be inserted between the electrodes of a decomposing cell it ought, except in extreme cases, to produce almost precisely the same result as a similar and equal slip of glass or mica. This was easily verified. Here we have the singular result of a marked diminution of the current by the insertion into the electrolyte of a substance which is in itself a much superior conductor. Even when the platinum completely closes the path from one electrode to the other, so as to form two decomposing cells instead of one, a comparatively small hole made in it at once changes its function from that of common electrode to each of two decomposing cells into that of a mere obstruction in one cell. It is an interesting experimental inquiry to trace the intermediate stages between these two states, as a pinhole in the platinum is gradually enlarged. Whatever, then, be the behaviour of the particles of an electrolyte, they do not behave like little pieces of platinum.

5. Note on Thermal Conduction. By Professor Tait.

Monday, 6th May 1878.

SIR C. WYVILLE THOMSON, Vice-President, in the Chair.

The following Communications were read:

1. On the Indications of Molecular Action in the Telephone. By R. M. Ferguson, Ph.D.

The accepted theory of the telephone represents that the vibrations of the sending plate to and from the pole of the magnet before which it is fixed is the origin of the currents generated in the pole bobbin of wire, and that these currents transmitted to the receiving telephone produce corresponding to-and-fro excursions of its plate. This theory, which is that of the inventor, may be shortly designated, in the happy words of Sir William Thomson for a kindred action, the push-and-pull theory. We have had in this session of the Society two communications of a practical nature, which seem directly confirmatory of this view. I refer to the lucid exposition of Gott's telephone experiment in the island of St Pierre, and

the beautiful and successful demonstration of the action of the phonograph, both by Professor Fleeming Jenkin. In the former of these, as we learned, one end of a thread was attached to the one side of the light suspended coil of a Thomson ink recorder, and the other to the paper disc of an ordinary mechanical telephone. This was done at the two communicating stations. When the sending disc was agitated by the voice, the coil to which it was attached twisted round in the powerful and uniform magnetic field in which it was placed, and dispatched corresponding electric current waves to the receiving instrument, the coil of which was thereby moved similarly in its field, and transferred its motion to its paper disc. A more beautiful manipulation of an exquisitely designed and executed apparatus it is not easy to conceive. In the phonograph we have as it were a mechanical telephone, with the string connecting the discs cut, and nothing left of it but the two ends stiffened into pricking pins. Instead of the sending disc dealing directly with the receiving one, its energy is employed in imprinting, by means of the pricker, its vibrations on the tinfoil, and this imprint, when again vivified by the energy of the rotating drum, reproduces the vibrations which originally stamped it.

After two such demonstrations, it may be held as proved that the electric telephone is equivalent to a mechanical telephone with an electro-magnetic intervening action instead of a mechanical one. It seems therefore a hopeless task to seek for indications of molecular action where mechanical action declares itself so manifestly. The mechanical action of the voice and of the membrane of the tympanum of the ear is above question, and that mechanical vibrations are dealt to the sending instrument, and emitted by the receiving one, is equally undoubted; but the intervening electric agency, how generated in the one and how transformed in the other, is a fair field for discussion. The action is novel, and it is surely a likely inquiry to investigate whether its explanation by the first principle that comes to hand, viz., the push-and-pull of the discs, fully covers the case. The question may be raised, for instance, whether the mere impact of the waves of air on the iron disc may not affect its magnetic condition by internal change or vibration,* so as to excite currents without vibrations of the push-and-pull kind, or whether in * Something like this was suggested by Professor Forbes.

the receiving disc the particles may not set up an action on their own account, independently of the displacement caused by the poles of the adjoining magnet. In a mechanical telephone, we do not find that it is made to sound only by the normal push or pull of the thread, the faintest rubbing on the irregularities of its surface, either on the disc or the tube to which it is attached, makes a sound loud enough to be heard, and we can easily admit that if an internal vibratory disturbance be set up in any direction in it, the same would be audible enough. In a discussion as to mechanical and molecular sounds, it may be safely admitted, where electricity or magnetism is concerned, that any action that is clearly traceable to disturbance within a body is molecular in its origin. It will, moreover, be granted that the mere smallness of any vibration does not necessarily give any clue to its origin. Infinitesimal vibrations are not necessarily molecular, nor are vibrations of molecular source free from external motion; and we can only say that a vibration comes from molecules if we can assign to it no outside cause. It may, however, be to the point that a vibration may be assumed to be molecular because of the difficulty in suppressing it, a vibration springing from within being more independent of direction than one produced from without from one quarter.

