iGO ON THE ASSAY OF COAL FOR CRUDE OIL, ETC. 
of Paris, or, still better, linseed meal, which, being some distance from the 
source of heat, is more effective than might be supposed. The opposite end 
of the iron tube is closed by a cork, which also, being from twenty-five to 
thirty centimetres from the furnace, is scarcely charred in several oj)erations. 
The cork of the condensing bottle should be perforated by a second hole for 
the insertion of a glass tube for carrying off gas, or for connection with a 
second condensing bottle. The second bottle may, in the distillation of coal 
yielding much very volatile matter, be connected with a Liebig’s condenser. 
Put if the operation be conducted slowly, extending over about two hours, the 
whole of the product will, in nearly every case, be found in the first bottle. 
The gas produced in the process may be conducted into a chimney, burned 
at a jet, or collected in a gas holder for measurement and experiment. 
I have had constructed other forms of retort than the long iron tube just 
described, but have not found them to possess any particular advantage. In 
one, the furnace being placed horizontally, the tube at the point where it 
emerged from the source of heat on the condensing side was bent upwards 
for about ten centimetres, and then downwards, so as to resemble the neck 
and beak of an ordinary glass retort. In some comparative experiments this 
form of retort yielded oil somewhat better in quality, but slightly smaller in 
quantity, than that given by the inclined straight tube. This result was pro¬ 
bably simply due to the exclusion of the heaviest hydrocarbons, an effect 
said to be sometimes desired by the manufacturer, but to be generally avoided 
by the experimentalist, who should be careful to obtain from a coal, etc., 
everything volatile that it can in any way be made to yield. Another retort, 
similar in form to the last mentioned, was provided with a tube for tlie ad¬ 
mission of steam, under the conjecture that the escape of oil-vapour as fast 
as produced might be thereby facilitated, as in the corresponding operation 
for the extraction of the ready-formed volatile oil of seeds, fruits, etc. But 
an experiment with the same coal as used in the other cases showed that the 
use of steam did not cause a better yield of oil, and was attended by some 
inconvenience, owing to the necessity of adopting adequate condensing arrange¬ 
ments. The use of superheated steam has been proposed, and is used to some 
extent, I am told, in the distillation of oil from coal on the large scale, no 
external heat being applied to the retorts. What direct or indirect advantage 
is thereby gained I am unable to state, not having performed any experiments 
in this direction. It may assist in the escape of the heavier vapours, but for 
this purpose the analyst will find carbonic acid gas preferable, as will be pre¬ 
sently seen. Should the use of steam at a temperature approaching redness 
ever be demonstrated to cause such an increase in the yield of oil as to coun¬ 
terbalance the increased cost of condensation, it would be easy for the analyst 
to imitate the method on the small scale, by passing steam generated in 
an ordinary flask through a heated coil of metal pipe before entering the retort 
containing the coal. 
The distillation of the coal completed, the source of heat is removed, and 
the weight of the condensing bottle and its contents noted ; from this weight 
that of the bottle, previously ascertained, is deducted, and the w'eight of the 
crude oil and water yielded by the coal thus determined. The separation of 
the oily from the aqueous portion of the distillate is generally easy of accom¬ 
plishment. The bottle is placed in a vessel of cold water, which is then 
heated to from 30° to 60° C. for an hour or two, and allowed to cool slowly. 
If complete separation of oil and water has apparently occurred, the latter 
may now be siphoned off, or, if the oil is semi-solid, the water may be simply 
poured off, and the weight of the two be respectively determined. If desired, 
the water can then be examined for ammonia, etc. Sometimes the separation 
'•of oil and water is not so decided as is necessary for analytical purposes. 
