This section is from the book "A Treatise On Beverages or The Complete Practical Bottler", by Charles Herman Sulz. Also available from Amazon: A Treatise On Beverages.
When a bottle of carbonated water contains air, a portion of the contents is ejected with violence when the bottle is opened; it also prevents good carbonating with the gas. By a careful method of charging and bottling this can be entirely avoided. The air-escape valve on the filling head, when properly adjusted, is an excellent device for letting escape the compressed atmospheric air from the bottle. But by the usual method of bottling, the bottle to be filled is placed under the filling head and the carbonated water forced in. The air-escape valve is seldom regulated, and when the pressure in the bottle prevents any more liquid flowing in, the treadle is raised several times, and the accumulated air and gas allowed to rush out; it is then replaced securely under the nipple and filled up. In this way the water is bound to contain a very great deal of air, and beverages containing ferrous compounds are sure to become turbid.
This method of bottling is decidedly a wrong one, and should be discarded, as there are only disadvantages connected with it and not a single advantage gained. The fast bottling which it is claimed this method affords is no advantage compared with the great errors included in it.
When the pressure-regulating valves, now attached to almost every apparatus, and the air-escape valve on filling head are properly regulated, and the proper attention is paid to them and to the whole process of bottling, tins careful method of bottling will allow to bottle just as fast, the product will be improved and no gas nor liquid be lost. A great error is also made by some manufacturers, who work with the English continuous system, and we saw it not long since in a leading establishment in New York, viz.: to connect the air-escape valve (or waste valve, as some call it) of the bottling machine by means of a rubber hose, with the gasometer, to save the waste of gas ! It is evident that, although some gas will unavoidably be mixed with it, the most part of that "waste" is compressed air forced out from the empty bottles or syphons while being refilled, and the presumably unadulterated and pure gas in the gasometer is thus spoiled and the balance of the liquid charged with air-laden gas. And this is done for the sake of saving a trifling waste of gas!