562 ALEXANDER MACFARLANE AND P. M. PLAYFAIR ON THE 
the same kind without melting the paraffin again. It will be observed 
from the table that the deflections for the succeeding four sparks are sensibly 
equal. The ratio of the electromotive force for the first discharge to that 
for the succeeding discharges is 3°6. The increased facility is of course due 
to the fact that the first discharge bores a hole through the solid. We 
observed the hole afterwards; it was zigzag as regards direction, and had 
a black surface. The passage of the sparks, and particularly of the first one, 
illuminated the whole cake of paraffin in a very beautiful manner. 
The paraffin was then taken out, and sparks taken for air under the same 
conditions as before. The mean reading for air compared with the reading 
for the solid paraffin gives the electric strength of the former to be five times 
that of the latter. 
Next we liquefied the paraffin and raised it to a pretty high temperature, 
so that it might remain liquid for some time though brought into contact with 
the cold dish and plates, and then poured it in to the same height as before. 
The mean reading gives its electric strength when liquid to be 2'4—scarce 
half of that which it possesses when in the solid state. The further obser- 
vations were taken to see how rapidly the electric strength changed. The 
deflection when the paraffin was quite clear was 230; this increased for the 
next three observations up to 390, when the paraffin was superficially solid; 
but the effect of this spark was to make a rent which allowed the discharge 
to pass more easily afterwards. These results are supported by other two 
sets of observations, which were preliminary in their nature. 
Discharge of Electricity through Olive Oil. 
We thought it of importance to ascertain whether that law of discharge 
holds for olive oil which we had already found to hold for the two dielectrics, 
paraffin oil and oil of turpentine, viz., that the electromotive force plus a con- 
stant is proportional to the distance between the parallel plates forming the 
electrodes. (Trans. Roy. Soc. Edin., vol. xxviii. p. 680). With that view, and 
also for the purpose of comparing its electric strength with that of air, we took 
the series of observations recorded in Table II. We found it possible to con- 
tinue the observations for a distance greater by one millimetre than in the case 
of the other two liquids. When the distance was increased to 6 mm. the 
index moved up pretty steadily to 0 divisions of the scale, when it suddenly 
left the scale on account of the great amount of electricity which then began 
to escape from the insulated wire. In the fourth column we have noted the 
position which the index took up after the passage of the spark and the 
