TORNADO. 



687 



gressive storm by whose side they arc developed, keep- 

 ing their average distance of several hundred miles 

 from its centre. Tliey occur must frequently in the 

 months of .May, April, June, and July, in the order 

 gi-en, and from 3.'M to 5 p. M., immediately after the 

 wannest part of the day, and when warm ascending 

 currents are most likely U> meet cool descending 

 currents. 



In the United States 3000 persons have been killed 

 and as many injured !>v these storms, while the lo^s ,if 

 property reaches score's of millions of dollars. They 

 occur most frequently in the States bordering the 

 northern Mississippi and the lower Missouri. Here 

 the warm and very moist winds from the (Julf meet 

 with currents of cold air from the north, exceedingly 

 unstable atmospheric conditions being produced. They 

 seldom or never occur west of the 100th meridian, a 

 region unvisited by severe cyclones. Of the various 

 tornadoes on record the most remarkable scries was that 

 of Feb. 9, 1881. On that, day. after II) A. M , mor.j 

 than 00 tornadoes occurred in tlieStites Iwtwcen Illi- 

 nois and Virginia on the north, an 1 Mississippi an 1 

 Georgia on the south, at distances of from .VM to ii . i > ) 

 miles from the cyclone centre. The damage was ex- 

 cessive; more than 10. o;0 buillinu's were destroyed, 

 800 persons killed, and 2.YK) wounded. 



Thunder and hail-storms occur in the sann advan ; 

 quadrant ot'tlu cyclone with the torna 1 >, an 1 possibly 

 all arise from the same general conditions. Tii-sj 

 .stunm are doubtless all du to the effort at equilibra- 

 tion in very unstable atmospheric con litions. Th.j 

 tornado is always preceded by warm an I linmil south- 

 erly winds, and followed by -old northerly ones. The 

 preceding atmosphere k, in tact, saturate 1 with vapor, 

 Its abnormally high temperature in-r.Msing its capacity 

 in this re-ip -t. There is probibly a very rapid dim- 

 inution of temperature with allit'i l.\ an 1 undniiht- 

 edly the larue temperature gradients vertically, an i th ; 

 marked contrast of vapor eon litions. have uniuli to do 

 with the development of the torn 1 1 >. which is natur- 

 ally propagated along the line and in the direction in 

 which the unstable condition of equilibrium prevails. 

 The gyration noticeably forms in the upper air, whore 

 the win 1< attain a much greater velocity than on the 

 surface, and gradually descends. Tne warmth and 

 high vapor conditions of the southern and eastern 

 re'.'ioiu of the cyclone are doubtless the reason that tor- 

 nadoes appear there and not in the northern and 

 western regions, where the air is comparatively dry 

 arid cold, and the equilibrium of the atmosphere much 

 less disturbed. 



Several theories as to the immediate cnnso of torna- 

 does have been advanced. (Juyot ascribe I them to 

 the meeting of opposite winds of different tempera- 

 tures in the upper atmosphere. According t > a moro 

 recent theory, the tornado is preceded by tins forma- 

 tion of a surface layer of abnormally warm air. above 

 which extends a layer of much cooler air. The former, 

 in its effort to ascend, pierces_ tho latter at some point, 

 and rushes upward in a spiral whirl or eddy which 

 quickly draws the surrounding warm air into its vortex. 

 This theory is based on tho analogy of the behavior of 

 water under circumstances similar to those assumed, as, 

 to adduce a familiar instance, when the stopper is 

 pulled from the escape-pipe of a bath-tub. In this 

 case the water rushes from all sides towards the vent, 

 and eddies round in a descending spiral as it passes 

 downward through tho pipe, a water vacuum existing 

 in the centre of the eddy. The analogy to the tornado 

 seems very close, yet this theory, though supported by 

 some eminent meteorologists, is hardly in accordance 

 with the facts. Instead of the preceding calm condi- 

 tions which it seen i -i to require, the tornado is preceded 

 by strong winds. Nor does it offer any explanation of 

 the cold north winds whicli follow the tornado, nor of 

 the rapid and somewhat regular progressive motion of 

 the latter. It is in agreement with the water analogy, 

 but hardly with the atmospheric conditions. 



