370 THOMAS GRAY ON THE EFFECT OF PERMANENT ELONGATION 
elongation. I find, however, that if the change of resistance due to stretching 
be measured when the stretching weight remains on the wire, the effect is 
considerably greater than if it be measured, for the same elongation, with the 
weight off. This shows that the resistance of the wire is increased by pulling 
it, altogether independently of the change of form due to the pull. This same 
result has been obtained by THomson (“ Electro-Dynamic Qualities of Metals,” 
Phil. Trans., 1856) and by Tomuinson (Proc. Roy. Soc., 1877, vol. xxvi.). 
The ratio of lateral to longitudinal change of dimensions was used by 
THoMSoN, and afterwards by ToMLINSon, in the deductions from their experi- 
ments. <A glance at the table of results for the elasticity constants given 
in this paper is sufficient to show that the same results follows from that 
method of treating my results. Although I have given the determinations 
which I made of the elastic constants, and of the ratios of lateral to longitudinal 
change of dimensions deduced from them, I do not consider it likely that the 
ratio is sufficiently accurate to base any calculation of the effect of pull upon. 
The constants are chiefly valuable as giving an indication of the quality of the 
wires used. 
The apparatus used in the electrical part of the experiment is shown in fig. 
1, which is a plan of the arrangement. In the figure, A represents an ordinary 
wire bridge, by ELLioT, consisting of a platinum wire w 1 metre long, and 
‘12 centimetres diameter, attached at its ends to thick copper bars furnished 
with binding screws. The effective length of this bridge wire was increased by 
the addition of a coil of German silver wire C to eachend. The total length of 
the wire was thus made equivalent to 22430 divisions of the scale 8S. Along 
this wire a contact key K can be moved, and the position of its contact with 
the wire read on a scale of millimetres S attached to the base of the instru- 
ment. This allowed a change of resistance of z94q5 Of the whole to be easily 
observed, a degree of accuracy beyond what could be expected in the elonga- 
tion readings. The wires under experiment, and the standard wire with which 
its resistance was compared, are represented at a and b. These two wires were 
in every case of exactly the same material, and were arranged side by side, so 
as to avoid variation due to change of temperature in the exposed wires. 
The galvanometer G, used in the bridge, was a THomson’s dead beat instru- 
ment, having an internal resistance of 0°325 Ohms. The galvanometer zero was 
controlled by a magnet m placed in front, and its sensibility by a second magnet 
m, placed at the side. The battery B consisted of a single small DANIELL’s 
element—this being quite sufficiently powerful when the magnet m, was pro- 
perly adjusted. K, is a key used in the ordinary way for making and breaking 
the battery circuit. The elongations of the wire a were measured on a scale 
of millimetres M, the length of the wire being reckoned in terms of the divisions 
of this scale. 
