Citation
An x-ray study of unrepaired, incomplete cleft palate oral- pharyngeal structures and their functioning during vowel phonation

Material Information

Title:
An x-ray study of unrepaired, incomplete cleft palate oral- pharyngeal structures and their functioning during vowel phonation
Creator:
Keenan, Joseph Sessions, 1928-
Publisher:
[s.n.]
Publication Date:
Language:
English
Physical Description:
vii, 108, [1] leaves. : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Cleft palate ( lcsh )
Speech disorders ( lcsh )
Speech thesis Ph. D ( lcsh )
Dissertations, Academic -- Speech -- UF ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 101-106.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
022002076 ( ALEPH )
18312799 ( OCLC )

Downloads

This item has the following downloads:


Full Text
An X-Ray Study of Unrepaired, Incomplete Cleft
Palate Oral-Pharyngeal Structures and Their
Functioning During Vowel Phonation
By
JOSEPH SESSIONS KEENAN
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA June. 1960




ACKNOWLEDGEMENT S
The writer wishes to express his appreciation and
indebtedness to Dr. McKenzie Buck, the chairman of his supervisory committee, for his constant assistance and encouragement throughout the course of this dissertation. As a member
of the national research group which conducted the parent study, Dr. Buck arranged for the use of the X-ray plates, the tape recordings, and the descriptive data utilized for analysis in this study.
The writer would also like to express sincere appreciation to Professor H.P. Constans, Dr. R.E. Tew, and Dr. L.L. Zimmerman of the Department of Speech, and to Dr. E.P. Home and Dr. J.C. Dixon of the Department of Psychology for their cooperation and encouragement as members of the writer's supervisory committee. In addition, the writer would like to thank Dr. G.11. Kurtzrock, who was an original member of his supervisory committee.
Special expression of appreciation is extended to Frank E. Law, D.D.S., Chief, Operational Research Branch, Division of Dental Public Health, Washington, D.C., who organized the study team which collected the raw data for the present study. The writer would also like to express special appreciation to the citizens of ?uerto Rico who participated in the study. Further special thanks are extended to Dr.
ii




Calvin Knobeloch, who made both his data and time generously available when they were needed. In addition, the writer wishes to especially thank Mr. Reid Holmes, Administrator of Baptist Hospitals, Inc. in Winston-Salem for his understanding and interest during the preparation of this dissertation.
These acknowledgements would be quite incomplete without an expression of gratitude to the writers wife. Without her courage and faith this task would not have been begun. Without her devotion and cooperation it could not have been completed. Finally, the writer is grateful to his children, Chris and Ceci, who have put up with him during times when it was a triel to have a writer for a father.
iii




TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS .. ... . . . ii
LIST OF TABLES . . . . . . . . . v
LIST OF FIGURES . . ...... ....... vi
Chapter
I. INTRODUCTION ................ 1
II. DESCRIPTION OF PROCEDURE . . . . . 14
Selection and Matching of Subjects . . o 16
Procedure for Recording and Analysis of
Speech Samples . . . .. . 18
Apparatus and Procedure for Obtaining
X-ray Plates . . . . 21
Measurements of the X-ray Plates . . . 25
III. RESULTS .. 35
Facial and Oral Cavity Dimensions . . . 35
Measurements Related to Tongue Size and
Horizontal Position of the High Point
of the Tongue . . . . .. . 41
Vertical Tongue Position . . . . . 49
Excursion of the Tongue from Rest to
Vowel Positions . . . a & a 55
Measurements of the Naso-pharyngeal
Structures .. . . 59
Lip and Mouth Opening- . . *..... 69
Analysis of Speech Samples .-. . . .& 73
Correlations between Acoustical and
Physical Measurements ..... . . . 78
IV. SUMMARY AND CONCLUSIONS. . *...... 84
V. DISCUSSION .. .... .. 92
LIST OF REFERENCES.o. . . . .1. o 101
iv




LIST OF TABLES
Table Page
1. Measurements of Maxillary, Mandibular, and
Oral Cavity Structures among Subjects with
Normal Development and Subjects with
Incomplete, Unrepaired Cleft Palates . . 37
2. Midsagittal Tongue Area and Horizontal
Position of High Point of Tongue Along
Incisal Reference Line among Subjects
with Normal Development and Subjects with
Incomplete, Unrepaired Cleft Palates . . 42
3. The Pharyngeal Openings On and Below the
Incisal Reference Line among Subjects with
Normal Development and Subjects with
Incomplete, Unrepaired Cleft Palates . . 46
4. Various Vertical Measurements of Tongue
Carriage among Subjects with Normal
Development and Subjects with Incomplete,
Unrepaired Cleft Palates . . . & . 50
5. Movement of the Tongue from Rest to Vowel
Positions among Subjects with Normal
Development and Subjects with Incomplete,
Unrepaired Cleft Palates. . . . . . 56
6. Area of the Posterior Pharyngeal Wall at
the Level of the Tubercle of the Atlas and
Opening of the Velo-pharyngeal Structures
among Subjects with Normal Development and Subjects with Incomplete, Unrepaired Cleft
Palates 60
7. Measurements of Movement of the Posterior
Pharyngeal Wall at the Level of the
Tubercle of the Atlas among Subjects with
Normal Development and Subjects with
Incomplete, Unrepaired Cleft Palates . . 64
8. Lip and Mouth Opening at Rest and During
Phonation among Subjects with Normal
Development and Subjects with Incomplete,
Unrepaired Cleft Palates ...... . 70
V




LIST OF TABLES
(Continue d)
Table Page
9. Coefficients of Correlation between Physical and Acoustical Measurements among
Subjects with Incomplete, Unrepaired Cleft
Palates. & o o.o e o 9 ao*oo 80
vi




LIST OF FIGURES
Figure Page
1. Master Tracing of a Typical Lateral X-ray
Plate Showing Reference Points and Location
of Measurements *..... ... .... 29
2. Master Tracing of a Typical Antero-posterior
X-ray Plate, Indicating the Structures and the Points from Which All Antero-posterior
Measurements Were Taken............ 33
vii




CHAPTER I
IN TRODUC TI ON
Various scholars have used the X-ray in speech investigations since its discovery in 1895, according to Parmenter and Trevino (28). Of primary significance, at least in terms of stimulating thought and further investigation, were the studies of Russell (30, 31, 32). He utilized X-ray techniques in his studies of vowel and consonant production in normal subjects.
Parmenter and Trevino (28) criticized Russell principally for his presentation of evidence against a physiological basis for vowel differentiation. They felt that Russell's interpretation of his X-ray plates was correct, but they were of the opinion that his technique for maintaining a fixed posture during exposure was inadequate. For their X-ray investigation of normal subjects, they devised a head positioner which appeared to control the head adequately in a horizontal plane. There is some question as to the control afforded the head in the antero-posterior plane. The results of their study supported the hypothesis of a progression of tongue positions as the vowels are produced in traditional order, and of a fairly definite tongue position for each vowel for each individual studied.
1




2
The criterion of a fixed and constant head position during X-ray exposure continued to receive attention in subsequent studies conducted on normal subjects. In 1934, Norris (27) utilized a head positioner developed by Higley in her study of vowel production in adults as it is related to voice. Norris' conclusions concerning nasality and denasality indicated a wide uvular opening or a thinning of velar tissue for the former, and a complete uvular closure or a thickening of the velum for the latter.
Further investigation of naso-pharyngeal closure in adults was done by Wolfe (41) in 1942. Recognizing the
limitations of lateral X-ray plates, he reported that his data seemed to reveal only a shortening of the anteroposterior diameter of the naso-pharynx. His data also indicated upward and backward movement of the velum, together with a slight forward movement of the posterior pharyngeal wall. This series of movements effected a nasopharyngeal closure which was not complete for most of his subjects. He found that greater anterior movement of the posterior pharyngeal wall was accompanied by less posterior
movement of the velum, and vice versa.
Comparing oral and pharyngeal structures of adults
with nasal voices and adults with superior voices, Hixon (20) found that the over-all vertical and antero-posterior dimensions of the face were very similar for both groups.




3
The measurements of the midsagittal diameter of the pharynx and the midsagittal measurements of the tongue indicated that the tongue was carried farther posteriorly in the nasal group while at rest and for the vowel (a], with a tendency in this direction for [ae]. Hixon also commented on his findings rewarding naso-pharyngeal structure, but he reported that these structures were not clearly defined on his X-ray plates. The study contained his recommendations to clarify this problem; i.e., the use of slower speed cassettes, the use of a lead focusing plate, and painting the entire inferior margin of the soft palate in order to differentiate the uvula from the soft palate proper.
While the study conducted by Brodie (2) was also
concerned with subjects with normal structures, it differed from the preceding investigations in that it was longitudinal in aspect. Brodie indicated that the use of an X-ray technique in growth studies had been predicted as early as 1922, but that its development was not perfected for this purpose until 1931. In that year Broadbent published the full details of a roentgenographic cephalometer and began
a long range investigation of the head of the growing child. Brodie summarized his own investigation with the statement that a longitudinal study of growing children indicated that the morphogenetic pattern of the individual was established
at a very early age. Once this pattern was attained, he




4
found that it did not change. Herzberg and Holio (19) reported that their findings confirmed Brodie's results by anthropology methods. Their findings were based on measures obtained from three hundred and twenty-six adult human skulls.
Growth studies have also been conducted using
cleft palate subjects. Graber (11) reported his analysis of the developmental patterns and facial morphology in cleft palate. As a group, his subjects showed deficient patterns of maxillary growth, laterally, antero-posteriorly, and vertically. He indicated that patients waose palatal clefts had not been manipulated surgically showed a pattern of maxillary growth that was essentially normal. Mandibular growth on the whole appeared normal. The lack of development noted in mandibular growth in Graber's subjects was considered to be no more than would be found in the same distribution of normal subjects. Graber's chief contention, it would seem, was a serious consideration of the propriety of early closure of cleft palates. Postponement of surgery until the end of the fourth year of life was indicated, at which time the majority of total maxillary width had been accomplished. A necessary further precaution was pointed
out due to a continuation of downward and forward growth of the maxilla until twenty years of age.
Krogman (25) offered some support of Graber's (12)




5
findings regarding the growth of unrepaired structures in his report of cumulative X-ray studies done at the Philadelphia Center for Research in Child Growth. He stated that the postnatal growth patterns of unrepaired
cleft palate and normal faces did not differ, except that growth in the former proceeded in keeping and in harmony with the basic deviation initially introduced by the cleft. He felt that an optimum time for surgical procedure would be somewhere between four and six years of age, and possibly as early as two to three years of age if the
growth tempo of the individual child was such that there was evidence of advanced growth.
Harrington (17) confirmed Wolfe's (41) questioning of the nature of velo-pharyngeal closure with the results of his study of ten cadavers and X-rays of three living subjects. The various aspects of his investigation led him to the conclusion that mesial movement of the pharyngeal walls occurred over considerable vertical
extent, rather than as a continuation of Fassavant's cushion. Consequently, the mechanism acting to produce velo-pharyngeal closure was not of the nature of a simple sphincter. A similar phenomenon was suggested earlier by Townshend (39) in his critique of the formation of
Passavant's bar.
Using X-rays, plaster models, and photographs,




6
Slaughter and Brodie (35) concluded their study with statements that further supported Graber's (11) contention regarding the impropriety of early surgical repair of cleft palates. They also presented statements to support Graber's (12) and Krogman's (25) remarks on the essentially normal growth rate of congenitally deformed parts which were not permanently damaged.
Buck (4,5) conducted an X-ray study of oral and pharyngeal structures in patients, ranging in age from seven to nineteen years, who had surgically repaired cleft
palates. His findings regarding over-all vertical and antero-posterior dimensions of the face further substantiated Graber's (12) premise of the interference of early surgical closure with growth. Buck found smaller midsagittal measurements of the tongue in his cleft palate subjects, as did Hixon (20) in his non-cleft nasal subjects.
Concerning midsagittal pharyngeal diameter on and below the incisal reference line, Buck found measurements for his two groups to be very similar. Proportional measurements of the horizontal position of the high point of the tongue, which took into consideration the structural differences between the two groups, were also very similar. These measurements failed to support Hixon's conclusion on posterior tongue carriage. Lower tongue carriage in the oral cavity was indicated by Buck when he found a tendency for the cleft




7
palate subjects to show a greater vertical distance between the hard palate and the tongue. Regarding velo-pharyngeal movement, Buck reported significantly larger openings for his cleft palate subjects, smaller horizontal and predominantly vertical movement of the velum in the cleft palate group, and very small posterior pharyngeal wall movement for both groups. He also found no significant differences between the two groups on measures of mouth and lip opening.
The error of Gustav Passavant was commented on by Calnan (6), who based his contentions on clinical and radiological evidence. Passavant had observed his ridge in a normal subject, but he failed to explain why it was observed in only one in fifty patients. Calnan felt that Passavant's phenomenon could be seen in unrepaired cleft palates, but with an occurrence in less than 25 per cent of cases. He explicitly denied that the phenomenon was a factor in normal speech.
Hagerty and Hoffmeister (13) felt that the cleft palate was closed primarily to improve speech. They offered a quantitative measure of speech as an indicator of success in palatal closure, if no other damage was incident to the closure. It was hypothesized that measurements of the minimal velo-pharyngeal distance taken during prolongation of [s] enabled them to predict




nasality from the X-ray with about 75 per cent accuracy.
Laminagraphy, a recent development in X-ray studies, has been utilized in some of the more recent cephalometric investigations. This is a technique for obtaining measurements of tissues which lie in planes from the surface to the midline. The tube and film of the laminagraph move during exposure, and the X-ray penetrates deeper each time it passes over a body. Brader (1) pointed out that this technique permits projection of any selected plane of a body to the exclusion of all other planes. To illustrate, structures lying in the plane of the vertical process of the maxilla, in the plane of the
nasal turbinates, and in the plane of the nasal septum could be viewed on successive laminagraphic plates.
Static X-ray plates present a view only of the mid-sagittal plane. Brader further stated that laminagraphy made possible the visualization of certain anatomical structures or pathologic lesions which are not susceptible to clear projection by any other method. Brader conducted preliminary tests to determine the accuracy with which images in laminagraphic projection might be measured. These tests indicated that such measurements lay within the limits of scientific accuracy, i.e., within one-half millimeter of direct measurements.
In 1953, Subtelny (37) reported his findings of a




9
laminagraphic investigation of the width of the naso-pharynx and related structures in normal and pre-operated cleft palate children. The subjects were all under three years of age and ninety-one unoperated cleft palate patients were included in the study. Perhaps the most striking finding reported was the indication of an abnormally wide naso-pharynx in the cleft palate individuals. The bi-hamular measurement, which is the distance between the inferior terminations of the medial pterygoid plates of the sphenoid bone, was found to be larger for this group. The outward
flare of these medial plates was found to increase as they descended. Subtelny hypothesized that certain muscles pulled on the open embryonic structures to create this wider dimension. He introduced an element of doubt as to the validity of this hypothesis when he commented on certain of these muscles which could have the opposite effect on the opening. The lateral pressures on the cleft structures created by the tongue might augment the widening effect. In the case of a repaired lip, the buccinator musculature is such that it should have a restraining effect. Subtelny commented that there appeared to be a levelling off in nasopharyngeal width during the second year of life.
Ricketts' (29) laminagraphic study stressed the need for a more complete understanding of the basicranial structures. He cautioned that inferences from height and




10
depth dimensions projected on lateral X-ray films were not conclusive. In view of Subtelny's (37) finding on nasopharyngeal dimensions and the findings of his ow study, he concluded that the actual size of the adenoid was not as important in velo-pharyngeal closure as the available space that it filled in the naso-pharynx. He was in agreement with Graber (11) in stressing the importance of the descending growth of the hard palate in evaluating the individual condition. Without an increased range of activity in the soft palate, this descent seemed to explain two circumstances: the frequently experienced increased nasal emission in postoperative cleft palate patients after the age of six to eight years, and the almost spontaneous disappearance of mouth breathing in some patients in this same age group.
A discussion of the significance of early orthodontia in cleft palate habilitative planning by Subtelny (38) included references to a laminagraphic study he had completed on repaired cleft lip and cleft palate children prior and subsequent to surgical reconstruction of the lip. He found that the molding action of the lip was not restricted to the alveolar border alone, but might cause an architectural rearrangement involving each maxilla. In many cases that demonstrated this movement, Subtelny found that the maxillary constriction was severe enough to bring




about contact between the inferior nasal turbinate and the nasal septum. This condition he offered as a possible explanation for the distortion of nasal sounds in cleft palate speech, which has been described as a hyponasality superimposed on the hypernasal quality.
Hagerty, Hill, Pettit, and Kane (15) reported a laminagraphic study concerned with posterior pharyngeal wall movement in normal adults. They concluded that it was doubtful whether the actual extent of excursion of the posterior pharyngeal wall was significant enough to make important differences in speech. Exception to this conclusion would occur when the individual possessed a Passavant's bar large and vigorous enough to participate actively in closure. These authors also reported on soft palate movement in normal adults (16). In the absence of a vigorous Passavant's bar, it seemed evident that the soft palate participated far more aggressively than the posterior pharyngeal wall in the regulation of the nasopharyngeal aperture during speech. These observations were supported by Hagerty and Hill (14) in their investigation of individuals with repaired cleft palates. Passavant's bar was seen in only nine of eighty subjects, and was functional in only three. These findings support those of Wolfe (41) on the participation of the velum in velo-pharyngeal closure.




