364 
Proceedings of the Royal Society of Edinburgh. [Sess. 
An average of the three gives for 
(L) = 31,800 
(2) f 2 — 28,200 
and correcting (2) for the value of m by multiplying by T01, we have: 
(1) J\ = 31,800 
(2) f 2 = 28,500 
The theory, of course, only holds good up to the elastic limit of the material 
of the plate, but the author has found no perceptible breakdown in the law 
up to pressures much beyond those which first produce overstrain. 
Now the natural yield point is 27,800 or thereabouts, and therefore 
the second of these formulae, namely /= A 1 ,p , seems to be the correct one. 
t 
It may be objected that it is impossible to tell where such curves as those 
given above cross the abscissa, owing to their tangential form when near 
it. This, of course, is true, and the only guide is to take as much of the 
curve as possible and endeavour by visual means to produce it back 
correctly. The probability is that the results are correct within two or 
three pounds per sq. in. pressure 
Diagram 6 consists of the permanent set curves of three plates, J (6 in. 
diameter, -063 in. thick), K (5 in. diameter, -0645 in. thick), L (4 in. diameter, 
*067 in. thick). 
J. The curve cuts the horizontal ordinate aty> = 19. 
(1) = 32,800 
(2) > 2 = 28,700 
K. The curve cuts the horizontal ordinate at p = 29. 
(1) = 32,700 
(2) / 2 = 29,000 
