One day an old Florentine pump-maker came to Galileo[1] to inquire why he could not make a pump work effectively when it was more than thirty-four feet long. The philosopher could not answer, nor did he solve the problem during his lifetime, but bequeathed it to his pupil, Torricelli.[2] This famous Italian succeeded in partially answering the question in 1643. He performed the following experiment: Taking a glass tube thirty-six inches long and one-fourth of an inch in diameter, closed at one end, he filled it with mercury, and holding his finger over the open end, inverted and placed it in a cup of mercury, then removing his finger, discovered that the quicksilver settled in the tube six inches, leaving a column of the shining metal thirty inches high. He thus demonstrated that air has weight, equal to that of a column of mercury thirty inches in height. On this supposition it was argued that if the whole height of the air should be _lessened_ the column would fall. Such was the opinion of Blaise Pascal,[3] who in 1646 requested M. Périer, his brother-in-law, to ascend the Puy de Dome, a summit near Clermont, and repeat the experiment of Torricelli. To his delight, upon reaching the top of the mountain, the column stood three inches lower. Pascal then used a tube fifty feet long, which he filled with water, and found that this liquid could be supported by the air to the height of thirty-four feet. Water is 13.6 lighter than mercury, and it will be observed from the foregoing statement, that it was supported 13.6 higher than quicksilver. Here was a full answer to the pump-maker’s query! A column of water, thirty-four feet long and one square inch at the base, weighs fifteen pounds. A column of mercury, thirty inches long and one square inch at the base, weighs fifteen pounds. A column of air the whole height of the atmosphere, one square inch at the base, weighs fifteen pounds. Any influence, therefore, which varies the weight of the air, will vary the height of a column of quicksilver; and the reverse will of course be true, that any fluctuation in the column of mercury indicates a change in the condition of the atmosphere. Thus, a “falling barometer” predicts foul weather, for it shows that the air is becoming lighter, and will therefore rise, while other air will rush in, with varying speed, to take its place, producing breezes, gales, and possibly tornadoes. The warm air rising may come in contact with a cold stratum above and its moisture be condensed into rain or snow. We shall presently refer to this again. [Illustration: SHOWING DENSITY OF ATMOSPHERE AT DIFFERENT HEIGHTS.]