April 15, 1909] 



NATURE 



in an inch, or in the fraction of an inch. Thus, 

 suppose ten marks in one inch, then the frequency 

 would be more than 300 vibrations per second. A diffi- 

 culty arises when we find a complete vibratory period 

 not represented by one wave, but possibly by three, 

 one, the first, large, and the other two smaller. 

 Then, to ascertain the real pitch, only the large 

 marks must be counted. 



Many attempts have been made to obtain an 

 enlarged record of the wave-forms on the phono- 

 graph and gramophone, and much success was 

 attained long ago by Fleeming Jenkin and Ewing, 

 and, in later times, by Hermann, Scripture, and 

 myself. Still, none of these were facsimile tracings. 

 It seems to me that the gramophone, in its present 

 condition, holds out the hope of an experimentalist 

 being able to obtain from records tracings on a 

 smoked glass circular plate travelling at the same 

 rate as the record. Then, by placing the plate in a 

 lantern, we should see a representation of the waves 

 amplified, but amplified in 

 all proportions. This I 

 have succeeded in accom- 

 plishing, with a consider- 

 able measure of success, 

 after a trial of a good 

 many methods. 



(i) Place a smoked glass 

 plate, 10 inches in diame- 

 ter, with a hole in the 

 centre, on the platform of 

 the gramophone (the re- 

 cording of waves on a 

 smoked glass plate was a 

 method employed by Ber- 

 liner in his early experi- 

 ments). As the swinging 

 arm (or taper arm, as it 

 is technically called) of 

 the gramophone tends to 

 swing outwards, and as 

 the outer lip of the groove 

 on the record tends to 

 draw it inwards, and thus 

 to follow the spiral, the 

 needle, if placed on the 

 smoked plate, will not 

 move inwards. To over- 

 come this I attached by a 

 thread the swinging arm 

 to a very slow-moving 

 train of wheels, driven by 

 an old phonograph, and 

 thus I gradually drew the 

 swinging arm inwards, so 

 that a spiral was made on 



the smoked glass having a thread of about fifty 

 to the inch. I was unable with my apparatus to 

 obtain a slower motion. Then I removed from 

 the gramophone the large horn, and sang, spoke. 



disks might be prepared. These light and dark 

 bands were got rid of by connecting the arms of 

 the two gramophones by a rigid rod. The finest 

 gramophone needles were used for recording. 



I also, by this method, caused a loud phonograph 

 :o act, bv a connecting tube, on the mica disk of the 

 gramophone, through the tube of the swinging arm, 

 and I obtained tracings. 



(2) Another method was unsuccessful, and had to 

 be abandoned, both owing to dilTiculties of adjust- 

 ment and because it gave incorrect results. It con- 

 sists of elongating the vertical rod in the centre of 

 the gramophone platform. A glass plate, smoked, 

 is then fixed to the rod by passing the rod through 

 a hole in the centre and through a collar, that can 

 be clamped. The plate is of the same diameter as 

 the gramophone disk record ; it is smoked on its 

 under surface, so that, when matters are adjusted, 

 the disk of the gramophone is directed upwards and 

 the smoked surface downwards, and about 2 J inches 



above the record. A firm upright bar of steel_ is 

 soldered to the outer surface of the end of the spring 

 attached to the needle of the sound-box, and this 

 bar carries vertically a very fine needle. The point 

 or shouted "into the tube at the end of the swinging ! of this needle is brought with a minimum of friction 

 arm. The disk of the gramophone vibrated, and 1 against the under surface of the smoked disk. The 



the needle described minute waves of various forms 

 on the glass plate. I found, however, that when 

 a cord connected the taper arms of the two gramo- 

 phones there was not a continuous pull, but a 

 rhthymic oscillation, producing a tracing showing 

 light and dark bands, as the oscillations of the arms 

 of the gramophones had not the same period. A 

 beautiful tracing was thus obtained, showing, at 

 regular distances, light bands owing to the lines 

 becoming very close together, giving a figure such 

 as one has seen for the illustration of waves of sound, 

 or such as occurs in Crova's disks. By this method 

 and the use of an eccentric arrangement, Crova's 



NO. 2059, VOL. 80] 



gramophone is then started, and the platform with 

 the central rod carries both the record disk and the 

 smoked glass disk. The tracing so obtained was 

 not satisfactory. 



(3) I obtained the best tracings by causing one 

 gramophone to sing or play at the usual rate, and, at 

 the same time, by a rigid rod connecting the two 

 arms, to draw towards itself the arm of a second 

 gramophone on which was placed a circular smoked 

 glass plate. The needle of the second gramophone 

 described a spiral with intervals between successive 

 spirals of i/ioo inch, exactly similar to the spirals 

 on the record of the first gramophone; but, before 



