"It might be difficult to use snowballs to boil the tea-kettle, but the heat given out in the formation of the snowflakes is doubtless employed quite as usefully for the poor as if used in preparing their tea. You have all noticed that before a snow-storm, or perhaps during the early part of the storm, the temperature generally becomes milder, and you have often heard the remark, 'It is too cold to snow.' Men have learned that the coming of a snow-storm is attended by a warming of the air. This popular impression is philosophical, yet few understand its philosophy. A foot of snow falls, equal to two or three inches of water. In the condensation of the vapor which formed this snow one thousand degrees of latent heat become sensible, and then in the congelation of the clouds into snowflakes one hundred and forty degrees of heat are evolved. This softening of the rigors of winter is, I think, as great a blessing to the poor as the heating of the tea-kettle. Let us make an estimate of the amount of heat set free in the production of one great snow-storm. Two feet of snow falls, equal, we will suppose, to five inches of water. In the condensation of the watery vapor one thousand degrees of heat are evolved, and in the congelation one hundred and forty degrees--an amount of heat which would boil three feet of cold spring water. In every square mile there are 27,878,400 square feet, and a square mile of water three feet in depth would contain 83,625,200 cubic feet. The production of such a snow-storm sets free for every square mile of surface heat which would boil more than 80,000,000 of cubic feet of spring water. Such a storm sometimes extends over a region of country a thousand miles square, that is, over a million of square miles. In the production of one such storm--a very heavy and extensive storm, I have supposed--heat is generated which would boil eighty millions of millions (80,000,000,000,000) of cubic feet of spring water--an amount altogether too vast for our comprehension. To accomplish this result by combustion would require more than 500,000,000 of tons of anthracite coal--an amount at least three times as great as the yearly product of all the coal-mines of the world. And this is but one heavy storm. The amount of rainfall in the United States may be thirty-six inches or forty or forty-five inches. Supposing the average rainfall of the whole earth to be twenty-four inches--an estimate very far below the truth--we have this result: There are, in round numbers, two hundred millions of square miles of surface, more than five and a half quadrillions (5,575,680,000,000,000) of square feet and more than eleven quadrillions of cubic feet of water. The condensation of this amount of vapor would boil more than sixty quadrillions of cubic feet of ice water. One pound of anthracite coal burned under the most favorable circumstances will boil sixty pounds of ice water. To boil sixty quadrillions of cubic