Musical sounds are therefore quite fixed as soon as we know their loudness, pitch, and quality. But not all sounds are musical. There are such things as noises, and many forms of human speech are anything but musical. As a matter of fact, speech consists of a very rapid succession of all kinds of notes, which last only a very short time, and rapidly change from one to the other. Of these notes, the highest in pitch are the hissing sounds, such as _s_ and _sh_, and these are most difficult to send through the telephone. Shortly after the electric telegraph was invented, the idea occurred to many electricians that it ought to be possible to send speech along the wire conducting the electric current. They knew that sound consisted of waves, and they thought that if an electric current could be interrupted a great many times per second, a musical note would be produced at the other end of the wire. An observation made in America in the year 1837 confirmed this idea. It was found that when a current is sent suddenly through a long coil of wire wound round a piece of soft iron, a little sound is heard. Such a coil is called an electromagnet, because the soft iron is a powerful magnet so long as a current is passing through the coil. Another sound is heard when the current is broken suddenly. This observation suggested that if the electromagnet were at the other end of a long wire, and the current at one end were made and broken rapidly, the electromagnet would give a musical note at the other end. In the year 1854, a French telegraphist proposed to send speech by electricity on that principle. He proposed to stretch a skin over a drum, and fix one end of a fine wire to the middle of the skin. The end of the wire should dip into a drop of quicksilver connected with another wire, and this contact should establish an electric circuit. On speaking into the drum the skin would vibrate, the wire would move in and out of the drop, and the electric circuit would be made and broken in turn. It would be made and broken very rapidly when a high note was sung into it, and less rapidly in the case of a low note. It would follow all the various notes of ordinary speech, and would produce the necessary vibrations at the receiving end. The Frenchman never carried out his proposal. In the year 1861 it was again taken up by a German professor of the name of Reis. His instrument was able to transmit musical notes, but not spoken language. It was a very wonderful instrument at the time, but it could not give the correct quality to notes. The sudden induction of the electric current was too jerky to give the very fine ripples on the sound waves, and especially to render the delicate shades of human speech. The inventor died poor and neglected in 1874, and it was not till 1876 that the telephone problem was solved really.