EXTRACTION OF MOISTURE.--In all large plants the extraction of the moisture may take place in two stages. Immediately after the generator, and before the washer if the generator requires such an apparatus to follow it, a condenser is placed. Here the gas is made to travel somewhat slowly through one or more pipes surrounded with cold air or water, or is made to travel through a space containing pipes in which cold water is circulating, the precise method of constructing the condenser being perfectly immaterial so long as the escaping gas has a temperature not appreciably exceeding that of the atmosphere. So cooled, however, the gas still contains much water-vapour, for it remains saturated therewith at the temperature to which it is reduced, and by the inevitable law of physics a further fall in temperature will be followed by a further deposition of liquid water from the acetylene. Manifestly, if the installation is so arranged that the gas can at no part of the service and on no occasion fall to a lower temperature than that at which it issues from the condenser, the removal of moisture as effected by such a condenser will be sufficient for all practical purposes; but at least in all large plants where a considerable length of main is exposed to the air, a more complete moisture extractor must be added to the plant, or water will be deposited in the pipes every cold night in the winter. It is, however, useless to put a chemical drier, or one more searching in its action than a water-cooled condenser, at so early a position in the acetylene plant, because the gas will be subsequently stored in a water- sealed holder, where it will most probably once again be saturated with moisture from the seal. When such generators are adopted as require to have a specific washer placed after them in order to remove the water- soluble impurities, _e.g._, those in which the gas does not actually bubble through a considerable quantity of liquid in the generating chamber itself, it is doubtful whether a separate condenser is altogether necessary, because, as the water in the washer can easily be kept at the atmospheric temperature (by means of water circulating in pipes or otherwise), the gas will be brought to the atmospheric temperature in the washer, and at that temperature it cannot carry with it more than a certain fixed proportion of moisture. The notion of partially drying a gas by causing it to pass through water may appear paradoxical, but a comprehension of physical laws will show that it is possible, and will prove efficient in practice, when due attention is given to the facts that the gas entering the washer is hot, and that it is subsequently to be stored over water in a holder.