Producible by Attrition and Contact of Metals. 15 



the inequality of temperature diminishes, the effect of attrition becomes more 

 effectual in producing the natural deflection on the opposite side of the magnetic 

 meridian, which iu some instances is very powerful. 



Numerous cases range themselves under this class of phenomena, in which 

 the direction of the needle is thus suffered to reverse itself by attrition. Instead 

 of being examples of an active reversal, they are in fact so many instances 

 opposing the alleged universality of this effect of attrition. 



In the foregoing experiments I used the hemispheres already described. 

 In order to obviate any doubt concerning the interference of the rheophores, 

 or the shape of the metals employed, the experiments were varied by laying 

 aside the hemispheres, and substituting rods of the two metals, connected with 

 the binding screws by means of capillary copper wire. The bismuth rod, con- 

 nected with the zinc side of the galvanometer, was totally immersed in hot 

 water, along with that part of the copper wire which was attached to one of 

 its extremities: the antimony rod, on the silver side, lay on the table at the 

 temperature of the air. The end of the bismuth rod, namely, that end which 

 was not connected with the copper wire, was raised about an eighth of an inch 

 out of the water, by its wooden handle, and the antimony rod, held also by its 

 wooden handle, was instantly rubbed against the extremity of the bismuth; there 

 was immediate eastern deflection. From the nature of the experiment, the only 

 way in which both metals could now be brought to the same temperature, and 

 the direction of the deflection changed to the natiu-al state (Law xiv.), was to 

 plunge both iuto the water, the attrition being still continued. This done, the 

 needle speedily went round many degrees to the west permanently. 



In this trial, the point of junction of the rheophore wire with the bismuth 

 rod was submerged in hot water ; and the other rheophore required no artifi- 

 cial means to maintain it at the same temperature as the antimony rod, because 

 both were exposed in the open air. Thus the temperature of each rheophore 

 and its rod being equal in the respective cases (Law iv.), there could be no 

 interference of the rheophores in the deflections produced by the attrition of the 

 metals. 



As the Law (vii.) which these trials were intended to illustrate is impor- 

 tant to the explanation of several phenomena, I varied the experiment in the 

 following manner, in order to discover whether mass, or mode of attrition, has 



