184 ’ H. A. Hazen— Tornadoes. 
west side low temperature and cold and dry northwest winds, 
due to an advancing high area. On the southeast side, however, 
now 
tracks upon the map we shall find them in the region to the 
southeast, and generally hundreds of miles from the storm- 
center, where there are no northerly winds. A critical study 
of tornado tracks upon a weather map will develop these points 
in nearly every instance. , eae cae there are in the tornado 
region higher temperatures in the north than in the south, and 
very seldom will there be a lowering of temperature from sout 
to north any greater than the constant difference due to the 
difference in latitude. 
Again, there seems to be a disposition to advance some gen- 
eral and sweeping theories about cold or hot winds underrun- 
ning hot or cold winds, ete. It may said that, in a condition 
of equilibrium, the cooler air must flow beneath and, at all 
times there would be a more or less insensible diffusion of the 
two masses. It would seem difficult to asp niga in this way for 
the definite formation of a tornado aacied ae advance dip- 
c 
selected for this paper pele -one of the tornadoes Rescrited in 
Sergeant (now Lieut.) Finley’s paper, “Characteristics of 600 
tornadoes,” also published by the Weather Bureau. Before 
the tracks were projected upon the maps, every tornado having 
special destructive action and at the same time a fairly well 
determined note of time, was tabulated and of these none were 
pe out in the final result. Tornadoes between Sepiecober, 1872, 
nd September, 1879, have alone been considered. 
ae ae rie 
