Amid these and such-like curious experiments, we must not forget the “Law” that the state of a substance depends on its temperature—solid when it is frozen hard enough, liquid under sufficient pressure, gaseous when free from pressure and at a sufficiently high temperature. But though first Faraday, and then the various inventors of refrigerating-machines—Carré, Tellier, Natterer, Thilorier—succeeded in liquefying so many gases, hydrogen and the two elements of the atmosphere resisted all efforts. By plunging oxygen in the sea, to the depth of a league, it was subjected to a pressure of four hundred atmospheres, but there was no sign of liquefaction. Again, Berthelot fastened a tube, strong and very narrow, and full of air, to a bulb filled with mercury. The mercury was heated until its expansion subjected the air to a pressure of seven hundred and eighty atmospheres—all that the glass could stand—but the air remained unchanged. Cailletet managed to get one thousand pressures by pumping mercury down a long, flexible steel tube upon a very strong vessel, full of air; but nothing came of it, except the bursting of the vessel, nor was there any more satisfactory result in the case of hydrogen. One result, at any rate, was established—that there is no law of compression like that named after Boyle or Mariotte, but that every gas behaves in a way of its own, without reference to any of the others, each having its own “critical point” of temperature, at which, under a certain pressure, it is neither liquid nor gaseous, but on the border-land between the two, and will remain in this condition so long as the temperature remains the same. Hence, air being just in this state of gaseo-liquid, the first step towards liquefying it must be to lower its temperature, and so get rid of its vapor by increasing its density. The plan adopted, both by Cailletet in Paris, and by Raoul Pictet (heir of a great scientific name) in Geneva, was to lower the temperature by letting off high-pressure steam. This had been so successful in the case of carbonic acid gas as to turn the vapor into snow; and in 1877 Cailletet pumped oxygen into a glass tube, until the pressure was equal to three hundred atmospheres. He then cooled it to four degrees Fahrenheit below zero, and, opening a valve, let out a jet of gaseous vapor, which, while expanding, caused intense cold, lowering the temperature some three hundred degrees, and turning the jet of vapor into fog. Here, then, was a partial liquefaction, and the same was effected in the case of nitrogen. Pictet did much the same thing. Having set up at Geneva a great ice-works (his refrigerating agency being sulphurous acid in a boiling state), he had all the necessary apparatus, and was able to subject oxygen to a pressure of three hundred and twenty atmospheres, and by means of carbonic acid boiling in vacuo, to cool the vessel containing it down to more than two hundred degrees Fahrenheit below zero.