510 PROF. JACOBL ON THE APPLICATION OF ELECTRO-MAGNETISM 
mercury is soon covered with a coating of oxide, which either entirely 
prevents the inetallic contact, or at least weakens it. In employing 
amalgamated surfaces this effect is produced still more rapidly. Be- 
sides I have by incontestible proofs arrived at the conviction, that 
the simple contact of metals witha clean surface is quite sufficient to 
conduct the electric current, even of the weakest tension. The con- 
tact by means of mercury adds nothing to the energy of this cur- 
rent. It is erroneous to judge of this energy by the brilliancy of the 
spark, proceeding only from the combustion of the mercury.. I have 
thought right to mention these circumstances, though apparently trifling. 
In a motor, from which we look to obtain an infinitely accelerated mo- 
tion, the smallest details should not be disregarded; the most trivial are 
ultimately of importance. 
7: 
Fig. 3. represents the commutator, adapted to the magnetic apparatus, 
so as to produce the inversion of the poles: a, 6, c, d are four discs of 
copper fixed upon the axis of rotation ee. The discs a, 6 and e, d are 
united by copper tubes f, f, entirely insulated from the axis by the in- 
terposition of a tube g, of varnished wood or any other insulating sub- 
stance. 
The periphery of each disc is divided into eight exactly equal parts, 
of which four # are cut into sectors and filled afterwards by pieces of 
ebony, forming with the metal an accurate and smooth surface. The dises 
are arranged upon the axis of rotation, so that the sectors of wood and 
of metal alternately correspond, as represented by the shaded parts of the 
figure. ZZ, CC are bars of copper, formed as levers, very move- 
able in their supports: they are intended to conduct the current. 
The arm of the longest lever forms at its extremity an edge, which rests 
on the periphery of the corresponding disc. The other arm is bent and 
plunged into a litile jar filled with mercury, &. The jars kk and k! k' 
are united by plates of copper, as represented in figure 1. The action of 
this commutator will easily be understood. The levers are always in 
contact with the discs, and are alternately so with the metallic and 
insulating parts. By their mobility in their supports they yield to the 
slightest inequality of the surface, and the friction they occasion is very 
trifling. The helices which surround the moveable bars are united so 
as to form a continuous wire, the ends of which J, m are soldered 
respectively to the systems of the discs a, 6 and c,d. The other 
helices, wound round the fixed bars, are also united, and the ends » and 
o immersed, the one in a jar of mercury 7, attached to the voltaic appa- 
ratus, and the other in the jar k of the commutator. Thus all the six- 
teen helices form only one connecting wire, through the medium of the 
commutator. The voltaic apparatus consists of four troughs of copper, 
in which plates of zinc are immersed, all being united as in a pile. The 
