In the case of ordinary mechanical power it had been long known that the intervention of machinery did not create force, but only transformed it. If a weight of 1 lb., A, just balances a weight of 2 lb., B, by aid of a pulley, and by the addition of a minute fraction, such as a grain, raises it 1 foot, it will be invariably found that A has descended 2 feet. In other words, 1 lb. working through 2 feet does exactly the same work as 2 lbs. working through 1 foot. And whatever may be the intervening machinery the same thing holds good, and the work put in at one end comes out, neither more nor less, at the other, except for a minute loss due to friction and resistance of air. If a force equal to 1 lb. is made, by multiplying the intermediate machinery, to raise a ton a foot from the ground, exactly as much force must have been exerted as if the ton had been divided into 2,240 parts of 1 lb. each, and each part separately lifted. But although energy cannot be created, at first sight it seems as if it might be destroyed, as when the ton falls to the ground and seems to have lost all its energy, whether of motion or of position. But here science steps in and shows us that it is not destroyed, but simply transformed into another sort of motion, which we call heat. Some connection between mechanical work and heat had long been known, as in the familiar experiment of rubbing our hands together to warm them; and the practice known to most primitive races of obtaining fire by twirling a stick rapidly in a hole drilled in a block of wood; a practice described by the old Sanskrit word ‘pramantha,’ which means an instrument for obtaining fire by pressure or friction, and which, translated into Greek, has been immortalised by the legend of Prometheus. But it was reserved for recent years, and for an English philosopher, Dr. Joule, to give scientific precision and generality to this idea, by actually measuring the amount of heat produced by a given amount of work, and showing that they were in all cases convertible terms, so much heat for so much work, and so much work for so much heat. He did this by measuring accurately by a thermometer the heat added to a given amount of water by the work done by a set of paddles revolving in it, set in rapid motion by a known weight descending through a known space. The unit of work being taken as that sufficient to raise 1 kilogramme through 1 metre, and that of heat as that required to raise the temperature of one kilogramme of water by 1° Centigrade, the relation between them, as found by a vast number of careful experiments, is that of 424 to 1. That is, one unit of heat is equal to 424 units of work.