In some microphones a rattling noise is frequently occasioned, which
borne along with the sound of the human voice causes an audible
disturbance in the telephone. The chief cause of these disturbances
may be ascribed to the fact that the carbon rollers in their journals,
rest loose in the flutings of the beam, which is fastened to the sound
plate. Owing to the shocks given to the entire apparatus, and
independent of the oscillations of the sound plate, they are set in
motion and roll to and fro in their bearings.

In microphones in which the sound plates are arranged vertically (as
shown in Fig. 2), these disturbances assume such a character that
there is no possibility of understanding the speaker, for in this case
the horizontally directed oscillations of the sound plate, _m_, cause
themselves a backward and forward motion on the part of the carbon
rollers without increasing or decreasing at the same time the lying-on
pressure of the roller journals, and by doing so bring the places of
contact one on the other, and thus occasion a conducting resistance of
greater or less force. This circumstance serves as an explanation of
the reason why the sound plates in Ader's microphones are not arranged
vertically, although this way of arranging them offers many advantages
over a horizontal or slightly inclined arrangement of the sound
plates. Speaking is more convenient in the vertical arrangement, and
moreover the plates can be fitted on to instruments better in this
way.

All the drawbacks just enumerated and found in Ader's microphones are
avoided in the apparatus made by Messrs. Mix & Genest. A sort of
braking contrivance operates on the carbon rollers in such a way as to
prevent their journals from lying on the lower points in the flutings