PROFESSOR THOMSON ON THE ELECTRO-DYNAMIC QUALITIES OF METALS, 709 
in Part 1 ., that the vitreous electricity carries heat with it in copper 54), or, as it 
may be expressed, the electric convection of heat is positive in copper. From the dia- 
gram we infer that it is greater, and consequently positive, in Brass. That it is 
positive in brass has been proved also by direct experiment (§§ 67 and 77)» We 
infer also with certainty from the diagram, that the electric convection of heat 
(whether positive or negative) is greater in Zinc than in Gold, and greater in Gold 
than in Silver ; that it is greater in Brass, Tin, Lead, Copper, Zinc, Gold, Silver, and 
Cadmium than in Platinum ; that it is greater in Brass, Copper, Gold, Silver, and 
Cadmium than in Iron ; that it is greater (that is to say, since it has been proved, 
§ 76, to be negative, less negative) in Platinum than in Mercury ; and that it is 
greater in Nickel than in Palladium. In Cadmium, as we may judge by the eye 
from the diagram, the convection is probably greater than in Copper; and in Palla- 
dium probably less (that is, greater negatively) than in Platinum. 
103. These conclusions, certain and probable, are collected in the following Table 
of Convections, in which the different metals are arranged in order of the amounts of 
the electric convection of heat which they experience, or in the order of the values of 
“ the specific heat of electricity in them.” 
Electrical Convection of Heat 
Order doubtful. 
In Cadmium 
Brass . 
Copper 
Lead 
Tin . .j 
Zinc . 
* \ equal . 
Order doubtful. - 
Probably nearly! 
equal. 1 
Gold . . 
Silver . , 
Iron 
Platinum 
Nickel 
C Palladium 
[ Mercury 
Positive. 
Positive. 
Positive. 
Positive. 
Positive, Zero, or Negative. 
Positive, Zero, or Negative. 
Positive, Zero, or Negative. 
Negative. 
Negative. 
Probably Negative. 
Probably Negative. 
Negative. 
PART III. EFFECTS OF MECHANICAL STRAIN AND OF MAGNETIZATION ON THE 
THERMO-ELECTRIC QUALITIES OF METALS. ^ 
104. Physical agencies having directional attributes and depending (as all physical 
agencies we know of except gravitation appear to do) on particular qualities of the 
substance occupying the space across or in which they are exerted, are transmitted 
or permitted with different degrees of facility in different directions if the substance 
is crystalline. The phenomenon of crystallization, exhibiting different chemical 
5 A 2 
