m 
Proceedings of the Royal Society 
If, oh the same plane with the line of simple flexure ABODE, 
a circle he described round S, the point of suspension of a supposed 
pendulum, and if the points, P in the curve and n in the circle, be 
so related that the directions of the two curves be always parallel to 
each other, then a uniform motion of P along the curve ABODE 
(PI. III. figs. 10, 11) will be accompanied by an oscillatory motion 
of n (from N to N' and back), exactly in imitation of the motion 
of a pendulum. 
In the twenty-fourth and twenty-sixth volumes of the Society’s 
Transactions^ I have described a very expeditious process for com- 
puting the motion of a heavy body along the circumference of a 
circle, which process, by an obvious modification, enables us to get 
the integral 
^^(cos a - cos A)~^~da 
in any proposed case. Thus we may obtain the length of the curve 
intercepted from the vertex B to any proposed horizontal line. 
It remains for us to investigate the relation of the abscissa y to 
the inclination. Substituting for x its value, we obtain from the 
above equation 
xdy = c ^ . cos a.da , 
whence 
^cos «(cos a-cos A) da 
c r "" 
y = —j^J ^ a. da, 
and thus the problem to determine the form of the curve of flexure 
by help of its co-ordinates resolves itself into a transcendental 
integration. 
2. On the Measurement of Eesistance to Electrolytes. By 
Cargill G. Knott, D.Sc., E.R.S.E. 
The difficulties attending the measurement of the electrical resist- 
ance of electrolytes are well known, the rapid growth of the 
polarisation of the electrodes especially preventing the application 
of the ordinary Wheatstone Bridge method. The polarisation may 
be kept down by using alternating currents, as was done by Kohl- 
