On the Rotary Action of Storms. 67 
deviates eastward from the meridian, and ultimately describes 
over the earth’s surface a curved line bearing towards the east. 
A current of air from latitude 45° north, having a due south di- 
rection, soon reaches regions moving faster to the east, falls behind 
them and describes a curve. to the west. Winds oblique to the 
meridian are similarly affected. ‘These familiar matters are re- 
ferred to here, and illustrated by figure 1, to elucidate what fol- 
lows. 
The influence of the figure and revolution of the earth upon 
east and west winds, must also be considered. A parallel of lat- 
itude, being a lesser circle of the globe, and at all points equally 
distant from the pole, necessarily describes upon the earth’s sur- 
face a curved line. But a direct course, due east at the com- 
mencement, follows a great circle, and parting from the parallel 
reaches a lower latitude. ‘The due east course continued ina 
right line describes a tangent to the curve of the latitude. The 
velocity of the earth’s surface at any place, by virtue of the di- 
urnal revolution, has for its direction the line of that tangent ; 
and when the air reposing over any spot is transferred to a region 
of diverse motion, the direction, as well as the degree, of its pre- 
vious force is to be taken from that of the soil on which it pre- 
viously rested. Hence a wind from due west, if in our hemi- 
sphere, will soon be found pursuing a southeasterly course, and 
crossing successive parallels of latitude. 
The labors of Mr. Espy have been directed to the hypothesis 
of a central ascending column of rarefied air, and centripetal cur- 
rents from every side rushing towards its base. Without pursu- 
ing his reasoning, it will be safe to assume that his collection of 
facts established the existence of a qualified central tendency of 
the air, in both the general storms and the smaller tornadoes. 
