278 M. Melloni on a new Electroscope. 



of deviation. The electrical charge, communicated to the fixed 

 system EADD, will then begin to diminish. 



This diminution, however, takes place much more slowly than 

 in the ordinary electroscopes, in consequence of the masked 

 electricity which is gradually disengaged from the central part, 

 and whicli replaces on the cup A its appendages DD, the wire 

 of communication and the disc E, a portion of the free electricity 

 lost by dispersion. The double or inductive electrization of the 

 moveable system BCC will exactly follow the successive phases 

 of the simple electrization of the fixed system, its two principles 

 will be re-compounded gradually in proportion to the loss of the 

 charge, and after a certain time the whole will return to the 

 natural condition. All this is independent of the method em- 

 ployed for charging the conductor E, and will consequently 

 apply equally to cases of direct charge by contact, and to cases 

 of indirect or opposite charge obtained by means of induction. 



To resume, the moveable part of the instrument is always 

 electrified by induction and never by communication ; the differ- 

 ence in the form of the centres and extremities of the fixed and 

 moveable portions renders the distribution of the motive forces 

 the most advantageous that can possibly be for the rotation of 

 the index, and the inductive action of the central surfaces 

 masking a portion of electricity which is afterwards set free in 

 proportion to the loss undergone, prolongs the duration of the 

 charge. 



The small size of the pieces constituting the essential part of 

 the apparatus accelerating the loss of electricity by diffusion in 

 the surrounding medium, it is necessary that they should be 

 enclosed in a case, in which the air must be kept very dry by 

 means of some substance which possesses a great attraction for 

 moisture. A dry state of the internal air is especially necessary 

 to prevent variation in the torsion of the silk thread which sup- 

 ports the reversed cup, so as to enable the index CC to return 

 constantly to the same azimuth, when the appendages DD have 

 lost their electrical charge. 



The case must be of a convenient form, and as the observa- 

 tions to be made require the knowledge of the angles of deviation 

 formed by two rods superposed without contact, and kept at a 

 certain distance from a dial plate placed below them, the most 

 favourable arrangement is evidently to suspend the free extremity 

 of the silk thread from the interior of the apex of a vertical tube, 

 opening in the centre of a horizontal glass disc, the circumference 

 of which rests upon a cylindrical metallic receiver, but little 

 wider than the moveable lever and the subjacent appendages of 

 the fixed cup. The upper edge of this receiver must be flattened 

 and lined with leather, to intercept the communication between 



