The progress and duration of heat in metals does not then follow the order of their density, except in tin, which being the least dense, is also that which soonest loses its heat; but the order of the five other metals demonstrates that it is in relation to their fusibility that they all receive and loose heat; for iron is more difficult to melt than copper, copper more than gold, gold more than silver, silver more than lead, lead more than tin; and therefore we may conclude that it is only by chance if the density and fusibility of tin be found so united as to place it in the last rank. Nevertheless, it would be advancing too much to pretend that we must attribute all to fusibility, and nothing to density. Nature never deprives herself of one of her properties in favour of another in an absolute manner; that is to say, in a mode that the first has not any influence on the second. Thus, density may be of some weight in the progress of heat; but we may safely affirm, that in the six metals it has very little comparatively with fusibility. This fact was neither known to chemists nor naturalists; they did not even imagine that gold which is more than twice as dense as iron, nevertheless loses its heat near a third sooner. It is the same with lead, silver, and copper, which are all more dense than iron, and which, like gold, heat and cool more readily; for though the object of this, second memoir was only refrigeration, yet the experiments of the one that preceded it demonstrate, that there is ingress and egress of heat in bodies, and that those which receive it most quickly also lose it the soonest. If we reflect on the real principles of density, and the cause of fusibility, we shall perceive, that density depends absolutely on the quantity of matter which Nature places in a given space; that the more she can make it enter therein, the more density there will be, and that gold, in this respect, is of all substances, that which contains the most matter relatively to its volume. It is for this reason that it has been hitherto thought, that more time is required to heat or cool gold than other metals; and it is natural enough to suppose, that containing double or treble the matter in the same volume, double or treble time would be required to penetrate it with heat; nay this would be true, if in every substance the constituent parts were of the same figure and ranged the same. But in the most dense the molecules of matter are, probably, of a figure sufficiently regular not to leave very void places between them; in others which are not so dense, and their figures more irregular, more vacuities are left, and in the lightest, the molecules being few, and most likely of a very irregular figure, a thousand times more void is found than plenitude; for it may be demonstrated by other experiments, that the volume of even the most dense substance contains more void space than full matter.