middle of November 1879, he connected a fender to the interrupter "to act as a radiator," and afterwards, instead of the fender, he used wires (answering to the "wings" of Hertz) on both transmitting and receiving apparatus, the wires being stiffened with laths to hold them in place.

The use of an "earth" connection led him to try the effect of joining the telephone to a gas-pipe of lead, and the microphone to a water-pipe of iron, as shown in fig. 5. The result was an improved sound in the telephone, and he concluded that the different metals formed

a weak "earth battery," from which a permanent current ran through the circuit. On this supposition he reasoned that the electric waves influencing the microphone, and perhaps changing its resistance, would rapidly alter the strength of this current, and so account for the heightened effects in the telephone. Acting on this idea, he included an E.M.F.

M

E

Fig. 5.

T

in the receiving circuit. A single cell was more than enough, and had to be reduced to as little as th of a volt in order not to permanently break down the contact resistance of the microphone.

"Thus," says Mr Munro, "Prof. Hughes had step by step put together all the principal elements of the wireless telegraph as we know it to-day, and although he was groping in the dark before the light of Hertz arose, it is little short of magical that in a few months, even weeks, and by using the simplest means, he thus forestalled the great Marconi advance by nearly twenty years!"

In the fifty years (just completed) of a brilliant professorial career at Cambridge, Sir George Stokes has given, times out of number, sound advice and helpful suggestions to those who have sought him; but in this case, as events show, the great weight of his opinion has kept back the clock for many years. With proper encouragement in 1879-80 Prof. Hughes would have followed up his clues, and, with his extraordinary keenness in research, there can be no doubt that he would have antici

pated Hertz in the complete discovery of electric waves, and Marconi in the application of them to wireless telegraphy, and so have altered considerably the course of scientific history.

As a recent commentator pithily says: "Hughes's experiments of 1879 were virtually a discovery of Hertzian waves before Hertz, of the coherer before Branly, and of wireless telegraphy before Marconi and others." The writer goes on to say, "Prof. Hughes has a great reputation already, but these latter experiments will add enormously to it, and place him among the foremost electricians of all time”1—praise which, knowing the learned professor as I do, I consider none too great.

APPENDIX E.

REPRINT OF SIGNOR G. MARCONI'S Patent.

No. 12,039, A.D. 1896.

Date of Application, 2nd June 1896. Complete Specification Left, 2nd Mar. 1897; Accepted, 2nd July 1897.

PROVISIONAL SPECIFICATION.

IMPROVEMENTS IN TRANSMITTING ELECTRICAL IMPULSES AND SIGNALS, AND IN APPARATUS THEREFOR.

I, Guglielmo Marconi, of 71 Hereford Road, Bayswater, in the county of Middlesex, do hereby declare the nature of this invention to be as follows:

According to this invention electrical actions or manifestations are transmitted through the air, earth, or water by means of electric oscillations of high frequency.

At the transmitting station I employ a Ruhmkorff coil having in its primary circuit a Morse key, or other appliance

1 The 'Globe,' May 12, 1899. Prof. Hughes died, full of honours, on January 22, 1900, aged sixty-nine. See, amongst other obituary notices, the 'Times,' January 24, and the 'Electrician,' January 26.

for starting or interrupting the current, and its pole appliances (such as insulated balls separated by small air spaces or high vacuum spaces, or compressed air or gas, or insulating liquids kept in place by a suitable insulating material, or tubes separated by similar spaces and carrying sliding discs) for producing the desired oscillations.

I find that a Ruhmkorff coil, or other similar apparatus, works much better if one of its vibrating contacts or brakes on its primary circuit is caused to revolve, which causes the secondary discharge to be more powerful and more regular, and keeps the platinum contacts of the vibrator cleaner and preserves them in good working order for an incomparably longer time than if they were not revolved. I cause them to revolve by means of a small electric motor actuated by the current which works the coil, or by another current, or in some cases I employ a mechanical (non-electrical) motor.

The coil may, however, be replaced by any other source of high tension electricity.

At the receiving instrument there is a local battery circuit containing an ordinary receiving telegraphic or signalling instrument, or other apparatus which may be necessary to work from a distance, and an appliance for closing the circuit, the latter being actuated by the oscillations from the transmitting instrument.

