PROFESSOR THOMSON ON THE ELECTRO-DYNAJHIC QUALITIES OF METALS. 715 
Fisr. 23. 
developes reverse thermo-electric qualities in the axial direction and in directions 
perpendicular to it ; for there can be little doubt but that a lateral traction would 
produce the reverse effect of a lateral pressure, or that a portion of a linear con- 
ductor of iron pulled out on two opposite sides in a direction at right angles to its 
length, would acquire such a thermo-electric quality as to give rise to currents from 
stretched to free through hot. But in the former experiment (§ 108) it was demon- 
strated, that when part of an iron conductor is pulled out longitudinally, the thermo- 
electric effect gives currents from free to stretched through hot. The crystalline cha- 
racteristic is therefore established for the thermo-electric effect of mechanical stress 
applied to iron, if it be true that traction produces the reverse temporary effect to that 
of pressure in the same direction. There seems so strong a probability in favour of 
this supposition, that it may almost be accepted without experimental proof ; but I 
intend, notwithstanding, to make experiments, for the purpose of explicitly testing it, 
as soon as some preparations at present in progress enable me to do so. In the 
mean time I have made the following decisive experiment on the difference of thermo- 
electric quality in different directions in iron subjected to stress. 
116. A piece of sheet-iron 36 inches long and 16 inches broad, 
was rolled round two thick iron wires (|^-inch diam.), along its 1 ° 
breadth at its two ends, and soldered to them. It was cut into | 
narrow slips, each about \ of an inch broad and of different | 
lengths, as shown in the diagram, so as to prevent electric con- | 
duction, except along a band about half an inch broad running | 
across the sheet at an angle of 45° through its centre. The ends j 
of the slips on each side of this band were clamped (as shown in | 
the annexed sketch) between two flat iron bars, but insulated ^ 
from them by thin pieces of hard wood and from one another, where necessary, by 
pieces of cotton cloth. These bars were each \ an inch thick, 3 inches broad, and 
30 inches long ; and the two at each side clamped together upon the pio-. 24 . 
pieces of hard wood, with the iron slips between them, formed a firm 
beam, by means of which a considerable stress would be brought to 
bear on the sheet iron to stretch it in the direction of the slips. The 
upper of these beams was laid resting with its two ends on the tops of 
stout wooden pillars, supported below on a very strong wooden bar 
laid on the stage of a Bramah’s press. The lower double iron beam 
hanging down and straightening the sheet iron by its weight had strong 
iron links put over its ends, and an iron bar of about If-inch square 
section slipped through them below, so as to hang down a small 
distance below the roof of the press. Thus, when the press is worked, the upper 
double iron beam is forced up, and the sheet iron is stretched between it and 
* The thinner the better, I believe, as a partial failure was experienced from these pieces of wood breaking 
at one side and allowing the ends of the iron slips to get drawn in between the iron bars. 
MDCCCLVI. 5 B 
