MEAN PRESSUEE OF THE ATMOSPHEEE. 233 



servations treated with, sagacity that we are able to distinguish that 

 which, in slight barometric oscillations, ought to be attributed either 

 to the pressure of the pure air, or to that of the watery vapour. As 

 to the abrupt variations in respect to which we cannot be mistaken, 

 they are sometimes enormous ; there are even some which are mark- 

 ed in the column of mercury by a difference of 2 or 3 inches, one- 

 fifteenth of the total height. A tempest in the ocean of the air is the 

 cause of this agitation of the liquid in the instrument.'^' 



The pressure of the atmosphere varies over all the earth, and we 

 cannot yet indicate it witli exactitude for the entire globe. It is pro- 

 bable, however^ that at the surface of the sea it exceeds on an average, 

 by a slight fraction, the amount of 29*90 inches. Towards the equator 

 the ordinary pressure is only 29*84 inches ; but from the 10th degree 

 of latitude in the two hemispheres the pressure increases little by little, 

 and towards the 30th or 35th degree it attains its maximum, 30 or 

 30 '08 inches. From thence, in the direction of the poles, the pressure 

 diminishes ; towards the 50th degree it is 29*92 inches, and further 

 north 29*76 inches only. Thus it is at about an equal distance between 

 the pole and the equator that the air exercises on an average its 

 greatest pressure on the barometric column ; nevertheless, there 

 being much more watery vapour in the aerial strata of the temperate 

 zone than in those of the polar zone, it is probable that, the air being 

 perfectly dry, its pressure would continually increase from the equator 

 to the poles more or less regularly in proportion to the sinking of 

 the temperature. This is moreover a phenomenon rendered very 

 probable by the rise in the barometer, which is ordinarily produced 

 by the transition from heat to cold. However it may be, the re- 

 searches of Sir James Ross and Wilkes in the southern seas establish 

 the fact that on an average tlie barometer is slightly higher in the 

 northern than in the southern hemisphere. We must necessarily con- 

 clude from this, that a greater quantity of air is accumulated over 

 that half of the earth where the continents are grouped. Thus, 

 as Sir John Herschel remarks, the current of a river is always rip- 

 pled above an unequal and stony bed ; and in the same way the 

 atmosphere must swell in weaves above the continental masses. This 

 best explains the astonishing contrast between the two hemi- 

 spheres.! 



If the normal pressure of the atmospheric strata varies at the level 



of the ocean under different latitudes, it varies also all over the earth 



* See below the section entitled^ Hurricanes. 

 t W. Fcnel, Motions of Fluids and Solid.', p. 39. 



