In order to reconcile these experiments with his preconceived ideas, Mr. Hughes thought that, since the differences of resistance proceeding from the vibrations of the conductor were only produced when it was broken, the molecular movements were arrested by the lateral resistances which were equal and opposite, but that if one of these resistances were destroyed, the molecular movement could be freely developed. He considers that an imperfect contact is equivalent to the suppression of one of these resistances, and as soon as this movement can take place, the molecular expansions and contractions which result from the vibrations must correspond to the increase or diminution of resistance in the circuit. We need not pursue Mr. Hughes’s theory further, since it would take too long to develope it, and we must continue our examination of the different properties of the microphone.[17] Carbon, as we have said, is not the only substance which can be employed to form the sensitive organ of this system of transmission. Mr. Hughes has tried other substances, including those which are good conductors, such as metals. Iron afforded rather good results, and the effect produced by surfaces of platinum when it was greatly subdivided was equal, if not superior, to that furnished by the mercurised carbon. Yet since the difficulty of making instruments with this metal is greater, he prefers the carbon, which resembles it in being incapable of oxidation. We have already said that the microphone may be used as a thermoscope, in which case it must have the special arrangement represented in fig. 43. Under these conditions, heat, reacting on the conductivity of these contacts, may cause such variations in the resistance of the circuit that the current of three Daniell cells will be annulled by approaching the hand to the tube. In order to estimate the relative intensity of the different sources of heat, it will be enough to introduce into the circuit of the two electrodes A and B, fig. 43, a battery P, of one or two Daniell cells, and a moderately sensitive galvanometer G. For this purpose one of 120 turns will suffice. When the deviation decreases, it shows that the source of heat is superior to the surrounding atmosphere; and conversely, that it is inferior when the deviation increases. Mr. Hughes says that the effects resulting from the intervention of sunshine and shadow are shown on the instrument by considerable variations in the deviations of the galvanometer. Indeed it is so sensitive to the slightest variations of temperature that it is impossible to maintain it in repose.