12
The preceding discussion of static X-ray investigations forms the background for the present study. Graber (12), Krogman (25), and Slaughter and Brodie (35) have suggested that the growth pattern of the structures of individuals with unrepaired cleft palates proceeded as
in normally structured individuals. Krogman offered the qualification that such growth would be in keeping and in harmony with the deviation introduced by the anomaly. The study by Buck (4), using individuals with repaired cleft palates, offered support to their contentions. A special case in Buck's investigation revealed that the cleft palate individual whose speech was judged as having relatively little nasality differed chiefly from the other cleft palate subjects in only two respects: (1) the relatively normal size of his facial and oral structures, and
(2) in his being able to achieve a fairly complete closure of the nasal port. It is of interest to note that Buck reports this individual to have undergone only lip repair prior to his seventeenth year. At the age of seventeen surgical repair was performed on the hard and soft palates.1
Several questions, based on suggestions derived
from the foregoing review of literature, constitute the basis
iThe background analysis and procedure chapters are essentially the same as heretofore reported (24).




for the present study:
1. How does a group of physically mature individuals with incomplete, unrepaired cleft
palates compare with a similar group of
normally structured individuals of the
same origin?
2. How does a group of physically mature individuals with incomplete, unrepaired cleft
palates compare with a similar group of
individuals with complete, unrepaired
cleft palates?
3. How does a group of physically mature individuals with incomplete, unrepaired cleft
palates compare with other cleft palate individuals who have undergone surgical
repair prior to maturation?
4. What effect does the presence of the unrepaired anomaly have on the speech of the
adult?




CHAPTER II
DESCRIPTION OF PROCEDURE
The present study may be described as a detailed analytical aspect of' a parent study sponsored by the U. S. Department of Health, Education, and Welfare. The motivating factor which led to the formulation of the parent study was a desire to accumulate data on physically mature individuals who had untreated cleft palates. The criterion for choosing a site for the study was the location of a suitable group of cases. Through the interest and cooperation of the Insular Department of Health, a sizable group of such cases was located on the Island of Puerto Rico. A study team of dentists and a speech pathologist visited the island in June, 1957, and June, 1958, to collect the raw data.2 Dentists, social workers, and other personnel from the Insular Department of Health and Puerto Rican
2The members of the study team included Frank E.
Law, D.D.S., Division of Dental Public Health, Public Health Service, U.S. Department of Health, Education, and Welfare, Washington, D.C.; John T. Fulton, D*D.S., Professor of Dental Epidemiology, University of North Carolina, Chapel Hill, N.C.; McKenzie Buck, Ph.D., Head,, Speech and hearing Clinic, University of Florida, Gainesville, Fla.; and others. (The complete listing of the research group members is available upon request). None of the results of the parent study are to be published until all aspects of it, such as this study, are completed. A portion of the raw data gathered by the team (X-rays, tape recordings, descriptions) were made available to the writer to enable him to fulfill a part of the requirements for his advanced degree in speech pathology.
14




15
health centers augmented the study team while the investigation was in progress. Prior to the team's arrival on the island, case finding was accomplished by the Bureau of Crippled children and the Bureau of Dental Health of the Puerto Rico Department of health, assisted by the Puerto Rican Office of Vocational Rehabilitation. Medical examninations were conducted on the 1957 study group but, as it was felt that the results were not significant for this research, they were not conducted on the 1958 study group.
The experimental design of the present study is
similar to that used by Buck (3) in his X-ray Investigation. he studied oral and pharyngeal structures and functioning In individuals with repaired cleft palates.
Two groups of adult Puerto Ricans were studied:
(a) an experimental group with incomplete, unrepaired cleft palates; and (b) a control group with normal structures, speech and hearing, selected at random. Tape recordings of both groups were judged for accuracy of articulation and excellence of voice quality.
Lateral X-ray plates were taken of each subject
during the phonation of the vowels [a], [1], and [u], and at physiologic rest. An artero-posterior plate was also taken of each subject at physiologic rest. As will be discussed in a later section, not all of the plates were available for
examination.




16
Each lateral X-ray plate was traced and the
tracings were measured to secure data related to the size, positioning, and movement of the oral and pharyngeal structures, and the cross-sectional diameters of the oral and pharyngeal cavities. Lateral measurements of the skeletal structures of the face were obtained directly from the antero-posterior plates.
The data obtained from the measurements and the judgments of the tape recordings were analyzed by various statistical techniques in order to permit comparison between the two groups of subjects.
Selection and Matching of Subjects
To be included in the study group, cases had to meet the following qualifications: (1) must have reached maturation (fifteen years of age or over); and (2) must have a palatal cleft which had (a) no surgical or prosthetic treatment, or (b) surgical treatment no earlier than twelve years of age, and which was not successful* A palate which was closed and which has since pulled apart indicates that the dynamic processes of facial growth may not have been retarded by the operation. One of the experimental group was classified under (b), above. The measurements taken on her will be discussed in detail in a later chapter of this study.




17
On the basis of descriptions that were available
for the individual subjects, the twenty individuals selected for inclusion in the experimental group have been classified as having incomplete, unrepaired cleft palates* Twelve of these subjects had normal hearing. One of the eight remaining had a bilateral high frequency hearing loss. The other seven subjects all exhibited some degree of hearing
loss which appeared to be conductive in kind, and which might be attributable to the cleft palate condition (10, 22, 33). Of this latter group, three showed unilateral and four showed bilateral losses.
Thirty-six normally structured adults were also
examined. The criteria for selection of these individuals were that they had achieved normal structural growth and possessed the normal speech and hearing of the Puerto Rican population. These individuals were selected at random from the various examination centers. Of the thirty-six, twenty were chosen for inclusion in the present study as a control group. This was the same control group used In a prior study (24).
The age range of the experimental group members was fifteen to forty-nine years. This group was composed of nine males and eleven females. The age range of the control group was seventeen to forty-six years. There were eighteen females and two males in this group. No attempt was made to




18
match individuals in the two groups as to age and sex, because all of the subjects had met the criterion regarding physical maturation, and optimum growth had been accompli shed.
Procedure for Recording and Analysis of Speech Samples
In the experimental group each person made a tape recording as he articulated a list of Spanish words. The list, which follows, was constructed to contain the consonants and vowels of the language of the subjects in all of the intraword positions in which they occur. Those words
selected were simple, everyday words which were representative of the common language usage among the subjects. Hence, the pronunciation of these words was probably highly stabilized in each subject, and unlikely to be altered by
another speaker's pronunciation.
Sound Words
Consonants: [p] pollo, copa
[t] taza, gato
[k] carro, vaca
[g] gato, guagua
[B] bote, libro
[d] diente, dedo
[If] cafe, funda
[m)] mesa, cama
(n] nene, mano, jabon
[ts] china, leche
[N] nilo, pifa
[a] silla, taza, ojos
[1] leche, bola, arbol
[r] carnet, dolar
[rr] rosa, carro




19
Sound Words
Consonants
(Continued): [y] llave, polo
[x] jarro, angel
Vowels: [a] agua, casa, pata
[e] eje, pesa, nene
[i] hilo, maguey, tina
[o] ojo, sopa, huevo
[u] uno, cuna
The following consonants are not the same as those used in English, and require explanation: [B], bilabial fricative; [d], voiced dental stop or fricative; [t], voiceless dental stop; [n], dental nasal; [N], palatal nasal; (1], dental lateral; [r], single tap lingua-alveolar; [rr], trilled lingua-alveolar; [y], voiced palatal fricative; and [x], voiceless velar fricative. None of the vowels are the same as those used in English. They represent approximately the same vowels that appear in Latin, in that they are pure and not diphthongized. There is a tendency for all of the vowels except [a] to be diphthongized in English.
Each subject was instructed to read the words in the order in which they appeared in the list. In several instances the subjects were unable to read and were asked to repeat the words after they were spoken to them by a Puerto Rican nurse. The control group subjects were recorded in
the same manner.
The recordings were made on a Wollensak tape
recorder; Model T 1500, high fidelity, dual speed. The instrument's portability was an important consideration in




20
its selection. Although the recording room conditions were far from ideal, the ambient noise level was kept at a
minimum.
Twenty recordings of the experimental group members and twelve recordings of the normal subjects were arranged in random order for analysis. The author and another speech pathologist acted as judges to determine phonetic accuracy of consonants and vowels, and quality differences as reflected in deviations from normal cro-nasal resonance. During the process of analysis, It soon became apparent that even the most nearly normal cleft palate speakers would be immediately distinguishable. Further attempts to disguise the identification of individual subjects were therefore terminated, and the judges made use of the control group to provide standards against which to rate the speech of the cleft palate individuals. To assure as equal as possible listening conditions, the two judges wore earphones of the same make and model and made simultaneous analyses. Thorough practice sessions were utilized over a four-week period prior to making the final analyses reported herein. During this time the judges examined each other's ratings at length, in an effort to control the variables being considered.
Resonance was rated on a five-point scale, with 1 representing normal cro-nasal resonance and 5 representing the greatest deviation from normal oro-nasal resonance.




21
A numerical score of articulation was attained by assigning values to the articulated sounds: 0 indicated correct production; 1 indicated a distortion; and 2 indicated either a substitution or an omission. In addition, the judges compared their analysis sheets in order to determine which consonants and vowels were found to be most frequently defective. Several deviations in articulation were noticed which, on the basis of the speech samples of the normal subjects, were apparently due to dialect differences. Such
deviations were not considered to be errors.
Apparatus and Procedure for Obtaining X-ray Plate!
Control of head Position
In order to minimize movement of the head during the X-ray exposures, a head positioner was utilized. The
device used in 1958 was designed by Howell (23) and was constructed at the University of Florida. This head positioner was not available in 1957, but a similar one was fashioned from photographer's equipment formerly used in portrait photography. One of the criteria used in the designing of the head positioner was that it should be portable without sacrificing sturdiness. This makes practicable the use of such a positioner in a field study. The head positioner utilizes ear rods, which insert into the external auditory meati, to control movement in the sagittal plane. An




22
armature which rests against the nasion of the subject is used in an attempt to prevent rotation of the head position around the axis created by the ear rods.
In addition to minimizing head movement, the head
positioner provides another advantage, that of aiding in the definition of the Frankfort line on the X-ray films. As in the studies by Buck (3) and Hixon (20), each subject was seated in a comfortable position in order not to disturb the habitual carriage of the oral and pharyngeal structures. The relative position of the chair with respect to the head positioner was adjusted to make possible this comfortable posture.
Marking of Structures
None of the structures were marked in the present study. The rationale behind this lack of marking may be attributed in part to Buck (3). The experimental group in his study had the midline of the tongue, the midline of the velum and hard palate, and the midline of the posterior
pharyngeal wall covering Passavant's pad marked with a mixture of barium and gum acacia. Severe gagging on the part of the control group members prevented marking of the midline of the posterior pharyngeal wall. With this exception, similar markings were applied to the control group members. In an effort to discover the possible error resulting from a lack of marking of structures, Buck conducted an additional




23
experiment in connection with his study. Eight normally structured individuals, who did not present severe gag reflexes, were X-rayed with and without the marking of structures. A comparison of measurements obtained from these X-rays revealed an average difference between the two sets of measurements of plus or minus 0.6 millimeters. X-ray Exposures Made
Four lateral cephalometric plates were acquired
on both the experimental and control groups: one at physiologic rest and one each while phonating the vowels [a], [i], and (u]. These vowels were selected because they are representative of extremes of tongue positions used in phonation, The low front vowel [ae], used in some X-ray studies, is not present in the language of the subjects of the present study. The fifth X-ray exposure was an antero-posterior view used to obtain lateral measurements of the skeletal structures of the face.
It was not possible to secure all five of the X-ray plates on every subject. For the experimental group there were fifteen antero-posterior plates, and of the lateral views: eighteen of the rest position, twenty of [a], eighteen of [i], and nineteen of [u]. Twelve rest exposures, twenty exposures each of the phonation of the three vowels, and twelve antero-posterior exposures were obtained for the control group. The remaining plates were not utilized due




24
to faulty exposure or developing techniques.
In addition, it must be remarked that of the one
hundred and seventy-four plates available for this study, many were unavoidably over-exposed or underdeveloped, obliterating much of the soft tissue. This condition was a result of variations in electric output, and in some instances the equipment appeared to be faulty. Various colored gelatins were used to clarify the structures illustrated by the under-developed or over-exposed plates.
Straw, amber, pink, and blue gelatins, or a combination of any two of these, were tried experimentally in viewing the plates. It was found that frequent changing of the gelatins was necessary in order to obtain the best results. This procedure tended to rectify the faulty exposure or development.
The accuracy of the tracings was checked in
various ways. Five subjects, randomly selected, were completely traced two times, and the tracings were tested by superposition of one upon the other. At various times during the tracing process the researcher and his supervisory chairman compared random tracings against the films from which they were drawn. In addition, all the tracings from the lateral exposures of each subject were superimposed to assure internal consistency in the outlining of the bony structures. Where discrepancies were found, the films were




25
re-examined and the faulty tracing was corrected. The small number of corrections necessary indicated a high degree of accuracy in the original tracings* Control of Vowel Phonations, Intensity, and Pitch
The subjects in each group were instructed to phonate the vowels indicated in the preceding section. Whenever possible, a practice period was held prior to the X-ray exposures. For the most part, the subjects had little difficulty producing good examples of these vowels.
There is some evidence in the literature to indicate that the positioning and movement of the oro-pharyngeal structures may be affected by vocal pitch and intensity (8, 21). Buck (3), however, reported that these two variables produced relatively minor effects on the positioning of the oral structures during vowel phonation, as long as the variations in pitch and intensity were not large. The subjects in the present study were instructed to maintain comfortable pitch and intensity levels during phonation when the X-ray exposures were being made. No instrumental
control was available to determine maintenance of the criteria.
Measurement of the X-ray Plates
The procedure for securing measurements from the X-ray plates was basically that used by Hixon (20) and




26
Buck (3). Each plate was placed on an X-ray tracing table especially constructed for this study. Flashed-opal frosted glass and varying sources of light of fifteen, twenty-five, sixty, and one hundred watts were used. A clear acetate tracing paper and a hard-lead drawing pencil were used to trace the outlines of the structures. The sella turcica, the nasal bone, and that portion of the frontal bone just superior to the nasion were traced to be used as orientation lines when comparing different lateral exposures of the same individual. The outline of the hard and soft palates, the anterior nasal spine, the central incisors, the symphysis of the mandible, the posterior wall of the pharynx, the tubercle of the Atlas, the midline of the tongue, and the lips were traced. To construct the Frankfort line, the inferior margin of the bony orbit and the outline of the ear rod were traced.
The Frankfort line was used as a baee line for the contruction of all perpendiculars and parallels constructed for the various measurements of the study. The incisal reference line was drawn tangent to the superior margin of the central incisors of the mandible. This reference line intersected the posterior pharyngeal wall and was parallel to the Frankfort line. When the superior surface of the ear rod was obscured by the armature of the head positioner on some of the plates, it was necessary to




