PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS. 665 
§§ 39 to 43. Decisive Experiments with Multiple Sheet Iron Conductor. 
39. These experiments seemed therefore on the whole to establish a probability in 
favour of the convection of heat by the so-called positive electricity, when a current 
is kept up through an unequally heated conductor. The convective effect, if of this 
kind, ought (§ 18) to be less in iron than in copper, I therefore had little expectation 
of finding an indication of it in the iron conductor which (§ 31) had been in the 
course of preparation ; but as soon as it was ready for use I made the following 
experiments, and was much surprised by the result, which became manifest before 
the first of them was finished. 
40. Conductor composed of thirty slips of sheet iron. 
Experiment III. November 12th, 1853. 
(Current six times eight minutes in 
differences of temperature after 
By end next A. 
each direction.) Temperatures and 
eight minutes of current entering 
By end next B. 
Augmentations of 
differences from 
middles to ends of 
periods. 
Periods. 
Ta- 
Tb- 
Tb-Ta=D. 
Ta- 
T 
Tb-Ta=D'. 
D'-D. 
I. 
51-43 
53-56 
2-13 
51-48 
53-49 
2-01 
0 
--12 
II. 
III. 
IV. 
V. 
VI. 
51-62 
51- 73 
52- 01 
51-30 
51-14 
53-30 
53-26 
53-80 
53-00 
52-98 
1-68 
1-53 
1-79 
1-70 
1-84 
51- 41 
52- 03 
51-32 
51-00 
50-69 
5321 
53-87 
53-42 
52-95 
52-80 
1-80 
1- 84 
2- 10 
1- 95 
2- 11 
-12 
-31 
-31 
-25 
-27 
Means for five periods.. 
51-56 
53-268 
1-708 
51-29 
53-25 
1-96 
-252 
Augmentation of difference during periods included ... -10 
Deduct average augmentation per half-period -010 
Effect due to reversal of current 0°-242, 
in favour of Resinous Electricity. 
