1906-7.] Variation of Young’s Modulus under Electric Current. 347 
was found that the modulus was approaching a turning value, the readings 
were taken at shorter intervals, so that the behaviour could be accurately 
represented in its neighbourhood. 
When the highest temperature to which it was thought necessary to 
carry the experiment had been reached, the current was gradually 
diminished and readings taken with a decreasing current. In all the 
graphs the readings with the increasing current are denoted by circles, and 
those with the decreasing current by crosses. 
The formula used in computing the modulus was — 
M = ^, 
a.L 
where 
P = stress in dynes. 
L = length of wire. 
a = area of cross-section. 
I = elongation. 
The results are given in the C.G.S. system of units, the value of g at 
Stranraer being 981 '4. 
In order to see if the friction of the pulley introduced an error, pieces 
of each wire from the same coil were suspended vertically. To make the 
conditions of this set of experiments as nearly similar as possible as when 
the wires were horizontal, the tube was clamped in a vertical position with 
the wire inside it. The ends of the tube were stopped with cotton- wool, 
and the ends of the wire outside the tube were protected from air-currents 
by pasteboard cylinders. The same weights were used as when the wire 
was horizontal, and the elongations were measured by the microscopes. It 
was fitted up in a corner of the room and protected from air-currents. The 
temperature was also taken by the platinum thermometer, three mercury 
thermometers being placed at different parts of the tube to make sure that 
the temperature throughout it had become uniform. 
Iron Wire. 
The modulus was first determined without any current, the wire being 
at the temperature of the room. It was then heated by a weak current by 
which the temperature was raised about 3 0, 5, and it was found that the 
modulus had fallen in value. The strength of the current was next 
gradually increased, the effect of this increase being to cause a steady rise 
in the modulus until the temperature had risen to about 53°. Beyond this 
there was a regular fall in the value, the rate of decrease being less than 
