Electrical Signals on Land Lines. 415 



after the contact with the battery at A has been made. The 

 curve of arrival is perfectly defined by using the time t and the 

 observed strength of current as coordinates (the time r remain- 

 ing constant) ; and quite similarly the curve showing the gradual 

 cessation of the current after the earth-contact has been made 

 at the sending- station, is obtained by making the galvanometer- 

 contact at F at intervals of time t after the connexion is made 

 between A and M. By this simple plan, although the time 

 during which the current varies may not exceed in all a few 

 hundredths of a second, the rate of increase or decrease can be 

 perfectly observed. It is to be regretted that M. Guillemin 

 used a simple detector instead of a measuring-galvanometer, by 

 the deflections of which the relative strengths of the current 

 might have been accurately compared. Data are also want- 

 ing as to the resistance of the lines on which the experiments 

 were made, also as to their insulation and some other points 

 of interest. These omissions prevent us from being able to cal- 

 culate with much accuracy the constant c required for the mathe- 

 matical theory, and render it uncertain whether slight discre- 

 pancies between the theoretical and the observed curves are 

 due to the imperfect method, or to necessary conditions on the 

 lines themselves not taken into account by the theory. 



M. Guillemin's results confirm the conclusion that the time 

 required to transmit a signal through a uniform conductor, or, 

 more exactly, to allow the received current to reach a certain 

 proportion of its maximum strength, varies as the square of the 

 length of that conductor. He believes that the electromotive 

 force of the battery does exercise a small influence on the rate 

 of transmission, which is contrary to theory ; but the effect ob- 

 served, if any, is very small, not amounting to 10 per cent, when 

 the battery is doubled, and may perhaps be accounted for by 

 considering the improvement which a powerful battery effects on 

 the insulation by producing polarization. 



The effect of the resistance was not tested, nor does M. Guil- 

 lemin seem to have apprehended the manner in which this ele- 

 ment enters into the calculation. No experiment on this point, 

 however, is required, as the truth of the theory in this respect 

 has been abundantly proved. 



The comparison between the theoretical and observed curves 

 may be made as follows : — M. Guillemin, with a wire 570 kilo- 

 metres long and 4 millims. in diameter, obtained the results given 

 in the three first lines of the following Table : — 