27
reconstruct the actual dimension of the ear rod on the
tracings. In a few instances, the tubercle of the Atlas was also partially or totally obscured by this armature. Whenever either of these effects was found, the obscured parts were constructed, by reference to the visible parts and by reference to the dimensions of unobscured ear rods and tubercles of the Atlas. Since the ear rods were of a standard eight millimeters in diameter, they were relatively simple to construct. The tubercle of the Atlas can be expected to vary in size according to a normal distribution. Therefore, ten of these structures which were clearly defined were selected at random and measured. The means of their heights and widths were taken, and those tubercles of the Atlas which were constructed, were constructed according to these means. Measurements on those subjects for whom the tubercle of the Atlas had to be constructed will be discussed in detail in a later section.
Hiagerty, et al. (15), used a series of lines on
the posterior pharyngeal wall to determine tissue movement at the level of the tubercle of the Atlas. The modification of this technique used by Knobeloch (24) was also utilized in the present study, I.e., a perpendicular line was constructed tangent to the most anterior projection of the tubercle of the Atlas. This line was extended to intersect the incisal reference line and served as the




28
posterior border of the area, A horizontal line, level 4, passed through the most anterior projection of the Atlas tubercle. At intervals of one-half centimeter, other horizontal lines were drawn superiorly and inferiorly to this level. The remaining lines were numbered in descending order, level 1 being the most superior line.
Figure 1 is a master tracing of an X-ray plate of a normal subject. This figure shows the reference points and location of measurements used for all of the tracings in the study. The two reference lines already discussed are designated as follows-, the Frankfort line, AfC', extending from the inferior margin of the bony orbit to the superior surface of the ear rod; and the incisal reference, LF, extending from the mandibular incisors to the posterior pharyngeal wall.
Measurements for the present study will be classified as follows: linear measurements, measurements of openings, area measurements, measurements of movement from rest to phonation, and skeletal measurements of the face. They will be discussed in that order, with reference to Figure 1.
Linear Measurements
The linear measurements that were made along the various lines include the following distances.- FG, the distance from the termination of the hard palate to the




29
I- tEAR ROD
W-ARD
PALATE
<--------- T -. ,M',TUBE P C t C5Ug5PINAUg Vr LU M OF cTH
7 ( ATLAS
!UPPfE( /
\ LJP \ \
I2'
I1
,Y01O BONE
Figure 1. Master tracing of a typical lateral X-ray plate, showing
reference points and locatio of measurements
A'C' Frankfort line LP Incisal reference line FG Termination of the hard palate to the posterior pharyngeal wall LF Oral cavity length
MO Tongue length on the incisal reference line BI Frankfort line to the high point of the tongue HI Hard palate to the high point of the tongue IN Tongue height above the incisal reference line LN Horizontal position of the high point of the tongue OP Pharyngeal opening on the incisal reference line
1-6 Levels used for obtaining measurements of movement of the
posterior pharyngeal wall




30
posterior pharyngeal wall; LP, the length of the oral cavity; MO, the length of the tongue on the incisal reference line; BI, the distance between the Frankfort line and the high point of the tongue; HI, the distance between the hard palate and the high point of the tongue; IN, tongue height aboye the incisal reference line; and LN, the horizontal position of the high point of the tongue in the oral cavity. Each linear measurement was made twice. In case of a discrepancy, the measurement was rechecked and recorded. Measurements of Openings
Measurements of openings are as follows: OP, pharyngeal opening on the incisal reference line; lip opening, the smallest opening between the lips; and the smallest pharyngeal opening, the smallest distance between the posterior border of the tongue and the posterior pharyngeal wall (above the body of the hyoid bone and below
the incisal reference line). Velo-pharyngeal opening is the smallest distance between the posterior surface of the velum and the posterior pharyngeal wall. Incisor opening
is considered to be the distance between the parallel lines drawn tangent to the maxillary and mandibular incisors. In an effort to substantiate this measurement, the distance BN,
extending from the Frankfort line to the incisal reference line, was also measured. The procedure used to check linear measurements was employed in the case of measurement of openings.




31
Area Measurements
Area measurements for the study were made by means of a planimeter. The three areas measured were the total tongue area above the incisal reference line, the tongue area anterior to the high point of the tongue, and the area of the posterior pharyngeal wall. Each area was traced three times and the average of the three measurements was recorded. This precaution was taken in an effort to control variations occurring while the areas were measured. The planimeter is a sensitive instrument, and it was difficult to trace the outline of an area without varying from the lines.
Measurements of Movement from Rest to Phonation
Measurements of movement from rest to phonation were made along the parallel and horizontal lines constructed in reference to the tongue and the posterior pharyngeal wall. The vertical movement of the high point of the tongue was made on line 3N, the distance IN being variable. Horizontal movement of the high point of the tongue was made on the incisal reference line, the variable distance being LN* Movement of the posterior pharyngeal wall was made on the lines constructed in the area of the tubercle of the Atlas. Unfortunately, the same area demarcated by the most superior and the most inferior of these lines was not visible on all of the lateral X-ray plates.




32
Skeletal Measurements of the Pace
JKO the length of the miaxilla was measured on the inoisal reference line. This measurement is the distance between perpendicular lines dropped tangent to the notch just inferior to the anterior nasal spine (the subspinale), and to the termination of the hard palate. LP', the length of the mandible, was also measured on the Incisal reference line. This is the distance between the mandibular incisors
and the perpendicular dropped tangent to the most anterior projection of the tubercle of the Atlas (line CF'), a measurement between stationary bony structures at rest.
Figure 2 is a master tracing of a typical anteroposterior X-ray plate. This view was used in the measurement of the width of the maxilla and the width of the mandible. These plates were not traced, but were placed on the illuminated tracing table, and the measurements were taken directly from them. There were two reasons for making these measurements directly from the X-ray plates: (1) since only two dimensions were required from them this was the simplest method of measurement, and (2) these dimensions were to be used as temporary estimates of maxillary width, until such time as the plaster casts made of the subjects should become available. Figure 2 is included only to remove any confusion as to what measurements were taken. With reference to Figure 2, the width of the maxilla was measured between




33
/ ...... :L .- =;,'W \.-.,''-MA X It( A
MAN 1)! L
// ,
Figure 2. Nster tracing of a typical antero-posterior X-ray
plate, indicating the structures and the points from
which all antero-posterior measurements were taken
XX' Width of the maxilla at the level of the horizontal plates YY' Width of the mandible at the angle of the ramus




34
points X and Xt. This represents the widest extent of its bony margin at the level of the horizontal plates of the maxilla. The mandibular width was taken as the distance between Y, the angle of the ramus on the left, and Y', the angle of the ramus on the right.
The accuracy of these skeletal measurements was
checked in the same way as were the linear measurements of the soft tissues. Each measurement was taken twice, and the two measurements were compared. When a discrepancy appeared the dimension was rechecked and correctly recorded.




CHAPTER III
RESULTS
The basic purpose of this study is to make a comparison between a group of normal subjects and an experimental group having incomplete, unrepaired cleft palates. A comparison will also be made between the experimental subjects of this study and the subjects with complete, unrepaired cleft palates reported by Knobeloch (24).3
Facial and Oral Cavity Dimensions
The measurement of the bony structures of the face and the oral cavity is an important concern of this study. Graber (12), Krogman (25), Slaughter and Brodie (35), and Buck (4) have all presented evidence which tends to indicate that early surgical repair of cleft palates may retard normal facial growth. The first section of this chapter will present a comparison of facial and oral cavity dimensions between the unrepaired cleft palate group and the normally structured control group. Other sections will
3Throughout tne remainder of this study it is to
be understood that the comparisons made with the measurements concerning complete, unrepaired clefts were those obtained from the analysis conducted by Knobeloch.
35




36
present data from measurements in the following order: tongue size, horizontal tongue carriage, pharyngeal openings, vertical tongue carriage, tongue movement, velo-pharyngeal opening, posterior pharyngeal wall measurements, and lip and jaw opening.
The statistics used in the tables in this chapter
include: (a) the arithmetic mean, (b) the standard deviation, and (c) the standard error of the mean. Also shown is the difference between the means for the two groups for each measurement, the standard error of this difference, and the critical ratio or "t" statistic. The values are given for "t" required for statistical significance at the 5.0 per cent and at the 1.0 per cent levels of confidence. The format for this presentation is t (df 38): l.0o' 2.711; 5.03 2.025. In this context df designates the degrees of freedom for a particular set of measurements. Differences below the 5.0 per cent level of confidence might be largely due to chance, and were not considered.
Maxilla Length and Width
Table 1 presents data related to the skeletal
measurements of the face. The mean length of the maxilla for the experimental group of this study was shorter than that for the control group by 5.04 millimeters, a difference which was significant beyond the 5.0 per cent level




37
TABLE 1
MEASUREMENTS OF MAXILLARY, MANDIBULAR, AND ORAL CAVITY STRUCTURES
AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES (All measurements in millimeters)
Normal Cleft Palate Differences
1. Maxilla length M 50.08 M 45.04 M 5.04
SD 5.02 SD 5.62 SE 2.01
SE 1.51 SE 1.33 t 2.507*
t (df 29): 1.0% 2.756; 5.0% 2.045
2. Maxilla width M 63.58 M 68.50 M 4.92
SD 5.94 SD 5.07 SE 2.24
SE 1.79 SE 1.36 t 2.196*
t (df 25): 1.0% 2.787; 5.0% 2.060
3. Mandible length M 80.92 M 77.19 M 3.73
SD 7.55 SD 7.98 SE 2.99
SE 2.27 SE 1.94 t 1.247
t (df 28): 1.0% 2.763; 5.0% 2.048
4. Mandible width M 97.17 M 102.43 M 5.26
SD 3.07 SD 7.69 SE 2.25
SE .92 SE 2.06 t 2.338*
t (df 25)
5. Oral cavity length M 83.66 M 73.72 M 4.94
on incisal reference SD 7.10 SD 6.10 SE 2.60
line SE 2.14 SE 1.48 t 1.900
t (df 28)
6. Termination of M 27.83 M 28.00 M .17
hard palate to SD 3.90 SD 4.86 SE 1.65
posterior phar- SE 1.17 SE 1.78 t .103
yngeal wall t (df 28)
*Significant beyond the 5.0% level of confidence




38
of confidence. That Is, there were fewer than five possibilities out of one hundred that this difference was due to chance. Graber (12) and Buck (4) found short maxillae among their subjects with repaired cleft palates. They felt this affect might be a result of early surgery. It should be noted, however, that the maxillary development of the present experimental group does not appear to be like that of the repaired cleft palate subjects reported by Graber (12). He noted a concave facial angle (from nasion to subspinale to gnathion) which resulted from a generally restricted maxilla. In contrast, the experimental group of the present study appeared to have flat facial angles like normal subjects. This would indicate that the shorter maxilla length of the group was probably not a result of a generally restricted development, but more likely a failure in posterior growth.
For the cleft palate group, the mean maxillary
width was 4.92 millimeters greater than that for the normal group, a difference which was significant beyond the 5.0 per cent level of confidence. Again, this indicates that there are fewer than five possibilities out of one hundred
that this difference is due to chance. This is in direct contrast to the smaller maxillary width measurements found among those with repaired cleft palates (12, 4). It appears to support Krogman's (25) statement to the effect that unrepaired cleft structures can be expected to follow




39
normal growth patterns, but modified by the anomaly. In other words, a hard palate with an unrepaired cleft may tend to grow to wider than normal dimensions. When the measurements of the incomplete, unrepaired cleft palates were compared with those of the complete, unrepaired cleft palates, there were no statistically significant differences. It was mentioned in Chapter 11 that one of the subjects in the experimental group had undergone surgical repair of the palate. She (Subject No. 206) was included because the surgical suture had pulled apart, indicating that the dynamic processes of facial growth had probably not been retarded by the operation. The measurements of maxilla width support this assumption. For the experimental group the mean maxilla width was 68.50 millimeters, with a standard deviation of 5.07. The maxilla of this subject was 67.0 millimeters, placing her in the mid-range of the unrepaired cleft palate group.
Mandible Length and Width
The subjects of the experimental group did not
differ significantly from the control group in mandibular length. This was also true of the subjects having complete, unrepaired palatal clefts. In contrast, the repaired cleft lip and palate subjects studied by Buck (4) were found to have significantly shorter mandibles. This difference may have been due to restricting pressure from the repaired




40
obicularis oris muscle which encircles the mouth (12).
In the measurement of mandibular width, the experimental group followed the trend seen in their maxillary width. The mandibles of the cleft palate group were wider, on the average, than those of the normal group by 5.26 millimeters, a difference which is significant beyond the 5.0 per cent level of confidence. Such measurement comparisons may be seen in Table 1, item 4. A similar, though
not statistically significant, trend was found among the mandible widths of the complete, unrepaired cleft palate subjects. It is a question whether the mandibles of the cleft palate subjects were wider due to mechanical pressures from the tongue and certain facial muscles (37), or due to neurological changes which may have been imposed by the cleft.
Oral Cavity Length
The oral cavity length is defined here as the distance from the mandibular incisors to the posterior pharyngeal wall, measured along the incisal reference line. This dimension would be directly affected by the mandible length. Among the cleft palate subjects of this study, mandible length was not markedly different from that of the normal group. As expected, then, the measurement of oral cavity length revealed no significant difference between the normal and the experimental group. This had also been true




41
of the complete, unrepaired cleft palate group.
Antero-_osterior Measurements of the Naso- harynx
The distance measured horizontally from the termination of the hard palate to the posterior pharyngeal wall was used as the antero-posterior dimension of the nasopharynx. For the two groups of the present study this dimension is strikingly similar. This similarity is especially remarkable when it is remembered that the cleft palate subjects had a significantly shorter maxilla. It might have been expected, then, that among the cleft palate subjects this measurement of the naso-pharynx would have been significantly larger. It is possible that the shorter maxillae provided a more anterior attachment for the palatopharyngeus and superior constrictor muscles. This would have displaced the posterior pharyngeal wall anteriorly. Thus, a nasopharyngeal opening of normal proportions might be maintained. There were no significant differences on this dimension between the experimental group and the group with complete, unrepaired clefts.
Measurements Related to Tongue Size and Horizontal
Position of the High Point of the Tongue
Two measurements of midsagittal tongue area were
made with a planimeter: (1) the total tongue area above the incisal reference line and (2) that part of the tongue area anterior to the high point of the tongue. As shown in Table 2,




42
TABLE 2
MIDSAGITTAL TONGUE AREA AND HORIZONTAL POSITION OF HIGH POINT OF
TONGUE ALONG INCISAL REFERENCE LINE AMONG SUBJECTS WITH NORMAL
DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES
Rest [ a] [i] [ u]
I. Total tongue Normal
area above M 11.02 10.78 9.28 9.26
incisal refer- SD 2.43 3.63 1.76 2.94
ence line; in SE .73 .83 .40 .67
square centi- Cleft Palate
meters M 10.01 12.79 8.64 10.96
SD 4.29 3.68 2.93 3.03
SE 1.04 .84 .71 .71
Differences
M 1.01 2.01 .64 1.70
SE 1.27 1.19 .81 .98
t .795 1.689 .790 1.735
2. Tongue area Normal
anterior to M 6.34 6.02 4.52 4.94
high point of SD 1.75 2.22 1.06 1.69
tongue; in SE .53 .51 .24 .39
square centi- Cleft Palate
meters M 5.41 7.15 4.11 6.11
SD 2.37 2.41 1.63 1.90
SE .58 .55 .40 .45
Differences
M .93 1.13 .41 1.17
SE 1.89 .75 .47 .59
t .492 1.507 .872 1.983
Rest t (df 28): 1.04 2.763; 5.01 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 Li] t (dr 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.718; 5.0% 2.027




