96 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



adduced in which it is possible to calculate the pressure ex- 

 erted by the wind. The most reliable results attained give 

 pressures varying from forty to two hundred and thirty 

 pounds to the square foot, which correspond to velocities of 

 from ninety to two hundred and fifteen miles per hour. 

 There were no rain-gauge measurements of the amount of 

 rain in the immediate path of the tornado; but, from the ac- 

 counts given, Mcintosh estimates that two inches of rain 

 fell over an area of ten miles in width. No cases were 

 found of unmistakable explosion from the expansion of air 

 contained within closed areas; and, indeed, the observer 

 shows that, in all the cases in which houses were burst asun- 

 der in every direction, this result is to be attributed to 

 the violent wind striking against their doors and windows, 

 and pressing inward. The whole interior of the building 

 thus became subjected to a strong pressure, due to the 

 pressure of the air ; the pressure was more than balanced 

 on the side struck by the wind, while on the other side there 

 was nothing to resist it, and the house was burst asunder. 

 Some houses were evidently thrown,.bodily, across the centre 

 of the tornado, and then struck by a contrary wind and shat- 

 tered into fragments. The temperature at the centre of the 

 storm is very generally described as having been decidedly 

 cold, and is assumed with some reason to have been about 

 13 below the average temperature on the surface of the 

 ground, as shown by the Signal Service charts. The baro- 

 metric pressure at the centre of the funnel is computed by 

 Mr. Mcintosh at 15.7 inches. This computation, however, is 

 based upon the well-known formula derived from the mechan- 

 ical theory of heat, and is subject to considerable uncertainty. 

 The sound made by the conflict of winds in the tornado was 

 tremendous, and drowned all other sounds. A man could 

 stand by his house as it was shivered to pieces and not hear 

 the noise of its breaking. A number of observers heard the 

 roar at a distance of sixty miles, like the roar of the ocean or 

 the thunder of the battle-field. Comparing this with the 

 sound of cannonading or the eruption of a volcano, the au- 

 thor writes that " the voice of the tornado is, therefore, one 

 of the strongest known to art or nature." The violent winds 

 experienced at the centre of the storm evidently prevailed to 

 a <n*eat distance from the centre, since numerous heavv ai> 



