June 1.3, 1890,] 



SCIENCE. 



355 



current is satisfactory, and we have seen that Professor Fer- 

 rel has done the same thing in his later studies. The fatal 

 objection to Faye's view is, that a downrush of air must 

 necessarily cause its heating; and it has been proved that in 

 such a case there can be no condensation of vapor and pre- 

 cipitation, but just the reverse. 



Objections. 

 I am well aware that the position of an objector is an un- 

 enviable one, and to be avoided if possible. It seems to me, 

 however, that all modern theories of tornadoes have in- 

 trenched themselves behind such a flimsy breastwork of fact 

 that they can be regarded as little more than the "baseless 

 fabric of a dream," and it is very important that the utter 

 valuelessness of these supposed experiments be demonstrated. 

 An attempt has been made to give, as nearly as could be 

 done in so short a space, all the theories that have had their 

 origin in Espy's experiments, and it is my purpose to exam- 

 ine these a little in detail. 



Nephelescope. 

 The earlier form of this apparatus just described was su- 

 perseded some years later by a double instrument, with 

 which very extensive researches were made. These later 

 experiments were much more complicated than the earlier, 

 but, when properly interpreted, do not seem to lead to any 

 different conclusions. It has already been shown that Espy 

 was entirely misled by his results, and that, if they show 

 any thing at all, they show that moist air did not behave 

 differently in his apparatus from dry. Somewhat the same 

 line of research, carried on by the writer with the best of 

 modern appliances, has shown conclusively that no different 

 result is had on the condensation of cloud from moist air 

 than from the expansion of air in which no cloud is formed 

 {see American Meteorological Journal, September, 1889). 



Summation of Theories. 



In tornado-generation there must be (1) a primitive im- 

 pulse or an unstable equilibrium, brought about (a) by a 

 local diminution of density,, or (6) by an abnormal heating of 

 the air near the earth, or (c) by a warm current from the south 

 underrunning a cooler from the north ; (2) a constantly act- 

 ing force furnished by the condensation of vapor in an up- 

 rushingcolumn of warm, moistair, which produces rarefaction 

 by means of the caloric given out of the vapor; (3) work per- 

 formed by the uprushiug air in pushing aside the atmos- 

 phere into which it expands; (4) gyrations of great velocity 

 from right to left set up in the uprushing air, which, from 

 the centrifugal effect, produce a great diminution of press- 

 ure (corks fly from empty bottles, cellar-doors are thrown 

 off, whole houses burst open, etc.) ; (5) a violent inrush of 

 air into the partial vacuum in the centre, the whole effect 

 being likened to that of heated air rushing into a flue or 

 -chimney; (6) an origin of these motions in the cloud region 

 from which they are gradually propagated to the earth by 

 the action through friction of the upper strata upon the 

 lower; (7) a progressive motion of the tornado in the direc- 

 tion of the general drift of the atmosphere from south-west 

 to north-east; (8) frequently a passing from a severe thun- 

 der-storm into a tornado. 



Let us consider each of these theories in turn. 



1. A Primitive Impulse. — It is a mistake to suppose that 



the sun heats any particular locality, whereby circumscribed 

 ascending currents are set up. The hottest part of the day, 

 that is, the time during which the heating effect is greatest, 

 is about one hour, so that the heat of the sun acts uniformly 

 upon a circle about a thousand miles in diameter. This 

 heat acts only upon the earth's surface, and that in turn 

 upon a layer of air only a few feet in thickness, as has been 

 proved by experiment. To avoid these difBculties, it has 

 been suggested that warm south wind underruns that which 

 is cooler from the north. This, however, is an impossible 

 condition, for the denser must always be beneath the lighter; 

 moreover, as we have seen before, the clouds are always in 

 the same direction as the lower wind. Finally, by (7) we 

 see, that, if the tornado drifts in the upper current from 

 south-west to north east, that current certainly cannot be 

 from the north. 



2, 3. Condensation op Vapor, and Work performed. — 

 There is nothing in the science of meteorology, or possibly in 

 any physical science, that has been developed from such a 

 worthless origin as this theory of the liberation of energy on 

 the condensation of moisture. We have already seen that 

 Espy's own researches contradicted themselves. Ought we 

 not to be allowed to theorize a little on this question? 

 The most important effect of the liberation of latent heat 

 is conceded to be the heating of the air, which produces a 

 marked diminution of density. In cooling any air, we must 

 consider the heating effect upon the surrounding air; that 

 is, if a mass of air is cooled, the heat must be used up either 

 in performing work or in heating other air. Take a cubic 

 foot of saturated air at 80°, and cool it down to 79° by 

 expansion : moisture is condensed, and latent heat becomes 

 sensible. For the sake of the argument, let us suppose that 

 a fraction of this heat is used in performing the work of ex- 

 pansion The remainder of the heat will warm up, let us 

 say, a cubic foot of air near by to above 81°, thereby ren- 

 dering it no longer saturated. We shall have, then, one 

 cubic foot of air at 79" and saturated, and another at above 

 81° and unsaturated. If we mix these, we shall have two 

 cubic feet of air at just above 80° and unsaturated, and this 

 must be cooled below 80° to saturate it ; so that the air in no 

 wise is heated by the liberation of latent heat. 



We may look at this from another standpoint. The lib- 

 eration of an infinitesimal amount of latent heat from the 

 condensation of moisture would be just sufficient to re- 

 evaporate the condensed moisture, so that no heat could be 

 spared for heating the air. Even if a portion of the heat is 

 used up in performing worJc, it is very plain that that could 

 not be used in heating the air. Surely, no one thinks for a 

 moment th'kt this heat from condensation can be used for two 

 purposes, each of which must take all of it at the same time. 

 This reasoning is so obvious, that it is difficult to see how the 

 force of it can be avoided. We have arrived at precisely the 

 same result both by most careful experiment and by an un- 

 answerable train of reasoning. The proof is overwhelming 

 that this great source of energy amounts to nothing. 



The following language, recently used by Professor Davis, 

 seems very remarkable. When the air is warm and satu- 

 rated, he says {Ajnerican Meteorological Journal, Decem- 

 ber, 1889), ''a given amount of cooling causes a larger 

 amount of condensation than the same cooling of cold satu- 

 rated air. This is thoroughly in accord with well-tried 



