COLUMBIA LIBRARIES OFFSITE 
 
 HEALTH SCIENCES STANDARD 
 
 HX641 37350 
 
 QP306 .A1 5 On a new method of r 
 
 RECAP 
 
 1 
 I 
 
 
 
 ' 
 
 
 '■.'■'' 
 
 
 A NEW 
 
 METHOD 
 
 
 
 
 OF RECOKDINO 
 
 
 I 
 
 THE 
 
 MOTIONS OF 
 
 HARRISON 
 
 THE SOFT 
 
 By 
 ALLEN, M.I). 
 
 PALATE. 
 
 
 
 
 
intljeCttpofiilftnl'ork 
 
 ^cfjool of Bcntal anb 0v&{ burger? 
 
 LIBRARY OF 
 
 Dr. carl F. W. BODECKER 
 
 1846-1912 
 
 The gift of 
 
 Dr. Henry and Dr. Charles Bodecker 
 
 1929 
 
/^x%>^- 
 
 ^ 
 
Digitized by the Internet Archive 
 
 in 2010 with funding from 
 
 Open Knowledge Commons (for the Medical Heritage Library project) 
 
 http://www.archive.org/details/onnewmethodofrecOOalle 
 
ON 
 
 A NEW METHOD 
 
 OF RECORDING 
 
 THE MOTIONS OF THE SOFT PALATE. 
 
 BY 
 
 HAKHISON ALLEN, M.D., 
 
 PROFESSOR OF PHYSIOLOGY IN THE UNIVERSITY OF PENNSYLVANIA. 
 
 EXTKACTED FROM THE 
 
 TRANSACTIONS OF THE COLLEGE OF PHYSICIANS OF PHILADELPHIA, 
 
 THIRD SERIES, VOLUME VII. 
 
 PHILADELPHIA: 
 
 P. BLAKISTON, SON & CO, 
 
 No. 1012 Walnut Street. 
 
 1884. 
 
V\-C C 
 
 COLLINS PRINTING HOUSK, 
 
 705 Jayne Street. 
 
The Palate- My ogi'aph in position. The left hand of the operator determines 
 the times at which the cylinder of the kymographion will revolve. The 
 right hand is so placed that the clock-work of the kymographion can be 
 wound up without interfering with an observation. The chair is brought 
 well up to the table in order that the elbows may rest easily upon it. 
 
ON 
 
 A NEW METHOD 
 
 OF RECORDING 
 
 THE MOTIONS OF THE SOFT PALATE. 
 
 In this paper I will describe a new method of 
 recording the motions of the soft palate. 
 
 When a straight rod is passed through the nose 
 from before backward in the living subject as far as 
 is possible, it will be found to impinge against the 
 roof of the naso-pharynx. In this position the rod is 
 not influenced by any motions of the soft palate ; 
 but if the end of the rod which remains without 
 the nostril be raised so that the rod be brought in 
 contact with the anterior border of the nostril, the 
 pharyngeal end of the rod will lie in a position which 
 will cause a decided motion to be transmitted to it 
 when the soft palate is raised. I have for a long 
 time been in the habit of relying upon this test to 
 determine when an instrument thrust through the 
 nose had reached the region of the naso-phaiynx. 
 
 Having observed that the sensations upon the 
 fingers when the rod was thus held in position varied 
 according to the character of the sound of the voice, 
 I instituted a series of careful comparisons, which led 
 me to infer that the elevation of the soft palate was 
 
greater in the sounds of a and e than in the sounds 
 of i, 0, and u, or in any of the short vowel sounds. 
 Substituting for the support of the fingers a flexible 
 copper wire, one end of which was attached to the 
 head-band, which is in common use for the support 
 of the laryngeal mirror, and the wire itself- bent 
 round the rod so as to maintain the apposition of the 
 rod against the anterior border of the nostril, I found 
 that the motions of the palate caused a perceptible 
 deviation of the free end of the rod. If, while the 
 wire was thus supporting the rod, a rhlnoscopic ex- 
 amination was made, the pharyngeal end of the rod 
 was found projecting from the posterior nares into 
 the naso-pharynx, at a point about midway between 
 the roof and the upper surface of the relaxed soft 
 palate. The rod being carefully adjusted so as to 
 secure the best eifects when the palate was raised, 
 and the subject seated in front of the Ludwig kymo- 
 graphion (see Frontispiece), which had been pre- 
 viously prepared with carbon covered paper, it was 
 ascertained that when the free end of the lever 
 touched the cylinder, at a time when the latter was 
 in motion from right to left and the soft palate was 
 raised, that a distinct tracing appeared upon the 
 cylinder and that the differences between the long 
 and short sounds of the vowels were found to corre- 
 late with the curves made by the rod upon the kymo- 
 graphion. 
 
