HOW KAIN LS FOKMKD. 301 



issued with much more certainty, and perhaps longer before the arrival 

 of tlie storm than at present. In the case of sucli storms as that which 

 reached onr islands on February 2, we often have such warnings from 

 America, but their tracks are often more to the northeast, in the di- 

 rection of Iceland, in which case they are not I'elt on our coasts, and 

 hence the frequent failure of these American warnings. 



It is the region of low pressure in the North Atlantic that is the 

 especial field of these storms. As they pass across it, they produce 

 considerable modifications in the distribution of pressure, but some of 

 its main features remain outstanding. Thus there is always a belt of 

 liigh barometer between the storm region and the trade-winds, and in 

 the winter there is almost always a region of high barometer over North 

 America, and another over Europe and Asia, however much they may 

 shift their places, and be temporarily encroached on by the great storm 

 eddies. 



These regions of high pressure are the places where the winds de- 

 scend, and, as I mentioned in the earlier part of this lecture, these winds 

 are dry, and generally accompany fine weather. On the contrary, the 

 eddies, where the air ascends, are damp and stormy, and especially that 

 part of the eddy that is fed by the southwest winds that have swept 

 the Atlantic since their descent, and so have become charged with va- 

 por. 



And now we are prepared to understand why east, and especially 

 northeast winds are generally so dry. They are air that has descended 

 in the area of high barometer that (especially in the winter and spring) 

 lies over Europe and Asia, and has subsequently swept the cold land- 

 surface, v/hicli does not furnish much vapor, and therefore they reach 

 us as dry cold winds. To begin with, the air comes from a considera- 

 ble height in the atmosphere, and in ascending to that height in some 

 other part of the world, it must have got rid of most of its vapor in 

 the way that has been already explained. In descending to the earth's 

 level it must, of course, have been dynamically heated by the compression 

 it has undergone, but all or nearly all this heat has been got rid of by radi- 

 ation into free space on the cold plains and under the clear frosty skies 

 of Northern Asia and Northern Europe, and it then blows outwards from 

 this region of high barometer over the land, towards the warmer region 

 of low barometer on the North Atlantic Ocean. 



Thus we see that, in all cases, rain is produced by the cooling of the 

 air, and that in nearly all, if not all, this cooling is produced by the 

 expansion of the air in ascending from lower to higher levels in the at- 

 mosphere, by what is termed dynamic cooling. This last fact is not set 

 forth so emphatically as it should be in some popular text-books on the 

 subject, but it is an undoubted fact. It was originally suggested by 

 Espy some forty years ago, but the truth is only now generally recog- 

 nized, and it is one of the results wlii<;h we owe to tlie great advance 

 in physical science eflected by Joule's discovery of the definite relation 

 of equivalence between heat and mechanical work. 



