Intelligence and Miscellaneous Articles. 393 



The following is au example of this class of experiments. 



The core is magnetized by four coils of 18 centims. external dia- 

 meter and 8 centims. depth; the iron core extends 5 centims. 

 beyond the extremities of the coils ; the copper wire is 2millims. in 

 diameter. 



The pile consists of eight rectangular Bunsen elements ; and the 

 intensity of the continuous current is 0*0218, the unit being that 

 which disengages in a voltameter 1 milligramme of hydrogen per 

 second. 



In the following Table, by the side of the values of n is set down 

 the corresponding duration in minutes, and the values H and h are 

 millimetres of water. 



No. of conca- ^ ^ n 



xnerations. < * ^ 



uiin. Observed. Calculated. 



4 3494 in 10 4-1 0-0011 0-0011 



2 3100 „ 10 13-5 1-3 0-0047 0-0044 



1 2003 „ 6 30-9 4-5 0-0177 0-0176 



The experimental law of which the above is an exposition is con- 

 formable to the law of magnetic energy which I communicated to the 

 Academy on Nov. 18, 1872. Let us designate by m the quantity 

 of magnetism received by the core when the current traverses the 

 four coils in the same direction without forming consequent points, 

 by I the interval between the two poles • I have shown that the 

 quantity of heat generated by the disappearance of this magnetism 

 is measured by mH. 



Supposing that the current changes direction once, so that there 

 is one consequent point in the middle of the core, this will behave 

 like the combination of two independent cores of which each pos- 

 sesses the quantity of magnetism <,, the polar interval „, and pro- 



2 7 27 



duces the quantity of heat -j- X ^. The total heat is therefore —r- ; 



and this is the experimental law. Other cases are demonstrated in 

 like manner. 



The differential thennomagnetic apparatus above described is sus- 

 ceptible of an improvement by which the core is completely screened 

 from the voltaic heat of the magnetizing coil. This purpose is ac- 

 complished by surrounding the compensating c} 7 linder of iron with 

 a wire like that of the magnetizing coil, with merely the difference 

 that the current traverses each layer of wire in alternately opposite 

 directions. In this way the compensating cylinder is submitted to 

 the same perturbing influence as the magnetized cylinder, but no 

 magnetism is developed in it ; the manometer indicates only the 

 effect of the heat, of magnetic origin, which is generated in the 

 magnetized core. 



I have thus constructed a very accurate apparatus, by means of 

 which I can deduce the value in calories of the thermic effect ob- 