 Fig. 1 represents this rod as completed, a is that 
 part of the rod, four inches in length, which is inserted 
 within the nose. The end to the left being furnished 
 with a bulb, and that to the right with a screw thread 
 for attachment to &, which is a delicate piece of vul- 
 
canite, marked by a number of annular depressions. 
 The portion of the rod marked c extends from h, 
 and represents that part of the instrument which is 
 brought in contact with the kymographion-cylinder : 
 it measures four inches iu length, and terminates in 
 a thin, flat, flexible end for the purpose of making a 
 
 tracuig. 
 
 Fig. 1. 
 
 b 
 
 Fig. 1. — The apparatus. The two arms of the rod should be of the 
 same length, a, the arm, or portion of the lever which is inserted 
 within the nose ; b, the vulcanite annular support for the wire- 
 loop; c, the arm or portion of the lever projecting from the 
 nostril (f). 
 
 When the instrument is in position (see Frontis- 
 piece) and the palate is raised (as in the voluntary 
 motion which constitutes the first act of deglutition) 
 a tracing is made (Fig. 2), which, when analyzed, is 
 
 Fig. 2. 
 
 Fig. 2. — Voluntary elevation of the soft palate illustrated in the first 
 act of deglutition. 
 
 found to be composed of a number of acute depressions 
 interrupting a horizontal line, and which yields for 
 examination three distinct parts : first, a long, con- 
 cave, slightly undulating line — which represents the 
 extent of downward deviation of the rod from its 
 horizontal position ; second, a short vertical line, 
 
which represents the descent of the palate, and 
 begins at the point at which the curved line ends 
 and terminates at the horizontal line; and, third, the 
 horizontal line itself, which represents the record 
 made by the end of the rod when at rest upon the 
 surface of the moving cylinder. It is evident that 
 the number of the notch-like depressions will answer 
 to as many acts of elevation as are recorded at a 
 single trial. The differences in the depressions will 
 be the differences in the acts of elevation themselves. 
 The degree of elevation and the abruptness of the 
 fall are constant, but the length of the curve will 
 depend upon the time taken by the palate to reach 
 its highest elevation and the time it is sustained in 
 that position before it falls to the position of rest.^ 
 
 The tracing of the motion of the palate in the first 
 act of deglutition is found to vary slightly from the 
 above when the act of swallowing is completed 
 (Fig. 3). This difference is expressed in the tracing 
 
 Fig. 3. 
 
 11 
 
 ill! 
 
 i 
 
 _^ 
 
 " """■" ^ !" "" ' '"" ' ' " ■""■? ■■ "■""■"! "■' 
 
 liiiliB 
 
 iiiiil 
 
 
 ■^V,-' 
 
 -'''" 
 
 
 ~~^Vy'i 
 
 
 
 J:xG. 3. 
 
 — The ti-acuii 
 
 • of the 
 
 act of 
 
 deglutition. 
 
 
 in a shorter curve of elevation, and a very gradual 
 instead of a precipitous descent. 
 
 The tracings of the acts of " exhaling,"^ of cough- 
 ing, of hawking, and of sniffing will be found to be 
 
 ' All tracings are to read from left to right. 
 
 2 By "exhaling" is meant the sound without voice which is made by sud- 
 denly depressine the palate from a moderately elevated position (the mouth 
 being closed), and by driving the air out through the nostrils. It is a sound 
 habitual with some persons who suffer from catarrh. 
 
suflS-ciently distinct by examination of the curves of 
 Fig. 4. 
 
 Fig. 4. 
 
 Exhaling — 
 
 Coughing^ 
 
 Hawking_ 
 (screatusj" 
 
 Sniffing-^ 
 
 The outlines, as a rule, are assumed to be due to 
 the upper surface of the soft palate striking the end 
 of the rod as it lies within the naso-pharynx. But it 
 is well to remember that the rod may be pushed from 
 right to left or from left to right in a liorizontal direc- 
 tion by the contraction of the two levator palati mus- 
 cles '^ for the elevation of the palate is dependent upon 
 
 • The motions of the levator muscles can easily be seen by direct inspec- 
 tion from in front, as has been shown by Zaufal (see p. 172) and by Hilton 
 (ibid.) from above. 
 
10 
 
 the force exerted by these muscles, and if the rod be 
 in the way of one of them it will be pushed to the 
 median side and to a point nearer the centre of the 
 naso-pharynx than it occupied prior to the time of 
 contraction. If the rod be thrust through the right 
 nostril it will be moved from left to right; if it be 
 thrust through the left nostril it will be moved from 
 right to left. These curves are best recorded by 
 placing the kymographion-cylinder in a horizontal 
 ]DOsition. 
 
 The following cut exhibits a few of the lateral 
 curves, so that an opportunity is afforded of contrast- 
 inof them with the vertical curves. 
 
