96 
MR. G. J. BURCH ON THE TIME-RELATIONS OF THE 
above described is necessary with every new electrometer, in order to determine its 
constants. When the equation of the curve has thus been determined once for all 
the process is much more simple. The position of the asymptotic circle is found by 
direct measurement upon a screen placed in the position of the sensitive plate, and a 
single normal excursion of known value is taken through the resistance used in 
the experiments to which the method of analysis is to be applied. Two or three 
ordinates with their corresponding subnormals are then measured, and the ratio of 
each ordinate to its subnormal is determined. 
The mean of these ratios is taken as the value of the constant multiplier. 
Measurement of the Subnormal of Normal Curve No. 289. 
The curve was spoilt for this purpose by the presence of small undulations in two 
places caused by some jar to the apparatus ; these only interfered with the measure¬ 
ment of y at one point, where, however, it was easy to take the mean position 
between them. But in order accurately to place the fine line upon the glass plate of 
the measuring rod as a tangent to the curve, it was necessary that this should be 
perfect for some little distance on each side of the point of contact. The two most 
favourable positions were near the beginning and the end of the curve, and these 
gave for the value of C respectively 
C = 8-50, 
C = 8-51. 
Two other positions, not so well situated, gave C = 8‘40 and C = 8‘43. The 
remaining readings were less reliable on account of the undulations referred to, and 
a short piece where the definition of the photograph was defective. The mean of the 
four measurements is C = 8‘46 ; but 8'50 is probably nearer the true reading. 
With this instrument, therefore, a difference of potential due to 83 centims. of the 
rheochord wire gave an excursion of 31'7 millims. on the sensitive plate. With a 
resistance in circuit equal to that of an ordinary physiological preparation, the sub¬ 
normal to the curve at its commencement was 26 '945 centims. 
Whence 
and 
83 
1 centim. on the subnormal = ———centims. = 30’8035 on the rheochord, 
2b - 945 
26-946 
1 centim. on the rheochord = — r—- centims. = ‘32464 on the subnormal. 
1 OO 
Comparison with other Normal Excursions. 
In order to show that the method may be relied on to give constant results, the 
following experiments are quoted. The circuit was the same as for Curve No. 289, 
