276 
Mr Sears, On the Longitudinal Impact 
easily accounted for by the back e.m.f. produced by the con- 
denser discharge passing between X and Y, when R. 2 is as small as 
300 ohms. Assuming this, we have > so that 
T=-C 2 R 2 \og e (l -y^) 
= - -05 x 10- 6 x 3 x 10 3 log^l - 
= 1-5 x 10- 4 log, 2-720 
= 1*501 x 10 -4 secs. 
Now, since there is no sparking effect, the equation (2) is true. 
*’• T= RlX T + R 1 
and ^ = 2‘945 x 10 -8 x = 2 945 x 10~ 8 x (a) ; 
T= 5000 x 2-945 x lO" 8 x 1-002 + ~ 
Xtl 
or 1-501 x 10- 4 = 1-476 x 10- 4 + 4 , 
R\ 
whence = -025 x 10 -4 secs., 
R\ 
which is the correction required. 
The work was carried out at the Engineering Laboratory, 
Cambridge, under the guidance of Prof. Hopkinson, whose practical 
interest contributed largely to its ultimate success. 
The tables of observations, from which the figures in this paper 
are deduced, here follow. It will be noted that below each table 
is a figure called the “Fuse key fling.” This was a small fling 
which was always produced in the galvanometer on depressing the 
fuse key, even when no impact of the rods took place. In the 
earlier experiments it was somewhat larger than the values here 
recorded, and it was found necessary to deduct it from the nett 
fling before consistent results could be obtained. The various 
columns in the tables are as follows : — 
V 
z 
F 
8 
M 
Voltage, 
Zero of Galvanometer, 
Fling of Galvanometer, 
Corrected fling (= F — Z — Fuse key fling), 
Mean, 
Time f = 
5000 x 2-945 8 
10 4 
V 
+ •025 
) < 10 - 
4 secs.). 
