The feebleness of the currents in submarine telegraphy makes it impossible to use the ordinary land telegraph receiver, and a more sensitive instrument known as the “mirror receiver” is used. This consists of a coil of very fine wire, in the centre of which a tiny magnetic needle is suspended by a fibre of unspun silk. A magnet placed close by keeps the needle in one position when no current is flowing. As the deflections of the needle are extremely small, it is necessary to magnify them in some way, and this is done by fixing to the needle a very small mirror, upon which falls a ray of light from a lamp. The mirror reflects this ray on to a sheet of white paper marked with a scale, and as the mirror moves along with the needle the point of light travels over the paper, a very small movement of the needle causing the light to travel some inches. The receiving operator sits in a darkened room and watches the light, which moves to the right or to the left according to the direction of the current. The signals employed are the same as those for the single-needle instrument, a movement to the left indicating a dot, and one to the right a dash. In many instruments the total weight of magnet and mirror is only two or three grains, and the sensitiveness is such that the current from a voltaic cell consisting of a lady’s silver thimble with a few drops of acidulated water and a diminutive rod of zinc, is sufficient to transmit a message across the Atlantic. The mirror receiver cannot write down its messages, and for recording purposes an instrument invented by Lord Kelvin, and called the “siphon recorder,” is used. In this instrument a coil of wire is suspended between the poles of an electro-magnet, and to it is connected by means of a silk fibre a delicate glass tube or siphon. One end of the siphon dips into an ink-well, and capillary attraction causes the ink to fill the siphon. The other end of the siphon almost touches a moving paper ribbon placed beneath it. The ink and the paper are oppositely electrified, and the attraction between the opposite charges causes the ink to spurt out of the siphon in very minute drops, which fall on to the paper. As long as no current is passing the siphon remains stationary, but when a current flows from the cable through the coil, the latter moves to one side or the other, according to the direction of the current, and makes the siphon move also. Consequently, instead of a straight line along the middle of the paper ribbon, a wavy line with little peaks on each side of the centre is produced by the minute drops of ink. This recorder sometimes refuses to work properly in damp weather, owing to the loss of the opposite charges on ink and paper, but a later inventor, named Cuttriss, has removed this trouble by using a siphon kept constantly in vibration by electro-magnetism. The ordinary single-needle code is used for the siphon recorder. CHAPTER XVIII THE TELEPHONE