182 : a REPORT—1862. 
is founded on a very limited number of experiments on wrought-iron columns 
of 15 to 30 diameters long as compared with others of 60 diameters, which, 
in my opinion, has been prematurely assumed as a general law. With wrought 
iron especially, the crushing-up of the rounded ends would soon bring pillars 
of that form into the condition of flat-ended pillars when the breaking weight 
approached the ultimate strength of the material—a conclusion confirmed by 
observing that the experiments in question are exactly those in Mr. Hodg- 
kinson’s table in which the breaking weights of the pillars are greatest. 
However this may be, the experiments I have given show that short cylinders 
with flat ends have twice the strength of similar cylinders with one end 
rounded. From this it would appear that the law for short cylinders is not 
the same, but altogether different from that obtained by Mr. Hodgkinson 
for long cylinders. 
The discrepancies which appeared to exist between my own experiments 
and those of Professor Hodgkinson induced me still further to inquire into 
the law which seems to govern short bolts of columns of two diameters 
in length. To account for those discrepancies, the experiments were extended 
to columns with both ends rounded; and what renders them interesting is, 
that in short columns with both ends rounded the powers of resistance are 
nearly the same as those with one end flat and one end rounded, and moreover 
they appear to follow a different law from that of Professor Hodgkinson’s long 
columns, which, in most cases, broke by flexure. 
The difference in strength between short columns with both ends rounded 
and those with one end flat and one end rounded is almost inappreciable, as 
will be seen by comparing their values as under :— 
Tons per square inch. 
Columns of two diameters long with flat ends crushed with 54:82 
Columns with one end rounded and one flat 4 3 26:86 
Columns with both ends rounded........ S 3 23°88 
So that the difference between them may be taken as the numbers 55, 27, 
and 24, or, in other words, in the ratio of 1: ‘49 with one end rounded and 
one end flat—that with both ends flat representing unity—and as 1 : :44 with 
both ends rounded ; a comparatively slight difference between those with one 
end flat and the others with both ends rounded. 
With regard to the dynamic effect, or work done, by round-ended shot as 
compared with flat-ended ones, it has already been shown that with dead pres- 
sure the indentations produced on wrought-iron plates by a round-ended shot 
are nearly 33 times greater than by those with the flat ends, and that the 
work done is twice as great in the case of the round ends as compared with 
that by the flat ends. This may be accounted for by rounded shot striking 
the plate with its; pointed end, and the force of the blow being given by a 
comparatively small area; the vs viva or 
the whole force is thus concentrated and 
driven into the target to a depth consider- 
ably greater than if spread over the whole 
area of the projectile. The flat-ended 
cylindrical shot, which indicates such 
powerful resistance to pressure, is gene- 
rally fractured by one or more of its sides 
being forced downwards in the direction 
of the line a, and hence its superior resist- 
ance when the whole area of the cylinder 
forms the base as the means of support. 
