STORMS A XD WEATIIKR FORECASTS 79 



Tlie three conditions essential to the formation of tornadoes 

 are clearh' as follows : (1) A cyclone or area of low pressure, tiic 

 center of which is to the north or northwest, with a l)aronietric 

 pressure not necessarily much l)elow the normal ; (2) a tempera- 

 ture of al)out 70 deurees on the mornin<>; map; (8) a great hu- 

 midity, and (4) that the time of year l)c jNIarch ]•"> to June 15. 

 These conditions may and often do exist sei)arately ; one or two 

 of them may he found coexisting ; hut so long as the third is alt- 

 sent, tornadic formation is not likely to occur. 



I am satisfied that the nnmher of these storms is not increas- 

 ing; that the breaking of the virgin soil, the planting or cutting 

 away of forests, the drainage of land surfaces by tiles, the string- 

 ing of thousands of miles of wire, or the laying of iron or steel 

 rails have not materially altered the climatic conditions or con- 

 tributed to the frequency or intensity of tornadoes. As well 

 might one b}' the casting of a pebble attempt to dam the mighty 

 waters of the majestic Mississippi as attempt the modification 

 or restriction by the feeble efforts of man of those tremendous 

 forces of nature which surround our earth and control our storms 

 and climate. To be sure, as towns become more numerous and 

 poi)ulation becomes more dense, greater destruction will ensue 

 from the same number of storms. 



It is not ])ossible with our ])rescnt knowledge of the mechan- 

 ism of storms to forewarn the exact cities and towns tbiitwill be 

 visited by tornadoes witlu)ut alarming some towns that will 

 w-holl}' escai)e injury; but we know that tornadoes are almost 

 entirely confined to the southeastern quadrant of the cyclone, 

 and that when the thermal, hygrometric. and other conditions are 

 favorable, the s[)ot 300 to 500 miles southeast from the cyclonic 

 center is in the greatest danger. 



Chart XV shows the conditions on the evening of the St Louis 

 tornado, two hours after its occurrence. The abnormal heat, 

 humidity, and other conditions of the rather small and weak 

 cyclonic system shown by the morning chart were suflicient to 

 justify the Weather Bureau in distributing at 10 a. m. danger 

 w^arnings throughout the whole of Missouri and eastern Kansas. 

 I am informed that the schools of St Louis were dismissed at 

 once on the receipt of the warning forecast. What is urgently 

 needed is a system by which weather signals may be sent simul- 

 taneously from telephone headquarters to all subscribers by a 

 stroke of a telegraph key; then a whole city couUl be warned in 

 a minute's time. 