 Fig. 5. 
 
 f\f\ [fill Iff if i P i f fi'i i|f ""fii f will!l|l| |'iil T i| f ! li|l l| i|li|i|i|liillll|if | 1P | f i' »' ^^ l'ii|"'H ' ' f| ri"Hfiii''iif F^ifyifH 
 
 oi, oi, el 
 u, 0, i 
 a, i, a 
 
 et 
 
 V . 
 
 lillaillESiiSil 
 
 Fig. 5. — A number of palate movements recorder! by the lever upon the 
 horizontal cylinder of the kymographion. 
 
 The correlation between the two kinds of curves is 
 sufficiently exact to permit either the one or the other 
 to be used, and since it is convenient to retain the 
 
11 
 
 cylinder in a vertical position the tracings will be 
 assumed in this paper to be made with the upright 
 cylinder unless it is otherwise stated. 
 
 In making the tracings in either of the positions 
 the head must be carefully supported. A dentist's 
 head-rest, such as is sold by the S. S. White Dental 
 Manufacturing Company, answers an admirable pur- 
 pose.^ Should it at any time be found necessary to 
 estimate the amount of variation in any curve due to 
 the motion of the head, a piece of wire attached to the 
 head-band maybe made to touch the cylinder a short 
 distance above the position of the index on the lever.^ 
 
 Inasmuch as it is desirable to have a name by 
 which the instrument can be designated I would 
 propose for it the term palate-myograph. In an exact 
 sense it furnishes the basis upon which the motions 
 of the levator palati muscles (and through these 
 structures the motions of the soft palate as well) can 
 be recorded, and is hence a myographion of a particu- 
 lar pair of muscles whose general function is already 
 known. 
 
 Literature of the metliods wliicli have heen proposed 
 to record the motions of the soft palate and pharyngeal 
 mechanisms generally. — The motions of the soft palate 
 have been studied in three ways: 1st, by direct ob- 
 servation; 2d, by the modification exerted by the 
 
 ' I desire to return acknowledgments to Dr. Jas. "W. White, President of 
 the company, for assistance kindly rendered in carrying out the plan of these 
 researches. 
 
 2 I am indebted to my friend, Dr. N. A. Randolph, for this suggestion. In 
 a few tracings, motions of the head are evident, see the tracing of sniffing 
 and the first series of Fig. 5, and in Figs. 23, 37, and 43. 
 
12 
 
 palate on the air of expiration; 3d, by the degrees 
 of force exerted by the muscles themselves. 
 
 Methods of Inspection. — These include the observa- 
 tion of J. Hilton,^ who, in a case of acquired defect of 
 the face, due to the former presence of an osseous 
 tumor, describes the " position of the palate during 
 the formation of letters and words" as follows: — 
 
 "Complete utterance occurs only during expira- 
 tion; and whilst repeating the alphabet the soft 
 palate rises at the beginning and again descends at 
 the termination of each letter or expiration; the sides 
 of the pharynx also slightly approximate. The louder 
 the voice, the more apparent are these actions in the 
 palate and the pharynx. The extent of this adapta- 
 tion in the ordinary voice resembles that which occurs 
 in the mildest oral expiration; with this difference 
 only that the palate is rather more steady in the pro- 
 cess of articulation or formation of the letter than in 
 simple expiration. 
 
 "The articulation of each of the vowels and the 
 letter Y produces about the same extent of elevation 
 of the palate. ... In the formation of the letters 
 H, M, and K, the soft palate is employed actively, 
 and especially so in the letter M." — Schuh^ has re- 
 corded a somewhat similar case. — The motions of the 
 salpingo-palatal fold, according to E. Zaufal,^ are most 
 pronounced in the letter i and in e (German), but less 
 well marked in o and u, and weakest in m and n. — 
 In progressive facial hemiatrophy Eulenberg found 
 
 ' Guy's Hosp. Rep., i, 1836, 504. 
 
 * Wiener Med. Wochenschr., 1858; see Med. Times and Gaz. xix. 
 1859, 357. 
 3 Archiv fur Ohrenheilkunde, 1875, ix. 137. 
 
13 
 
 the soft palate involved as well as the region termed 
 by Grusset^ (in quoting Eiilenberg) " peri-laryngeal." 
 The letter r together with other letters was pro- 
 nounced with difficulty. — Duchenne^ has studied the 
 effects of electrical stimulation of the muscles of the 
 palate. — Lennox Browne and Emil Behnke' have 
 examined the shapes of the soft palate from the oral 
 aspects, and have recorded their appearances by 
 photography. 
 
 Of the meiJiods of recording by the modification in 
 the outgoing of air upon a flame or tambour, refer- 
 ence may be made to the researches of Konig and 
 Hensen.^ 
 
 A. Gentilli'^ has measured the force of the expired 
 air which passed out of the nose in the formation of 
 the sounds m and n by placing in front of the nostril 
 a disk or fan upon which the expired air impinged. 
 Accessories were attached thereto by means of 
 which tracings were made upon a recording surface. 
 
