PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 725 
Fig. 33. 
Fig. 34. 
an iron conductor of the shape indicated in the diagram magnetized, with perfect 
uniformity everywhere, in the direction shown by the lines of 
shading, and if, when the two ends kept at the same temperature 
are put into the circuit of a galvanometer, the corner is heated, 
a current would be found to set in the direction shown by the 
arrow-heads, that is, from transversely magnetized to longitudi- 
nally magnetized through hot. 
138. To test and illustrate this conclusion, I took a piece of 
sheet iron, cut to the shape shown in the diagram, and wound it 
spirally on a wooden cylinder, prepared with spiral grooves and 
pipes for steam and cold water, as described below. The oblique 
edge of the iron, shown on the left boundary in the diagram, 
being cut at angles of 45° and 135° to the long edges conterminous 
with it, was bent in a plane perpendicular to the axis of the cylinder, and thus the 
long edges of the iron, and the cut separating it into two branches, formed spirals, 
each at an angle of 45° to the axis of the cylinder. 
The two long edges themselves came very nearly to 
coincide, the circumference of the cylinder being a 
little greater in length than the oblique edge of the 
iron which thus nearly met round it. These two 
edges, as well as the two edges on each side of the 
cut between the branches, were prevented from touch- 
ing one another by being, one at least in each of the 
contiguous pairs, bound with cotton tape. The pro- 
jecting slips (shown on the right in the diagram) 
came to positions parallel to the axis of the cylinder, through two diametrically 
opposite parts of its circumference. Their 
ends had copper wires soldered to them, and 
were arranged with the usual precautions 
92) to ensure electric insulation and equality 
of temperature between them. The wooden cylinder had two diametrically opposite 
spiral grooves, each at the same inclination of 45° to the axis, and spiral sheet copper 
tubes, prepared of the proper shape, were slipped into these grooves, and nearly filled 
up the spaces to the surface of the cylinder. The outsides of these tubes were coated 
with paper, so as to maintain electric insulation between them, and the sheet iron 
wound on outside. 
The wooden cylinder bearing the spiral tubes, and the sheet iron arranged in the 
manner described, was slipped into the hollow of an electro-dynamic helix, steam 
was sent through one of the spiral tubes and water through the other, and the 
copper wires soldered to the ends of the iron slips were connected with the electrodes 
of a galvanometer. No current was at first indicated. The galvanometer circuit 
5 c 2 
Fig. 35. 
