870 
iVIE. HODGKIXSOX’S EXPEEIMEXTAL EESEAECTIES 
respectively, or as 23 to 21 nearly; if then ra be assumed as 23, the falling off for 
different values of d will be, as in the example above, becoming successively 22’954, 
22’492, 21’608, 20‘907, 20‘366 ; and the value of d, which gives 21, is about 
Example 3. Again, suppose — —=z-, and the rest as before, 
if fZ=|D, w=m(^D® ®-^x-242 D^-®)=m(D^ “--048 D^-*)=mx-952 D^-^=21-896 D' *; 
if fZ=^D, — ^X '366 -073 /ux '927 D®®=21-321 > 
iid=^I), -458 -0916 D^^)=//iX -9084 D^-^=20-893 D^^ 
4 
and the value ot d, which gives 21, is about 
The pillars in the following summary of results were all cast sohd from models 24- 
inches diameter, and made 10 feet long, with their ends flat. Those of iron, denomi- 
nated No. 1, were chosen for comparison from some of the leading irons in the king- 
dom. They vary in strength from 29-50 tons to 20-05 tons, or nearly as 3 to 2. 
Formulse for the strength of pillars of all the irons tried, thii’teen m number, are given, 
both on the supposition used in the former research, and also on that arrived at in the 
present. The weights which would crush short pillars of the same diameter are hkewise 
given. 
i/' 1)1. Ik' iiNNiiiued 
