PHYSICS. fi21 



Fossati has shown that the common imiiression that the strength ol' 

 a horseshoe steel magnet is increased by hanging it np and gradually 

 increasing the weights hung on its keeper is entirely an error, the in- 

 crease in portative force being the same whether the magnet be weighted 

 or not. This increase is in all cases much less than is supposed ; and 

 its variation is accompanied by a corresponding variation in distribu- 

 tion. {11 Niiovo Cimento., XV, 158, 232 ; J. Phys., December, 1885, II, IV, 

 565.) 



Barus and Strouhal have published, as the fourteenth Bulletin of the 

 U. S. Geological Survey, an extended memoir on the physical charac- 

 teristics of the iron carburets; more particularly on the galvanic, ther- 

 moelectric, and magnetic properties of wrought iron, steel, and cavSt 

 iron in different states of hardness, together with a physical diagram for 

 the classification of iron carburets. With reference to steel, the authors 

 say : " The diflerence between the logarithms of the respective values of 

 thermoelectric hardness for the same carburation passes through a 

 pronounced maximum defining a carbide, the mechanical jiroperties of 

 which are those of a type steel and may be fully given thus: Let each 

 member of the whole series of iron carburets be subjected successively 

 to the followingoperations : I, A process of very slow cooling from a given 

 temperature in red heat ; II, a jirocess of most rapid cooling possible from 

 the same temperature. If now the carburets be examined with reference 

 to the hardnes.s produced in the two istances there will be found among 

 them a certain unique member whose properties are such that while 

 process I has more nearly identified it with pure soft iron, process II will 

 have moved it farther away from this initial carburet than is simulta- 

 neuosly the case with any other iron-cjjrbon product; a unique member, 

 in other words, which is capable of occurring in the greatest number of 

 states of hardness relative to the soft state possible. To the said prod- 

 uct the term ' steel' is to be applied." 



Perkins has made an investigation on the variation of the magnetic 

 permeability of nickel at different temperatures and has plotted the re- 

 sults obtained. The curves show that at the first all the magnetism is 

 temporary. The permeability rises to a maximum at about the same 

 place as that of total magnetization, then falls less rapidly and approaches 

 the total, though it is evident that it can never reach it, since the jjerma- 

 uent magnetism cannot be less for a high magnetizing force than for a 

 lower at any given temperature. {Am. J. Sci., Septembei", 1885, III, 

 XXX, 218.) 



Hurion has examined and confirmed the statement of Eighi that the 

 electric resistance of bismuth increases when this metal is subjected to 

 the action of an electro-magnet. From his experiments he concludes 

 that the variation in the resistance of bismuth under these circumstances 

 arises in great part from the mechanical action exerted b^' the magnetic 

 lield on the naet^l J^owever;, it seems tbat^ the variiitiop of tl]f> resist 



