CUERENT-CURVES AND TENSION-CURVES. 177 



The further nature of the arch does not matter. 

 It may consist of one or more wires, and it may or 

 may not include moist conductors. Only one condition 

 must be fulfilled : no electric actions must be caused 

 by the contact of the diverting arch with the conductor 

 which is to be examined. Now, we have already seen 

 that this is unavoidable when metallic wires are ap- 

 plied to moist animal substances. The ends of the 

 wire of the arch must, therefore, be connected with the 

 zinc diverting-vessels described above (fig. »^8). In 

 this arrangement the clay shields, saturated with a 

 salt-solution, represent the feet of the divertiug arch. 

 Such an arch, which neither in itself nor by its appli- 

 cation to the conductor under examination affords any 

 cause for the generation of currents, is an homogeneous 

 arch. 



In order to attain a thorough knowledge of the 

 distribution of tensions in a conductor, it would ap- 

 parently be necessary to touch all points of the latter 

 in turn with the feet of the diverting arch. This is 

 easily done in the case of the surface of the body, but 

 as regards the inner parts it is hard and often imprac- 

 ticable. We must therefore rest satisfied with an 

 examination of the surface ; but it may be shown that 

 trustworthy conclusions as to the character of the inner 

 parts may be drawn from this study of the sm'face. 



2. Two cases must be distinguished. Either the 

 body to be examined is in itself incapable of electric 

 action, and the electric currents, the internal distri- 

 bution of which is to be examined, are imparted to 

 it from external sources ; or electromotive forces are 

 situated within the body itself, and it is the currents 

 generated by these which form the object of research. 

 9 



