The term reversible is not here used in the popular sense in which a mere reversal of the direction of motion of each part is contemplated, _i.e._ what would be more properly termed ‘backing,’ it is used in the higher sense of taking an engine which converts a certain quantity of the heat spent on it into work, while it lets the rest down from the boiler to the condenser, and then spending upon it the same amount of work with the result of taking back the heat from the condenser, adding thereto the heat-equivalent of the work so spent, and thus restoring the whole of its original loss in heat to the boiler; simply in fact reversing all the results of the direct action.[38] * 107. Sir W. Thomson, in 1848, was the first to recall attention to the work of Carnot, after Colding and Joule had published their experimental discoveries; and he pointed out that the action of the reversible engine gave what had been up to that time vainly sought, an absolute definition of temperature—a definition, that is, altogether independent of the properties of any particular species of matter. In fact it is obvious that as reversibility in the sense we have just explained is the stamp of perfection in a heat-engine, all reversible engines, whatever be the working substance, will, under the same circumstances, that is to say, _with the same temperatures of boiler and condenser_, convert the same fraction of the heat spent on them into work. This, of course, still leaves wide scope for choice of a definition of temperature: but that finally determined on by Thomson was chosen (in consequence of a hint from some experimental results of Joule) so as to make the absolute measurement agree nearly with that of the long-familiar air-thermometer. It therefore stands as follows:— _The heat taken in by a perfect engine is to the heat given out by it in the same proportion as the absolute temperature of the boiler to that of the condenser._ Of course it is hardly necessary to state that it is only the excess of the heat taken in over that given out by any engine that can have been converted into available work. This follows at once from the conservation of energy. Experiments carried on by Joule and Thomson[39] together have shown that the absolute zero of temperature is nearly 274° below zero of the centigrade scale; so that on the absolute scale the temperature of melting ice is 274°, while that of water boiling under the standard pressure is 374°.