 Among the methods dependent upon the degrees 
 of force exerted by the palatal muscles may be men- 
 tioned that of Arloing and Carlet,*^ who placed in the 
 mouth of a horse compressible flasks filled with 
 
 ' Traits Pratique des Maladies du Systfeme Nerveux, 1881, 627. 
 
 2 Selections from Clinical Studies, Trans, in New Sydenham Soc, Series, 
 1883, 353. 
 
 3 Voice, Song, and Speech, London, 1883, fig. 215. 
 
 * For general account and literature, see P. Grutzner in Hermann's 
 Handbuch der Physiologic, I., Bd. II., Theil, 1879, 185-189. See, also, 
 C. Blake, Archives of Otolog}', vii. 
 
 5 The Glossograph; see abstract by C. Blake in Amer. Journ. Otology, 
 iv. 1882, 193. 
 
 6 Comptes Kendus, 1874 (Nov. 2). 
 
14 
 
 water, which were in communication with a flexible 
 tube, which in turn was united with a manometer. 
 
 It will be seen that no attempt appears to have 
 been made to record the degrees of elevation of the 
 palate. Marey^ declares that any method of explo- 
 ration of the palate by direct means b}-^ which the 
 movements can be computed would be difficult, but 
 makes no suggestion how the difficulties can be over- 
 come. 
 
 I will now proceed to briefly analyze the tracings 
 made by the palate-myograph in a variety of palate- 
 motions in my own throat, in order to indicate the 
 uses to which the method here proposed may be put. 
 These descriptions will be placed under the following 
 heads : — 
 
 Vowel-Curves. 
 
 Consonant-Curves. 
 
 Syllables ending in Yowels and those ending 
 
 in Consonants conti'asted. 
 Word-Curves. 
 Test-Phrases. 
 Variation. 
 The presumed Order of Development of the 
 
 Palate-Curves. 
 Inspiritse. 
 ISTegations. 
 Tlie Yowel Curves. — The lever is found to descend 
 to the most pronounced degree in pronunciation of the 
 
 ' "Aiitant il serait difficile d'explorer d'une manifere directe les mouve- 
 ments du voile du palais, autant il est facile de siqualer ces mouvements 
 d'apres I'echappement d'air qui en est la consequence." La Methode Gra- 
 phique, etc., Paris, 1878, 396. 
 
15 
 
 vowel e. 'Next in order to e may be ranged a. In 
 Fig. 6 a comparison between these two sounds is 
 shown. 
 
 Fig. 6. 
 
 1st Series. 
 
 a e a e a e 
 
 a e 
 
 2d Series. 
 
 3d Series. 
 
 e a 
 
 It will be seen that while a slight variation is de- 
 tected between the curve of the same sound, when 
 made a number of times in the same sitting, that the 
 general character of each curve is distinct. In the 
 third line the exaggerated character of the curve is to 
 be ascribed to the rod having been thrust too far 
 back. 
 
 So in Fig. 7, where d (as in awe), and a (as in 
 father) are contrasted. 
 
 The a-sound of r is difficult to render since the pa- 
 late lies so far below the plane of the rod in making 
 
16 
 
 this sound. These curves are found to be subject to 
 considerable variation. 
 
 Fig. 8. 
 
 Variants of the sound of r. 
 
 Among other sounds of the vowels may be men- 
 tioned e (as in ''''fern''''), a (as in ""ask). This is illus- 
 trated in (Fig. 9). 
 
 Fig. 9. 
 e e e e 
 
 'OXr^j^vj 
 
 i^ 
 
 Fig. 10. 
 
 a-:k a^k a a a a 
 
 In Fig. 10 the first two tracings represent the entire 
 word "ask," and the remaining tracings the sound 
 of a in the same word. 
 
 In Fig. 11 the curves of the sounds of a and e are 
 contrasted. 
 
 Fig. 11. 
 
17 
 
 In Fig. 12 the curves of u and o and o (as in liot), 
 and 66 (as iwfood) are contrasted. 
 
 Fin. 12. 
 
 U Q\ 
 
 and 86 i 
 
 The Consonant Curves. — The tracings of the con- 
 sonants will resemble those of the vowels more or less 
 exactly according to the quality of the vowel sound 
 entering into the consonant. Thus in Fig. 13 are 
 
 Fig. 13. 
 
 4 
 
 4 _^; .- ^ 
 
 4 --' - _ 
 
 b "^--^ ■■ 
 
 4 __ r: 
 
 
18 
 
 seen the value of the curve of &, v, and z, based upon 
 the curve of e (Fig. 6)/ 
 
 In Fig. 14 the curves of the consonants j and Jc 
 based on a are presented. 
 
 Fkj. 14, 
 
 In Fig. 15 the curves of/* and I are seen based 
 upon e. 
 
 Fig. 15. 
 
 Fig. 16. 
 
 r 
 
 1 _. 
 
 \» 
 
 In Fig. 16 a contrast of the letter S in I, m, and ti^ 
 is sfiven. 
 
 ' The variations in the curves, and their limitations, are discussed on 
 pajjes 185, 191. 
 