43
TABLE 2 Continued
Rest [a) Li] [u]
3. Horizontal Normal
distance from M 43.16 48.30 32.65 49.15
high point of SD 4.76 4.81 4.52 4.48
tongue to man- SE 1.43 1.10 1.04 1.03
dibular central Cleft Palate
incisors; in M 38.94 49.43 28.72 52.55
millimeters SD 5.25 5.83 5.02 6.24
SE 1.27 1.34 1.22 1.47
Differences
M 4.22 1.13 3.93 3.40
SE 1.92 1.73 1.60 1.79
t 2.198* .653 2.456* 1.899
4. Tongue length Normal
on incisal M 66.25 59.20 52.90 49.50
reference SD 6.17 7.92 6.28 6.94
line; in SE 1.86 1.82 1.44 1.59
millimeters Cleft Palate
M 62.55 56.30 55.17 49.82
SD 8.29 8.18 6.15 7.22
SE 2.01 1.88 1o49 1.70
Differences
M 3.70 2.90 2.27 .32
SE 2.74 2.61 2.08 2.33
t 1.350 1.111 1.091 .137
Rest t (df 28): 1.0% 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 Li] t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.718; 5.0% 2.027
*Significant beyond the 5.0% level of confidence




44
there were no statistically significant differences found between the control group and the experimental group regarding measurements of midsagittal tongue area. This result is essentially the same as reported for the complete, unrepaired cleft palate group. As Knobeloch (24) pointed out, this is in contrast to the smaller midsagittal tongue areas found by Hixon (20) for normally structured nasal
speakers, and by Buck (4) for repaired cleft palate speakers. Tongue Length and Horizontal Position
Table 2 further shows that comparisons of the
linear measurements of tongue length on the incisal reference line between the control group and the experimental group revealed no statistically significant differences. However, in the horizontal carriage of the tongue it was found that the experimental subjects, like the subjects in the complete, unrepaired cleft palate group, carried their tongues more anteriorly at the rest position and during the vowel (i]. Both of these differences were significant beyond the 5.0 per cent level of confidence. Also, like the complete, unrepaired cleft group, the cleft palate subjects of the present study exhibited a slight tendency toward a more posterior tongue position during the vowel (u]. The horizontal position of the tongue during the vowel [a] was quite similar in the experimental and the control groups of the present study. The incomplete, unrepaired cleft palate




45
group and the complete, unrepaired cleft palate group were similar in measurements of horizontal tongue carriage, tongue length, and midsagittal tongue area. Pharyngeal Openings
The mideagittal dimension of the oro-pharynx is a further indication of the horizontal carriage of the tongue at rest and during phonation. This can be understood when it is realized that the oro-pharyngeal opening is defined here as the distance between the posterior pharyngeal wall and the posterior border of the tongue. Thus, it would be expected that as the tongue moves backward this opening
would grow smaller, and conversely, when the tongue moves forward this opening would grow larger. Table 3 gives two measurements of the oro-pharynx: the first is taken along the incisal reference line; the second is taken between the incisal reference line and the superior border of the body of the hyoid bone.
For both measurements of oro-pharyngeal openings the experimental subjects exhibited smaller dimensions on the vowel [u] than did the normal subjects. For this vowel the two measurements were actually identical among the experimental subjects, because the smallest opening of the oro-pharynx was never below the incisal reference line. The mean differences between the experimental group and the control group were approximately four and one-half




46
TABLE 3
THE PHARYNGEAL OPENINGS ON AND BELOW THE INCISAL REFERENCE LINE AMONG
SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES
(All measurements in millimeters)
Rest [a] Li] Lu]
1. Pharyngeal open- Normal
ing between the M 11.33 8.35 22.10 12.20
tongue and the SD 3.73 3.03 4.54 4.24
posterior wall, SE 1.12 .70 1.04 .97
on the incisal Cleft Palate
reference line M 11.11 7.90 22.72 7.76
SD 3.90 3.85 4.79 5.62
SE .95 .88 1.16 1.32
Differences
M .22 .45 .62 4.4
SE 1.47 1.12 1.56 1.64
t .150 .402 .397 2.707*
2. Smallest phar- Normal
yngeal opening M 8.41 7.15 19.90 12.05
between the SD 4.58 3.20 5.21 4.25
incisal refer- SE 1.38 .73 1.20 .97
ence line and Cleft Palate
the superior M 8.11 7.85 16.58 7.76
border of the SD 3.12 3.89 4.93 5.62
body of the SE .76 .89 1.20 1.32
hyoid bone Differences
M .30 .70 3.32 4.29
SE 1.50 1.16 1.69 1.65
t .200 .603 1.964 2.600*
Rest t (df 28): 1.0% 2.763; 5.0% 2,048 La] t (df 38): 1.0% 2.711; 5.0% 2.025 [ i] t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.718; 5.0% 2.027
*Significant beyond the 5.0% level of confidence




47
millimeters. In both instances this difference was significant beyond the 5.0 per cent level of confidence. No
other significant differences were found among the measurements presented in this table. The control and the experimental groups had very similar oro-pharyngeal openings for both the rest and the (a] positions. This appears to confirm a part of the data discussed in the previous section, that the experimental group tended to maintain a more posterior placement of the tongue during the vowel [u], as well as a placement similar to that of the normal group during the vowel [a].
Like the experimental subjects of the present study, the complete, unrepaired cleft palate subjects tended to
have smaller oro-pharyngeal openings for the vowel [u) than did the control group, although the tendency was not great enough to reach statistical significance. For the vowel (i], these subjects had a significantly larger oro-pharyngeal
opening than did the control group.
It is interesting to compare the measurements of oro-pharyngeal opening of the two unrepaired cleft palate
groups during the phonation of [i]. In examining the inferior pharyngeal opening measurements, a difference significant beyond the 5.0 per cent level of confidence was found between the two cleft palate groups. The incomplete cleft palate group was more like the normal group than like the group with




48
complete, unrepaired clefts. There appears, then, for the first time a real difference between the unrepaired cleft palate groups. They differ from the mean of the controls in opposite directions. An apparent incongruity may be observed here, since it was previously noted that the high point of the tongue was significantly farther forward among the incomplete cleft palate subjects on the vowel [i]. It would be expected that the pharyngeal opening for this vowel might be wider rather than narrower. The probable explanation is that during the phonation of (i], the subjects with incomplete clefts characteristically maintained two concentrations of tongue tissue. There was an anterior concentration beneath the hard palate, and a posterior concentration projecting horizontally at the tongue base. Inspection of the X-ray tracings supports this statistical interpretation. It might be speculated that the function of this particular tongue shape was an involuntary result of anatomical and linguistic adjustments.
It must be stated that conclusions drawn from these measurements can only be tentative. Mention has already been made of the poor definition in some of the X-ray plates. At times the tracings could only be a compromise among two or more possibilities. It appears, however, that the tongue carriage cannot be described by a single measurement. For
the vowel [a] no significant mean difference in tongue




49
placement appeared between the groups in either Table 2 or
Table 3. Finally, there was a definite tendency toward a more posterior tongue carriage among the experimental group for the vowel (u].
Vertical Tongue Position
In order to adequately describe the vertical position of the tongue at rest and during vowel phonation.. the high point of the tongue was related to three different lines. First, a measurement was made of the maximum tongue height above the incisal reference line. Second, the distance was measured between the high point of the tongue and the inferior surface of the hard palate. Third, the distance from the Frankfort line to the high point of the tongue was measured, These three dimensions were expected to yield relatable, though not identical, information. Tongue Heighit Above the Incisa. Reference Line
As seen in item 1 of Table 4, there was a tendency for thie incomplete cleft palate group to display somewhat lower than normal tongue positions at rest and during the production of the vowel [ii, and somewhat higher than normal tongue positions during the production of vowels [a] and (uJ. These measurements for item 1 are measured vertically above the incisal reference line. The mean differences are significant beyond the 1.0 per cent level of confidence for the




50
TABLE 4
VARIOUS VERTICAL MEASUREMENTS OF TONGUE CARRIAGE AMONG SUBJECTS WITH
NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES
(All measurements in millimeters)
Rest [La] [i] [u ]
1. Distance of Normal
high point of M 23.00 24.80 23.80 24.70
tongue above SD 3.13 5.78 2.98 5.14
the incisal SE .94 1.33 .68 1.18
reference line Cleft Palate
M 21.58 30,48 21.61 27.63
SD 7.30 7.06 5.06 6.06
SE 1.77 1.62 1.23 1.43
Differences
M 1.42 5.68 2.19 2.93
SE 2.03 2.09 14O0 1.85
t 700 2.718** 1.564 1.584
2. Distance of Normal
high point of M 6.66 12.85 3.60 5.05
tongue below SD 2.77 3.18 1.57 2.87
the hard SE .83 .73 .36 .76
palate Cleft Palate
M 9.47 15.40 11.50 9.08
SD 5o18 5.70 5.12 5.24
SE 1.26 1.31 1.24 1.24
Differences
M 2.81 2.55 7.90 4.03
SE 1.51 1.50 1.29 1.40
t 1.861 1.700 6.124** 3.474**
Rest t (df 28): 1.04 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 [i] t (df 36): 1.0% 2.718; 5.0% 2.027 lu] t (df 37): 1.0% 2.718; 5.0% 2.027




51
TABLE 4 Continued
Rest [a] [i] [u]
3. Distance froa Normal
the Frankfort M 30.08 41.05 31.45 33.40
line to high SD 3.40 5.61 3.97 5.10
point of SE 1.02 1.29 .91 1.17
tongue Cleft Palate
M 34.39 39.78 37.00 33.76
SD 5.46 6.01 6.02 5.71
SE 1.32 1.38 1.46 1.35
Differences
M 4.31 1.27 5.55 .36
SE 1.68 1.88 1.72 1.78
t 2.566* .148 3.099** .202
Rest t (df 28): 1.0% 2.763; 5.0% 2.048 (a] t (df 38): 1.0% 2.711; 5.0% 2.025 [i] t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.718; 5.0% 2.027
*Significant beyond the 5.0% level of confidence
**Significant beyond the 1.0% level of confidence




52
vowel (a]. Such results are in essential agreement with those reported for the complete cleft palate group. It is interesting to note here that Hixon's (20) nasal subjects
with normal structures had a tendency toward greater tongue height. In addition, the repaired cleft palate subjects studied by Buck (4) maintained a lower tongue than did the normal subjects with whom they were compared. Although all
of these experimental groups may possess nasal voices, the nasality apparently cannot be explained solely by tongue usage without a careful description of the oral cavity. Distance of the High Point of the Tongue Below the Palate
Item 2 of Table 4 represents the distance of the
high point of the tongue below the hard palate. This probably should not be accepted as a reliable measure of absolute tongue height. Towards the front of the mouth, the curve of the palate as it extends posteriorly would be expected to complicate the measurement of tongue height. And for the back vowels, [a] and [u], the high point of the tongue was frequently behind the termination of the hard palate. The distance measured for (a] and [u] was an
approximation therefore acquired by extending the line of the inferior palatal surface. This dimension is, however, useful as a statement of the relationship between palatolingual tissues. The present experimental group, as well as those with complete, unrepaired clefts, maintained a




53
wider separation between tongue and palate than did the control group at rest and during all three vowels. For the experimental group, the difference from the control group was significant for the vowels [i] and Lu) beyond the 1.0 per cent level of confidence. It is interesting to note that Buck (4) reported similar findings in his study of subjects with repaired cleft palates. At rest, and for the vowels [as] and Lu], his subjects also maintained a greater than normal separation between the tongue and the hard palate. Distance from the Frankfort Line to the Higha Point of the Tongue
Tongue height may also be measured by taking the distance between the Frankfort line and the high point of the tongue. This is perhaps the most reliable of the three dimen3ions outlined In Table 4. The Frankfort line does not move up and down as does the incisal reference line and it is not curved as is the hard palate. Measurements of this dimension indicate that the cleft palate group of this study maintained a greater than normal distance between the Frankfort line and the tongue at rest and during the vowel [i]. As will be
demonstrated in a later section, there was a slight tendency for this group to maintain a wider jaw opening for all positions studied. The effect of this wider jaw opening on the measurement of tongue height 'was offset by the tongue's being raised for (a] and [u], and increased by the tonguets being held lower for rest and [i], as shown in Table 4. item 1.




54
Thus it may be seen that the measurement of tongue height is not a simple one. It appears to be the confounding factor of jaw opening which accounts for the discrepancies between
item 1 and item 3 of Table 4.
The two cleft palate groups were also compared on
measurements of tongue height. During phonation of the vowel [u), a difference significant beyond the 5.0 per cent level of confidence was noted between the cleft palate groups. The subjects with incomplete clefts tended to maintain a greater distance between their tongues and their hard palates. For the vowel [i] the experimental group, with incomplete clefts, maintained approximately five and one-half millimeters greater separation between their tongues and their hard palates than did the members of the group with complete clefts. This difference was significant beyond the 1.0 per cent level of confidence. This supports the observation made previously that the members of the experimental group tended to have both an anterior and a posterior concentration of tongue tissue during the production of (i]. It seems that because of the tissue concentration formed near the base of the tongue, less tissue would be available to form a tongue elevation beneath the hard palate. Hence, the high point of the tongue was lower for the incomplete cleft palate group*
Similarly, the distance between the Frankfort line




55
and the high point of the tongue was greater among the experimental group than among the group with complete, unrepaired clefts for the vowel [i]. This difference was significant beyond the 5.0 per cent level of confidence. No
other statistically significant differences were found between these groups on these measurements.
Excursion of the Tongue from Rest to Vowel Positions
The measurements of tongue movement, shown in
Table 5, were secured by computing the means of the movement from rest to phonation for each subject. It will become immediately apparent upon examination of the table that relatively large standard deviations have been computed for
the means. Buok (4) had observed a similar effect in his measurements of tongue movements. Hie felt that the measurements were complicated by variations in the tongue positions
at rest and by variable jaw positions during the production of vowels. The same complicating factors would be expected to increase the range of values for the measurements of both vertical and horizontal tongue movement. That is, these values vary in direction as well as in magnitude. Vertical Towu Movement
Item 1 of Table 5 indicates that the two groups in the present study varied little in the extent of vertical tongue movement from rest to the vowel [i]. For the vowels




56
TABLE 5
MOVEMENT OF THE TONGUE FROM REST TO VOWEL POSITIONS AMONG
SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES (Negative numbers indicate a downward direction to vertical movement
and a posterior direction to horizontal movement. All measurements
in millimeters)
La] (i) Lu]
1. Linear Normal
measure- M 2.16 1.00 2.16
ments of SD 5.68 3.16 3.88
vertical Range -5.0 to 13.0 -6.0 to 6.0 -5.0 to 10.0
movement SE 1.71 .95 1.17
of the Cleft Palate
tongue M 7M83 -.62 5.65
SD 6,28 6.60 6.55
Range -3,0 to 22.0 -15.5 to 10.5 -6.0 to 17.5
SE 1.52 1.70 1.64
Differences
M 5.67 1.62 3.49
SE 2.29 1.95 2.01
t 2.476* .831 1.736
2. Linear Normal
measure- M -4.91 10.66 -5.41
ments of SD 5.37 3.94 4.23
horizon- Range -15.0 to 7.0 5.0 to 17.0 -11.0 to 2.0
tal SE 1.62 1.19 1.27
movement Cleft Palate
of the M -1C.58 9.91 -13.91
tongue SD 5.68 4.89 5.53
Range -21.5 to 4.0 3.0 to 18o0 -21.0 to -4.0
SE 1.38 1.26 1.38
Differences
M 5.67 .75 8.50
SE 2.13 1.73 1.88
t 2o662* .434 4.521**
[a] t (df 28): 1.0% 2.763; 5.0% 2.048 [i] t (df 26): 1.0% 2.799; 5.0% 2.056 [u] t (df 27): 1o0% 2.771; 5.0% 2.052
*Significant beyond the 5.0% level of confidence
*Significant beyond the 1.0% level of confidence




517
(u] and [a] the experimental group tended to have greater vertical movement than the normal group. For [a) the difference was significant beyond the 5.0 per cent level of confidence. These results very closely agree with those reported for the complete cleft group for this measurement. The greater upward tongue movement seen in the cleft palate subjects for the vowel [a] may be understood in relation to other movements and positions. It will be recalled from the previous section that the two groups began at the rest position with approximately equal tongue height above the incisal reference line. It will also be recalled, however,, that at the rest position the tongues of the experimental subjects were significantly farther below the Frankfort line. As has been mentioned, there was a tendency for the experimental subjects to maintain a wider jaw opening for all positions studied. Thus the experimental subjects had to raise their tongues higher to approach the normal palato-lingual division of the oral cavity which they achieved. In absolute terms, the greater amount of vertical tongue movement displayed does not appear to be induced so much by the void in palatal structures, as Knobeloch (24) suggested, but by a tendency toward a wider jaw opening.
An examination of the range of movements for the two groups of the present study reveals a consistently greater variability among the experimental subjects. This,




