Memoirs presented to the Cambridge philosophical society on the occasion of the jubilee of Sir George Gabriel Stokes, bart., Hon. LL. D., Hon. SC. D., Lucasian professor.

DISCHARGE OF AN AIR CONDENSER, WITH A DETERMINATION OF "v." 141
The correctness of absolute time was secured by comparing the chronometer every
day with a standard clock which was rated from the Observatory. The error in the rate
of the chronometer was thus found negligible, being certainly not more than one or two
seconds a day.
Although it was possible to keep the speed constant for ten minutes or so, it was
rather wearying and was really unnecessary, two or three minutes being quite sufficient,
on this method of observing. Table I. gives a set of readings taken on the 23rd July,
1889, the "eight" pattern being kept steady and every other tap, or every 200th revolution, being timed:
TABLE I.
h. m. s. h. m. s.
xi 32 3'5 xi 33 43'0
9.5 49-5,,,, 9'5.9
15'5., 56'0
22-0, 34 2'0.. 285, 8-5
345, 14-5
410, 20-5
47-0 27-0,,),, y? o,,,, 00..535. ". 33'5... 59.5. 9 39.5
33 5'5, 45'5
120, 52-0
Analysing these figures it will be found that the average time for 16 "taps" of
200 revolutions each is 100 seconds; as it happens exactly. And this corresponds to
32 revolutions per second; appropriate to the steadiness of the "eight" pattern.
After this the speed could be increased till the "four" pattern was steady, with the
certainty that the plate was then revolving 64 times a second with extreme accuracy.
Thus the fork was used merely as an intermediary time-keeper to the chronometer, the
media of comparison being the counter and the stroboscopic disk.
PROCESS OF TAKING A SERIES OF SPARK PHOTOGRAPHS.
The room being thoroughly darkened one of the sensitive plates was extracted
from its case, and by the light of an exceedingly dim red glimmer fixed into the
rotating frame holder. The spark knobs had previously been focussed on a dummy plate
so that the spark length would be exactly radial, and near its outer margin. The
packing-case cover being well covered, light was admitted to the room so as to make
visible the stroboscopic disk which was watched between the jaws of the vibrating fork,
and the turbine was turned on. The patterns were seen steadying themselves one after
the other until the 4 pattern was reached and just passed; the water was regulated
close to the point by past experience; the cord was then gripped by the observer and
the escaping pattern brought back steady.