460 SHOKT MEMOIRS ON METEOROLOGICAL SUBJECTS. 



clear that since the equatorial current moves in a northeasterly direc- 

 tion, it will be affected by the force of rotation; and therefore this cur- 

 rent, as already remarked, cannot pursue the given direction unless the 

 air-pressure increases from west to east right across the current, as for- 

 mula (5) requires. But, on the other hand, it is also evident from what 

 has been said that the polar current running east of the same is like- 

 wise unable to pursue the given direction unless the air-pressure in this 

 current similarly increases from east to west, according to the law laid 

 down in formula (5). For these two currents, therefore, the level sur- 

 faces have an ascent toward each other, and hence it results that both 

 currents are kept braced against each other by the force of rotation, 

 while each of them seeks to crowd the other out of its place. If cur- 

 rents of air are of different specific gravities, the one having the greater 

 specific gravity must, as in the foregoing case, penetrate into the other; 

 and since it is generally the polar current whose specific gravity is the 

 greater, this stream will usually penetrate from the southeast into the 

 equatorial current, and give rain from the southeast or south. An 

 essential difference between the conditions upon the east side of the 

 equatorial current and the condition upon the west side of the same is 

 that while the air-pressure upon the west side of the equatorial current 

 is always relatively low, the pressure upon the east side of the same 

 current is always relatively high. The most remarkable difference, how- 

 ever, is perhaps this, that although in the region between the equatorial 

 current and the polar current which runs east of it calms may occur 

 similar to those which the great air-whirls produce upon the west side 

 of the equatorial current, yet a whirlwind of any magnitude can never 

 be formed upon the east side of the equatorial current, because the force 

 of gravity here works in the same direction as the centrifugal force, thus 

 tending to scatter the masses of air, while upon the west side it works 

 against the centrifugal force and holds the masses of air together. If 

 we now further consider that, in the very nature of the case, the rota- 

 tory movements that take place upon the east side of an equatorial cur- 

 rent must be "against the sun,"' while those which occur upon the west 

 side of such a current must be "■ with the sun," the reason is at once 

 seen why all revolving tornadoes and hurricanes must rotate against 

 the sun in the manner which experience has shown. 



If we now imagine ourselves to be in a polar air-current, and moving 

 under normal conditions across it in a westerly direction toward the 

 equatorial current, we shall, according to the preceding, from a relatively 

 cold and dry northeasterly wind, with a high barometric pressure, grad- 

 ually find the wind, under a rising air-pressure, changing to an easteHy 

 direction, and afterward, under a slightly diminishing air pressure, 

 changing into a rainy but somewhat milder southeasterly wind. Under 

 a constantly diminishing air-pressure, with milder and moister air, the 

 wind passes through the south over to the southwest, and we then find 

 ourselves in the southwesterly equatorial current, in which the state of 



