tars, and other mixtures of 
oils obtained by the distillation, under varied circumstances, of bituminous, vegetable, 
and animal substances, — of which it may at least be said that we have but very imper- 
fect knowledge, —I might almost say no knowledge, except such as could be derived 
from the study of very impure materials, — still mixtures of different bodies, — with 
which, instead of the pure substances sought for, chemists have felt compelled to 
content themselves, as the best results which they were able to obtain by the means 
at their command. 
In repeated instances, apparently after persevering and protracted efforts, investiga- 
tors have been forced to assert either the impossibility, or their inability, to obtain, 
from such mixtures, bodies of constant boiling-point, — a property which is generally 
received as a test of purity for liquid bodies. 
I may here specify a few recent instances of this kind. 
* Atinales de Chimie et de Physique, 3° Série, XLII. 132. 
VOL. IX. 20 
* 
