ELECTRO-MAGNETISM. 



applying heat, the temperature of one 

 or more of the other angles of the hex - 

 asjon was reduced by means of ice. 

 When the action of the ice was com- 

 bined with that of the flame, applied at 

 the same time to the alternate angles 

 all round the hexagon, the effect was 

 still more considerable, the deviation of 

 the needle amounting to sixty degrees. 



(320.) By continuing these experi- 

 ments with more numerous alternations, 

 it was found that the total effect of a 

 compound thermo-electric circuit is very 

 inferior to the sum of the effects which 

 the same elements could produce when 

 employed in the formation of simple 

 circuits, so that the electro-magnetic 

 forces called into action increase in a 

 much less ratio than the number of 

 alternations constituting the series *. 



(321.) The latest thermo-electric ex- 

 periments are those of Messrs. Nobili 

 and Melloni, of which an account was 

 read to the French Academy of Sciences 

 in September last (1831.)t A thermo- 

 electric pile, consisting of thirty- six pairs 

 of plates of bismuth and antimony, hav- 

 ing a galvanometer with two needles 

 attached to it, was found to be so sen- 

 sible as to be affected by the warmth of 

 a person at the distance of thirty feet. 

 A number of delicate experiments were 

 made with this apparatus on the permea- 

 bility of bodies to radiant heat, on the 

 temperature of insects, and on the pow- 

 ers of different bodies of emitting, re- 

 flecting, and absorbing heat. Consider- 

 able doubt is also thrown by these in- 

 quirers on the conclusions of Fourier 

 and Oersted with respect to the limited 

 effect of increasing the number of alter- 

 nations in augmenting the intensity of 

 the current. 



(322.) It would appear also, from the 

 observations of Professor Gumming, 

 that although the hydro-electric and 

 thermo-electric currents may both be 

 regarded as continuous, when compared 

 with those of common electricity, excited 

 by the electrical machine, which are 

 manifestly discontinuous (see Galvan- 

 ism, 21 and 94), yet when the hydro- 

 electric and thermo-electric currents are 

 compared with one another, the con- 

 tinuity of the latter is by far the most 

 complete. This will appear evident from 

 the consideration that, as it is necessary 

 that one of the three elements of a gal- 

 vanic circuit must be a fluid (see 



* Quarterly Journal of Science, xvi. 126. 

 t Bulletin des Sciences, No. II. for 1831. 



Galvanism, 69), it will, therefore, be a 

 more imperfect conductor than the me. 

 tals, and will oppose some degree of 

 resistance to the passage of the electric 

 currents circulating through the whole 

 assemblage. 



(323.) There are, therefore, strictly 

 speaking, three states of electricity. That 

 derived from the common electrical ma- 

 chine is in the highest state of tension, 

 and accumulates till it is able to force a 

 passage through the air, which is a per- 

 fect non-conductor. In the galvanic 

 apparatus the currents have a smaller 

 degree of tension ; because, although 

 they pass freely through the metallic 

 elements, they meet with some impedi- 

 ment in traversing the fluid conductor. 

 But in the thermo-electric currents the 

 tension is reduced to nothing, because 

 throughout the whole course of the cir- 

 cuit no impediment exists to its free and 

 uniform circulation. 



(324.) These considerations serve also 

 to explain why the latter species of 

 current is inadequate to effect any kind 

 of chemical decomposition, or even to 

 produce any degree of permanent mag- 

 netism. It has hitherto been found 

 impossible to magnetize steel bars by 

 means of thermo-electricity, although 

 the apparatus employed for that pur- 

 pose was capable of producing a strong 

 effect on the magnetic needle. 



(325.) It is probably owing to some 

 quality of this kind in the currents, which 

 theory assigns as the source of magne- 

 tism (namely, from deficiency in tension), 

 that all the endeavours which have so 

 many different times been made by va- 

 rious experimentalists to obtain from 

 magnetism any effects that may be con- 

 sidered as exclusively electrical, have 

 uniformly failed. Although a magnet 

 is powerfully affected by a current of 

 electricity proceeding from a voltaic 

 battery, it does not appear that the 

 magnet is capable of augmenting or 

 diminishing either the intensity or ve- 

 locity, or any other of the qualities of 

 the electric current. The only way in 

 which the reaction of the magnet is 

 shown, is by giving the electric current 

 a tendency to lateral deflexion, as if 

 urged by tangential forces proceeding 

 from the poles of the magnet. If the 

 wire of a galvanometer form part of the 

 voltaic circuit, and a powerful magnet 

 be applied to other parts of the circuit, 

 whether bent into a helix or not, no 

 indication of any action from the magnet 

 is afforded by the needle of the galvano- 



