EXCURSIONS OF THE CAPILLARY ELECTROMETER. 
95 
Analysis of Normal Carve No. 289, fig. 5, Plate 4. Upward excursion equivalent 
to the difference of potential given by 83 centims. of the rheochord wire with 
1 Callaud cell. Resistance in circuit, a gastrocnemius preparation with non- 
polarizable electrodes. Total length of the excursion = 31‘7 millims. 
Found. 
Calculated. 
Difference. 
millims. 
millims. 
millim. 
*y = 28-000 
28-800 
24-900 
24-855 
+ -045 
21-400 
21-451 
- -051 
18-500 
18-513 
- -013 
16-000 
15-978 
+ -022 
13-800 
13-789 
+ -Oil 
11-900 
11-901 
- -001 
10-200 
10-271 
- -071 
8-900 
8-864 
+ -036 
7-650 
7-650 
The column of calculated values is a series in geometrical progression, inserted 
between the first and the last terms. It will be seen that the error is in all cases 
well within the limit of accuracy of the measurements. 
Having thus determined the position of the asymptotic circle, i.e., the position of 
the meniscus when y — 0, on the assumption that the formula y = ae~ d was exact, 
the next step was to try whether any traces could be discovered of the influence of 
change of resistance, or of the sensitiveness of the capillary, during the excursion. 
In other words, I had to determine whether these errors were due to inaccuracies of 
measurement or of the formula assigned to the curve. If the latter were the case, 
then there must be some value of n which would make T increase or decrease with 
some approach to regularity from one end of the curve to the other, and we should 
have approximately 
log (n + y x ) — log (n + y 2 ) = log T, 
log ( n -f y 2 ) - log (n + y 3 ) = log (T + b), 
log (n + y s ) — log (n + 2 / 4 ) = log (T + 2b), 
and so on. On assuming trial values for n, however, the discrepancies seemed only 
to increase, and, after a somewhat laborious investigation, I concluded that in this 
particular instrument the effect of diminished resistance was practically balanced by a 
slight increase of sensitiveness towards the tip of the capillary—the one tending to 
increase, and the other to diminish, the velocity of the meniscus, so that the equation 
y = ae~ d represents very closely the time-relations of the excursion. The process 
* These ten points correspond to intervals of '01 sec., as given by the time-tracing. That is why the 
first one is some 3 millims. from the commencement of the curve. 
