732 PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 
Matteucci, and I believe Kirchhofp and others, have used in tracing equipotential 
lines on the surface of a conductor traversed by an electric current, occurred to rne. 
Six months later, I thought of the multiplying branch (first used in the experiment 
described in § 161 below) to render available the sensibility which a powerful current 
through the body to be tested, with the use of a moderately sensitive galvanometer, 
must obviously give to that method when applied to the investigation of differential 
effects on the electric conductivity of a body in different directions ; and I succeeded 
with great ease in making very satisfactory experiments 161 to 165 below) by 
means of it, which first decided the question as to whether or not the effects of mag- 
netization give different electric conductivity in different directions to a mass of iron. 
At first, however, I did not see this or any other way to render the method practicable 
with galvanometer electrodes, either moveable upon the sheet of metal to be tested 
(in which case a motion of of inch would drive the needle from an extreme 
deflection on one side to an extreme reverse deflection), or by electrodes soldered to 
points on an equipotential line (in which case a slight alteration in temperature in 
different parts of the plate might drive the needle irrecoverably to an extreme deflec- 
tion on one side or the other) ; but the experiments which 1 knew as having been made 
by Matteucci suggested to me the following very simple plan, which I immediately 
commenced trying, and which I have since found applicable with the greatest ease to 
a variety (I believe now to every variety) of experiments on electric conductivities *. 
149. Let AB be the conductor to be tested, and let CD be another of nearly equal 
resistance, either a piece of the same wire continuous with the other through an arc 
BC, or connected with it by a thicker arc of copper, or of another metal, as may 
appear convenient for the particular case treated. Sometimes the experiment is 
arranged to test differential effects experienced alternately or simultaneously by AB 
and CD. But when one of them, AB, alone is acted upon, with a view to varying its 
resistance, it alone maybe regarded as the conductor which is tested; and the other, 
CD, will then be called the reference conductor. Let a wire, AOPD, which will be 
* [Note added Nov. 1856.] An hour before the meeting of the Royal Society at which this paper was read, 
I learned that a method of testing resistances had been given by Mr. Wheatstone which would probably be 
found to be the same in principle as that to which I had been led in the manner described in the text. I have 
since ascertained that Mr. Wheatstone’s “ Differential Resistance Measurer” (described in § 15 of the Bake- 
rian Lecture for 1843, see Transactions, June 15, 1843) is an instrument founded on precisely the same prin- 
ciple as all the various arrangements by which, with great and necessary alterations of detail, I have continued 
the investigation of effects of magnetism and of other influences, on the electric conductivity of metals, to the 
present time, and of which some are fully described in Parts IV, and V. of the text. Mr. Wheatstone refers 
to “ Experimental Determinations of the Laws of Magneto-electric Induction,” printed in the Philosophical 
Transactions for 1833, “as containing the description of a differential arrangement of which the principle is 
the same as that on which” his own instrument has been devised, and adds, “ To Mr. Christie must therefore 
be attributed the first idea of this useful and accurate method of measuring resistances.” 
It is worth remarking, that the experiments of Matteucci and Kirchhoff, alluded to in the text, are stated 
to have been first suggested from Wheatstone idea of applying the two electrodes of a galvanometer to 
jioints in separate channels through which two parts of the whole current from one battery are conducted. 
