22 Mr. Donovan on Galvanometric Deflections 



temperature of the water ; yet if the rods be now connected with the galvano- 

 meter, there will be immediate deflection, although, by the conditions of Law iv., 

 it might be expected that none would follow. This deflection will continue 

 until the water has cooled to the temperature of the atmosphere, and then the 

 needle will have arrived at zero. During the continuance of the deflection, heat 

 was entering both rods of metal, but with difierent degrees of rapidity, because 

 of difference of conducting power, or of the quantity of heat necessary to pro- 

 duce the same temperature in both rods, and other causes. As fast as heat en- 

 tered, it passed out through the portions of the rods which were not immersed, . 

 and was continually dissipated in the air, while new portions entered from the 

 water. Hence there was a constant current of heat passing through the metals, 

 and hence they were continually in that condition, whatever it may be, which 

 produces deflection, until no more heat could be supplied or put in motion. 



It might appear that, although this experiment supports the law under con- 

 sideration, it militates with Law iv. In fact it might be conceived to prove 

 that metals, by being equally heated above the temperature of the surrounding 

 media, are capable of causing deflections, as Mr. Erman and others supposed. 

 By a little variation of the experiment it will be shown that no such inference 

 ought to be drawn. If the two rods, tied as before, be connected with the gal- 

 vanometer by capillary copper wire, and one-third plunged in boiling water, it 

 will be found, in conformity with what has been just explained, that the deflec- 

 tion will continue so long as the temperature is maintained above that of the 

 surrounding media. But if the rods, still tied, be entirely submerged in boiling 

 water, so that all parts be brought to the same temperature, and no current of 

 heat, as it may be expressed, is passing through them, the deflection will in a 

 minute or two cease. The needle having arrived at zero will there remain, and 

 there will not be the least variation unless circumstances should cause one of 

 the rods to change its temperature, such as unequal cooling of the water or 

 containing vessel ; the most trifling inequality will produce a deflection. 



In these experiments, if, instead of hot water, we use water cooled with ice, 

 the converse process will take place, provided the atmosphere at the time is at 

 a higher temperature than the cooled water. A deflection will result, but it 

 will be on the opposite side of the magnetic meridian ; this will continue until 

 the temperature of the metals and water rise to that of the surrounding media, 



