It results from all the laws that we have cited that by properly regulating the tensions of the vapors of a mixture of alcohol and water, and the temperature of the liquid, we shall be able to obtain a liquid of a desired richness by the condensation of these vapors.

III. It was likewise indispensable to make sure of one important fact: When the temperature of a liquid like alcohol is considerably lowered, can the distillation of a given weight of this substance be effected with sufficient rapidity for industrial requirements? Repeated experiments with a host of volatile liquids have demonstrated the following laws:

If we introduce a volatile liquid into two spherical receivers connected by a wide tube, and if these be kept at different temperatures after driving out all the air from the apparatus, the liquid distills from the warmer into the cooler receiver, and we ascertain that:

h. The weight of the liquid which distills in the unit of time increases with the deviation of temperature between the two receivers.

i. The weight of the liquid which distills in the unit of time is constant for a same deviation of temperature between the receivers, whatever be, moreover, the absolute temperature of the receivers.

k. The weight of the liquid distilled in the unit of time is proportional to the active surfaces of the receivers; that is to say, to the surfaces which are the seat of passage of heat through their thickness.

l. The least trace of a foreign gas in the vapors left in the apparatus throws the preceding laws into confusion, and checks distillation to a considerable degree, especially at low temperatures.

Thus, water distilling between 100° and 60° will pass over as quickly as that which is distilling between 40° and 0°. Absolute temperature is without influence, provided every trace of air or foreign gas be got rid of.

The distillatory apparatus should be provided with an excellent air-pump, capable of preventing all those entrances of air which are inevitable in practice.

The following is the industrial application that we have endeavored to make of these theoretical views: The rectification of alcohols is one of the most complex of operations; it looks toward several results simultaneously. Alcohol derived from the fermentation of grain, sugar, and of all starchy matters in general, contains an innumerable host of different products, which may be grouped under four principal heads:

1. Empyreumatic essential oils, characteristic of the source of the alcohol, and having a powerful odor which infects the total mass of the crude spirits. 2. A considerable quantity of water. 3. A certain quantity of pure alcohol. 4. A variable proportion of volatile substances, composed in great part of ethers, different alcohols, and bodies as yet not well defined. These latter affect the quality of the alcohol by an odor which is entirely different from that of the essential oils.

The object of rectification is to bring out No. 3 all alone; that is to say, to extract the alcohol in a pure state by ridding it of oils, water, ether, and foreign alcohols.

The alcohol industry never realizes this operation in an absolutely complete manner. All the rectifying apparatus in operation at the present day are based on the use of high temperatures varying between 78.5° and 100°. The successive condensation and vaporization of the vapors issuing from the spirits effect in the rectifying columns a partial separation of these liquids, and there are received successively as products of rectification:

1. Bad tasting alcohols, containing the majority of the ethers and impure alcohols.

2. Fine alcohol.

3. Alcohols contaminated by notable proportions of empyreumatic oils.

Industry knows only one means of obtaining an excellent product, and that is to diminish the quantity of fine alcohol which comes from a same lot of spirits, and to make a large number of successive distillations. Hence the large expenses attending rectification, which produce fine alcohols necessarily at an elevated price. We may remark, in passing, that the toxic action of commercial alcohols is in great part caused by the presence of essential oils, amylic alcohol, and ethers, absolutely pure alcohol, as compared with these, being relatively innocent.

Why is it that our present apparatus cannot produce good results in rectifying alcohol? Because they are limited by the temperature at which they must operate. Between 78° and 100° the tension of the vapors of all the liquids mixed in the spirits is considerable for each of them; they all pass over, then, in certain proportions during the operation of rectification.

We have been led, by examining the theoretical question, to ascertain that the proportion of alcohol which evaporates from a mixture is maximum at low temperatures; consequently, we should seek to establish some arrangement which can realize the following conditions: (1) Render variable, at will, the temperature of the boiling liquid; and (2), render variable the pressure of the vapors which act on the liquid.

Thus, to effect the rectification of alcohol it suffices to cause its ebullition at very low temperatures, and to keep up the ebullition without changing such temperatures when once obtained.

It is exactly these two conditions that we have fulfilled in the apparatus that we have just installed in our factory in Rue Immeubles Industriels, at Paris.

By their arrangement, which is shown in the opposite figure, they form a mechanical system permitting of the rectification of alcohols at temperatures as low as -40° or even -50°. They verify experimentally, by their operation, the theoretical deductions which precede. The boilers, A, which, in an industrial application, may be more numerous, receive their supply of spirits from the country distilleries in the vicinity of the factory. There may even be introduced directly into them vinasses, or washes, that is to say, liquids, such as are obtained by alcoholic fermentation.