June 13, i8go,] 



SCIENCE. 



353 



of the receiver, from the surface of the earth to the top of 

 the receiver, weighs 20 inches of mercury, the air on the 

 inside will weigh only 19 inches; and of course it will be 

 pressed upwards against the upper end of the receiver at a 

 with a force of about half a pound on the square inch, the 

 bottom of the receiver being open at c. 



"Also, if a barometer should be placed in the inside at the 

 top a, it would stand an inch higher than one on the out- 

 side at the same height: therefore, if a small hole should be 

 made in the top of the receiver, the air would spout out with 

 a velocity due to a head of pressure equal to one inch of 

 mercury. This is equal in weight to about 900 feet of air of 

 mean density at the earth's surface. The velocity with 

 which it would spout, on supposition of its having this den- 

 sity, will be found to be 8 4/9OO = 240 feet per second (164 

 miles per hour). But the air at the top a is only one-third 

 the assumed density, provided no allowance is made for tem- 

 perature; and as the velocity of spouting fluids under equal 



pressure is inversely as the square root of their densities, the 

 real velocity with which the air will spout out at a will be 

 240 \/B = 415 feet per second (283 miles per hour). 



"If, now, we suppose the whole top of the receiver to be 

 taken off, the velocity will be the same, if thei-e is no fric- 

 tion up the sides of the receiver, and the air gets freely in at 

 the bottom ; an allowance, of course, being made for the re- 

 action of the air in the upper part of the receiver on the air 

 below, in consequence of the velocity increasing all the way 

 up. If we remove the entire sides of the receiver, it is mani- 

 fest that the heated column of air, which we suppose to be 

 the same as before, would spread out laterally in ascending, 

 in the form of an inverted cone, or mushroom, as exhibited 

 by the dotted lines p, s, n, e. 



"When the air near the earth's surface becomes very much 

 heated, or very highly charged with aqueous vapor, such an 

 ascending column as is here imagined may actually take 

 place, and be kept up for a long time. The difference of 

 temperatui-e of the ascending column and that of the atmos- 

 phere through which it passes may be much greater than that 

 here supposed, partly caused by its greater temperature be- 

 low, but chiefly from the great quantity of latent caloric 

 evolved by the condensation of vapor into cloud." 



These quotations might be multiplied by the score; but 

 enough has been given to show that Espy relied upon his 

 experiments with the nephelescope for the facts upon which 

 to base his theory, and that, according to his view, the set- 

 ting-free of latent caloric by the condensation of moisture in 

 an uprushing current was the principal factor to be consid- 

 ered in tornado-generation. 



Ferrel's Views. 



Professor Ferrel began writing upon mathematical theo- 

 ries in meteorology more than thirty years ago, and it will 

 be of some interest to quote from both the earlier and later 

 works. According to these earlier views, we may consider 

 that there are two forces acting in the production of a tor- 

 nado or hurricane: 1. A primitive impulse, such as an ab- 

 normal heating of the air, thus giving rise to an upward 

 tendency; 2. A constantly acting force. "This force (2) 

 may be furnished by the condensation of vapor ascending in 

 the upward current in the middle of the hurricane, in ac- 

 cordance with Professor Espy's theory of storms and rains. 

 According to this theory, all storms are produced by an as- 

 cending current of warmer atmosphere saturated with moist- 

 ure, and this current is kept in motion by the continual rare - 

 faction of the atmosphere above by means of the caloric 

 given out of the vapor which is condensed as it ascends to 

 colder regions above. The violence of the hurricane, and 

 also its duration, depend upon the quantity of vapor sup- 

 plied by the currents flowing in below" (see Nashville Jour- 

 nal of Medicine and Surgery, 1856). 



Again: " The preceding condition, found in the unequal 

 distribution of temperature, must be regarded simply as » 

 primary cause of disturbance, giving rise merely to the ini- 

 tial cyclonic disturbances; for without other conditions, 

 depending upon the hygrometric state of the atmosphere, and 

 upon the rate of decrease of temperature with increase of 

 altitude in the atmosphere generally in which the cyclone 

 exists, we would have no cyclone of long continuance or of 

 much violence. If air is saturated with vapor, after ascend- 

 ing to only a moderate elevation, its tension and tempera- 

 ture are so much diminished that the vapor is condensed 

 into cloud and rain; and the heat given out in the conden- 

 sation of the vapor as the air ascends prevents the rapid 

 cooling which takes place in dry air, and the rate of cooling 

 with increase of altitude is reduced, in ordinary tempera- 

 tures and elevations, to less than half of what it is in dry air. 

 Tornadoes are simply very small cyclones, extending over 

 so small an area that the effect of the earth's rotation has no 

 sensible inflluence; and the gyrations arise from a disturbed 

 state of the atmosphere in which the tornado occurs, which 

 renders it impossible for the air to flow from all sides to- 

 wards a centre without running into gyrations around that 

 centre." An illustration of this principle is given in the 

 fl-ow of water from a basin by an opening at the bottom. 

 " In a tornado the diminution of pressure and tension in the 

 centre arises almost entirely from the centrifugal force. On 

 account of the rapidity of the gyrations near the centre, this 

 diminution of pressure may be very great there, while at a 

 very short distance from the centre it is imperceptible. 

 When these gyrations begin above, as they usually do, since 

 the air there is most frequently in the state of unstable 

 equilibrium (i.e., having a tendency to rush upward), they 



