The tape prepared in this way is passed through a Wheatstone transmitter, and long or short currents, according to the arrangement of the perforations, are sent out along the telegraph line. At the receiving station these signals operate a receiving perforator. This machine produces another perforated tape, which is an exact copy of the tape at the sending station, and it turns out this duplicate tape at the rate of from 150 to 200 words a minute. There are two forms of this receiving perforator, one worked entirely by electricity, and the other by a combination of electricity and compressed air, both forms serving the same purpose. The duplicate tape is then passed through an automatic printer, which reproduces the message in large Roman characters on a paper tape. The printer works at a speed of from 80 to about 100 words a minute, and the printed tape is pasted on a telegraphic form by a semi-automatic process, and the message is then ready for delivery. Plate XI. shows a specimen of the tape from the receiving perforator, and the corresponding translation as turned out by the printer. This message formed part of a leading article in the _Daily Mail_. Some idea of the wonderful capabilities of the Creed system may be gained from the fact that by means of it practically the whole contents of the _Daily Mail_ are telegraphed every night from London to Manchester and Paris, for publication next morning. One of the most remarkable features about present-day telegraphy is the ease with which two or more messages can be sent simultaneously over one line. Duplex telegraphy, or the simultaneous transmission of two separate messages in opposite directions over one wire, is now practised on almost every line of any importance. At first sight duplex telegraphy seems to be an impossibility, for if we have two stations, one at each end of a single wire, and each station fitted with a transmitter and a receiver, it appears as if each transmitter would affect not only the receiver at the opposite end of the wire, but also the receiver at its own end, thus causing hopeless confusion when both transmitters were in use at the same time. This actually would be the case with ordinary telegraphic methods, but by the use of a special arrangement all confusion in working is avoided. [Illustration: PLATE XI. _By permission of_ _Creed, Bille & Co. Ltd._ SPECIMEN OF THE WORK OF THE CREED HIGH-SPEED PRINTING TELEGRAPH.]