378. This period of ice-extension has been named the Glacial Epoch. In accounting for it great minds have fallen into grave errors, as we shall presently see.

379. The substance on which we have thus far been working exists in three different states: as a solid in ice; as a liquid in water; as a gas in vapour. To cause it to pass from one of these states to the next following one, heat is necessary.

380. Dig a hole in the ice of the Mer de Glace in summer, and place a thermometer in the hole; it will stand at 32° Fahr. Dip your thermometer into one of the glacier streams; it will still mark 32°. The water is therefore as cold as ice.

381. Hence the whole of the heat poured by the sun upon the glacier, and which has been absorbed by the glacier, is expended in simply liquefying the ice, and not in rendering either ice or water a single degree warmer.

382. Expose water to a fire; it becomes hotter for a time. It boils, and from that moment it ceases to get hotter. After it has begun to boil, all the heat communicated by the fire is carried away by the steam, though the steam itself is not the least fraction of a degree hotter than the water.

383. In fact, simply to liquefy ice a large quantity of heat is necessary, and to vaporize water a still larger quantity is necessary. And inasmuch as this heat does not render the water warmer than the ice, nor the steam warmer than the water, it was at one time supposed to be hidden in the water and in the steam. And it was therefore called latent heat.

384. Let us ask how much heat must the sun expend in order to convert a pound weight of the tropical ocean into vapour? This problem has been accurately solved by experiment. It would require in round numbers 1,000 times the amount of heat necessary to raise one pound of water one degree in temperature.

385. But the quantity of heat which would raise the temperature of a pound of water one degree would raise the temperature of a pound of iron ten degrees. This has been also proved by experiment. Hence to convert one pound of the tropical ocean into vapour the sun must expend 10,000 times as much heat as would raise one pound of iron one degree in temperature.

386. This quantity of heat would raise the temperature of 5 lbs. of iron 2,000 degrees, which is the fusing point of cast iron; at this temperature the metal would not only be white hot, but would be passing into the molten condition.