Australia, is probably due to a great destruction of the native species during the Glacial period, and the arrival since of several from the countries to the north-west. I do not know how else some of the facts can be explained, such as that of there being only eight butterflies, and amongst these some of wide distribution; and that of the Heteroptera there are thirteen known species belonging to thirteen different genera and nine distinct families.* Such gaps as these in the fauna of a country are as significant as the grooved and polished surfaces of its rocks, and the naturalist may as surely point to the evidences of the Glacial period as the geologist.

Mr. Wallace has also drawn attention to the large destruction of species of insects in the Chilian sub-region, evidenced by the great number of peculiar genera of beetles of extremely isolated forms, and I might multiply instances. from the faunas and floras of southern lands, all tending to the conclusion that the southern hemisphere has been glaciated as much as, or more than, the northern; but I could not do justice to this phase of the question within the limits of this article, and I have only glanced at some of its most salient points with the object of indicating that the physical evidence of glaciation does not stand alone, but is strengthened by that of the present distribution of animal and vegetable life.

Capt. HUTTON, Trans. New Zeal. Inst., vol. v., p. 227.

VOL. VII. (N.S.)

2 B

IV. RECENT ADVANCES IN TELEGRAPHY.

By J. MUNRO, C.E.

F recent years the most important development of practical telegraphy has been the duplex system of sending messages. The idea of transmitting more than one message along a single wire at once appears first to have been conceived by M. Zantedeschi, as early as 1829; but we may take it for certain that it was not until 1853, or about twenty-five years ago, that any serious attempts were made to carry it out in practice. From that time the attention of electricians, both in Europe and America, has been directed to this interesting problem, but, until the last ten years, with so little success that in the 1867 edition of Sabine's "History of the Electric Telegraph "we find the following sentence :-" Both these systems of telegraphing in opposite directions, and of telegraphing in the same direction more than one message at a time, must be looked upon as little more than feats of intellectual gymnastics -very beautiful in their way, but quite useless in a practical point of view."

Experience, however, shows that it is unwise to repudiate an electric novelty; and since these words were written the duplex system of telegraphing in opposite directions has become the ordinary means of communication over thousands of miles of land-lines in England and America, and of submarine cables in Europe and the East; while the multiplex system of sending several messages in the same direction is rapidly being brought into practical service by means of the Meyer instrument in France and the telephones of Elisha Gray in America and La Cour in Denmark. Indeed, by a future combination of the duplex and multiplex systems, we may yet have a single wire transmitting as many as twenty or more distinct messages, ten either way.

In order to explain the general principle of duplex or counter transmission, it is necessary to be perfectly familiar with the ordinary method of simplex or simple transmission. An ordinary telegraphic circuit invariably consists of the battery or source of the electric current; the key or sending instrument, by which the circuit is opened or closed

and the current admitted into the line; next, the line or wire stretching from the Station A, where the message is sent off to the Station B, where it is received; and next, the receiving instrument at Station B, actuated by the current from the line, so as to give sensible signals. These parts of the circuit are all connected together, and at

Station A one "pole" of the sending battery is put in con

nection with the ground by means of an "earth-plate," which is generally an unoxidisable metal, such as copper; while at Station B one "terminal" of the receiving instrument is connected, similarly, to an "earth-plate" there. A complete external circuit for the sending battery is thus formed through the line, the receiving instrument, and the

earth. This circuit is only interrupted when the lever or handle of the key is up. At such a time, of course, no current flows through the circuit. But when the sending clerk depresses the key the circuit is complete, a current traverses the line, and a signal is received at Station B. Combined according to the Morse code, such signals make up the message.

At first sight it seemed impossible that two messages could be sent, each in an opposite direction to the other, at the same time through the line. How was it possible that the messages could pass each other without interfering? If currents of equal strength were sent in opposite directions, would they not neutralise each other? In that case no current would traverse the wire at all, and what would become of the messages? Yet it is precisely on the ground that two currents of equal intensity do eliminate each other that duplex telegraphy is accomplished. The feat has been done, not by discovering or inventing a means whereby two currents could pass each other in a wire without disturbing one another, but by ingeniously joining up the sending and receiving apparatus at each end of the wire, so that the very interference of the currents, the one with the other, should cause the instruments to signal.

Such an arrangement was necessarily different from that used in ordinary transmission, as described. There was necessarily both a battery and a receiving instrument at each end; but the whole secret of the duplex system consists in the placing of the receiving instrument. All the methods hitherto invented agree in this-that the receiving instrument at either end is so placed that currents leaving the station where it is cannot cause it to signalise, whereas currents coming in from the distant station can. Again, should the currents leaving the station where it is be by any means stopped, this will have the effect of making it signalise. In describing the leading methods we shall show in each case how these conditions were obtained.

Dr. Gintl, a director of Austrian telegraphs, may be regarded as the founder of duplex telegraphy. Gintl, in 1853, described his system to the Academy of Sciences, Vienna, and practically tried it on the land-line between Vienna and Prague. Gintl's original system is shown in Fig. 1, which represents the arrangement of apparatus at Vienna and at Prague, with the line wire between. Here the receiving instrument is placed between the battery and the line, so that, when the circuit is completed, by depressing the key the current flows through the receiving instrument into the

line. With an ordinary instrument this current would produce a signal, but it is essential for duplex working that the sending from a station should not affect the receiving instrument at that station. Therefore, in order to counteract the effect of the signalling current on the instrument, Gintl employed a second smaller battery with its local or "compensating" circuit, and whenever the signalling current passed through the receiving instrument into the line he also caused a current from this second or compensating battery to pass in an opposite direction through the instrument, so as to neutralise or balance the effect of the signalling current. For this purpose the instrument was wound in opposite directions by two wires, one long and thin, represented by the full line, the other short and thick, represented by the dotted line. The long wire was in circuit with the sending battery and the line; the short wire was in circuit with the small battery and the compensating circuit. A peculiar key having double points was used for sending, so that when it was depressed it closed both the line and compensating circuits. The signalling current then rushed into the line by way of the long wire of the receiving instrument; but although it thus passed through it, it failed to make the receiving instrument signalise, because the current from the compensating battery passed simultaneously in an opposite direction through the short wire. These two currents were accurately adjusted to balance each other in their effects on the needle or indicator of the instrument. The essential condition of having the receiving instrument so placed as to be undisturbed by the sending was thus fulfilled, while at the same time it was in circuit with the line, and therefore free to receive signals from the distant station. Gintl's system had two serious defects :-In working the key in sending, the line was insulated for the moment when the lever was passing from the earth-contact, e, to the battery-contact, b. And the compensating battery was found to spend its strength quicker than the signalling battery. The harm arising from the first fault was that the received currents did not get freely to earth, and were therefore "broken up." The non-equivalence of the batteries, of course, destroyed the balance of currents on the receiving instrument, and "false" signals due to the sending were the result.

Gintl's plan, however, served as a stepping-stone to something better. Herr Carl Frischen, a telegraph engineer of Hanover, in the year 1854, greatly improved it by making the receiving instrument differential,-that is to say, instead of winding it with two dissimilar wires, as Gintl had done,