of the Providence Tornado. 45 
Left or North side of the Track. Right or South side of the Track. 
Direction Inclination inward Direction Inclination inward 
of first and backward from of first and backward from 
Case. prostration. course of tornado, |Case. prostration. course of tornado. 
2 S. 23°E. 67° 34 N. 568° id 
3 S. 45 E. 45 36 East. 0 
4 $.12 E. 78 40 N.65 E. 25 
5 S.35 W. (backward) 125 28 N. 63 E. 27 
6 S. 5 E. 85 31 N.75 E. 15 
yf S.40 E. 50 44 N. 63 E. 2 
8 S.11 E. 79 37 N. 87 E. 13 
9 S.10 E. 80 43 N. 30 E. 60 
10{ fell S.25W. aaa 115 Al S. 85 E —5 
toS.11 E. 42 Fast. 0 
11 S. 26°. 64 
12 ©§.55 E. 35 Mean direction of prostration on 
13 8.55 E. 35 the right side of the track, N. 73° E. ; 
14 first thrown N. 23 | o average inclination inward from 
W. (backward) Jf 247 
course of tornado, seventeen degrees. 
15 8. 45°R. 45 Mean direction of first prostrations 
16 §©6©8. 30 E. 60 on theleft side of track, S. 4° W.; ave- 
17 S. 55 E. 35 rage inclination inward and backward 
18 East. 0 from course of tornado, ninety-four 
19 S. 85 E. 5 degrees. 
20 S. 27 E. 63 Relative inclinations of the two 
21 N.55 W. (backward) 215 
first fell N. W. = 
22 turned to S.37E. } 225 
a N. 45°W. (backward) 225 
23 pee, of appie- \ 183 
sides to the line of axis, more than 
five to one. 
It is proper to mention, that the 
average inward inclination of all the 
prostrations on the right side of the 
track for a distance of four miles 
east of the river was thirty degrees*. 
This however does not affect the 
conclusions in favour of rotation to 
the left. 
tree thrown west 
24 S. 20 W. (backward) 110 
25 S.55 W. (backward) 145 
26 South. 90 
27 ~— firstthrownN.10W. 260 
These average results, on the two sides, together with the 
several particular observations already adduced, appear to me 
to afford decisive evidence of whirlwind rotation in this tornado, 
in the direction from right to left, or which is contrary to the 
hands of a watch. In reference to this evidence and that ex- 
hibited in my paper on the New Brunswick tornado, I add 
from my prepared sketches the following figure, as an approxi- 
mate illustration of the whirling action in these tornadoes, so 
fat as this may be shown hor2zontally and by a stationary 
figure. 
Let the involuted lines or arrows on this figure be supposed 
to represent the motion of the wind at or near the bottom of 
a vertically cylindrical portion of the centre of a tornado, com- 
* This larger average gives a relative degree of inclination on the two 
sides of three to one. Nearly the same difference is found in two out- 
side bands of prostration, of equal widths (Tables I. and V.), shown in my 
survey of the New Brunswick Tornado. See Silliman’s Journal, vol. xli, 
p. 78. 
