206 Mr. G. J. Burch. On the Interpretation of [Feb. 11, 



tn-h = (h-a+p> n -p{)jv, 



i.e., the duration is increased by an amount equal to the time required 

 by the wave to traverse the whole length occupied by the points of 

 ■origin. 



The rise of P.D. will be more or less gradual, and so will its sub- 

 sidence, although the actual changes in each constituent conductor are 

 still supposed to be sudden and definite in amount. The effect of a 

 gradual development of E.M.F. will be dealt with later. 



It is manifest that the most favourable conditions for studying the 

 distribution of the points of origin are when the leads are far enough 

 apart to separate the two phases of the response by a zero interval. 



(2.) With leads B and D, i.e., on both sides of a group of points of 

 origin, the case is different (see fig. 9). 



By the same formula as before, 



t{ = (d-p^lv; U n = {d-pn)lv; 

 U = (b-pi)/v; t.?=(b-p n )lv; 



.and the duration is comprised between the smallest and the greatest 

 value of t. 



Now if PJD < BPl then is BP~ > T\DT 



For letJPJD = a, BT\-P\D = A Pi-P^ = 



Then BP 7l = a + P + y, which is greater than ViD = a + y. 



Hence in this case the total duration of the first phase, due to the 

 wave-front, is governed by T n the point of origin nearest to either 

 lead, and the direction or sign of the P.D. depends on which lead it is 

 nearest to. 



From this it follows that if any two similar conductors, Pi and P„ j} in 

 the bundle have their points of origin of E.M.F. situated at equal 

 distances on either side of the middle point between B and D, they 

 will neutralise each other, not only as regards the P.D. resulting from 

 the development of the wave of E.M.F. in them, but also as regards 

 the effect produced by its subsidence. 



Similarly Po will neutralise P m _i, and P 3 will neutralise P m _2, so 

 that no difference of potential will result save from the portion 

 P m+ i .... T n , i.e., the points of origin not symmetrically situated 

 between the leads B and D. 



This consideration indicates a method which I have occasionally 

 employed of locating the mean position of a group of points of origin. 



(3.) With points of origin partly between and partly beyond the 

 leads, as in fig. 10. 



Here B.P n is the longest distance, and therefore governs the 

 •superior limit of duration. But D itself is over a point of origin, 