There are other water eddies, however, of more 

 normal character, that of which we have spoken need- 

 ing unusual conditions for its manifestation. The nor- 

 mal eddy requires moving water for its propagation, 

 and never appears when this water moves steadily and 

 regularly. But at the least disturbance of this regu- 

 larity, eddies are generated, and increase in width and 

 violence as the disturbance becomes greater. In short, 

 the ordinary eddy is due to opposition to the move- 

 ment of water currents, either from the meeting of two 

 currents more or less opposed in direction, or from the 

 uupingiuent of a current on an unyielding obstacle 

 and the hurling of the water back upon itself. The 

 latter is the case in the tornado-like whirlpool at Ni- 

 agara. In every case we have the meeting of two 

 bodies of water which cannot pass each other, and 

 whoso progressive million is converted into a whirl, 

 while the accumulating waters can only escape down- 

 wards, so that the whirl is converted into a descending 

 spiral. 



This form of eddy appears to be the trne analogue 

 to tho atmospheric eddy known by the various names 

 of whirlwind, tornado, etc. In the latter case wind is 

 always present, and apparently opposing winds. In 

 the little dust-whirl of the streets this seems evident. 

 It only appears during gusty winds, where the air 

 currents frequently vary their direction, and it seems 

 to indicate the meeting of two opposing puffs of wind, 

 which breal- ; nto an eddy as they me t, while the pro- 

 LM. ive motion of the whirl is due to the gradual 

 meeting ot the angularly opposed fronts of the 

 cuiTonts. In this phenomena we probably have, on a 

 small scale, what is repeated in 'he tornado on a large 

 scale. 



IJel'ore proceeding with the argument, however, a 

 more particularized statement of tornado conditions is 

 important, The disturbance, as already said, always 

 in the upper air. Lieut. Maury, in J882, at- 

 tributed the tortrado to an upper air current seen at 

 times to move from the south-west at a speed of JOO 

 miles per hour. Numerous descriptions show that the 

 first tornado indication is a cloud to the northward, 

 heavy, black, and comparatively slow ir. its southward 

 movement. This is met by a light, rather smoky, and 

 more rapid cloud from the south-west. These clouds, 

 borne by their respective winds, rush together, violent 

 conflicts of currents drive the clouds in every direction, 

 up and down, round and round; clouds, like great 

 sheets of white smoke, dash about in a frightful man- 

 ner. Finally, a black, threatening mass descends 

 slowly towards the earth, whirling violently, but still 

 manifesting confusion in form. This soon gives place 

 to the peculiar funnel-like shape, intensely black in 

 appearance. Aa white clouds appear and are drawn 

 into tho vortex, the funnel trunk sways like an elastic 

 column. Substances drawn up are seen flying in its 

 centre, gradually rising, to be finally flung out near the 

 top after the storm has progressed a mile or two. 

 Dark masses of cloud shoot down the sides of the fun- 

 nel, and are drawn into the' vortex near the ground. 

 At times the velocity appears to increase, and every- 

 thing is torn to splinters. A peculiar hollow, rumbling, 

 but very loud sound accompanies the storm. 



This description of the genesis of a tornado, abbre- 

 viated from Prof. Eddy's paper in the Popular Science 

 Monthly, is in accordance with the following descrip- 

 tion of the tornado of JMay 30, 1879, from the Signal 

 Service Report : 



"The cloud from which the funnel depended, seen 

 at a distance of 8 miles, appeared to be in terrible 

 commotion ; in fact, while the hail was falling, a sort 

 of tumbling in the clouds was noticed as they came up 

 from the north-west and south-west, and about where 

 they appeared to meet was the point from which tho 

 funnel was seen to descend. There was but one fun- 

 nel at first, which was soon accompanied by several 

 smaller ones. . . . Finally, one of them seemed to 

 expand and extend downwards more steadily than the 



