462 RIESS ON THE INCANDESCENCE OF 
remarkably short space of time, and indeed upon the first partial 
discharges. This affords a new confirmation of what I have 
endeavoured to establish, that the electrical fusion of the metals 
is not due to a gradual increase of temperature, but must be 
attributed to a more quickly acting mechanical agency. 
The Reduction to Powder. 
The first visible direct action of the electrical discharge upon 
a new wire consists, as was remarked at page 437, in the forma- 
tion of a cloud of vapour or smoke rising from the wire. It is 
probable that this vapour is composed of metallic particles forci- 
bly separated from the surface of the wire, and dependent there- 
fore in quantity upon the nature of the surface. By increasing 
the power of the discharge beyond what would be required to 
melt the wire completely, we are enabled to convert the whole 
mass of wire into vapour. This conversion is accompanied by 
a brilliant display of ight and a loud report. 
Experiment 44.—Through a platina wire, radius 0°0209, length 
15 lines, enclosed in a glass tube, which was made incandescent 
by a discharge in which s = 5 and g = 13, and was melted into 
globules when ¢ = 17, the quantity of electricity 22 was dis- 
charged. It disappeared with a brilliant light, and the tube was 
covered with a gray sublimate, which could be wiped off. 
The experiment was repeated in the air and a plate of mica 
fixed horizontally a few lines above the wire. The vapour into 
which the wire was converted covered the mica with grayish and 
blackish spots, which under the microscope, with a power of 
280, appeared to consist of metallic fragments of different sizes 
and shapes. Comparatively there were but few melted particles 
and globules. 
The dispersion of the wire evidently consisted in its reduction 
to powder, in a mechanical separation of the metal into small 
particles. All the metals can thus be reduced to powder, but 
the power of discharge requisite to produce this effect stands in 
no relation to that which is required to melt the metals. Thus 
tin is melted by a less powerful discharge than cadmium, but 
the discharge which completely reduced the latter metal to pow- 
der left the greater part of the tin in a fused state. The brittle- 
ness of the metal has evidently the greatest influence on its re- 
duction to powder. 
Finely-divided metal, many of the particles of which are heated 