58
as Knobeloch (24) indicated, is most probably directly af'. focted by the oral anomaly. It appears that persons must compensate for their abnormal structures In individual ways in order to produce acceptable speech results.
horizontal Tongue Movement
Item 2 of Table 5 indicates the amount of change in horizontal tongue position from rest to each vowel phonation. The experimental group and the normal group differ very slightly in this dimension for the vowel [i]. The experimental group shows greater than normal posterior movement for the other two vowels, the difference for (a] being significant beyond the 5.0 per cent level of confidence, and for [u] significant beyond the 1.0 per cent level of confidence. When the data presented earlier are recalled, these results are precisely what should have been expected, The cleft palate group carried their tongues farther forward at the rest position. Therefore,, a normal amount of movement should leave their tongues farther forward for the (i] position. A somewhat greater than normal amount of posterior movement for (a] should leave their tongues normally placed for that vowel. A still greater than normal amount of movement for Lu] should be necessary to achieve a more posterior position for that vowel. Such expectations assumed that the tongues of the two groups would tend to move in the same direction for each phonation, and this has




59
apparently occurred. The range of movements for the two groups did not differ so greatly in the horizontal dimension as they did in the vertical dimension.
The incomplete, unrepaired cleft palate group was compared with the complete, unrepaired cleft palate group in measurements of tongue movement. No significant differences were found here between these two groups.
Measurements of the Naso-pharyngeal Structures
No attempt was made to measure the vertical or horizontal shift of the velum from rest to phonation for two reasons. First, the exact outline of the velum was more poorly defined than that of most structures traced. Secondly, such movement as could be identified among the
experimental subjects was very slight and inconsistent from one vowel to another. The following discussion of velopharyngeal openings, therefore, should be considered as highly tentative.
Velo-pharyngeal Opening
Item 1 of Table 6 shows that at the rest position the two groups of this study did not differ statistically in velo-pharyngeal opening. For all three of the vowels, however, the cleft palate subjects had consistently greater openings, differing from the normal group well beyond the
1.0 per cent level of confidence. Knobeloch (24), who




60
TABLE 6
AREA OF THE POSTERIOR PHARYNGEAL WALL AT THE LEVEL OF THE TUBERCLE OF
THE ATLAS AND OPENING OF THE VELO-PHARYNGEAL STRUCTURES AMONG
SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES
Rest [a] [i] [u ]
1. Smallest velo- Normal
pharyngeal M 9.83 1.00 .55 .15
opening; in SD 2.17 1.62 1.19 .37
millimeters SE .65 .37 .27 .08
Cleft Falate
M 10.03 6.22 8.78 7.39
SD 2.96 302 3.67 3.58
SE .72 .69 .89 .84
Differences
M .20 5.22 8.23 7.24
SE .97 .78 .30 .85
t .206 6.692** 27.433** 8.518**
2. Area of the Normal
posterior M 1.67 1,78 1.60 1.58
pharyngeal SD 92 .24 .23 .23
wall anterior SE .28 .06 .05 .05
to the tubercle Cleft Palate
of the Atlas; M 1.59 1.58 1.51 1.30
in square SD .814 .79 .23 .18
centimeters SE .20 .18 .06 .04
Differences
M .08 .20 .09 .28
SE .35 .24 .24 .22
t .228 .833 .375 1.273
Rest t (df 28): 1.0% 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 [i) t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.718; 5.0% 2.027
**Significant beyond the 1.0% level of confidence




61
found quite similar results for this measurement, felt that any velar movement could be primarily due to contact with the tongue. He observed that among his subjects for the vowels La] and [u] the velum appeared to rest on the tongue resulting in a posterior displacement. For [i] the velum hung downward very closely approximating the rest position. An examination of the X-ray tracings of the experimental subjects of the present study -yields similar observations in some instances. Some tracings, however, revealed velar movement independent of the tongue, though following the lingual pattern as described by Harrington (17) and Williams (40). Apparently, among some of the subjects,. the muscular continuity of the palate was only partially disturbed. One exceptional subject in the experimental group was able to achieve complete velo-pharyngeal closure on all three vowels. This subject will be individually discussed in Chapter V. Area of the Posterior Pharyqngeal Wall
Hagerty, e.t al. (15) described a technique for
measuring the area of the posterior pharyngeal wall at the level of the tubercle of the Atlas. For the present study a very similar area was measured, with the chief exception being that the location of the posterior margin of the area was changed. Whereas H-agerty, et al. had measured forward of a perpendicular line through the mid-point of the tubercle of the Atlas, the present study closely follows Knobelochts




62
(24) design in placing the posterior margin of the area tangent to the most anterior projection of the tubercle.
Inspection of item 2, Table 6 reveals no significant differences between the two groups of the present study in regard to this measurement. hagerty, et al. (15) reported a small increase in the area measured during production of the vowel [at). The vowel [a) used in the present study is roughly comparable, and among the control group a small increase in area may be seen for this vowel. The experimental group, however, showed an insignificant decrease in area of .01 square centimeters. In general, for both groups the changes in posterior pharyngeal wall area were quite small. No statistically significant differences were revealed when the two unrepaired cleft palate groups were compared by this
measurement.
Posterior Phrnga Wall Movement
Measurements of the movement of the posterior
pharyngeal wall, like the measurements of its area, resulted from an adaptation of the technique employed by Hagerty, et al. The horizontal lines along which they measured were drawn 1.0 centimeter apart. They observed, however, that the greatest movements of the pharyngeal wall frequently appeared to fall between the lines. Therefore, in order to measure the movements more precisely, for the present study lines were drawn 0.5 centimeters apart horizontally across the area




63
to be measured. The most superior of these lines was designated level 1. The lines below were consecutively numbered down to level 6.
Table 7 displays the measurements made of the posterior pharyngeal wall movements, level by level. Because of the differences in measurement procedures, a detailed comparison with the findings reported by Hagerty, et al. will not be made. It should be mentioned,, however, that they found very little movement in an anterior direction and none in a posterior direction among their normal subjects. Movements in both directions were found among the subjects of the present study; therefore, negative numbers were used to indicate posterior movement.
No statistically significant differences were observed at level 1. For (I], the normal group showed no mean movement. All other mean movements In both groups were In an anterior direction. The range or variability of movements for all vowels was greater among the normal subjects.
Again, at level 2, there were no statistically signifioant differences observed. All mean movements among the experimental group tended to be in an anterior direction. Among the normal group the tendencies were towards posterior movement for Li], no movement for [u), and anterior movement for La]. As before, the range or variability of movements among the normal group was greater for all vowels.




64
TABLE 7
MEASUREMPNTS OF MOVEMENT OF THE POSTERIOR PIARYNGEAL WALL AT THE
LEVEL OF THE TUBERCLE OF THE ATLAS AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES
(Negative numbers indicate a posterior movement.
All measurements in millimeters)
(a] [ii [u]
1. Level 1 Normal
M .33 .00 .67
SD 3.88 3.90 4.32
Range -5.0 to 6,0 -5.0 to 5.0 -6.0 to 6.0
SE 1.73 1.74 1.93
Cleft Palate
M .37 .61 .10
SD 2.23 1.36 1.23
Range -3.0 to 6.0 -1.0 to 4.0 -3.0 to 1.5
SE .58 .38 .33
Differences
M .04 .61 .57
SE 1.83 1.78 1.96
t .022 .343 .291
(a] t (df 20): 1.0% 2.845; 5.0% 2.086 (i] t (dr 18): 1.0% 2.878; 5.0% 2.101 [u] t (Of 19): 1.0% 2.861; 50% 2.093
2. Level 2 Norml
M .83 -.50 o00
SD 3.41 2.58 2.41
Range -5.0 to 7.0 -3.0 to 6.0 -3.0 to 5.0
SE 1.03 .78 .73
Cleft Palate
M .53 043 34
SD 2.22 1.25 1.20
Range -3.5 to 4.5 -1.5 to 3.0 -3.0 to 2.0
SE .56 .33 .31
Differences
M .30 93 .34
SE 1.17 .84 .25
t .256 1.107 1.360
Ca] t (af 27): 1.0% 2.771; 5.0% 2.052 [i] t (df 25): 1.0% 2.787; 5,0% 2.060 Lu] t (df 26): 1.0% 2.779; 5.0% 2.056




65
TABLE 7 Continued
[a) [i] [u]
3. Level 3 Normal
M .67 -.42 .00
SD 2.02 lo44 1.35
Range -3.0 to 4.0 -3.0 to 2.0 -2.0 to 2.0
SE .61 .43 .41
Cleft Palate
M .33 .34 -.88
SD 2.14 1.40 1.78
Range -3.0 to 5.5 -2.0 to 4.0 -4.0 to 1.5
SE .52 .36 .44
Differences
M .34 .76 .88
SE 080 .18 .60
t .425 4.222** 1.467
4. Level 4 Normal
M .83 -.08 .08
SD 1.70 1.93 1.93
Range -3.0 to 3.0 -4.0 to 3.0 -3.0 to 3.0
SE .51 .58 .58
Cleft Palate
M .58 .31 -.32
SD 1.50 1.27 1.25
Range -1.5 -to 3.0 -2.0 to 2.5 -3.0 to 2.0
SE .36 .33 .31
Differences
M .25 .39 .40
SE .20 .21 .21
t 1.250 1.857 1.905
[a] t (df 28): 1.0% 2.763; 5.0% 2.048 [i] t (df 26): 1,0% 2.779; 5.0% 2.056 [u] t (df 27): 1.0% 2.771; 5.0% 2.052




66
TABLE 7 Continued
[ a] [ i] [u]
5. Level 5 Normal
M .82 .09 .27
SD 1.83 1.14 1.68
Range -3.0 to 4.0 -3.0 to 1.0 -4.0 to 2.0
SE .58 .36 .53
Cleft Palate
M -.06 -.56 -.88
SD 1.05 1.11 1.30
Range -2.0 to 2.0 -3.0 to 1.0 -4.0 to 1.0
SE ..26 .29 .32
Differences
M .88 .65 1.15
SE .20 .14 .20
t 4.400** 4.643** 5.750**
[a], [u] t (df 26): 1.0% 2.779; 5.0% 2.056 [i] t (df 25): 1.0% 2.787; 5.0% 2.060
6. Level 6 Normal
M .28 .14 .00
SD 2.22 1.46 1.97
Range -4.0 to 3.0 -3.0 to 1.0 -4.0 to 2.0
SE .91 ,60 .80
Cleft Palate
M -.25 -.61 -1.32
SD 1.36 2.02 2,00
Range -2.0 to 3.0 -5.5 to 3.0 -5.5 to 2.0
SE .41 .58 .63
Differences
M .53 .75 1.32
SE 1.00 .26 1.00
t .530 2.885** 1.320
[a] t (df 17): 1.0% 2.898; 5.0% 2.110 [i) t (df 18): 1.0% 2.878; 5.0% 2.101 [u] t (df 16): 1.0% 2.921; 5.0% 2.120
"Significant beyond the 1.0% level of confidence




67
At level 3 the experimental group displayed a
slightly greater range of movement for all vowels. For (1) the normal group again showed posterior movement while the experimental group showed anterior movement. The difference here was significant beyond the 1.0 per cent level of confidenee. No other significant differences were noted. For
(a) both groups showed anterior movement, while for (u] the normal group showed no movement and the experimental group showed posterior movement.
The trend of movement for both groups at level 4 is very close to that seen at the previous level, although no significant differences were observed. The means of both groups indicated anterior movement for (a]. For (i] the normal group showed slight posterior movement and the experimental group continued to show anterior movement. A slight tendency toward anterior movement was observed among the normal group for [u], while the experimental group continued to display mostly posterior movement. A wider range of movements was once again seen among the normal group members.
At level 5 the normal group showed anterior movement,, while the experimental group showed posterior movement for all vowels. For each vowel the group differences were significant beyond the 1.0 per cent level of confidence. The variability of movements tended to be slightly greater




68
among the normal subjects.
At level 6 the experimental subjects continued to
display posterior movement during all vowels. For the normal subjects the movement was anterior for [a] and [4), and tended to be static for [u]. For [1] the two groups were significantly different beyond the 5.0 per cent level of confidence. The variability of movements displayed by the experimental group was greater during the vowels [1] and [u].
The normal group showed more movement for the vowel (a).
In summary, there appeared to be a greater range of movements of the posterior pharyngeal. wall displayed by the normal subjects. A general tendency was observed for the normal subjects to narrow the velo-plaaryngeal opening by anterior movement of the pharyngeal wall, -while the experimental. subjects tended to widen the opening by posterior movement of the pharyngeal wall, Even where the mean differenoes between the groups were statistically significant# a large amount of overlap was seen In the ranges of movement. In general, the similarities between these two groups seem greater than their differences in these dimensions. This conclusion would seem to be supported by the previous table, which revealed no significant differences between the
groups in the area of the posterior ph~aryngeal wall.
An examination of the X-ray tracings revealed that only one or two of the normal subjects and only three of the




69
experimental subjects displayed an obvious bulging of the posterior pharyngeal wall which could be called Passavant's pad. In no instance was the phenomenon as marked as some of those pictured by Hagerty, et al. (15). Calnan (6) reported that Passavant's pad may be an abnormal phenomenon in phonation, and probably more appropriately associated with deglutition. Little or no pharyngeal wall movement, for most cases, has been reported by Buck (5), Hixon (20), Williams (40), Wolfe (41), and Hagerty, et al. (15). The results of the present investigation seem to be consonant
with their findings.
Comparisons were made between the complete, unrepaired cleft palate group and the present experimental group regarding measurements of the naso-pharyngeal structures. There were no significant differences between these two groups.
Lip and Mouth Opening
Li Opening
Item 1 of Table 8 presents the measurements of the smallest lip openings of the subjects at rest and during phonation. A general tendency was seen in all positions studied for the experimental group to display wider lip openings. This tendency reached statistical significance beyond the 1.0 per cent level of confidence for (a). The




70
TABLE 8
LIP AND MOUTH OPENING AT REST AND DURING PHONATION AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES (All measurements in millimeters)
Rest [ a] [i] [u]
1. Lip opening Normal
M 2.66 14.15 7.20 1.75
SD lo50 4.87 2.80 1.25
SE .45 1.12 .64 .29
Cleft Palate
M 2.80 19.42 7.72 2.76
SD 4.36 6.25 3.97 1.92
SE 106 1.43 .96 .45
Differences
M .14 5.27 .52 1.01
SE 1.15 1.82 1.16 .53
t .122 2896** .448 1.906
Rest t (df 28): 1.0% 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 [ i] t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.718; 5.0% 2.027
2o Incisor opening Normal M 1.00 12.50 2.15 4.65
SD 4.37 6.20 3.73 4.69
SE 1.32 1.42 .86 1.08
Cleft Palate
M -.38 13.57 2.32 3.72
SD 2.67 5.15 3o04 3.89
SE .67 1.22 .76 .94
Differences
M 1.38 1.07 .17 .93
SE 1.48 1.87 1.14 1.43
t .932 .572 .149 .650
Rest t (df 27): 1.0% 2.771; 5.0% 2.052 [a] t (df 37): 1.0% 2.713; 5.0% 2.027 [1i] t (df 35): 1n0% 2.728; 5.0% 2.032 [u] t (df 36): 1.0% 2.713; 5.0% 2.027




71
TABLE 8 Continued
Rest (a] Li] [u]
3. Distance from Normal
Frankfort line M 53.08 65.85 55.,30 58.10
to incisal SD 4.81 6.57 4.31 5.74
reference line SE 1.45 1.51 .99 1.32
Cleft Palate
M 55.97 70.25 58.61 61.39
SD 6.76 7.60 6.93 6.67
SE 1.64 1.74 1.68 1.57
Differences
M 2.89 4.40 3.31 3.29
SE 2.19 2Q30 1.95 2,05
t 1.320 1.913 1.697 1.605
Rest t (df 28): 1.0% 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 [i] t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2.713; 5.0% 2.027
**Significant beyond the 1.0% level of confidence




