6 Mr. W. Sturgeon on the Thermo-Magnetisjn of 



needle is more deflected than if the heat were confined to a 

 single point. When the whole side ab is uniformly heated, in 

 consequence of the superior energies attributable to the half 

 nearest to a, the current is directed from b to a; but the 

 energies of this resulting force are necessarily very feeble. 



TJiird Class of Experiments. 



22. Three rectangular frames of bismuth, which I shall 

 distinguish by the letters A, B, C, were cast in the same mould, 

 and from the same mass of metal. The longest side of each 

 frame measured ^'S inches, and the shorter sides 2 inches. 

 The weight of the whole was 21 ounces; so that the average 

 weight of each was 7 ounces. They were cast under different 

 circumstances, as regards the temperature of the mould, agita- 

 tion, position, &c. 



23. The rectangle A was cast whilst the mould was quite 

 cold; its plane horizontal, and its longest sides parallel to 

 the magnetic meridian. The metal was all poured into the 

 mould (which was simply a groove in one of the flat sides 

 of a Bath brick) at one particular point, which was in one of 

 the longest sides of the rectangle, and about 1-5 inch from one 

 of the angles. The metal consequently flowed from this point 

 to every other part of the mould. Two' powerful bar mag- 

 nets were applied to the two longest sides whilst the bismuth 

 remained in a fluid state ; they were drawn from the centre 

 of each long side of the rectangle in precisely the same man- 

 ner as is practised in magnetizing a needle by the double 

 touch. The process was continued till some time after the 

 metal was set. The mould was kept perfectly at rest all the 

 time. 



2+. The rectangle B was cast whilst the mould was quite 

 warm from the heat communicated to it by the last casting. 

 The magnetizing process was carried on as before, but the 

 metal was agitated as much as possible all the time. The 

 position of the mould was the same as whilst casting the rect- 

 angle A. The magnets in both cases were so applied, that 

 if the magnetism of the earth had any influence in arranging 

 the particles of the fluid metal*, those artificial magnets would 

 have tended to promote that arrangement. 



25. The rectangle C was cast whilst the mould was quite 

 hot, and with its longest sides at right angles to the magnetic 

 meridian. No magnet was applied to the bismuth, but con- 

 tinued agitation was kept up till long after the metal was set. 



* From some previous observations, I had formed an idea that this 

 might possibly be the case. 



26. Expert- 



