October 4, 19 17] 



NATURE 



97 



ittached to the tube; it can be fixed at various points. 

 FGH is a wire handle. 

 To use the instrument, it is placed in the mouth 

 ither in the manner shown in Fig. 4, or else so that 



Fig. 6. — X-ray photograph of cardinal vowel » (as in French). 



the tube is in contact with the teeth at the tooth-stop 

 and also in contact with some point of the palate 

 (the position of the apparatus depending on the nature 



Fig. 7. — X-ray pliotograph of cardinal a. 



of the sound to be analysed). The wire is then pushed 

 along until the end of it is felt to touch the tongue. 

 The instrument is withdrawn and applied to a pre- 



Fiii. 8. — X-ray photograph of cardinal u. 



viouslv prepared diagram of the shape of the observer's 

 palate. The position of the end of the wire is then 

 marked on the paper. 



NO. 2501, VOL. 100] 



Further observations are then taken with the tooth- 

 stop fixed at other points. In this way the positions 

 of other points of the surface of the tongue are ascer- 

 tained. In the end we get on our paper a series of, 

 say, ten or more points which show with fair accuracy 

 the shape of the most important part of the tongue. 



Fig. 5 shows specimens of results obtained by this 

 means. They were prepared by Mr. Atkinson, and are 

 reproduced here by his kind permission. 



A third method of obtaining sectional diagrams of 

 tongue-positions is X-ray photography. In order to 



Fig. 9. — X-ray photogr.iph of the scund X- as in cnvt-. 



get good results by this process it is necessary to make 

 use of some opaque substance to show the outline of 

 the tongue. The plan which has given the most suc- 

 cessful results is to place on the tongue a little chain 

 of small lead plates. (This plan was originally devised 

 by Dr. E. A. Meyer.) 



Figs. 6 to 10 are photographs of this description 

 taken by Dr. H. Trevelyan George, of .St. Bartholo- 

 mew's Hospital, who has displayed much ingenuity and 

 patience in getting over the numerous difficulties which 

 present themselves in the course of work of this nature. 



Fig. 10.— X-ray photograph of Welsh o, as in ton, " wave. ' Said 

 by Mr. Stephen Jones, Assistant for F.xperiniental Phonetics 

 at University College, London. Tongue-position shown by 

 lower chain. The upper chain passe* through the DO«e, aixl 

 shows the shape of the upper side of the soft palate. 



Another element of speech which can be successfully 

 studied by the methods of experimental phonetics is the 

 vibration' of the vocal chords. Some speech-sounds 

 {e.g. normal v or z) are accompanied by vibration of 

 the vocal chords, others (e.g. /, 5) are not ; others, again, 

 are accompanied by vibration during a part of their 

 length. It is important for linguistic purposes to ascer- 

 tain with accuracy the precise points where vibration 

 of the vocal chords begins and ends in connected 

 speech. 



