Suppose, for instance, we heat two globes of iron, one an inch in diameter, the other seven inches, to a white heat. The surface of the larger is forty-nine times that of the smaller, and thus it gives out at the beginning, and at each corresponding stage of cooling, forty-nine times as much heat as the smaller. But it possesses at the beginning three hundred and forty-three (seven times seven times seven) times as much heat. Consequently, the supply will last seven times as long, precisely as a stock of three hundred and forty-three thousand pounds, expended forty-nine times as fast as a stock of one thousand pounds only, would last seven times as long. In every case we find that the duration of the heat-emission for globes of the same material equally heated at the outset is proportional to their diameters. Now, before applying this result to the case of the moon, we must take into account two considerations:--First, the probability that when the moon was formed she was not nearly so hot as the earth when it first took planetary shape; and secondly, the different densities of the earth and moon. The original heat of every member of the solar system, including the sun, depended on the gravitating energy of its own mass. The greater that energy, the greater the heat generated either by the process of steady contraction imagined in Laplace's theory, or by the process of meteoric indraught imagined in the aggregation theory. To show how very different are the heat-generating powers of two very unequal masses, consider what would happen if the earth drew down to its own surface a meteoric mass which had approached the earth under her own attraction only. (The case is of course purely imaginary, because no meteor can approach the earth which has not been subjected to the far greater attractive energy of the sun, and does not possess a velocity far greater than any which the earth herself could impart). In this case such a mass would strike the earth with a velocity of about seven miles per second, and the heat generated would be that due to this velocity only. Now, when a meteor strikes the sun full tilt after a journey from the star depths under his attraction, it reaches his surface with a velocity of nearly three hundred and sixty miles per second. The heat generated is nearly fifty times greater than in the imagined case of the earth. The moon being very much less than the earth, the velocity she can impart to meteoric bodies is still less. It amounts, in fact, to only about a mile per second. The condensing energy of the moon in her vaporous era was in like manner far less than that of the earth, and consequently far less heat was then generated. Thus, although we might well believe on _a priori_ grounds, even if not assured by actual study of the lunar features, that the moon when first formed as a planet had a surface far hotter than molten iron, we must yet believe that, when first formed, the moon had a temperature very much below that