72
same general tendency vwas found among the subjects with complete, unrepaired cleft palates. Knobeloch (24) felt that this tendency might be explained by the presence of unrepaired lips among a majority of his cleft palate subjects. This explanation obviously will not hold, however, for any of the subjects of the present investigation. Furthermore, since the two experimental groups were so nearly alike in lip opening, it might be speculated that the same factors were operating in each of them. Mouth Opening
Two dimensions were used to acquire data on mouth or jaw opening. The most direct measurement, the vertical distance between the mandibular and maxillary incisors, was apparently complicated by dental irregularities. The other dimension introduced was the vertical distance from the Frankfort line to the incisal reference line. It was expeoted that this second measurement would be only minimally affected by such irregularities, and consequently more valid.
Neither item 2 nor item 3 of Table 8 shows any significant differences between the groups regarding mouth opening. But if the above assumption can be made, that item
3 is the more valid measure, than a slight tendency may be observed for the cleft palate group to have wider mouth openings in all positions. This tendency, however slight, would seem to be definite when consideration is given to




73
observations made earlier on tongue position and movement (Tables 4 and 5).
The data from item 2 indicate a smaller mouth
opening for the experimental subjects at rest and for [u), and larger mouth openings for the other two vowels. It would be difficult to attempt an explanation of the differences between items 2 and 3 other than on the basis of dental irregularities.
Analysis of Speech Samples
Tape recordings were acquired of the members of
both the normal and the cleft palate groups. The recordings of the normal subjects were used to provide a standard against which to analyze the recordings of the cleft palate group. With this standard in mind, the experimental group subjects were judged first, as to phonetic accuracy or articulation, and secondly, as to deviations from normal oronasal resonance.
The articulation of each sound was scored as: 0, correctly produced; 1, distorted; and 2, substituted or omitted. The total possible score, representing the maximum number of phonetic errors was 44. Any sound which was not correctly produced in all intraword positions was counted an error. The rationale behind this either-or sorting procedure was the judges' feelings that the sounds would be




74
correct or incorrect due to the organic factor involved. Clinical observation indicates that there is a great deal of consistency in the misarticulations of individuals with
palatal inadequacies. The possibility remains, however, that if the errors had been noted in the specific positions in which they occurred, inconsistencies might have jovn up indicative of functional errors.
ratings of vowel and consonant articulation by the judges did not differ with statistical significance. A high positive coefficient of correlation, .99, was calculated for the two sets of ratings. The articulation scores of the experimental group ranged from 3 to 20. Their mean score was 11 and their median score was 10.5. For this same measurement, Knobeloch (24) reported a range of from 10 to 36, with a mean of 20. It is immediately apparent that the subjects with complete, unrepaired clefts tended to have more phonetic errors than did the subjects with incomplete, unrepaired clefts.
The following symbols were used to represent the sounds peculiar to Spanish: [B], Bilabial fricative; [d], voiced dental stop or fricative; [t], voiceless dental stop; [n], dental nasal; [b], palatal nasal; [1], dental lateral; [r], single tap lingua-alveolar; [rr], trilled linguaalveolar; [y], voiced palatal fricative; and [x], voiceless velar fricative. The Spanish vowels approximate those of




75
Latin. They are pure vowels, not diphtlhongized. In English there is a tendency to diphthongize all vowels but [a). An actual count of the number of times errors were noted yielded the following sequence, in decreasing order of frequency of occurrence:
Consonant Sound Frequency of Error
[rr] 20
[k] 13
[r] 13
It] 12
[g] 11
It!] 10
Ip] 10
Is) 8
[d] 7
IB] 6
If] 6
[m] 5
IN) 3
In] 1
Ix] 1
Ill 0
Iy] 0
Although some vowel sounds seemed to border upon distortion, they all appeared to be sufficiently accurate phonetically not to be scored as errors.
A rank order correlation of the preceding consonant error count with a comparable list reported for the complete, unrepaired cleft palate subjects yielded a coefficient of .85. That is, there was a high degree of correspondence between the cleft palate groups in the relative frequency of misarticulations per sound.
The trilled lingua-alveolar, [rr], heard in the




76
speech of all the control subjects, was missing in the cleft palate subjects. The sounds of [k] and [g], which normally require a functional velu, were frequently misarticulated. It was somewhat surprising to find fr) and It] more frequently in error than [.S], since they do not depend so directly upon velar action. Part of the explanation may be that these sounds require a greater implosion of air in the mouth than the experimental subjects were able to achieve,
or that the contact required between the tongue and hard palate was difficult to achieve because of missing palate tissue. A partial explanation may be that some of these and other errors were functional.
The deviations from normal oro-nasal resonance were rated on a five-point scale. A score of 1 represented normal resonance, and a score of 5 represented the greatest deviation from it. The judges observed other kinds of quality deviations among the experimental group such as harshness, hoarseness, and breathiness. However, they tried to consider only what might be called oro-nasal resonance in their ratings. The deviations which were heard could be classified as open or hypernasality, denasality, and a combination of the two which might be termed cul de sac nasality. A tabulation of the quality ratings follows:
Rating Number of Subjects
1 0
2 1
3o
4 8
5 5




77
The tapes were listened to continuously at several different times. No marked differences in numerical rating were noted. The final judgments yielded a positive coefficient of correlation of .92. The judgments did not reveal
any statistically significant differences. The mean rating for the group was 3.85. None of the cleft palate subjects were judged to have normal quality, and only one of this group approached a normal quality rating. Six of the experimental subjects received a rating that was half way between the two extremes. Thirteen, or over half, of the experimental group subjects were judged to have extreme deviations from normal oro-nasal resonance. These ratings are much closer to those assigned the complete, unrepaired cleft palate group than were the scores for phonetic accuracy just discussed. The mean rating for the group with complete clefts was 3.7. Like the experimental group members of the present study,, 65.0 per cent of the complete, unrepaired
cleft palate subjects were assigned ratings of 4 or 5, and none of then were assigned ratings of 1. The ratings of articulation and voice for these two cleft palate groups are believed to be comparable because the same two judges were used for both studies and the ratings were made under identical conditions within a few weeks of each other.
Spriestersbach (36) and Sherman (34) have suggested that tape recordings of speech samples should be played




78
backward to give more valid ratings of quality. They found that articulation factors, which tend to influence quality judgments, are thus eliminated. Since neither of the judges was conversant with Spanish, it was felt that there was no need to reverse the tape recordings of the Spanish-speaking individuals. It should be said, however, that during the course of the analysis the judges did grow familiar with the Spanish words to which they listened. For this reason it may have been better to have played the tapes backwards. In order to investigate the influence of articulation on the judges' ratings of quality, a coefficient of correlation was computed between the ratings of articulation and quality. This correlation was -.35, which indicates that the articulation of the Spanish-speaking subjects had no apparent effect
on the judgments of quality.
Correlations between Acoustical and Ihysical measurements The data to be presented in this section are intended to demonstrate some of the correlations found between the acoustic measurements of articulation and voice quality and the physical measurements made from the X-ray tracings of the experimental group. No correlations were computed with physical measurements uhich were not statistically different from normal. A few other physical measurements were not used in the computations. These were:




79
(1) the mandible width, which has been seen to
follow the trend established by the maxilla, hence
a correlation with maxilla width should suffice;
(2) the pharyngeal opening between the incisal
reference line and the superior border of the body
of the hyoid bone nn [u], because this was identical with the measurement made along the incisal
reference line during the production of that
vowel;
(35) movements at certain levels of the posterior
pharyngeal wall, those movements having so much
variation that It was felt they would probably
not produce significant results.
Among the experimental group the velo-pharyngeal opening
did not appear to vary in a demonstrably consistent pattern from one vowel position to another. Therefore, for each subject the velo-pharymgeal opening measured for all three vowels was averaged, and this average was correlated with the obtained acoustical scores.
The significance of the correlation coefficients was tested by a method suggested by Edwards (7). The criterion of significance was established at the 5.0 per cent level of confidence.
An examination of Table 9 reveals a significant, positive correlation between the Index of Articulation and




80
TABLE 9
COEFFICIENTS OF CORRELATION BETWEEN PHYSICAL AND ACOUSTICAL
MEASUREMENTS AMONG SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES
Index of Rating of Voice
Articulation Quality Deviation
Coefficients of Correlation
1. Maxilla length .39 .02
2. Maxilla width .57* -.25
3. Horizontal distance from lower
incisors to high point of
tongue: Rest .37 .22
4. Horizontal distance from lower
incisors to high point of
tongue: [i] -.25 .47*
5. Pharyngeal opening along
incisal reference line: [u] -.26 -.01
6. Distance of high point of
tongue above incisal
reference line: [a] -.04 -.23
7. Distance of high point of
tongue below hard palate: [i] .18 .53*
8. Distance of high point of
tongue below hard palate: [u] .28 .16
9. Distance of high point of
tongue below Frankfort
line: Rest .24 .34




81
TABLE 9 Continued
Index of Rating of Voice
Articulation Quality Deviation
Coefficients of Correlation
10. Distance of high point of
tongue below Frankfort
line: [i] .21 .45
11. Velo-pharyngeal opening,
average for vowels [a,i,u] .18 .26
12. Lip opening: [a] -o19 .17
*Significant at or beyond 5.0% level of confidence




832
mnaxilla width. That Is, a large phonetic error count appears to be associated with maxillae which are wider. The correlation with maxil.la width is not surprising, since presumably the width of the maxilla is directly associated with the severity of the cleft. It must be clearly pointed out, however,, that no implication of causation is here intended. Factors which yield a coefficient of .57 are presumably related, but both may be caused by some more basic factor.
Further examination of Table 9 reveals significant correlations between voice quality deviation ratings and two measures of the tongue position for the vowel [i). That is, those subjects who, during the vowel (i]$ tended to place the high point of the tongue more posteriorly, and those who tended to carry the tongue a greater distance below the hard palate also tended to be those subjects whose voices were judged to have the most abnormal oro-nasal resonance. it was shown in Table 2 that the experimental group subjects tended to carry their tongues more anteriorly on [I]. The correlation of .47 between a more posterior tongue placement and a higher rating of voice quality deviation would appear to indicate that those subjects who did not follow the trend toward more anterior placement tended to be those whose voices were judged to be more abnormal.
Table 4 indicated that the cleft palate subjects




83
tended to carry their tongues a greater distance below the hard palate for [i]. This tendency showed a .53 correlation with judgments of increased voice quality deviation. This should not lead one to conclude that a higher tongue on [i] will necessarily result in more normal voice quality. Without more evidence it can only be stated that these factors are correlated. It seems unlikely that there exists any cause-effect relationship between them.
Insignificant correlations were found between the average velo-pharyngeal opening for the vowels and both of the acoustic scores. This does not mean that the velopharyngeal opening was of no importance in contributing to these scores. It does mean that the velo-pharyngeal openings of this group were probably well beyond the critical point as described by kcDonald and Koepp-Baker (26).
Further correlatives of the physio-acoustical relationships of these subjects and of those with complete, unrepaired clefts should be of interest, and will be made as the parent study continues. Direct measurements have still to be made of the palatal casts, and these may possibly become the most valuable measurements yet considered for these subjects.




CHAPTER IV
SUMMARY AND CONCLUSIONS
The experimental group of the present study was
composed of twenty natives of Puerto Rico who possessed congenitally cleft palates. To be included in this group subjects had to have clefts which extended no farther forward than the median border of the alveolar ridge, have had no prosthetic or successful surgical repair, and have reached physical maturity. A control group was made up of twenty adult Puerto Ricans who had normal facial structures, and who had the normal speech and hearing of Puerto Ricans. The two groups ranged in age from fifteen years to forty-nine years. The inclusion of the youngest members as adults is based on the study by Fleagle (9) which supports the popular conception that people tend to mature sooner in the tropics.
Lateral X-ray exposures were made of all subjects at physiologic rest and during the phonation of the vowels [a], [i], and [u]. For the control group, twenty exposures of each vowel and twelve exposures of the rest position were obtained. The films available on the experimental group included eighteen of the rest position, twenty of [a], eighteen of [i], and nineteen of [u]. These films were all traced and measurements were taken from the tracings. Antero-posterior views of the subjects at physiologic rest were also made.
84




85
Twelve of these were obtained for the control group and fifteen for the experimental group. These films were not traced, but measurements were made directly from them. All measurements were treated statistically to compare the control group with the experimental group. The experimental group was also compared with Knobeloch's (24) experimental group of individuals with complete, unrepaired cleft palates.
Tape recordings were made of all experimental group member and some control group members. The recordings of the experimental group members were analyzed for phonetic accizracy of consonants and vowels and for deviations from normal oro-nasal resonance. Correlations were computed between acoustical and significant physical measurements.
The following conclusions may be drawn from the data:
1. Differences significant beyond the 5.0 per cent level of confidence occurred between the control and experimental groups on measurements of maxilla length and width and mandible width. Measurements of mandible length, oral cavity
length, and width of the naso-pharynx revealed no statistically significant differences (Table 1). No statistically significant differences were found to exist on the above measurements between the experimental group and the subjects with complete, unrepaired cleft palates.
2. In measurements of the horizontal position of the




86
tongue at rest and for Li], the experimental group differed from the control group beyond the 5.0 per cent level of confidence, their tendency being toward a more anterior tongue placement. Other measurements related to horizontal position of the tongue and tongue size revealed no statistically significant differences between the groups (Table 2). The experimental group did not differ significantly from the subjects with complete, unrepaired cleft palates on any of the above measurements.
3. P~haryngeal opening measurements on and below
the incisal reference line differed with statistical significance for the vowel [u], but not for any other vowel position (Table 3). For the measurement below the incisal reference line on Li], the two unrepaired cleft palate groups differed from the control group in opposing directions, and were significantly different from each other beyond the 5.0 per cent level of confidence. A posterior as well as an anterior concentration of tongue tissue during the sounding of this vowel among the subjects in the experimental group seems to
account for this difference.
4. At rest the experimental group tended to hold
their tongues somewhat lower below the hard palate and significantly lower below the Frankfort line than did the control group, but their tongues were similarly placed above the incisal reference line. For La] the experimental group raised




87
their tongues significantly higher above the incisal reference line, but their tongues remained normally distant from the hard palate and the incisal reference line. For [i] the experimental group showed a slight tendency to raise their tongues less above the incisal reference line than the control group. This difference was amplified to significance beyond the 1.0 per cent level of confidence in measurements from tne tongue to the hard palate and to the Frankfort line. Differences for [u] were not statistically significant except in the measurement from the tongue to the hard palate. For this measurement the experimental group appeared to maintain a greater distance between the palato-lingual surfaces (Table 4). The apparent discrepancies in the various measurements of tongue height may be accounted for by variations in jaw opening and by the fact that for [u] the measurement from the tongue to the hard palate was an approximation, actually made to a line extended from the hard palate. The experimental group and the group with complete, unrepaired clefts did not differ significantly on any but two of the above measurements. These were measurements during the vowel Li] of the distance from the tongue to the hard palate and from the tongue to the Frankfort line. These differences were significant beyond bhe 5.0 per cent level of confidence. They appear to be accounted for by the tendency already noted among the




88
subjects of the experimental group to maintain a posterior as well as an anterior concentration of tongue tissue during this vowel.
5. Measurements of tongue movements (Table 5)
follow the pattern which would have been hypothesized from the previous measurements of tongue position. Also, the greater variation seen in the vertical tongue movements of the experimental group may reflect the presence of the oral anomaly. The experimental group tended to move their tongues higher for [u] and [a], the difference on [a] reaching significance beyond the.5.0 per cent level of confidence. More posterior movement was seen on these same two vowels for the experimental group. The differences were significant beyond the 5.0 per cent level of confidence for [a] and beyond the
1.0 per cent level of confidence for Lu]. No significant differences were noted for the vowel fi]. Comparisons between the experimental group and the group with complete, unrepaired clefts on the above measurements revealed no statistically significant differences.
6. measurements of velo-pharyngeal opening during phonation revealed very large differences between the two groups of this study (Table 6). The experimental subjects had consistently larger openings. The mean group differences ranged from 5.22 millimeters to 8.23 millimeters, all of
which were significant well beyond the 1.0 per cent level of




89
confidence. It was felt to be unnecessary to statistically compare the two unrepaired cleft palate groups on this measurement, because of the similarity of the findings in these groups.
7. The experimental group did not differ significantly from the control group in measurements of the area of the posterior pharyngeal wall (Table 6). Since this had also been true of the complete, unrepaired cleft group, the two cleft palate groups were not statistically compared on
this measurement.
8. Measurements along the lines constructed across
the area of the posterior pharyngeal wall revealed only small amounts of movement for both groups. The only significant differences found were: at level 3 for [i], the control subjects showed posterior movement, whereas the experimental subjects showed anterior movement; at level 5 for all vowels, the control group showed anterior movement, whereas the experimental group showed posterior movement; at level 6 for Ii], the control group showed anterior movement, whereas the experimental group showed posterior movement. All differences noted were significant beyond the 1.0 per cent level of confidence (Table 7). A comparison between the unrepaired, cleft palate groups on these measurements revealed no statistically significant differences.
9. A larger lip opening, significant beyond the 1.0




90
per cent level of confidence, was found for the experimental group on the vowel [a]. ho other significant differences were found for measurements of lip opening, incisor opening, or distance from the Frankfort line to the incisal reference
line. There appeared, however, to be a general tendency among the experimental group to maintain a slightly wider lip and jaw opening (Table 8). Comparisons between the unrepaired, cleft palate groups for these measurements revealed no statistically significant differences.
10. The speech of the experimental group subjects
exhibited a great many phonetic errors, including some which were inconsistently made. Those sounds inconsistently misarticulated do not appear to be explained by the presence of
the oral anomaly. The subjects with complete, unrepaired clefts produced a great many more errors than the experimental group subjects with incomplete, unrepaired clefts.
11. The voice quality analyses of the tape recordings indicated that the experimental subjects differed markedly from the control subjects in ratings of oro-nasal resonance. However, the pattern of ratings for the two unrepaired cleft palate groups appeared to be quite similar.
12. The number of phonetic errors made by subjects of the experimental group was positively correlated with
maxilla width (Table 9).
13. Among the experimental group subjects,




91
deviations from normal oro-nasal resonance were positively correlated with the distance of the high point of the tongue below the hard palate for [i] and also with the horizontal distance from the lower central incisors to the high point of the tongue for [i] (Table 9).




