But what is true for an atom is also true for a planet or a sun. Supposing our earth to be brought to rest in her orbit by a sudden shock, we are able to state the exact amount of heat which would be thereby generated. The consequence of the earth's being thus brought to rest would be that it would fall into the sun, and the amount of heat which would be generated by this second collision is also calculable. Helmholtz has calculated that in the former case the heat generated would be equal to that produced by the combustion of fourteen earths of solid coal, and in the latter case the amount would be 400 times greater. [Sidenote: SHIFTING OF ATOMS.] Whenever a weight is lifted by a steam-engine in opposition to the force of gravity an amount of heat is consumed equivalent to the work done; and whenever the molecules of a body are shifted in opposition to their mutual attractions work is also performed, and an equivalent amount of heat is consumed. Indeed the amount of work done in the shifting of the molecules of a body by heat, when expressed in ordinary mechanical work, is perfectly enormous. The lifting of a heavy weight to the height of 1000 feet may be as nothing compared with the shifting of the atoms of a body by an amount so small that our finest means of measurement hardly enable us to determine it. Different bodies give heat different degrees of trouble, if I may use the term, in shifting their atoms and putting them in new places. Iron gives more trouble than lead; and water gives far more trouble than either. The heat expended in this molecular work is lost as heat; it does not show itself as temperature. Suppose the heat produced by the combustion of an ounce of candle to be concentrated in a pound of iron, a certain portion of that heat would go to perform the molecular work to which I have referred, and the remainder would be expended in raising the temperature of the body; and if the same amount of heat were communicated to a pound of iron and to a pound of lead, the balance in favour of temperature would be greater in the latter case than in the former, because the heat would have less molecular work to do; the lead would become more heated than the iron. To raise a pound of iron a certain number of degrees in temperature would, in fact, require more than three times the absolute quantity of heat which would be required to raise a pound of lead the same number of degrees. Conversely, if we place the pound of iron and the pound of lead, heated to the same temperature, into ice, we shall find that the quantity of ice melted by the iron will be more than three times that melted by the lead. In fact, the greater amount of molecular work invested in the iron now comes into play, the atoms again obey their own powerful forces, and an amount of heat corresponding to the energy of these forces is generated.