730 PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 
Table II. — Effects of Magnetism on the Thermo-electric Qualities of Iron and Nickel. 
Description of 
Conductor. 
Thermo-electric Order reckoned from Bismuth towards Antimony. 
Iron 
Under transverse magnetizing force ... 
Free 
Under longitudinal magnetizing force. 
Unmagnetized 
Retaining longitudinal magnetization. 
Nickel 
Under longitudinal magnetizing force. 
Free. 
PART IV. METHODS FOR COMPARING AND DETERMINING GALVANIC RESISTANCES. 
ILLUSTRATED BY PRELIMINARY EXPERIMENTS ON THE EFFECTS OF TENSION AND 
OF MAGNETIZATION ON THE ELECTRIC CONDUCTIVITY OF METALS. 
145. In endeavouring to discover the effects of magnetization and of mechanical 
strain on the electric conductivity of iron and other metals, I was led, from trying 
various more or less obvious methods for testing resistances, to use a differential galva- 
nometer of a very simple kind, which I constructed for the purpose. I shall give no 
description of this instrument, as I now (Nov. 1856) find it in one important quality 
inferior to the differential galvanometer first constructed and used by M. Becquerel*, 
and I do not know that its peculiarity has compensating advantages. I mention it 
only because it was with it that I made nearly the first of my trials to find the effects 
of magnetism on the electric conductivity of iron, and the very first by which I 
obtained a decided result. 
146. In these experiments I used two covered iron wires, each several yards long, 
coiled into circles about 4 inches diameter, as the two resistance branches in the 
divided channel through the two conductors of the galvanometer. Magnetizing one 
of them tangentially by means of a coil of covered copper wire wound on a copper 
sheath soldered round it as an electric drain, I ascertained, on the 23rd of April, 
1855, that the electric conductivity of iron wire is diminished by longitudinal mag- 
netization. The arrangement however proved, as I anticipated, to be of a very 
unsatisfactory kind ; and the needle kept moving across the field in one direction 
almost steadily, during the whole time the current was sustained through the tested 
conductors, which was for several hours. Continually more and more resistance 
had to be added to the conducting channel containing the iron wire round which 
there was no magnetizing coil, to keep the needle within range. After the magneti- 
zing current had passed for somq time, this variation of the needle went on more 
rapidly, and called for more frequent adjustment by the additions to the other branch. 
All this was just as must be expected; and my reason for not introducing currents 
of cold water round the two iron coils, to maintain them in precisely similar thermal 
circumstances, was that the tubular systems required for the purpose could not be 
easily made, and that I thought I might find out the nature of the result in the first 
* Annales de Chimie et de Physique, tome xvii. 1846. 