 2 The results here recorded atrree with those of Zaufal, but differ from those 
 of Hilton see (p. 172), wlio found the motions of the palate especially marked 
 
19 
 
 In Fig. 17 the curve of e in s is figured in a 
 single group of variants. 
 
 Fro. 17. 
 
 1 
 
 -A 
 
 In Fig. 18 the curve of the letter Ti in the conso- 
 nant g is figured. 
 
 Fig. 18. 
 
 qp 
 
 The SyllahJes ending in Vowels and tJiose ending in 
 Consonants contrasted. — Since words are purely of con- 
 ventional character it is evident that the division of 
 the curves into the phonetic values of vowels and con- 
 sonants, and the arbitrary values assigned to the word- 
 curve is one of convenience only. A series of curves 
 is herewith presented (Fig. 11)) which contrast the 
 vowel sounds ending on consonants with the conso- 
 nantal sounds ending on vowels. It is interesting to 
 note that the mechanism of the palate is distinct in 
 the two instances. 
 
 The Word-curves. — For practical purposes it is con- 
 venient to use the curves made by the pronunciation 
 of entire words instead of the separate value of vowels 
 
 in the pronunciation of the letters m and n. It must be remembered that 
 this observer looked down upon the palate. The palate would appear to 
 move in the formation of the letters, but the palate is not sufficiently raised 
 either to the degree sufficient to contract the levators or to raise the velum to 
 a point at which a horizontal lever thrust into the naso-pharynx is raised. 
 
20 
 
 Fig. 19. 
 
 1 ad, (la, ad, da, 
 
 2 d, ed, d, ed. 
 
 3 t, et, t, et 
 
 4 da, ad 
 
 or consonants. The analysis of these word-curves 
 show that there are four kinds — the single, the doable, 
 the multiple, and the composite. 
 
 Of the single word-curves the following (Fig. 20) 
 may be taken as examples. 
 
 Fie 20. 
 
 Of the double curves the following (Fig. 21) may 
 be taken as examples. 
 
 Fin. 21. 
 
 Sin, sins ; tin, tins 
 Sins, sin ; sins, sin 
 
 ~\r 
 
 
 
 'J V 
 
 \i \] 
 
 "^j-^r 
 
 ~'\J \r\.r 
 
 ~^:. -:^'v/: 
 
21 
 
 Fig. 22 represents the curves belonging to the 
 letter w, which, of course, has the value when pro- 
 nounced of " double yu." 
 
 Fig. 22. 
 
 In Fig. 23 a multiple word-curve is seen in the 
 word "kingdom." See, also. Fig. 24. 
 
 Kingdom, kingdoms 
 
 not 
 
 snap 
 
 Fig. 23. 
 
 Fig. 24.' 
 
 A 
 
 once 
 
 A 
 
 chanire 
 
 c 
 
 trundle 
 
 snarl 
 
 quinsy quinsy quinsy 
 
 kingdom. 
 
 A series of double curves is represented in Fig. 
 24 showing the combination of curves more or less re- 
 sembling one another. In each instance the palate 
 comes to a state of rest for an appreciable interval be- 
 fore a second time descending. Three variants of 
 "quinsy" are presented, for it was found to vary in 
 Mr. Cohen and myself. The curves with the largest 
 excursus was evidently influenced by the levator 
 palati-muscle. The curves of "kingdom" are re- 
 peated in Fig. 23. 
 
 ' Tlie words marked "A" and " C" indicate the names of the recorders, 
 Allen and Cohen. 
 
22 
 
 A series of separate word-curves from the foregoing 
 is represented in Fig. 25. 
 
 but 
 
 Three variants of " Wcis" { ^^''^ 
 
 was 
 down 
 
 Fig. 25. 
 
 This group is intermediate between tliat of the 
 single and the doable curves. They all belong to 
 words of a single syllable, yet are distinguished 
 from those seen in Fig. 25 by showing but a slight 
 disposition on the part of the palate to descend before 
 it a second time ascends. 
 
 A second series of the same is seen in Fig. 26. 
 
 Fig. 2G. 
 
 
 --^••-.^-^v.^-X-^-^'— 
 
 m 
 
 - "^^-vT^-vi^v-j^.. 
 
 J. 
 
 ^^=^-,-=v^..-_--=^=^^ _-^^^_-^^^.^^- -^- ^^^^ ..r=-=,^=-=.-^^.-.- 
 
 .is™fj 
 
 back 
 clock 
 
 A third series, showing the greatest amount of 
 secondary curving before the palate falls to rest, is 
 seen in Fig. 27. 
 
 Test- phrases. — Since much of the interest which 
 will obtain in the use of the instrument here de- 
 scribed is clinical, it is likely that the curves described 
 at the time of the utterance of a phrase will have the 
 same value as those correlating to phonetic values or 
 
23 
 
 Fig. 27. 
 
