352 



SCIENCE. 



[Vol. XV. No. 384 



ing was very much greater than can ever occur in nature. 

 It is also probable that his air was far from saturation, and 

 that the cloud he saw was due to dust which was pumped in. 

 Experiment has shown that such a cloud can be formed in 

 quite dry air, provided the expansion is quick enough. 



One very important element does not seem to have been 

 considered, and that is the velocity of escape of the com- 

 pressed air needed to form a cloud. It is probable that this 

 was more than five hundred feet per second, or at least fifty 

 times as great as can ever take place in the formation of 

 cloud in our storms. It is a little singular that Espy him- 

 self did not discover more than he did. He found, that, 

 while the results he obtained with dry air were very regular 

 and constant, yet with moist air they were just the con- 

 trary; and the irregularities seemed to depend in part upon 

 the interval after the compression, before the escape of the 

 air. The following table is of much interest, and shows, in 

 the first column, the number of minutes after compression, 

 before opening the stop-cook, and, in the second column, the 

 ratio between the amount of compression and the final read- 

 ing of the gauge. In dry air this ratio was found to be from 

 four to five; so that if Espy's statement just given is cor- 

 rect, that the temperature with dry air is reduced twice as 

 much as with moist, the ratios should be eight to ten in 

 this second column, according to his reasoning. 



Minutes. Ratio. Minutes. Eatio. 



13,160 4.3 360 6.9 



10,080 7.U ' 180 7.3 



7,200 4.5 180 6.9 



5,760 4.5 180 7.4 



5,760 4.3 180 7.0 



5,040 4.3 90 6.9 



4,320 4.4 60 7.4 



4,330 4.5 60 7.4 



1,800 4.4 15 5.5 



1,680 4.5 10 7.8 



1,300 4.7 10 5.1 



1,080 5.5 



Espy considers this result, "so contrary to all our notions 

 since the experiments of Dalton on the subject of the dew- 

 point," very remarkable. 



We have not all the data needed to give a complete ex- 

 planation of Espy's results, but this much we do know : if 

 Espy had allowed a slower, escape of the moist air', there 

 would have been no cloud, and he would have had no dif- 

 ference between dry and moist air. When the explosion 

 was very sudden, a cloud was formed, and the conditions 

 inside the apparatus were very different from those with dry 

 air, but not because of the liberation of latent caloric. At 

 the end of ten minutes he found the ratio much smaller, ap- 

 parently, than at the end of an hour or two; and after that 

 the ratio diminished until it became exactly what it was in 

 dry air. It is plain that this final result could not be at- 

 tributed to the gradual n on -saturation of the air, as Espy 

 thought. There are no other experiments made by Espy 

 under exactly similar conditions of temperature in moist and 

 dry air. The evidence is conclusive, from these and other 

 more recent experiments by the writer, that practically no 

 different result will be produced, whether moist or dry air 

 be introduced into the nephelescope. 



I have dwelt upon these experiments at some length for 

 the reason that they form the sum and substance of all ef- 

 forts in this line up to very recent times. 



Upon these experiments depend all of Espy's theories as to 

 tornado-formation, and he has been quite closely followed in 

 all views as to the essence of the forces underlying our most 

 violent tornadoes. Mr. Espy gives the following resumS of 

 his theories as ascertained and promulgated by the French 

 Academy of Sciences: — • 



"If a very extended stratum of warm and humid air at 

 rest covers the surface of a region of land or sea, and by any 

 cause whatever (for example, a less local density) an as- 

 cending current is formed in this mass of humid air, the as- 

 cending force, instead of diminishing in consequence of the 

 elevation of the rising column, will increase with the height 

 of the column exactly as though a current of hydrogen was 

 rising through the common air, which current would be 

 pushed towards the top of the atmosphere with a force and 

 velocity in proportion to its height. This column of heated 

 air may also be compared to that in chimneys and stove- 

 pipes, of which the draught is in proportion to the height of 

 the pipe containing the warm air. What, then, is the cause 

 which renders the warm and humid ascending current lighter 

 in each of its parts than the air which is found at the same 

 height with these different portions of the ascending col- 

 umn? 



"This cause, according to the sufficiently exact calculations 

 of Mr. Espy, is the constantly higher temperature which the 

 ascending column retains, and which proceeds from the heat 

 furnished by the partial condensation of the vapor mixed with 

 the air, making this ascending column a true column of 

 heated air, that is to say, of a lighter gas ; for the weight of 

 the water which passes into the liquid state is far from com- 

 pensating the excess of levity which proceeds from the more 

 elevated temperature which the air preserves. 



"Thus the higher the column is, the greater is the ascend- 

 ing force; and the rushing-in of the surrounding air on all 

 sides will be produced with more energy. To understand 

 this effect better, let us consider a mass of warm and dry air 

 rising in the midst of a colder atmosphere. In proportion 

 as this air rises it will expand, because of the less pressure 

 which it will experience, and consequently become colder; 

 it will arrive then quickly at an equilibrium both of tempera- 

 ture and pressure with a layer more or less elevated, which 

 it will soon reach, and in which it will remain ; but if this 

 only cause of cold, expansion, is overbalanced by a cause of 

 heat (for example, the heat furnished by the vapor which is 

 condensing), this ajr will remain constantly warmer than 

 would have been necessary to attain the same temperature 

 and pressure as the surrounding air. It will then be con- 

 stantly lighter; and the higher the column, the greater the 

 ascending force." 



This statement contains, perhaps, as clear a view of Espy's 

 theories as can be obtained. There are, however, one or two 

 additional opinions, regarding tornadoes specifically, that 

 should be mentioned. Mr. Espy says, "Suppose a receiver 

 (Fig. 4) only a few hundred yards in diameter, but so lofty 

 that its top would reach to where the barometer would stand 

 at 10 inches, and that it contained air about 25.25°, for ex- 

 ample, hotter than the air on the outside, this latter being 

 at a mean, 32°. The column of air then in the inside of the 



25.25 

 receiver would be expanded -^^ „- of the whole, or one- 

 twentieth of the whole bulk. Now, as the air on the outside 



