1874.] The Iowa and Illinois Tornado. 379 
it reached only a few miles to the south of the tornado, and 
that very little rain or hail fell there. 
There is also a general agreement among the witnesses that 
a little hail fell some minutes before the advent of the funnel, 
and a little rain immediately after its passage. The whole 
appears to have occupied at least thirty minutes in passing. 
This would give the cloud at the funnel a width of at least 
15 miles in the line of its progress. Probably it was con- 
siderably greater. We may, therefore, safely conclude that 
we are within the limits where we assign to the centrifugal 
cloud an average horizontal diameter of 30 miles. 
I have no data for determining the elevation of the under 
side of the centrifugal cloud beyond what is involved in those 
for measuring the altitude of the centripetal. There must, 
however, have intervened a very considerable space between 
the two. For if there was nothing intervening there could 
be nothing to prevent the under, and specifically lighter, 
air from taking the shortest way up. There could be no 
reason why it should first rush to a centre, ascend there, 
and then rush from it. 
Dimensions of the Centripetal Current.—The data for deter- 
mining these are still less precise than the preceding, but 
yet sufficiently so as to give a more definite idea than any 
mere description in general terms. 
The witnesses generally testify that hail fell from fifteen 
to thirty minutes before the whirlwind, and rain for as long 
a period after. They further almost unanimously say that 
the wind was easterly with the hail and westerly with the 
rain. If we assume twenty minutes as the duration of each 
of these winds, we obtain 20 miles as the diameter of the 
inrushing winds at the earth’s surface. 
If the disturbing influences were not much greater at the 
earth’s surface than at a higher altitude, the dimensions of 
the whirl should, within the limits of the centripetal wind, 
owing to the absence of friction, vastly increase as we 
ascend. Moreover, since the funnel was first formed at a 
considerable elevation, and since it touched the ground with 
a narrow point, and merely incidentally, as it were, since it 
for long distances ceased to strike the earth, and yet pro- 
ceeded with undiminished energy; since, in short, both the 
originating and sustaining sources of its power seem to 
have mainly existed high in the atmosphere, we have the 
Strongest reasons for concluding that the horizontal 
dimensions of the centripetal current was very much greater 
at its more elevated portions than at its base. It would 
therefore appear to be not unlikely that the diameter of the 
