But now another phase of the problem presents itself to the experimenter. Oxygen has assumed the quiescent liquid state, to be sure, but in so doing it has fallen below the temperature of its cooling medium; hence it is now receiving from that medium more energy of vibration than it gives, and unless this is prevented very soon its particles will again have power to kick themselves apart and resume the gaseous state. Something, then, must be done to insulate the liquefied gas, else it will retain the liquid state for too short a time to be much experimented with. How might such insulation be accomplished? The most successful attack upon this important problem has been made by Professor Dewar. He invented a receptacle for holding liquefied gases which, while not fulfilling the ideal conditions referred to above, yet accomplishes a very remarkable degree of heat insulation. In consists of a glass vessel with double walls, the space between which is rendered a vacuum of the highest practicable degree. This vacuum, containing practically no particles of matter, cannot, of course, convey heat-impulses to or from the matter in the receptacle with any degree of rapidity. Thus one of the two possible means of heat transfer is shut off and a degree of insulation afforded the liquefied substance. But of course the other channel, ether radiation, remains. Even this may be blocked to a large extent, however, by leaving a trace of mercury vapor in the vacuum space, which will be deposited as a fine mirror on the inner surface of the chamber. This mirror serves as an admirable reflector of the heat-rays that traverse the vacuum, sending more than half of them back again. So, by the combined action of vacuum and mirror, the amount of heat that can penetrate to the interior of the receptacle is reduced to about one-thirtieth of what would enter an ordinary vessel. In other words, a quantity of liquefied gas which would evaporate in one minute from an ordinary vessel will last half an hour in one of Professor Dewar's best vacuum vessels. Thus in one of these vessels a quantity of liquefied air, for example, can be kept for a considerable time in an atmosphere at ordinary temperature, and will only volatilize at the surface, like water under the same conditions, though of course more rapidly; whereas the same liquid in an ordinary vessel would boil briskly away, like water over a fire. Only, be it remembered, the air in "boiling" is at a temperature of about one hundred and eighty degrees below zero, so that it would instantly freeze almost any substance placed into it. A portion of alcohol poured on its surface will be changed quickly into a globule of ice, which will rattle about the sides of the vessel like a marble. That is not what one ordinarily thinks of as a "boiling" temperature.