CHAPTER V
DISCUSSION
The two groups of this study were found to differ significantly in the following skeletal measurements: the experimental group had maxillae which were shorter, and both maxillae and mandibles which were wider than those of the control group. The cause of the greater mandibular width among the unrepaired cleft palate subjects of this study cannot be readily ascertained. It may be that the laterally spreading maxilla exerts an outward force on the
mandible. But would not the more solid bony frame of the mandible tend to resist this force? It is possible that the mandibles of this group grew wider as a result of neurological changes which may have been imposed by the cleft condition. When the experimental group of this study, who had incomplete, unrepaired cleft palates, was compared with Knobeloch's (24) group who had complete, unrepaired cleft palates, no significant skeletal differences were
found. The differences between the complete, unrepaired cleft palate group and the normal group, however, failed to reach significance except in maxilla width. It would appear that when a fetus fails to form a complete palate, whether that failure is partial or total makes little difference in the potential effect it has on the facial skeletal growth 92




93
of the person. That effect, left undisturbed by surgery, appears to be most noticeable in the measurement of maxillary width. A question arises as to how directly related the width of the cleft is to the extra width of the maxilla in the adult. To what extent is the cleft condition exaggerated by a subnormal amount of bone tissue? The question provides a beginning point for future valuable research.
When the incomplete, unrepaired cleft palate subjects are compared with the repaired cleft palate subjects studied by Buck (4) and Graber (12) important differences are seen in skeletal dimensions. The repaired cleft palate subjects had significantly shorter vertical facial dimensions, together with concave facial angles at physiologic rest. Their faces were also shorter in antero-posterior width than the normal subjects with whom they were compared, and their maxillary and mandibular dimensions were smaller. In general, it can be said that the unrepaired cleft palate subjects were much more like normal subjects in facial skeletal development than they were like repaired cleft palate subjects.
The most outstanding functional differences discovered between the experimental and the control groups was in the amount of velo-pharyngeal opening during phonation. As a group, the cleft palate subjects consistently showed a




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EW9H4B425_J9AOFY INGEST_TIME 2018-07-19T21:16:03Z PACKAGE AA00063477_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

An X-Ray Study of Unrepaired, Incomplete Cleft Palate Oral-Pharyngeal Structures and Their Functioning During Vowel Phonation By JOSEPH SESSIONS KEENAN A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA June. 1960

PAGE 2

ACKNOWLEDGEMENTS The writer wishes to express his appreciation and indebtedness to Dr. McKenzie Buck, the chairman of his super visory committee, for his constant assistance and encourage ment throughout the course of this dissertation. As a member of the national research g roup which c onducted the parent study, Dr. Buck arranged for the use of ~1e X ray plates, the tape recordings, and the descriptiv e data utilized for analysis 1n this study. The writer would also like to express sincere ap preciation to Professor H P Constans, Dr. R.E. Tew, and Dr. L.L. Z immerman of the De p artment of Speech, and to Dr. E. P H orne and Dr. J. C Dixon of the Department of P sychology for their cooperation and encouragement as members of the writer's supervisory committee. In addition, the writer would like to thank Dr. G. H Kurtzrock, who was an original member of his supervisory committee. Special expression of a pp reciation is extended to Frank E e Law, D. D .S., Chief, Operational Research Branch, Division of Dental Public H ealth, Washington, D.C., who or ganized the study team which collected the raw data for the present study. The writer would also like to express special ap p reciation to the citizens of P uerto Rico who participated in the study. F'urther s pe cial thanks are extended to D r. 11

PAGE 3

Calvin Knobeloch, who made both his data and time g enerously available when they were needed. In addition, the writer wishes to especially thank Mr. Reid Holmee, Administrator of Baptist Hos p1 tals, Inc. in Winston-Salem .for his under standing and interest du~1ng the preparation of this dis sertation. These acknowledgements would be quite incompla te without an expression of eratitude to the writer's wife. Without her courage end faith this ta.sk would not have been begun. Without her devotion and coopere.tion it could not have been completed. Finally, the writer is g rat~ful to his children, Chris and Cec1, who have put up with him during times when it was a trial to have a writer for a father. 111

PAGE 4

TABLE OF CONTENTS AC KN OWLEOOEMENT S LIST OF TABLES. . Page 11 V LIST OF FIGURES vi Chapter I. INTRODUCTION l 14 II. DESCRIPTION OF PROCEDURE Selection and Matching of Subjects 16 Procedure for Recording and Analysis of Speech Samples 18 Apparatus and Procedure for Obtaining X-ray Plates 21 Measurements of the X-ray Plates 25 III. RESULTS 35 Facial and Oral Cavity Dimensions. 35 Measurements Related to Tongue Size and Horizontal Position of the High ~oint of the Tongue 41 Vertical Tongue Position 49 Excursion of the Tongue from Rest to Vowel Positions. 55 Measurements of the Nasop haryngeal Structures 59 Lip and Mouth Opening 69 Analysis of Speech Samples 73 Correlations between Acoustical and Physical Measurements. 78 IV. SUMMARY AND CONCLUSIONS. V. DISCUSSION 84 92 101 LIST OF REFERENCES. iv

PAGE 5

Table l. 2. 4. LIST OF TABLES M easurements of M axillary, M andibular, and Oral Cavity Structures among Subjects with Normal Development and Subjects w1 th Incomplete, Unrepaired Cleft Palates Midsagittal Ton g ue Area and H orizontal ~ osition of H igh P oint of Tongue Along Incisal Reference Line among Subjects with Normal Development and Subjects with Incomplete, Unrepaired Cleft P alates The Pharyngeal Openings On and Below the Incisal Reference Line among Subjects with Normal Development and Subjects with Incomplete, Unrepaired Cleft P alates Various Vertical M easurements of Tongue Carriage among Subjects with Normal Development and Subjects with Incomplete, Unrepaired Cleft P alates Page 37 42 46 50 5. M ovement of the Tongue from Rest to Vowel 6. 7. 8. P ositions among Subjects with N ormal Development and S ubjects with Incom p lete, Unrepaired Cleft P alates. 56 Area of the Posterior P haryngeal Wall at the Level of the ru bercle of the Atlas and Openin g of the Velo-pharyn g eal Structures among Subjects with N ormal Develo pm ent and Subjects with Incomplete, Unre p aired Cleft Palates M easurements of M ovement of the Posterior Ph aryngeal W all at the L evel of the Tubercle of the Atlas among Subjects with Normal Development and Subjects with Incomplete, Unre p aired Cleft P alates Lip and M outh Opening at Rest and During Fhonation among Subjects with N ormal Development and Subjects with Incom p lete, Unrepaired Cleft P alates V 60 64 0 70

PAGE 6

.rable 9. LIST OF rABLES (Continued} Coefficients of Correlation between P hy sical and Acoustical M easurements among Subjects with Incom p lete, Unre p aired Cleft Pa lates vi Page 80

PAGE 7

LIST OF FIGURES Figure l Master Tracing of a Typical Lateral X ray Plate Showing Reference Points and Location Page of M easurements 29 2. Master Tracing of a rypical An.taro posterior X ray P l ate, Indicating the Structures and the P o ints from Which All Antero poaterior Measurements Were Taken. 33 vii

PAGE 8

CHAPTER I I N TRODUCTIO N Various scholars have used the X-ray in speech investigations since its discovery in 1895, according to P armenter and Trevino (28). Of primary significance, at least in terms of stimulating thou g ht and further investi gation, were the studies of Russell (30, 31, 32). H e utilized X-ray techniques in his studies of vowel and consonant p roduction in normal subjects. Parmenter and Trevino (28} criticized Russell principally for his presentation of evidence against a physiological basis for vowel differentiation. They felt that Russell's interpretation of his X -ray plates ,,as correct, but they were of the o p inion that h is technique for maintaining a fixed posture during ex p osure was inadequate. For their X-ray investigation of normal subjects, they devised a head positioner which appeared to control the head ade '-i uately in a horizontal plane. There is some question as to the control afforded the head in the antero-posterior plane. The results of their study sup p orted the hypothesis of a progression of tongue positions as the vowels are produced in traditional order, and of a fairly definite tongue position for each vowel for each individual studied. l

PAGE 9

2 The criterion of a fixed and constant head position during X-ray exposure continued to receive attention in subsequent studies conducted on normal sub jects. In 1934, Norris (27) utilized a head positioner developed by Higley in her study of vowel production in adults as it is related to voice. Norris' conclusions concerning nasality and denasality indicated a wide uvular opening or a thinning of velar tissue for the former, and a complete uvular closure or a thickening of the velum for the latter. Further investigation of naso pharyngeal closure in adults was done by W olfe (41) in 1942. Recognizing the limitations of lateral X ray plates, he re p orted that his data seemed to reveal only a shortening of the antero posterior diameter of the naso pharynx. His data also indicated upward and backward movement of the velum, together with a slight forward movement of the posterior pharyngeal wall. This series of movements effected a naso pharyngeal closure which was not complete for most of his subjects. He found that greater anterior m ovement of the posterior pharyngeal wall was accom pa nied by less p osterior movement of the velum, and vice versa. Com par ing oral and pharyngeal structures of adults with nasal voices and adults with superior voices, Hix on (20} found that the over-all vertical and antero-posterior dimensions of the fa c e were very similar for both groups.

PAGE 10

The measurements of the midsagittal diameter of the pharynx and the midsagittal measurements of the ton g ue indica ted that the tongue was carried farther posteriorly in the nasal group while at rest and for the vowel [a], with a tendency in this direction for (ae ]. H ixon also commented on his findings re g arding naso-pharyngeal structure, ~ ut he reported that these structures were not clearly defined on his X-ray plates. The study co ntained his recommendations to clarify this problem;!~, the use of slower speed cassettes, the use of a lead focusing plate, and p a inting the entire inferior margin of the so.ft palate in order to differentiate the uvula from the soft pala te proper. While the study conducted by Brodie (2) was also concerned with subjects with normal structures, it differed from the preceding investigations 1n that it was longitudinal in aspect. Brodie indicated that the use of an X-ray technique 1n growth studies had been predicted as early as 1922, but that its development was not perfected for this purpose until 1931. In th at year Broadbent published the full details of a roent g enographic cephalometer and began a long range investigation of the head of the growing child.o Brodie sunnnarized his own investigation with the statement that a longitudinal study of growing children indicated that the morphogenetic pattern of the individual was established at a very early age. Once this pattern was attained, he

PAGE 11

found that it ctld not change. Herzberg and H olic (19) reported that their findings confirmed Brodie's results by anthropologic methods. Their findings were based on measures obtained from three hundred and twenty-six adult human skulls. Growth studies have also been conducted using cleft palate subjects. Graber {11) reported his analysis of the developmental patterns and facial morpholo g y in 4 cleft palate. As a group, his subjects showed deficient patterns of maxillary growth, laterally, antero-poateriorly, and vertically. He inctlcated that patients whose palatal clefts had not been manipulated surgically showed a pattern of maxillary growth that was essentially normal. M andibular growth on the whole appeared normal. The lack of development noted in mandibular growth in Graber's subjects was consi dered to be no mor e than would be found in the same distri bution of normal subjects. Graber's chief contention, it would seem, was a serious consideration of the propriety of early closure of cleft palates. P ostponement of surgery until the end of the fourth year of life was indicated, at which time the majority of total maxillary width had been accomplished. A necessary further precaution was pointed out due to a continuation of downward and forward g rowth of the maxilla until twenty years of a g e. Krogman (25) offered some sup p ort of Graber's (12)

PAGE 12

5 findings regardin g the growt h of unrepaired structures in his report of cumulative X-ray studies done at the Philadelphia Center for Research in Child Growth. H e stated that the postnatal growth patterns of unrepaired cleft palate and normal faces did not differ, exce p t that growth in the former proceeded 1n keeping and in harmony with the basic deviation initially introduced by the cleft. He felt that an optimum time for sur g ical procedure would be somewhere between four and six years of age, and possibly as early as two to three years of age if the growth tempo of the individual child was such that there was evidence of advanced growth. Harrington (17) confirmed Wolfe's (41) questioning of the nature of velo-pharyngeal closure with the results of his study of ten cadavers and X-rays of three living subjects. The various aspects of his investigation led him to the conclusion that mesial movement of the pharyngeal walls occurred over considerable vertical extent, rather than as a continuation of Passavant'a cushion. Consequently, the mechanism acting to produce velo-pharyngeal closure was not of the nature of a simple sphincter. A similar phenomenon was sug g ested earlier by Townshend (39) 1n his critique of the formation of Passavant 1 s bar. Using X-rays, plaster models, and photo g raphs,

PAGE 13

6 Slaughter and Brodie (35) concluded their study with statements that further su p ported Graber's (ll) contention regarding the impropriety of early surgical repair of cleft palates. They also presented statements to support Graber's {12) and Krogma n 's (25) remarks on the essentially normal gro w th rate of congenitally deformed parts which were not permanently damaged. Buck {4,5) conducted an X-ray study of oral and pharyngeal structures in patients, ranging in age from seven to nineteen years, who had surgically repaired cleft palates. His findings regarding over-all vertical and antero-posterior dimensions of the face further substan tiated Graber's {12) premise of the interference of early surgical closure with growth. Buck found smaller mid sagittal measurements of the tongue in his cleft palate subjects, as did H ixon {20) in his non-cleft nasal subjects. Concerning midaagittal pharyngeal diameter on and below the incisal reference line, Buck found measurements for his two groups to be very similar. Proportional measurements of the horizontal position of the high point of the tongue, which took into consideration the structural differences between the two groups, were also very similar. These measurements failed to support Hixon's conclusion on posterior tongue carriage. Lower tongue carriage in the oral cavity was indicated by Buck when he found a tendency for the cleft

