but by the inertia of the upper contact-piece which by its own weight pressed gently upon the lower contact-piece. In every one of these forms, except the last, there was moreover an adjusting-screw to determine the exact degree of initial pressure between the contact surfaces. Doubtless the difficulty of adjusting this screw to give the exact degree of contact, enhanced as that difficulty was in consequence of the liability of the membraneous tympanum to become flaccid by the moisture of the breath, induced Reis to think that the later form of the apparatus in which this adjustment was no longer retained would be more easy to use, or, as he says in his Prospectus, more accessible to others. Yet undoubtedly the absence of the spring at the contacts led some persons to fancy that the instrument was intended to be shouted or sung to so loudly that every vibration should make the upper contact-piece jump up from the lower, and as Professor Müller even suggests (p. 98), produce a spark! But such a manner of using the instrument would entirely defeat Reis’s most fundamental principle, that the interruptions should be such as to correspond to the _undulating curve_ which represents the pressure due to vibration of the sound-wave; the possibility of representing the degree of pressure by a curve being one of the two principles set forth in his paper “on Telephony” (p. 55), in which he remarks, that “Taking my stand on the preceding principles, I have succeeded in constructing an apparatus by means of which I am in a position to reproduce ... even to a certain degree the human voice.” Reis was perfectly well aware, as his curves show, that a complicated sound-wave does not consist invariably of _one_ condensation followed by _one_ rarefaction, but that there are all sorts of degrees of condensation which may follow one another, and all capable of being represented by a curve. If all sounds consisted of one rarefaction following immediately after each one condensation there might be some propriety in proposing that after each “make” of contact there should be a “break” in the sense of an abrupt or complete breach in the continuity of the current. But, obviously, the fact that one condensation may follow another without a rarefaction between (which Reis’s curves show that he knew) must be amply sufficient to prove that on Reis’s own principle _his interruptor was meant to produce variations in the degree of contact in exact correspondence with the variations in the degree of pressure_, whatever these might be. Had he not meant this, he could not have talked about “taking his stand” on the principle of representing varying pressures by an undulatory curve. Now, from what has been adduced, the following points are clear:-- _Firstly_, that the contact-regulator which Reis combined with the tympanum was meant to interrupt the current, more or less, according to the varying movements imparted to it by the voice.