192 PROCEEDINGS OF THE AMERICAN ACADEMY 



These observations show that no large change in the Hall effect is 

 produced in copper by a rise of more than 300°, and leave it an open 

 question whether any change is produced. 



Phosphor-Bronze. 



The Hall effect in a strip of this substance was about one half as 

 large as in the copper just described, the thickness of the two strips 

 being nearly the same. Mr. Clough found no increase of this effect 

 in raising the temperature from 20° to 360°. He found the tempera- 

 ture coefficient of the electrical resistance of the specimen examined 

 to be about .00045. 



Cold-Rolled Steel. 



A number of strips of this substance, about 0.016 cm. thick, mounted 

 or supported in various ways, were tested. For satisfactory results 

 at high temperatures it was found necessary to make the side branches 

 long enough to extend outside the air bath. Otherwise very trouble- 

 some thermo-electric effects were likely to occur at the junctions with 

 the wires leading to the mirror galvonometer. The following table 

 gives, in divisions of the galvanometer scale, the results of the series 

 of observations that ran through the widest range of temperature. 



t H. 



18 5.26 



116 15.0 



199 27.2 



263 39.5 



319 49 



The second column assumes a constant value of the primary current 

 in the strip, and a constant value, about 1400 c. g. s., for the strength 

 of the magnetic field and for the magnetic induction per square centi- 

 meter through the steel. 



Curve A in Figure 1 is plotted to correspond with the quantities 

 just given, the abscissas representing temperatures and the ordinates 

 values of the Hall effect. For comparison, curve a, Figure 1, is in- 

 tended to show the relation of temperature to magnetic permeability 

 in " Whitworth mild steel," with a constant value, 1500, for the mag- 

 netic induction per square centimeter through it. This curve is plot- 

 ted from data obtained from curves given by Dr. John Hopkinson in 

 the Transactions of the Royal Society for 1889. It will be seen that 

 the permeability curve of the Whitworth steel is concave upward, even 



