towards the Theory of Thermo-electricity. 213 



silver wire ; and the full electrical effect appeared to be re- 

 stored when the obstruction was overpassed. 



13. Becquerel found (ubi supra) that a circle of platinum 

 wire, of equal diameter throughout, produced no current when 

 heated in any part; but a knob being made in the circle, and 

 heat applied near it, on either side, a current set always from 

 the heated point towards the knob, where the heat had room 

 to diffuse itself. 



Also, that on soldering together the ends of his magnetest 

 wire, no current ensued, when the point of junction was heated 

 even to redness ; but on touching the wire on either side of 

 the heated point with a cold piece of the same metal, the 

 current immediately set towards the part so touched. 



14. It is true that the tables of conduction for heat and elec- 

 tricity, contrasted by Cumming with that of thermo-electricity, 

 deny the presumption of any connexion between them. But 

 there would appear to be two properties of conduction for ca- 

 loric, — promptitude and final efficacy. For although we handle 

 platinum wire, at two inches distance from a red-hot point, in 

 operations with the blowpipe, which hardly any other metal will 

 admit of, yet M. Despretz found that a rod of platinum even- 

 tually became hotter, at a given distance from a given source 

 of heat, than any other metallic rod of equal dimensions; and 

 if thermo-electricity be influenced by conduction at all, it must 

 be by promptitude of conduction, for very short distances: 

 whilst the tables hitherto published refer to a property com- 

 pounded of the two, or intermediate between them. 



A copper and an iron wire, drawn so that equal lengths 

 corresponded to the atomic weights, were bound with thread 

 for an inch each, cut off at \ an inch above the binding, — this 

 •J-inch bent over in the form of a hook, and the end filed flat. 

 The two wires, thus prepared exactly alike, were fixed to a 

 little wire frame, to keep them parallel and separate, and the 

 filed ends dipped into melted lard. A couple of orange peas, 

 of equal weight and size, were also dipped in the lard. When 

 cold, the cohesion of the lard was sufficient to support the 

 little balls against the ends of the wires. The bound part of 

 the wires was now plunged into warm mercury, and the ball 

 soon fell from the copper by fusion of the lard. The wires 

 were then cooled by plunging the bound part into cold mer- 

 cury, and the balls interchanged. On replunging into warm 

 mercury, the ball fell again from the copper; and this always 

 took place on repeated interchange of the balls, and alterna- 

 tion of cold and heat; the ball falling every time from the cop- 

 per, until the mercury was cooled so far that the lard would 

 not melt, and so that neither ball would fall. This was decisive 



