NATURAL PHILOSOPHY. 139 



\vhich magnetic oxide had been prepared, and surrounded by a coil ; 

 this exhibited to the spectator looking through it, an increase of the 

 transmitted light when the coil was electrized. All these experi- 

 ments, he considers, go to prove that, whenever magnetization takes 

 place, a change is produced in the molecular condition of the sub- 

 stances magnetized ; and it occurred to him that, if this be the case, by a 

 species of molecular friction heat might be produced. In proving the 

 correctness of these conjectures, many difficulties presented them- 

 selves, the principal of which was, that with electro-magnets, the 

 heat produced by the electrized coil surrounding them might be ex- 

 pected to mask any heat developed by the magnetism. This interfer- 

 ence he eliminated by surrounding the poles of an electro-magnet 

 with cisterns of water," and by this means, and by covering the keeper 

 with flannel and other expedients, he was enabled to produce, in a 

 cylindrical cast-iron keeper, when rapidly magnetized and demagnet- 

 ized, a rise of temperature several degrees beyond that which obtain- 

 ed in the electro-magnet, and which therefore could not have been due 

 to the radiation of "heat from it. By filling the cisterns with water 

 colder than the electro-magnet, the latter could be cooled, while the 

 keeper was being heated by magnetization. Subsequently, distinct 

 thermic effects were detected in a bar of soft iron, placed opposite to a 

 rotating permanent steel magnet. To separate the effects of magneto- 

 electrical currents, the author then made experiments with non-mag- 

 netic metals and with silico-borate of lead, substituted for the iron 

 keepers, but no thermic effects were developed. He then tried the 

 magnetic metals, nickel and cobalt, and obtained thermic effects with 

 both, in proportion to their magnetical intensity. The author then 

 concludes by saying that he considers that these experiments prove 

 that, whenever a bar of iron or other magnetic metal is magnetized, 

 its temperature is raised. Mechanic's Journal. 



ON THE POLARIZATION OF HEAT. 



THE polarization of heat, first announced by Berard, has been 

 established by various experiments by Forbes and Melloni. Provost- 

 aye and Desains have lately announced to the Academy of Sciences, 

 at Paris, new investigations, showing, 1st. That heat, traversing 

 Iceland spar, is divided into two pencils, completely polarized in the 

 plane of the principal section or a perpendicular plane. 2d. That the 

 law ascertained by Malus, according to which the intensity of a ray- 

 completely polarized is divided between the ordinary and extraordi- 

 nary images to which it gives origin in traversing the spar, is appli- 

 cable to heat as well as light. 3d. That the variations of intensity 

 which polarized heat experiences in its reflection from glass at differ- 

 ent incidences, are exactly represented by FresnePs formulas deter- 

 mined for light, only allowing that the solar heat traversing the prism 

 has a little different index/ 1.5. 4th. That there is a most perfect 

 correspondence between the phenomena presented in the reflection 

 from polished metals of polarized heat and polarized light. 



