It must not be forgotten that T_{2} and T_{1} are reckoned not in the ordinary scales of temperature such as Fahrenheit and Centigrade, but on the absolute scale, absolute zero being that temperature at which a body has no molecular motion. Calculations based upon various considerations point to the fact that absolute zero corresponds to about -273° C. We have just pointed out that in a perfectly efficient engine T_{2} - T_{1} = ————————————— = 1. T_{2} In order that this may be so, we must have T_{1} = 0, the absolute zero. In practice it is impossible to make the temperature of the exhausted gases as low as this, and so the only way to obtain more efficient engines is to make T_{2} as large as possible, that is to say, the initial temperature of the gases must be high. It is, however, just as possible to turn all the heat supplied to a heat engine into work as it is to use up all the energy of a waterfall in a turbine, because the level from which the zero of the potential of the energy water is measured is the centre of earth, which is as inaccessible as absolute zero of temperature. It therefore behoves us to make the ratio of the initial and final temperatures of the gas which does work in a gas engine as large as possible, and it is for this reason that gas engines can be made more efficient than steam engines, for in the former a momentary initial temperature of 1500° C. may be obtained by the combustion, whilst steam at 200 lbs. on the square inch is at about 1/10th of that temperature. There are practical difficulties which prevent higher initial temperatures being used, residing chiefly in the fact that at 400° C. iron is red-hot, so that any lubricant coming into contact with it is decomposed and loses its lubricating properties. Even at 300° C. most lubricating oils in contact with the air become oxidized and destroyed. This difficulty of lubrication, by limiting the temperature, at the same time limits the efficiency, and not till some new lubricant is discovered which defies heat will there be much improvement in this direction. Even as it is, it is necessary to cool the sides of the vessel or cylinder in which the gases expand, and in doing so we lose a great deal of heat. Hot-air engines using ordinary air as the expansible gas have been devised from time to time, but they have not met with much success owing to their weight and the large amount of space they take up, neither are they as efficient as a good modern gas engine. We will not, therefore, study the theory of hot-air engines, but further consider the details of gas engines, whose superiority over all other heat engines we think we have sufficiently pointed out.