With reference to these atmospheres, it should be explained that air compressed to half its volume acquires twice its ordinary force, and to a quarter of its volume quadruple this force--hence the phrase of two, four, or more atmospheres. The exact degree of pressure is readily ascertained by means of a manometer, an instrument resembling a pressure-gauge, with a hollow screw at the base, which is driven through the cork of the bottle. A pressure of 5-3/4 atmospheres constitutes what is styled a 'grand mousseux,' and the wine exhibiting it may be safely conveyed to the coolest subterranean depths, for no doubt need be entertained as to its future effervescent properties. Should the pressure, however, scarcely exceed four atmospheres, it is advisable to keep the wine in a cellier above-ground, that it may more rapidly acquire the requisite sparkling qualities. If fewer than four atmospheres are indicated, it would be necessary to pour the wine back into the casks again, and add a certain amount of cane-sugar to it; but such an eventuality very rarely happens, thanks to the scientific formulas and apparatus, which enable the degree of pressure the wine will show to be determined beforehand to a nicety. Still mistakes are sometimes made, and there are instances where charcoal fires have had to be lighted in the cellars to encourage the latent effervescence to develop itself.[412] [Illustration: THE TIRAGE OR BOTTLING OF CHAMPAGNE.] The bottles are first placed in a horizontal position, the side to be kept uppermost being indicated by a daub of whitewash, and are stacked in rows of varying length and depth, one above the other, to about the height of a man, with narrow laths between them. Thus they will spend the summer, providing all goes well; but in about three weeks' time the process of gas-making inside the bottles is at its height, and a period of considerable anxiety to the Champagne manufacturer ensues, through his dread lest an undue number of them should burst from the expansion of the carbonic acid gas generated in the wine. The glucometer notwithstanding, it is impossible to check a certain amount of breakage, especially when a hot season has caused the grapes, and consequently the raw wine, to be sweeter than usual. Moreover, when once _casse_ or breakage sets in on a large scale, the temperature of the cellar is raised by the volume of carbonic acid gas let loose, which is not without its effect on the remaining bottles. Not only does the increased temperature unduly accelerate fermentation, but the mere shock of one bottle exploding often starts such of its neighbours as are predisposed that way, in addition to the direct havoc wrought by the heavier fragments of flying glass. The only remedy is the instant removal of the wine to a lower temperature whenever this is practicable. [Illustration]