HALL AND CAMPBELL. — MAGNETIC EFFECTS IN SOFT IRON. GGl 



large positive value for the coenioieiit in question, which may be de- 

 fined as follows : 



A^' de„ , . 



A^' being the increase of temperature-difference between two points 

 d cm. apart along the plate. 



Critical examination of the case, however, showed that in using, as 

 we must with Plate 1, iron wires connected with points (1) and (8), 

 we had brought in a source of error so great as to make the result 

 entirely uncertain: 



We hope to try the experiment with Plate 2, using now the con- 

 stan tan-copper junctions (1) and (3).^° 



Summary. 



1. A new and descriptive nomenclature is proposed and used for 

 the various transverse and longitudinal coefficients. (See equations 

 (1), (2), (3), (4), (23), (24), (25), (26).) For example, the Hall-effect 

 coefHcieut is called ^ Te, the initial subscript e indicating that the longi- 

 tudinal current is electrical, the T indicating that the effect observed 

 is transverse, the final subscript e indicating that this transverse effect 

 is electrical. The Ettingshausen coefficient is called ^ T^, the Nernst co- 

 efiicient ^ Te, the Leduc coefficient ^ 7"^. Each of the four transverse 

 effects is called + when the current to which it gives rise in the exterior 

 part of a transverse circuit leaves the metal plate in the direction of 

 the " ponderomotive " force acting on this plate or in what would be the 

 direction of this force if longitudinal heat-flow were replaced by longi- 

 tudinal electric flow. This accords with the usual convention regard- 

 ing the sign of the Hall effect, but is opposite to the usual convention 

 regarding the sign of the Nernst effect. 



2. Measurements of gTe, eTn, nTg, n^n-, and of their temperature- 

 coefficients through a small range, have been made in a single plate of 

 soft iron, for which the electrical conductivity, the Thomson-effect 

 coefficient, and the thermo-electric height relative to copper, w'ere 

 already well known through previous study of pieces from the same bar 

 that furnished the plate. (See Table I.) 



3. The Moreau formula gT^e -i- p = ^ T^e "^ ■'') in which p is the specific 

 resistance, absolute, of the iron, and s is the number of ergs of heat- 

 energy absorbed by ten coulombs in rising through one centigrade de- 

 gree in iron (the Thomson effect), is discussed theoretically and is 



2" See footnote, p. 664. 