I propose in this communication to raise such questions in regard to the telephone, and though the results obtained may not be decisive, they may be some little contribution to the discussion.

I would begin with a case where internal action seems wholly absent. I refer to the action of a tuning-fork on the telephone. It has been mentioned in more than one communication to the Society, that a tuning-fork acts best without the disc. We find that the loudest sounds are sent to the listener at the receiving telephone, when one prong is brought with its flat vibrating end in front of the core or pole pin, and next to that when the prongs, if they are not too far apart, are laid with their flat sides vertical at an equal distance on each side of the pin. When the handle of the fork is laid on the core, and held upright, the resonance of the wooden frame of the telephone and the table on which it rests becomes loud, but only a faint trace of this is sent to the distant hearer. If we magnetise two like forks, one which we may call A, to be like a bar magnet having the end of the handle as one pole, and the other

pole split in two in the two prongs; and another fork B, the two prongs of which are made like the poles of a horse-shoe magnet, with the handle an excrescence between, we find that while the fork A produces sounds alike with both prongs when held near the core, the two prongs of the fork B show a marked difference. The like pole to that of the core sounds much weaker than the other. All this is indicative of the ordinary magneto-electric induction at work.

If we detach the coil from the magnet, we have still further illustrations of the same. Both forks, A and B, sound loudest when placed with one prong on its flat side over the hollow at the centre (fig. 1), and both continue to sound, but with diminished force, as they are withdrawn in the same position from the middle to the margin of the coil. When laid with the plane of the prongs horizontal (fig. 2), they act differently. The A fork has its best sounding position when each prong lies symmetrically to the hollow axis, and it has a position of silence at a point between the middle and

Fig. 1.

།་

Fig. 2.

Fig. 3.

Fig 4.

the outside, whilst the B fork in these positions acts in the opposite way. There are two positions that the forks may occupy at the side of the coils, where their similar and dissimilar actions are again shown. The first is when the plane of the fork is perpendicular to the axis (fig. 3), where both forks transmit no sound when held in the middle of the coil, but are heard when vibrating on either side. The second is when the plane is parallel to the

axis (fig. 4), where A is silent when its prongs are equidistant from the middle of the coil, and the fork B loudest. All this is, as we have said, simply an illustration of a well-known action, and at the same time a beautiful demonstration of the way in which a tuning-fork vibrates. The coils I used were of fine copper wire, 1 inch diameter and thick, but smaller coils would do equally well, and the forks were the ordinary small A and C forks sold by the musicsellers. It is perhaps worthy of note that a coil, a magnetised fork, and a telephone form a handy combination for testing the completeness of a circuit, as the sound of the fork coming directly to the ear is immensely below that heard in the telephone in the operator's hand. When the telephone does not sound, there is a break in the circuit. In these various performances of the fork, we have evidence enough to prove that the cause assigned by Bell for the sending action of the telephone covers at least the greater part of that action. At the same time, it must be borne in mind that the vibrations of the fork, and the sounds produced by them, are immensely greater than any connected with the telephonic effect of the voice, and that it is possible that the conditions of iron vibrating under the energy treasured up in it may be different from what they may be when the iron is beaten by the air.

But even this tuning-fork performance is not quite free from ambiguity. To find whether there might not be some change of magnetic condition due to internal vibration, able to generate currents, I cemented two coils (fig. 5.) to the vibrating ends of a large tuning-fork. It was a C (256 vibrations), with prongs upwards of 6 inches long,

broad, and more in average than thick. The distance between the prongs at the end inside was inch. The coils weighed oz.; they were

inch diameter, 1 inch long, and were of 007 inch copper wire. They reduced the pitch from C to

Fig.5

A. The cement was the hard and tough black tarry compound