C. M. Warren on a Process of Fractional Condensation. 337 
would be required than if the material were but slightly vola- 
tile; as the waste in the former case, from evaporation, would 
be much greater. 
But in many cases it will be found that highly volatile bodies 
are present only in very small proportion,—e. g., in visci 
oleum like Rangoon tar, and in the products of distillation of 
some species of asphalt. In such cases, the requisite quantity 
to be operated upon, to obtain the most volatile constituents in 
sufficient quantity for anything like a complete study of their 
chemical relations, would be extremely large,—too large to be 
conducted in the laboratory,—and one would have to resort to 
the manufactory for the first distillation. I have dwelt at some 
length on this point, having experienced the disappointment 
which one feels, after months of labor, on finding the products 
insufficient for his requirements, when the expenditure of a little 
more time, comparatively, might have given double the quan- 
tities obtained. 
les, and each carefully labelled with the temperatures between 
which it was obtained. The fractions for each fresh portion of 
eend of the worm into the retort. 1 then carefully raise the 
temperature of the bath until the vapors from the retort pass 
through the heated worm so freely that the liquid, in tea 
ing from them, shall drop with tolerable rapidity into the colc 
teceiver. In order that this dropping may be continuous, Pll 
hecessary that the temperature of the bath should rise aes 
ually as the more volatile constituents of the mixture are ta 
a a is easily effected by carefully regulating the flame under 
€ Dath. 
sed is 
ebullition, so that a steady stream of liquid shall flow back from 
It is advisable to boil the retort as rapidly as possible without 
choking the ome i: of the heated worm with the returning 
quid. As this choking would give rise to additional pressure 
M the retort, and consequently occasion @ 