The appliance I employ consists of a tube containing conductive powder, or grains, or conductors in imperfect contact, each end of the column of powder or the terminals of the imperfect contact or conductor being connected to a metallic plate, preferably of suitable length so as to cause the system to resonate electrically in unison with the electrical oscillations transmitted to it. In some cases I give these plates or conductors the shape of an ordinary Hertz resonator consisting of two semicircular conductors, but with the difference that at the spark-gap I place one of my sensitive tubes, whilst the other ends of the conductors are connected to small condensers.

I have found that the best rules for making the sensitive tubes are as follows:

1st. The column of powder ought not to be long, the effects being better in sensitiveness and regularity with tubes contain

ing columns of powder or grains not exceeding two-thirds of an inch in length.

2nd. The tube containing the powder ought to be sealed.

3rd. Each wire which passes through the tube, in order to establish electrical communication, ought to terminate with pieces of metal or small knobs of a comparatively large surface, or preferably with pieces of thicker wire, of a diameter equal to the internal diameter of the tube, so as to oblige the powder or grains to be corked in between.

4th. If it is necessary to employ a local battery of higher E.M.F. than that with which an ordinarily prepared tube will work, the column of powder must be longer and divided into several sections by metallic divisions, the amount of powder or grains in each section being practically in the same condition as in a tube containing a single section. When no oscillations are sent from the transmitting instrument the powder or imperfect contact does not conduct the current, and the local battery circuit is broken; but when the powder or imperfect contact is influenced by the electrical oscillations, it conducts and closes the circuit.

I find, however, that once started, the powder or contact continues to conduct even when the oscillations at the transmitting station have ceased; but if it be shaken or tapped, the circuit is broken.

I do this tapping automatically, employing the current which the sensitive tube or contact had allowed to begin to flow under the influence of the electric oscillations from the transmitting instrument to work a trembler (similar to that of an electric bell), which hits the tube or imperfect contact, and so stops the current and, consequently, its own movement, which had been generated by the said current, which by this means automatically and almost instantaneously interrupts itself until another oscillation from the transmitting instrument repeats the process. Whilst for certain purposes I prefer working the trembler and the instruments on the same circuit which contains the sensitive tube or contact, for other purposes I prefer working the trembler and the instruments on another circuit, which is made to work in accordance with the first by means of a relay. It is by means of actions from the current, which the sensitive tube or contact allows to pass when the oscilla

tions influence it, that I prefer starting the apparatus that has to interrupt automatically the same current.

In order to prevent the action of the self-induction of the local circuits on the sensitive tube or contact, and also to destroy the perturbating effect of the small spark which occurs at the breaking of the circuit inside the tube or imperfect contact, and also at the vibrating contact of the trembler or at the movable contact of the relay, I put in derivation across those parts where the circuit is periodically broken a condenser of suitable capacity, or a coil of suitable resistance and self-induction, so that its self-induction may neutralise the selfinduction of the said circuits; or preferably I employ in derivation on different parts of the circuit conductors or so-called semi-conductors of high resistance and small selfinduction, such as bars of charcoal or preferably tubes containing water or other suitable liquid, in electrical communication with those conductors of the local circuits which are liable in course of self-induction to assume such differences of potential as to transmit jerky currents such as would influence the sensitive tube or contact so as to prevent its working with regularity.

In some cases, however, I find it suitable to employ an independent trembler moved by the current from another battery. This trembler is prevented from generating jerking or vibrating currents by means of the appliances which I have described. This trembler is kept going all the time during which one expects oscillations to be transmitted, and, as already described, the powder or imperfect contact closes the circuit of a local battery, in which are included the instruments which one desires to work, for the time during which the electrical oscillations are transmitted, breaking the circuit in case of the mechanical vibrations as soon as the oscillations from the transmitting machine cease. When transmitting through the air, and it is desired that the signal or electrical action should only be sent in one direction, or when it is necessary to transmit electrical effects to the greatest possible distance without wires, I place the oscillation producer at the focus or focal line of a reflector directed to the receiving station, and I place the tube or imperfect contact at the receiving instrument in a similar reflector directed towards the transmitting instrument.