 Shoulder 
 
 Clothes 
 
 Constable 
 
 Constable 
 (horizontal) 
 
 Papa 
 
 Brobdignag 
 
 to single words. A few phrases are herewith ap- 
 pended in illustration of this subject. (Figs. 28-3o.) 
 It is evident the palate has fewer difficulties to over- 
 come during the pronunciation of the legend of Fig. 
 30 than that of any of the other legends. Fig. 33 is 
 interesting in showing how exactly the curves of the 
 words "^o" and "&e" repeat themselves in the delibe- 
 rately pronounced phrase " to he or not to ^e." 
 
 Fig. 28. 
 
 An empty bag cannot stand upright. 
 
 Fig. .50. 
 
 A bird in the hand is worth two in the bush. 
 
24 
 
 Fig. 31. 
 
 Early to bed, early 
 to rise makes a man 
 healthy, wealthy, and 
 wise. 
 
 Fig. 32. 
 
 A rolling stone gathers no moss. 
 
 Fig. 33. 
 
 To be or not to be. 
 
 to 
 
 be 
 
 not to be 
 
 The question that has often been asked why words 
 are pronounced in a manner different from that which 
 would be premised from the spelling, and why the 
 street pedlers of cast-off clothing persist in crying 
 "old clo'' when "old clothes" is known by them to 
 be the correct pronunciation, may receive an answer 
 when the curves of the word as really pronounced 
 are compared to the curves of the word as they should 
 be pronounced if every syllable were given its full 
 value. Such words as Magdalen, Ispahan, Treheran, 
 and Beauchamp may serve as examples. Fig. 34 
 
 Fig. 34. 
 
 W^-'^-'-'^r-^'^- 
 
 — ~ 
 
 / 
 
 — ^ 
 
 i '^ 
 
 
 
 -m 
 
 Beecham (repeated.) 
 Beauchamp. 
 
 illustrates the difference between Beauchamp and 
 *' Beecham." 
 
25 
 
 Variation. — It is not to be supposed, however, that 
 the palate will make the same curves for the same 
 sounds in all individuals. A margin for individual 
 variation must be made if any conclusion can be 
 drawn from the manner in which my assistant, Mr. 
 Cohen, will form certain curves as contrasted to those 
 of other observers, including myself. 
 
 Thus Mr. Cohen in pronouncing " awe?" secures 
 two curves (see Fig. 35), whereas in my own throat but 
 a single curve of a different shape results.^ 
 
 Fig. 85. 
 
 Variations in " and" in two individuals. 
 
 The character of the curve is here of less import- 
 ance than the number, for in a separate series of stu- 
 dies of "a?^(?" in my own throat (see Fig. 36) I have 
 
 Fig. 36. 
 
 Variations in " and" in the same individual. 
 
 secured two distinct types of curves from this word 
 but never more than one curve for each word. The 
 difference in the shapes of the curves may be attri- 
 buted to the difference in the position of the rod with 
 respect to the palate during the different sittings. 
 
 ' The first two of the first series of lines have unfortunately been muti- 
 lated in Fig. 35. The last two (placed to the right of the cut) are intact. 
 
26 
 
 TJie presumed Order of Development of the Palate- 
 Curves. — The palate-myograph may prove useful, it 
 is thought, in studying the times at which the different 
 motions of the palate in phonation came into play. 
 If the articulate sounds made by a child be imitated 
 when the palate-myogi'aph is in position it will be 
 seen that the sound of mamma is the simplest of them 
 all. I have under this heading simply contrasted the 
 curves of "mdmma" and ^^pfdpd^^ (see Fig. 27), to 
 show that it is probable that the reason for the sound 
 "• mamma" (Fig. 37) being first used by the infant is 
 that the palate is moved least in producing the 
 sound. 
 
 Fig. 37. 
 
 The curves of " mamma. 
 
 TJie InspirifcB.—S'mce the sounds of speech for the 
 most part are made upon the out-going breath the 
 movements of the soft palate in assisting in their for- 
 mation will be co-related to the respiratory act. The 
 palatal mechanism might be compared to a second 
 sphincter placed between the laryngeal sphincter and 
 the lips. When the proposition is differently stated 
 and the speech sounds are known to be found not on 
 the out-going but on the in-going air, it has been 
 thought worth while to notice if the palate mechanism 
 is in any way modified on account of the air passing 
 through the pharynx. 'No sound of the character just 
 named is normal to English or indeed to any of the 
 European languages. Dr. D. G. Brinton, who first 
 culled my attention to the subject, informs me that in 
 
27 
 
 the Bushman dialect as many as four of the inspiritse 
 (as they are termed) can be enumerated. The Esqui- 
 maux have at least one, the "cluck," and the Ara- 
 bians some. 
 
 In experimenting with the inspiritse difficulty was 
 experienced in the unpleasant drying of the vocal 
 cord attending the act of articulating upon the in- 
 going breath. I have contented myself to reproduce 
 the "cluck" sound (Fig. 38) both on inspiration 
 
 Fig. 38. 
 
