— 
TRANSACTIONS OF SECTION G. OD 
referred to must flow in that manner, because the cleavage planes between the 
discs produce discontinuity in the material in the direction of the axis of the 
cylinder, and thus prevent flow taking place in that direction, if does not 
follow that the flow of a solid cylinder of lead must be the same. 
With the object of determining the nature of the flow of solids, the author 
carried out a series of experiments, which he believes prove that the flow in 
every case below a certain critical velocity consists of molecular irrotational 
and rotational motion, the direction of the flow being stream-line, and that the 
flow is steady, straight-line motion until the stream changes its course in con- 
verging towards the orifice, and in the case of the extrusion of solid lead from 
an orifice there is no radial flow or concentric contraction as has been assumed 
until the stream-lines begin to converge towards the orifice. In fact, the solid 
material behaves exactly as a noncomprehensible viscous fluid. When the 
motion is very slow the flow is perfectly stable, and when above a certain 
eritical velocity eddies tend to form, and turbulent motion results, and the 
temperature of the material undergoing extrusion rises. 
The method of carrying out the experiments was as follows: The various 
die-blocks were made in two halves, definitely registered, and strongly bolted 
together, and fitted with pistons, the sections of the die-blocks being parallel 
to the direction of flow. The lead cylinders for the experiments were similarly 
formed in two halves, which were compressed in the die-blocks by the pistons, 
with the orifice plugged so as to render the lead perfectly homogeneous and 
free from the possibility of local motion. The face of one of each pair of semi- 
cylindrical blocks thus formed was then grooved at regular distances apart by 
grooves parallel to their axes. Tin wires were then placed in the grooves, and 
the pairs of semi-cylinders again compressed in the die-block with the orifice 
plugged, which had the effect of fixing the wires securely in the grooves by 
compressing the lead round them, after which the plug was removed from the 
orifice, and the extrusion of the lead cylinder carried out. The compression and 
extrusion of the lead cylinders was done under the compression heads of a 
fifty-ton testing machine, which allowed stress-flow curves to be taken at various 
speeds of extrusion. 
The tin wires retain their position in the lead blocks, and show clearly the 
direction and general conditions of flow. 
The flow of lead in a cylinder with sudden concentric enlargement in the 
direction of flow, also the flow past a solid of revolution, was investigated, and 
in both cases the motion was found to be of stream-line order, as in the case of 
extrusion from an orifice. The effect of punching a solid block with and without 
shearing out the wad was also investigated, and the stream-line flow determined. 
The behaviour of other metals under suitable conditions was investigated, 
and the same results were found as with the lead. 
4. Note on the Slrength of Free-ended Struts. 
By Anprew Ropertson, M.Sc, 
This paper gave an account of experiments made on mild steel struts to 
ascertain the law of variation of strength with length when special precautions 
were taken to secure (a) axial loading, (6) straightness, (¢c) freedom of the ends 
from constraint. 
The conclusions are :— 
1. That Euler’s law is followed down to the length for which the load per 
square inch given by this law is equal to the stress at yield. 
2. That below this limit collapse occurs when the load per square inch is equal 
to the yield stress. 
3. That for round struts of length less than five diameters there is no definite 
ultimate load. The transition from the elastic to the plastic state is marked 
not by collapse of the strut, but merely by a sudden deformation of appreciable 
magnitude. 
Tn all the cases in which the collapsing load per square inch was equal to the 
yield stress the specimens were bent, and on relieving the load and again testing 
them they failed under a load smaller than the first collapsing load. This is 
