321 Mr. U. Sabine on Measuring the Contour oj 



between two waves of opposite sign takes to arrive at c may 

 be determined, and at the same time the contour of the vanish- 

 ing and increasing waves. For this purpose it is necessary to 

 furnish the key (e) at the end (a) of the conductor with a second 

 contact-stud, "g' (fig. 2), in connexion with a battery, B', which 



Fier. 2. 



1 <* i 9 "<* 



\j c \ 



7> 



1 AI 



^B' ^B 



E 



<$>r 



E' 



supplies opposite electricity to that supplied by B. The cur- 

 rent of this battery (B 7 ) can either be kept flowing through 

 until the wave from B is transmitted, or it may be inserted for a 

 short interval only, as in working a line with reversed currents. 



When with this arrangement e is depressed, the current of 

 B flows into the conductor, and as it passes onwards neutralizes 

 the electricity left in it by W and recharges the line in the 

 other direction. If the accumulator be discharged before the 

 neutral point has reached c, the galvanometer shows electricity 

 left by B'. If the neutral point has passed beyond c, it shows 

 the opposite electricity — that of B. The neutral point having 

 arrived at c, at the instant of separating I from m, no move- 

 ment of the galvanometer is observed. The dotted curves in 

 fig. 2 give a rough idea of the contour of waves produced in 

 this manner. 



To obtain results with this method, it is of course necessary 

 to be provided with an apparatus which can be exactly adjusted 

 to open the contact between I and m at any desired small in- 

 terval of time after it has closed the contact between e and g. 



The apparatus which I use for this purpose, and, I believe, 

 the only one which is applicable for the purpose, is the same 

 that I had from Sir C. Wheatstone, and used for controlling 

 the time-intervals given by means of an accumulator dischar- 

 ging through a known resistance, as described in the May 

 Number of the Philosophical Magazine. 



This apparatus, with the modifications which have been re- 

 cently made in it, appears likely to be extremely useful for a 

 variety of physical inquiries involving short definite intervals of 



