414 Mr. F, Jenkin on the Retardation of 



method of observing the phenomena was therefore wanted, and 

 that employed by M. Guillemin leaves little to be desired. 



Figs.' 3 and 4 will assist in rendering these experiments intel- 

 ligible. Fig. 3 represents the series of changes produced in the 

 electric current at the receiving-end of a wire by a succession of 

 uniformly long signals, where each contact used to charge and 

 discharge the line is continued for a sufficiently long time to 

 allow the current in each case to rise to its maximum and fall 

 again to zero ; c, c p c lt represent the commencement of the 

 charging contacts, and b, b t , b u the commencement of the dis- 

 charging contacts. 



M. Guillemnr's experiments have for their object the defini- 

 tion of the curves c b and b c, each of which is identical with that 

 given in fig. 2; and this object. he obtains as follows : — 



In fig. 4, A L D represents his circuit, formed by a wire leaving 

 Paris and passing through say Tours and Mans back to Paris 

 by a different route. 



E and Ej represent the earth-plates, two in number, G a gal- 

 vanometer, B the battery. The receiving-end of the line J) was 

 kept constantly in connexion with the earth and with one end 

 of the galvanometer-coil. The end A of the line, by means of a 

 cylinder suitably arranged and revolving at a given uniform speed, 

 could be alternately connected with M and N (that is to say, 

 with the battery and earth) for short periods, variable- at will, but 

 generally equal to ^th or 20th of a second. The length of 

 these contacts corresponds to the times indicated by the distance 

 separating c and b in fig. 3. These contacts sent a succession of 

 signals round the line to be received at D, where they produced 

 a series of changes in the electric current, such as are represented 

 in fig. 3. This current, it will be observed, does not pass 

 through the galvanometer ; but the end F of the galvanometer- 

 coil is so connected with the revolving cylinder, that a contact 

 at F can be established for a very short time, represented by the 

 short space r in fig. 3 (equal, say, to yo^th of a second), and 

 commencing at any given interval t after the contact with the 

 battery at A has been made. Every time such a short contact 

 at D has been completed, a fraction of the received current pass- 

 ing at the moment is diverted, and passes through the galvano- 

 meter, tending to deflect the needle. The galvanometer being- 

 sensitive, a small fraction of the whole current deflects it suffi- 

 ciently for the purpose of observation ; this derivation exercises 

 very little influence on the whole resistance of the circuit, and 

 does not sensibly disturb the rate of variation in the current 

 received at D. The constant succession of impulses thus given 

 to the galvanometer-needle, produce a constant deflection indi- 

 cating the strength of the varying received current at a time t 