 Inspired. 
 Expired. 
 
 Inspired. 
 Expired. 
 
 ■■^. 
 
 ,r-^-r^'^- 
 
 r-, — :. ,-i 
 
 
 ~\ V"\ \ 
 
 -~x vm-r^ 
 
 
 
 ■n '"^^'.J — ^ 
 
 _ ^ 
 
 ^ i' — ^ \~ 
 
 -^x r-\ ' ■ (■ "i 
 
 
 V, i V,\ 
 
 
 The curves of "cluck." 
 
 and on expiration in order that the two can be con- 
 trasted; it will be noticed that the difference between 
 the two sets of curves is very slight if any can be 
 said to exist. 
 
 Fig. 39. 
 
 In Fig. 39 the curves of the vowel sounds a, e, i, 
 0, u are represented as formed in inspiration.^ 
 
 ' Dr. Berens. 
 
28 
 
 The inspiritae of low pitch cause larger curves to be 
 described than those of high pitch. 
 
 The amount of elevation of the soft palate in enun- 
 ciation is independent or nearly so both of volume 
 and of pitch. In Fig. 40 the vowels a and e are 
 
 contrasted in the head and chest registers. The 
 curves show some influence from the action of the 
 levator palati muscles, but, with the exception of 
 being a little too deep, are normal curves for these 
 sounds. It will be seen that there is little or no effect 
 of register upon the formation of these curves. 
 
 In two series the following falsetto notes (Fig. 
 41) were formed on the same vowel sounds. A more 
 
 Fig. 41. 
 e a 
 
 marked difference is here observed between the curves. 
 The palate appears to be tremulous, and makes two 
 undulations in the curves of e. This record is from 
 
29 
 
 my own throat. I have no vocal training, and it is 
 probable that the palate of a vocalist may give dif- 
 ferent results. 
 
 Negations. — While it has been seen that the palate 
 moves in every consonantal and vowel sound, it 
 nevertheless does not assist in the formation of many 
 sounds. Thus the motions of the palate are the same 
 in the word " souf'' as in " souW (Fig. 42), and 
 
 Fig. 42. 
 
 " Souf" and south. 
 
 in ''■nuffiyvf as in ^' nothing^'' (Fig. 43). The diffi- 
 culty the negro acknowledges in pronouncing 
 
 Fig. 43. 
 
 " Nuffinsr" and notliinfr. 
 
 '^ souW and '^notJiing^^ (converting the words into 
 '' soiif'^ and '' nuffing^') lies in the relations of the tip 
 of the tongue to the teeth, hard palate, and lips (in a 
 word to his prognathism) and not to any peculiarities 
 of the soft palate. 
 
 In like manner those persons who pronounce luhat 
 as though spelt " vaP (Fig. 44) make precisely the 
 
 Fig. 44. 
 
 What 
 
 'Vat. 
 
30 
 
 same palate curves for the two sounds. The curves, 
 indeed, are those of d with the terminal consonant t, 
 and tlie distinction between the ivk and the v are in 
 the lips and teeth. 
 
 The distinctions between that and " ?a^" (Fig. 45) 
 are of the same character. Most children are slow in 
 
 Fig. 45. 
 Lat and that. 
 
 pi'ocuiing the " ^A" sound but acquire the " Z" sound 
 easily. The same remark applies to the Chinese. 
 
 The same comments ai-e not applicable t the dis- 
 tinction between ivell and ^^veV^ (Fig. 46); for a 
 
 Fig. 46. 
 Well and "veil. 
 
 simple curve is seen to be recorded for "i(;eZZ," but a 
 sinuate curve for "reZZ." 
 
 In Fig. 47 the curve of the words one, two, three, 
 eight, yiine, ten are recorded with the object of show- 
 ing that the word "one" makes scarcely any curve, 
 the word "nine" none. 
 
 Remakks. — Such in brief is an outline of this new 
 method. The fact that the soft palate is raised in 
 aiticuhition of all articulate sounds, that it is raised 
 in the acts of swallowing, of coughing, of hawking. 
 
31 
 
 can be i-eadily demonstrated. The length of time at 
 which the palate remains elevated, and the duration 
 of the time of ascent and descent, can also be easily 
 
 Fig. 47 
 1 2 
 
 ], 2, 3 
 8, 9, 10 
 
 8 9 10 
 
 The curves of "one, two, thi-ee, eight, nine, ten." 
 
 ascertained. The palate is seen to be raised once 
 only for some words ; twice for others, and three 
 times for others. The numbers of these motions are 
 invariable within a narrow range of individual varia- 
 tion. The size of the curves, however, is not fixed 
 owing to the difficulty of always reaching precisely 
 the same spot of the broad upper surface of the soft 
 palate. Bat the curves, whether deep or shallow, hold 
 the same proportions one to another, as is seen in 
 the diflPerent series of Fig. 6, and in the various 
 curves of the letter I in Fig. 16, and of the letter s 
 in Fig 17. By proper care in freeing the rod from 
 contact with the salpingo-palatal fold, and in placing 
 the rod about the centre of the palate, the exaggera- 
 tion of curve can be in great measure prevented. It 
 is evident that variation in the size of the bulb at the 
 pharyngeal end of the rod will cause corresponding 
 variations in the size of the curves. Hence it is re- 
 commended to preserve the form and size of the bulb 
 described on p. 167, and to keep it free from mucus. 
 
