472 RIESS ON THE INCANDESCENCE OF 
and designates, disruptive discharge, glow discharge, dark dis- 
charge, and carrying discharge. These discharges belong essen- 
tially to the discontinuous kind of discharge; whilst the gra- 
dual dispersion of electricity into the air, established by the laws 
of Coulomb, must be considered as the continuous discharge of 
an electrified body through the substance of the air. The dif- 
ferent modes of propagation of discharges through fluids are no 
less evident. A battery can be discharged completely and with- 
out the least noise through a tube filled with water, alcohol, or 
some saline solution; but by slightly increasing the power of 
the discharge, a spark is seen to pass in the tube and it is forcibly 
shattered to pieces. Fluids are decomposed into their compo- 
nent parts by a noiseless discharge, and these appear singly at 
certain points which may be at any distance from each other ; 
whilst the explosive discharge leaves the component parts of the 
scattered fluid everywhere in a combined state. Although these 
effects of not very different quantities of electricity sufficiently 
prove the different modes of propagation, still I have further 
shown them to exist by observations on a thermometer placed 
at the same time with an imperfect conductor in the connecting 
circuit. A moist piece of wood, 1% line in thickness, was 
clamped between two points in the connecting circuit, or a damp 
card was used to connect two points at 10 lines distance from 
each other; the discharges up to a certain intensity passed 
through without noise, and the thermometer did not show the 
least signs of being heated. But at a certain point, when the 
discharge was only slightly increased, it pierced the interposed 
substance with an emission of light and a violent report, and 
raised the temperature of the thermometer considerably. In 
these experiments however the battery was only in the last case 
completely discharged, whilst in the former no inconsiderable 
quantity of the electricity remained behind. I shall therefore 
only call attention to the following more accurate experiments, 
in all of which a complete discharge of the battery was effected. 
Experiment 49.—At an interruption in the connecting circuit 
two platina wires, 0°224 line in thickness, were placed perpen- 
dicularly and bent at an angle, so that their points were at } line 
distance from each other: they were thus immersed in a vessel 
of distilled water. A thermometer placed in another part of the 
circuit indicated on the different discharges the following tem- 
peratures :— 
