REDUCTION OF TERRESTKIAL TEMPERATURE BY STORMS 523 



perate regions, air is drawn in from warm southerly latitudes. On the 

 opposite side air is drawn in from relatively cold regions, especially in the 

 main storm belt. In the center of the area of low barometric pressure, 

 which is the fundamental feature of a cyclonic storm, the cold, heavy air 

 frdm the poleward side flows under the warm, light air from the equator- 

 ward side. Thus, in the center, where there is a strong tendency for the 

 air to rise because of the surrounding high pressure, it is the warmer air 

 which rises. The rising air is carried far aloft. As it reaches higher alti- 

 tudes, where the atmospheric pressure is less, it expands little by little 

 and grows cool, according to the usual adiabatic laws. At the same time 

 it grows cool by radiation, but this last process is much slower than the 

 adiabatic cooling. Accordingly, air which is unusually warm when it 

 leaves the earth^s surface is still relatively warm when it reaches high 

 altitudes. Manifestly such air carries heat away from the earth^s surface. 

 Having once reached a high altitude, it stays there a long time, for there 

 are no rapid downward movements of the atmosphere at all comparable 

 to the strong upward currents in the centers of cyclonic storms. During 

 this period it grows cool by radiation, and much of the heat thus sent out 

 is a complete loss, so far as the earth^s surface is concerned. 



During periods of many sun-spots the increase in the total intensity 

 of the earth's storminess must cause a corresponding increase in the 

 amount of air carried from low to high altitudes. The average rate of 

 movement also increases in all probability, for the storms are likely to be 

 more severe. When the amount of warm air that goes upward increases, 

 and still more if it rises at a more rapid rate than formerly, it is evident 

 that the amount of heat which is carried away from the earth's surface 

 and is dissipated in th^ upper air must increase to a corresponding de- 

 gree. As the air which thus rises is derived from regions to the equator- 

 ward of the storms, there would naturally be a distinct tendency to lower 

 the temperature of the entire equatorial and subtropical area between the 

 storm belts of the two hemispheres. Apparently the fact that at times 

 of maximum sun-spots the storms tend to increase not only in a boreal, 

 but a subtropical belt nearer to the equator than the present storm belt, 

 increases the extent to which the equatorial regions are cooled by the 

 draining away of their warmer air. Such a process of cooling the regions 

 to the equatorward of the storm belt must apparently occur whenever the 

 number of storms increases, no matter whether the amount of heat re- 

 ceived from the sun changes or remains constant. In the storm belt a 

 similar, but less noticeable, effect seems to be produced. More air than 

 formerly is drawn in from the south, but more also comes from the north, 

 and the two perhaps balance one another. The warm air, however, rises 



