Tornado, April 29, 1909. 29 



It is the passage of these cyclones across the Mississippi Val- 

 ley which generally brings about our changes of weather. Here 

 in Tennessee the center of the cyclone generally travels on an east- 

 ward or northeast track lying to the north of us. The winds 

 flow in toward the low center; so after a spell of clear, cool 

 weather, as a cyclone approaches from the west the winds set in 

 toward its center, thus passing over us from the east, changing 

 to southeast and south winds as the center gets more nearly 

 north of us, bringing moisture from the ocean and Gulf of Mexi- 

 co, forming clouds and rain. As the center passes on, the winds 

 whirl round, blowing from the west and northwest, the barome- 

 ter rises and the temperature falls and we come under the in- 

 fluence of a high-pressure area or "anti-cyclone." 



Now in the spring and early summer, when the sun is rapidly 

 coming north, we find from experience that these atmospheric 

 changes are intensified, and then it is that the dreadful "Tornado" 

 is most likely to occur. It is a notable fact that tornadoes are 

 not most frequent very close to the center of the cyclone where 

 the pressure is lowest, but from two hundred to eight hundred 

 miles south or southeast of that point — most frequently about six 

 hundred miles distant. Usually, when conditions are favorable, 

 a number of tornadoes, following approximately parallel paths, 

 occur over the same general region on the same day. On Feb- 

 ruary 19, 1884, there were as many as forty. The tornado has a 

 narrow and comparatively short track, usually from five to 

 twenty miles long and a quarter of a mile or less broad, and last- 

 ing altogether generally less than an hour, while its duration at 

 any one point may be less than a minute. 



The tornado near Clarksville, Montgomery County, on April 

 29, 1909, about 8:30 p. m., illustrated very clearly, in its general 

 features, the laws just noted. On the morning of that day, as 

 shown by the weather map published at Nashville, there was a 

 marked cyclonic center in Eastern Kansas, close to Kansas City. 

 By the next morning it had advanced to Northern Indiana and 

 Southern Michigan, so that at the time of our tornado it must 

 have been about central Illinois — some five hundred miles north 

 of us. Also the weather records show there were several other 

 tornadoes in our general region near the same time. Moreover, 

 on the day of the storm, which occurred about 8 :30 p. m., the tem- 



