410 Variations of the Wind and Barometric Pressure. [Dec. 7, 



curve of barometric pressure. This implies that the greater height of the 

 morning maximum of the barometer is due chiefly to an influx of air 

 from the eastward, coming from the sun. 



rig. 12, which is formed simply by the addition of the ordinates of figs. 

 10 and 11, so closely corresponds to those derived from actual observation 

 of the barometer, and its range approximates so nearly to the actual 

 diurnal range of the barometer in low latitudes, as to leave little room to 

 doubt that the true explanation of the large features of the diurnal 

 variation of the barometer is to be found in the diurnal variation of the 

 wind. 



This explanation requires that in high latitudes a reversal of the baro- 

 metric variations should be found to compensate that part of the diurnal 

 barometric variation of low latitudes which is due to the variation of the 

 north component of the wind. It also leads to the anticipation that 

 that part of the barometric variation which is due to the variation of the 

 east component of the wind will, speaking generally, have the same 

 character from equator to poles, having a maximum range at the equator 

 and vanishing at the poles, like the variation of that component of the 

 wind on which it depends. A reason is herein found why the reversal 

 of the actual barometric variation should not occur until very high lati- 

 tudes are reached. But it will probably be found, when other wind- 

 observations of higher latitudes than Berijiuda are similarly discussed, 

 that the part of the barometric variation which is due to the variation of 

 the north component only, is reversed in character at a much lower lati- 

 tude than the actual total variation. The latitude in which the variation 

 may be expected to be nil will be that where the reversed barometric 

 variation, due to the variation of the north component of the wind, is 

 equal to and therefore neutralized by the direct variation due to the vari- 

 ation of the east component. 



Other causes, such as the variation of vapour pressure and the dyna- 

 mical reactions pointed out by Espy, which are the equivalents of the 

 energy expended by or upon the atmosphere in expanding and contract- 

 ing under the influence of gravity, doubtless affect the actual observed 

 variations of the barometer to some extent ; but, since the diurnal varia- 

 tions of the wind are sufficient to account for nearly the whole of the 

 barometric variation, the inference that other causes really exercise but 

 a minor influence seems inevitable. 



If we could suppose the centre of gravity of each atmospheric column 

 to remain vertically undisturbed while the increasing or decreasing ten- 

 sion of the air on different parts of the earth's surface, caused by the 

 diurnal variation of temperature, was being equalized solely by lateral 

 expansions and contractions, it could be shown that a very small lateral 

 velocity of the air would suffice for such equalization, and that the dyna- 

 mical reactions resulting from the creation of these lateral movements 

 would be small compared with those which would result from a vertical 