 Even if the same spot could at all times be reached, 
 would it be reasonable to expect the curves to have 
 
32 
 
 uniform values for the same sound in different indi- 
 viduals, or in the same individual at different times, 
 or even in the same individual at different times in 
 a single sitting? The soft palate is but partially 
 under the control of the will, and this control varies 
 in different persons. The palate, as other muscular 
 structures, is subject to fatigue, or may act irregu- 
 larly from prolonged contact with the metallic 
 rod. Again, the personal element that enters into the 
 variation in the transit of volitional impulses must be 
 as appreciable in the case of the soft palate as any 
 other muscular apparatus in the economy. 
 
 On the whole, it may be said that the sources of 
 inconstancy of the curves are not sufficient to invali- 
 date the claim of accuracy for the method. 
 
 The palate-myograph may be of use in studying 
 the mechanism of the soft palate, in disease as well 
 as in health. It is evident that in pai-alysis of the 
 palate the motions would be absent, and that by this 
 method a means of detection of this condition is 
 available. It may probably prove of use in studying 
 stammering, and in determining the degree of de- 
 generation of the levator-palati muscles in progres- 
 sive dry aural catarrh. 
 
 As the soft palate ascends, the oral surface becomes 
 concave.^ This concavity is occupied by the convex 
 
 ' It would be an ungracious act to examine critically the figures in the 
 numerous works upon the throat and mouth in which the palate is represented, 
 as seen in sagittal (anterio-posterior) section during the formation of the 
 Towels and consonants ; without exception they are based upon erroneous 
 conceptions of the movements of the velum in speech. 
 
33 
 
 dorsum of the tongue in some sounds as in the con- 
 sonant quantities, Ted and qu, etc., but may be free from 
 the tongue in other sounds, as that of o. The elevation 
 of the palate is not the result of the pushing up of 
 the relaxed velum by the lingual basi-dorsum, but of 
 the active elevation by means of its own muscles. 
 The correlations of the tongue with the palate are, 
 as a rule, notwithstanding, exact enough to warrant 
 careful comparison of the two factors in the mechan- 
 ism of speech. 
 
 The palate-myograph may be employed in the 
 study of phonetics, together with the different forms 
 of glossographs as an accessory, if the premises of 
 the preceding paragraph are correct. 
 
 On the action of palato-pharyngeal and azygos mus- 
 cles little can be here said. I have made a number of 
 observations with a long curved rod passed through 
 the nose, and have ascertained that the approximation 
 of the two palato-pharyngei may be recorded, but the 
 retention of the rod in this position is painful, excites 
 a little bleeding, and soon throws the pharynx into 
 an excited state. 
 
 The plan of holding the rod in any desired position 
 by the flexible wire and head band, can be utilized in 
 keeping instruments in position such as the snare, 
 the syringe, etc., so that the head-band may be em- 
 ployed as an accessory in the removal of growths 
 from the naso-pharynx and the nose.^ 
 
 ' I desire to express my acknowledgments to Dr. N. A. Randolph, Dr. C. 
 Berens, and Mr. N. Cohen, for assistance rendered during the coiu'se of the 
 investigation. 
 3 
 
34 
 
 The foregoing account is designed to indicate the 
 medical uses of this method of studying palatal 
 motions. The method may have a wider range of 
 application than the one delineated ; and it may be 
 well to suggest at this place that, in the judgment of 
 competent critics, it may be made available for the 
 comparative study of language, for the instruction of 
 the deaf, and for the formation of a system of logo- 
 graphy or short-hand writing. 
 
 I have thought it best, however, to confine my text 
 to the limits which naturally define the horizon of 
 medical observation, rather than delay publication by 
 awaiting the results of laborious efforts needed to ex- 
 tend these limits over a wider and untried field. 
 
COLUMBIA UNIVERSITY LIBRARY 
 
 This book is due on the date indicated below, or at the 
 expiration of a definite period after the date of borrowing, 
 as provided by the rules of the Library or by special ar- 
 rangement with the Librarian in charge. 
 
 DATE BORROWED 
 
 DATE DUE 
 
 DATE BORROWED 
 
 DATE DUE 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 C28(23S)M100 
 
 
 
 
QP306 
 
 
 
 
 A] 5 
 
 Allen 
 
 
 
 
 cop • 2 
 
 On a 
 
 new 
 
 method 
 
 of 
 
 recording the