398 Intelligence and Miscellaneous Articles. 



unpolished surface diminishes as the angle of incidence increases, and 

 passes finally into regular reflexion. 



2. Between polished plates, and plates roughened on their two 

 surfaces, there are intermediate states such that a calorific coloration 

 of the transmitted rays is produced, quite independently of the effect 

 produced in the interior of the substance, and solely in consequence 

 of the mechanical properties of the surface. 



3. Hence in reference to unpolished or turbid media, it is neces- 

 sary to discriminate between a diffusion which always takes place, 

 and an elective absorption which sometimes occurs. 



4. Fused common salt produced a diffusion, but no calorific colo- 

 ration. 



5. A piece of old rock-salt, found to be chemically and mechani- 

 cally impure, exerted a diffusive action as well as an elective absorp- 

 tion. Analogous conditions explain the divergent observations which 

 have been made in various experiments with rock-salt. — Poggen- 

 dorff's Annalen, vol. cxx. p. 177. 



ON THE VARIATION OF MAGNETIC FORCE WITH THE TEMPERA- 

 TURE. BY M. MAURITIUS. 



The subject of the author's researches is, in the first place, the 

 variation of permanent magnets by repeated changes of tempera- 

 ture ; and secondly, the variation of the temporary magnetism of iron, 

 starting from a very elevated temperature. 



Permanent Magnets. — The method of experimenting is that em- 

 ployed by M. Wiedemann in researches on the same subject : the 

 magnetized bar is placed perpendicularly to the magnetic meridian 

 in the horizontal plane of a moveable magnet whose angular devia- 

 tions are observed by reflexion. The deviation of the moveable 

 magnet from the magnetic meridian gives the measure of the force 

 of the magnetized bar. 



The magnet, the temperature of which is to be varied, is enclosed 

 in a zinc tube which can be put in communication either with a 

 steam-boiler, or with a vessel of water cooled to 0°. After having 

 caused the current of steam or cold water to pass through the tube 

 for five minutes, the angular position of the moveable magnet is 

 observed. 



The experiments consisted, in the first place, in producing alter- 

 nately the temperatures of 100° and 0° until the magnetism of the 

 bar became sensibly constant on the return of the same tempera- 

 ture ; and secondly, in submitting the same bar to this process 

 several times successively, magnetizing it afresh after each series of 

 observations. Five magnets of tempered steel were successively 

 used ; they were parallelopipedons in shape, and of different sizes, 

 and were magnetized by friction. 



Results. — When the magnetism of a magnet becomes constant 

 after having been submitted to a great number of changes of tempe- 

 rature, the diminution of its magnetism from 0° to 100° is propor- 



