an 的小不

By drawing this strip beneath a contact lever, thelatter was actuated so as to control the circuits andsend the desired signals over the line. It was observedthat when the strip was moved very rapidlythe vibration of the lever resulted in the productionof an audible note. With these facts before him,Edison reasoned that if the paper strip could be imprintedwith elevations and depressions representativeof sound-waves, they might be caused to actuate adiaphragm so as to reproduce the correspondingsounds. The next step in the line of developmentwas to form the necessary undulations on the strip,and it was then reasoned that original sounds themselvesmight be utilized to form a graphic record byactuating a diaphragm and causing a cutting or indentingpoint carried thereby to vibrate in contactwith a moving surface, so as to cut or indent therecord therein. Strange as it may seem, therefore,and contrary to the general belief, the phonographwas developed backward, the production of the soundsbeing of prior development to the idea of actuallyrecording them.Mr. Edison's own account of the invention of thephonograph is intensely interesting. "I wasexperimenting,"

 he says, "on an automatic method ofrecording telegraph messages on a disk of paper laidon a revolving platen, exactly the same as the disktalking-machine of to-day. The platen had a spiralgroove on its surface, like the disk. Over this wasplaced a circular disk of paper; an electromagnetwith the embossing point connected to an armtravelled over the disk; and any signals giventhrough the magnets were embossed on the disk ofpaper. If this disk was removed from the machineand put on a similar machine provided with a contactpoint, the embossed record would cause thesignals to be repeated into another wire. The ordinaryspeed of telegraphic signals is thirty-five toforty words a minute; but with this machine severalhundred words were possible."From my experiments on the telephone I knewof the power of a diaphragm to take up sound vibrations,as I had made a little toy which,

when yourecited loudly in the funnel, would work a pawl connectedto the diaphragm; and this engaging a ratchet-wheel served to give continuous rotation to a pulley.This pulley was connected by a cord to a little papertoy representing a man sawing wood. Hence, if oneshouted: `Mary had a little lamb,' etc., the paperman would start sawing wood. I reached the conclusionthat if I could record the movements of thediaphragm properly, I could cause such record toreproduce the original movements imparted to thediaphragm by the voice, and thus succeed in recordingand reproducing the human voice."Instead of using a disk I designed a little machineusing a cylinder provided with grooves around thesurface. Over this was to be placed tinfoil, whicheasily received and recorded the movements of thediaphragm. A sketch was made, and the piece-workprice, $18, was marked on the sketch.