316 Royal Institutton. 



tion of heat to the point of junction of two different metals (as 

 bismuth and antimony, or bismuth and copper) creates a galvanic 

 action, as is shown by connecting wires with the two ends of the 

 united metals and forming a circuit; and observed that here we 

 seemed to have in nature a cause which might explain the origin of 

 terrestrial magnetism. Attention was then called to the general 

 similarity of Sabine's lines of equal magnetic intensity with Hum- 

 boldt's lines of equal temperature, the lecturer remarking, that a 

 much greater similarity would have been seen if he had been able 

 to display a chart of lines perpendicular to the direction of horizontal 

 magnetism, as proposed by Professor Christie. Allusion was then 

 made to the very remarkable experiment by Professor Christie, in 

 which a disc of bismuth being surrounded by a ring of copper, and 

 heat being applied to the edge of the copper, an extraordinary 

 amount of magnetism was developed ; two poles, austral and boreal, 

 being produced at certain points on one surface, and poles of 

 opposite character (separated from these by the thickness of the 

 bismuth only) on the opposite surface. Professor Christie had en- 

 deavoured to extend this experiment to the case of a spherical 

 copper shell filled with bismutli, and heated generally at the equator, 

 but more particularly at one point ; and the results appeared as far 

 as they went to correspond well with the state of terrestrial mag- 

 netism, but the difficulty of insuring a good union between the 

 copper and the bismuth (a difficulty which perhaps might now be 

 overcome by electrotyping) had made the results somewhat un- 

 certain. 



The lecturer then remarked that, for the advancement of the truly 

 scientific part of this inquiry, it does not appear that we have need 

 of any new Expeditions or of any further accumulation of observa- 

 tions made on tlie present plan. We have already vast collections 

 of observations which will be useless till they are published, and 

 which cannot be properly considered in a few years. But it is 

 probable that the discussion of them will suggest new instruments 

 of observation, more especially if (as is his own opinion) great im- 

 portance shall be thought due to the small disturbances. Already 

 he had thought that in every fixed observatory eye-observations 

 ought to be abandoned, and photographic self-registration to be 

 substituted for it: and he now thought that it would be necessary 

 so to improve the m:!gnets that they may be sensible to more rapid 

 disturbances, and so to improve the photographic paper that a mo- 

 mentary beam of light may make an impression upon it. But any 

 suggestions as to the course to be pursued in tracing the causes of 

 magnetism must be guided by the opinion of the person who under- 

 took the inquiry. The lecturer's own belief is, that thermo-elec- 

 tricity is the fundamental cause : and in this belief he expressed his 

 opinion that the importance of experimental investigation of the 

 laws of thermo-electric magnetism, where broad surfaces of dilferent 

 metals are in contact, is paramount to every other. An experimenter 

 might commence with Christie's valuable experiments, and extend 



