The Weather. 101 



mercury (in round numbers), at freezing temperature, approach 

 so nearly to the average pressure, that this pressure is called the 

 standard pressure of the air. In rising to a greater height 

 above the sea, or if air becomes lighter at any place, or a 

 portion of it flows off from the top of the atmosphere, the 

 pressure becomes less, and the column of mercury becomes 

 shorter. From this circumstance an instrument constructed 

 on these principles, and known after its inventor as the 

 " Torricellian tube," is called a barometer (fiapw;, heavy; 

 fierpov, a measure), because it measures the weight of the 

 atmosphere (whether in rest or motion) over any place where 

 it is used. 



Air at standard pressure, and at the temperature of freezing 

 water, is 10,486 times lighter than the fluid mercury by which 

 its weight is measured. The height of the atmosphere, it 

 might be supposed, is therefore just so many times greater than 

 thirty inches ; or, in other words, five miles all but sixty-two* 

 yards. This height of a "homogeneous atmosphere," as 

 it is called, is, however, only imaginary, because it is well 

 known that air is an elastic fluid, which presses on everything 

 (at constant temperatures) in proportion to its density. The 

 densest of the atmosphere is, therefore, below, and the rarest^ 

 part above. So far as barometric observations have extended, 

 the standard pressure of the atmosphere remains the same, 

 and air does not appear to leave the earth. A definite limit 

 to the atmosphere, therefore, probably exists, but at what dis- 

 tance from the earth, whether at 100 or 150 miles, or upwards, 

 is not known, although certain circumstances lead us to suspect 

 about that height. 



The law of rarefaction in that part of the atmosphere which 

 need be considered in meteorology is, however, this, that 

 while the height increases in arithmetical, the pressure and the 

 density diminish in geometrical progression. Consider, for 

 example, the decimal fraction 0'999, etc., continued ad infi- 

 nitum, to represent by unity the whole weight of the atmo- 

 sphere in any column from the level of the sea. The amount 

 of it contained, by weight, in the first ten, twenty, thirty 

 miles, etc., above the sea, is represented by the first one, two, 

 three, etc., figures of this decimal. This is, in fact, not merely 

 an illustration, but a near approximation to the truth. Thus at 

 a height of ten miles above the sea nine-tenths of the atmo- 

 sphere are left below, and one-tenth part of it only remains 

 above. Ten miles compared to the diameter of the globe is 

 no thicker than the roughness of an egg, but even in this 

 small depth the lower half contains twice as much atmosphere 

 as the half above. The sensible atmospheric covering of the 

 globe, therefore, transcends very little the tops of the highest 