PAGE 14

7 palate subjects to show a greater vertical distance between the hard palate and the tongue. Regarding velo-pharyngeal movement, Buck reported significantly larger openings for his cleft palate subjects, smaller horizontal and predominantly vertical movement of the velum in the cleft palate group, and very small posterior pharyngeal wall movement for both groups. He also found no significant differences between the two groups on measures of mouth and lip opening. The error of Gustav Passavant was commented on by Calnan (6), who based his contentions on clinical and radiological evidence. Passavant had observed his ridge in a normal subject, but he failed to explain why it was observed in only one in fifty patients. Calnan felt that Passavants phenomenon could be seen in unrepaired cleft palates, but with an occurrence in less than 25 per cent of cases. He explicitly denied that the phenomenon was a factor in normal speech. Hagerty and Hoffmeister (13) felt that the cleft palate was closed primarily to improve speech. They offered a quantitative measure of speech as an indicator of success in palatal closure, if no other damage was incident to the closure. It was hypothesized that measurements of the minimal velo-pharyngeal distance taken during prolongation of [s] enabled them to predict

PAGE 15

nasality from the X-ray with about 75 p er cent accuracy. Laminagraphy 1 a recen t develop m ent in X-ray studies. has been utilized in some of the more recent cephalometric investigations. This ia a technique for obtaining measurements of tissues which lie in planes from the surface to the midline. The tube and film of 8 the laminagraph move during exposure 6 and the X-ray pene trates deeper each time it passes over a body. Brader (1) pointed out that this technique permits projection of any selected plane of a body to the exclusion of all other planes. To illustrate. structures lying in the plane of the vertical process of the maxilla 6 in the plane of the nasal turbinates. and in the plane of the nasal septum could be viewed on successive laminagraphic plates. Static X-ray plates present a view only of the mid-sa g ittal plane. Brader further stated that laminagraphy made possible the visualization of certain anatomical structures or pathologic lesions which are not susceptible to clear projection by any other method. Brader conducted prelimi nary tests to determine the accuracy with which images in lamina g raphic projection might be measured. These tests indicated that such measurements lay within the limits of scientific accuracy, 1.~ within one-half millimeter of direct measurements. In 1953 6 Subtelny (37) reported his findings of a

PAGE 16

9 laminagraphic investigation of the width of the naso-pharynx and related structures in normal and pre-operated cleft palate children. The subjects were all under three years of age and ninety-one unoperated cleft palate patients were included in the study. Perhaps the most striking finding reported was the indication of an abnormally wide naso-pharynx in the claft palate individuals. The bi-hamular measurement, which is the distance between the inferior terminations of the medial pterygoid plates of the sphenoid bone, was found to be larger for this group. The outward flare of these medial plates was found to increase as they descended. Subtelny hypothesized that certain muscles pulled on the open embryonic structures to create this wider dimension. He introduced an element of doubt as to the validity of this hypothesis when he commented on certain of these muscles whloh could have the opposite effect on the opening. The lateral pressures on the cleft structures created by the tongue might augment the widening effect. In the case of a repaired lip, the buccinator musculature is such that it should have a restraining effect. Subtelny commented that there appeared to be a levelling off in naso pharyngeal width during the second year of life. Ricketts' (29) lam1nagraph1c study stressed the need for a more complete understanding of the basioranial structures. He cautioned that inferences from height and

PAGE 17

10 depth dimensions projected on lateral X-ray films were not conclusive. In view of Subtelny's (37) finding on naso pharyngeal dimensions and the findin g s of his own study. he concluded that the actual size of the adenoid was not as important in velo-pharyngeal closure as the available space that it filled in the naso-pharynx. He was in agreement with Graber (11) 1n stressin g the importance of the descending growth of the hard palate in evaluating the individual condition. Without an increased range of activity 1n the soft palate. this descent seemed to explain two circumstances: the frequently ex p erienced increased nasal emission in p ostoperative cleft p alate p atients after the age of six to ei g ht y ears. and the almost spontaneous disa pp earance of mouth breathing in some patients 1n t h is same age group. A discussion of the significance of early ortho dontia 1n cleft palate habilitative plannin g by Subtelny (38) included references to a lamina g raphic study he had com pleted on repaired cleft lip and cleft palate children prior and subsequent to surgical reconstruction of the li p H e found that the molding action of the lip was not restricted to the alveolar border alone. but might cause an architectural rearrangement involving each maxilla. In many cases that demonstrated this movement. Subtelny found that the maxillary constriction was severe enou g h to brin g

PAGE 18

ll about contact between the inferior nasal turbinate and the nasal septum. This condition he offered as a possible explanation for the distortion of nasal sounds in cleft palate speech, which has been described as a hyponasality superimposed on the hypernasal quality. Hagerty, Hill, Pettit, and Kane (15) reported a laminagraphio study concerned with posterior pharyngeal wall movement in normal adults. They concluded that it was doubtful whether the actual extent of excursion of the posterior pharyngeal wall was significant enough to make important differences in speech. Exception to this conclusion would occur when the individual possessed a Passavant 1 s bar large and vigorous enough to participate actively in closure. These authors also reported on soft palate movement in normal adults (16). In the absence of a vigorous Passavant's bar, it seemed evident that the soft palate participated far more aggressively than the pos terior pharyngeal wall in the regulation of the naso pharyngeal aperture during speech. These observations were supported by Hagerty and Hill (14) in their investi gation of individuals with repaired cleft palates. Passavant's bar was seen in only nine of eighty subjects, and was functional in only three. These findings support those of Wolfe (41) on the participation of the velum in velo-pharyngeal closure.

PAGE 19

12 The preceding discussion of static X-ray investi gations forms the background for the present study. Graber (12), Krogman (25), and Slaughter and Brodie (35) have suggested that the growth pattern of the structures of individuals with unrepaired cleft palates proceeded as 1n normally structured individuals. Krogman offered the qualification that such growth would be in keeping and in harmony with the deviation introduced by the anomaly. The study by Buck (4), using individuals with repaired cleft palates, offered support to their contentions. A special case in Buck's investigation revealed that the cleft palate individual whose speech was judged as having relatively little nasality differed chiefly from the other cleft palate subjects in only two respects: (1) the rela tively normal size of his facial and oral structures, and (2) 1n his being able to achieve a fairly complete closure of the nasal port. It is of interest to note that Buck reports this individual to have undergone only lip repair prior to his seventeenth year. At the age of seventeen surgical repair was performed on the hard and soft palates. 1 Several questions, based on suggestions derived from the foregoing review of literature, constitute the basis lThe background analysis and procedure cha~ters a r e essentially the same as heretofore re p orted {24).

PAGE 20

for the pre s en t study: 1. How does a grou p of physically m atur e in d viduals with i ncomplete, unre p aired cleft palates compare wi th a similar gr oup of normally structured individuals of the same origin? 2. Ho w does a group of physically mature indi viduals with incomplete, unrepaired cleft palates compare with a similar group of individuals with complete, unrepaired cleft palates? 3. How does a group of physically mature indi viduals with incomplete, unrepaired cleft palates compare with other cleft palate individuals who have undergone sur g ical repair prior to maturation? 4. What effect does the presen ce of the unre p aired anomaly have on the speech of the adult? 13

PAGE 21

CHAPTER II DESCRIPTION OF PROCEDURE The present study may be described aa a detailed analytical aspect of a parent study sponsored by the u. s. Department of Health, Education, and Welfare. The moti vating factor whioh led to the formulation of the parent study was a desire to accumulate data on physically mature individuals who had untreated cleft palates. The criterion for choosing a site for the study was the location of a suitable group of cases. Through the interest and cooper ation of the Insular Department of Health, a sizable group of such cases was located on the Island of Puerto Rico. A study team of dentists and a speech pathologist visited the island in June, 1957, and June, 1958, to collect the raw data.2 Dentists, social workers, and other personnel from the Insular Department of Health and Puerto Rican 2 The members of the study team included Frank E. Law, D.D.S., Division of Dental Public Health, Public Health Service, U.S. Department of Health, Education, and Welfare, Washington, D.C.; John T. Fulton, D .D.S., Professor of Dental Epidemiology, University of North Carolina, Chapel Hill, N.C.; McKenzie Buck, Ph.D., Head, Speech and H earing Clinic, Uni versity of Florida, Gainesville, Fla.; and others. (The com plete listing of the research group members is available upon request). None of the results of the parent study are to be published until all aspects of it, such as this study, are completed. A portion of the raw data sathered by the team (X-rays, tape recordings, descriptions} were made available to the writer to enable him to fulfill a part of the re quirements for his advanced degree in speech pathology. 14

PAGE 22

15 health centers augmented the study team while the investi gation was in progress. Prior to the tea.m's arrival on the island, case finding was accomplished by the Bureau of Crippled children and the B ureau of Dental H ealth of the Puerto Rico Department of H ealth, assisted by the Puerto Rican Office of Vocational Rehabilitation. Medical exami nations were conducted on the 1957 study group but, as it was felt that the results were not significant for this research, they were not conducted on the 1958 study group. The experimental design of the present study is similar to that used by Buck (3) in his X-ray investigation. He studied oral and pharyngeal structures and functioning in individuals with repaired cleft palates. Two groups of adult Puerto Ricans were studied: (a) an experimental group with incomplete, unrepaired cleft palates; and (b) a control group with normal structures, speech and hearing, selected at random. Tape recordings of both groups were judged for accuracy of articulation and excellence of voice quality. Lateral X-ray plates were taken of each subject during the phonation of the vowels [a], [i], and [u], and at physiologic rest. An an.tarop osterior p late was also taken of each subject at physiologic rest. As will be discussed in a later section, not all of the plates were available for examination.

PAGE 23

16 Each lateral X-ray plate was traced and the tracings were measured to secure data related to the size, positioning, and movement of the oral and pharyngeal structures, and the cross-sectional diameters of the oral and pharyngeal cavities. Lateral measurements of the skeletal structures of the face were obtained directly from the antero-posterior plates. The data obtained from the measurements and the judgments of the tape recordings were analyzed by various statistical techniques in order to permit comparison between the two groups of subjects. Selection and Matching of Subjects To be included in the study group, cases had to meet the following qualifications: (l) must have reached maturation (fifteen years of age or over); and (2) must have a palatal cleft which had (a) no s urgical or prosthetic treatment, or (b) surgical treatment no earlier than twelve years of age, and which was not successful. A palate which was closed and which has since pulled apart indicates that the dynamic processes of facial growth may not have been retarded by the operation. One of the experimental group was classified under (b), above. The measurements taken on her will be discussed in detail in a later chapter of this study.

PAGE 24

17 On the basis of descriptions that were available for the individual subjects, the twenty individuals selected for inclusion in the experimental group have been classified as having incomplete, unrepaired cleft palates. Twelve of these subjects had normal hearin g One of the eight re maining had a bilateral high frequency hearing loss. The other seven subjects all exhibited some degree of hearing loss which appeared to be conductive in kind, and which might be attributable to the cleft palate condition (10, 22, 33). Of this latter group, three showed unilateral and four showed bilateral losses. Thirty-six normally structured adults were also examined. The criteria for selection of these individuals were that they had achieved normal structural growth and possessed the normal speech and hearing of the Puerto Rican population. These individuals were selected at random from the various examination centers. Of the thirty-six, twenty were chosen for inclusion 1n the present study as a control group. Thia was the same control group used 1n a prior study (24). The age range of the experimental group membe r s was r1fteen to forty-nine years. This group was composed of nine males and eleven females. rhe a g e ran g e of the control group was seventeen to forty-six years. There were ei g hteen females and two males 1n this s ro up N o attempt was made to

PAGE 25

1 8 match individuals in the two groups as to age and sex, because all of the subjects had met the criterion regarding physical maturation, and optimum growth had been accom plished. Procedure for Recording and Analysis of Speech Samples In the experimental group each person made a tape recording as he articulated a 11st of Spanish words. The 11st, which follows, was constructed to contain the conso nants and vowels of the language of the subjects in all of the intraword positions 1n wb i ch they occur. Those words selected were simple, everyday words which were represen tative of the common language us a ge among the subjects. Hence, the pronunciation of these words was probably highly stabilized in each subject, and unlikely to be altered by another speaker's pronunciation. Consonants: Sound [p) [t] [k] [g] [B] [d] [f] [m] [n] [ tS] [N] [s] [l] (r] [rr] Words pollo, copa taza, gato oarro, vaoa gato, guagua bote, libro diente, dedo oafe, funda mesa, oama nene, mano, jabon china, leche niffo, p1f'ta silla, taza, ojos leohe, bola, arbol carne, dolar rosa, oarro

PAGE 26

Consonants (Continued): Vowels: Sound [y] [x] [a] [e] [ 1] [o] [u] Words llave., pollo jarro., angel agua., casa., pata eje., pesa., nene hllo., ma g uey., tina ojo., aopa., huevo uno., cuna 19 The following consonants are not the same aa those used in English., and require explanation: [BJ., bilabial fricative; [d]., voiced dental stop or fricative; [t]., voice le~s dental stop; [n]., dental nasal; [N]., palatal nasal; [l], dental lateral; [r]., single tap lingua-alveolar; [rr]., trilled lingua-alveolar; (y]., voiced palatal fricative; and [x], voiceless velar fricative. N one of the vowels are the same as those used in English. They represent approxi mately the same vowels that appear in Latin., in that they are pure and not diphthongized. There is a tendency for all of the vowels except [a] to be diphthongized 1n English. Each subject wa a instructed to read the words in the order in which they appeared in the 11st. In several instances the subj e c ts were unable to read and were asked to repeat the words a.fter they were spoken to them by a Puerto Rican nurse. The control g roup subjects were recorded in the same manner. ;l 1 he r eco rdings we re made on a Violle n sak tape recorder: Model T 1500., high fidelity., dual speed. The instrument's portability was an important consideration in

PAGE 27

20 its selection. Although the recording room conditions we re far from ideal, the ambient noise level was kept at a minimum. Twenty recordings of the experimental group members and twelve recordings of the normal subjects were arranged in random order for analysis. The author and another speech pathologist acted as judges to determine phonetic accuracy of consonants and vowels, and quality differences as re flected in deviations from normal oro-nasal resonance. During the process of analysis, it soon became a pp arent that even the most nearly normal cleft palate speakers would be immediately distinguishable. Further attempts to disguise the identification of individual subjects were therefore terminated, and the judges made use of the control group to provide standards against which to rate the speech of the cleft palate individuals. To assure as equal as possible listening conditions, the two judges wore earphones of the same make and model and made simultaneous analyses. Thorough practice sessions were utilized over a four-week period prior to making the final analyses reported herein. During this time the judges examined each other's ratings at length, i n an effort to control the variables being considered. Resonance was rated on a five-point scale_ with 1 representing normal oro-nasal resonance and 5 representing the greatest deviation from normal oro-nasal resonance.

PAGE 28

21 A numerical score of ar t i c u l a tio n W'dS attained by assi gni n g values to the ar t icul at e d sounds: 0 indicated correct pro duction; 1 indicated a di stortio n ; and 2 indicated either a s ub stitution or an omi ss ion. In addition, the judges com pared their analysis sheets in order to determine which consonants and vowels were found to be mos t frequently defective. Several deviations in articulation were no t iced which, on the basis of the speech sam p les of the normal subjects, were a pp are n tly due to dialect differences. Such deviations were not considered to be errors. A ppar atus and. Pr oce d ure fo r O b t ain i ng X -r ay P la t e s Control .2_ P osition In order to minimize movement of the head during the X-ray exposures, a head positioner was utilized. The device used in 1958 was designed by H owell (23) and was con structed at the Universit y of Florida. This h e ad positioner was not available in 1957, but a similar o n e was fashioned from photo g rapher's equi pm ent formerly used 1n portrait p h oto g raphy. O n e of the crite r ia used in the desi gn in g of the head p ositioner was that it shoul d be p ortable w ithout sacrificing sturdiness. This make s practicable the use of such a posit1oner 1n a field study. The head pos1tioner utilizes ear rods, which ins e rt into the external auditory meati, to control movement in the sa g ittal p lane. An

PAGE 29

22 armature which rests against the nasion of the subject is used in an attempt to prevent rotation of the head position around the axis created by the ear rods. In addition to minimizing head movement, the head positioner provides another advantage, that of aiding in the definition of the Frankfort line on the X-ray films. As in the studies by Buck (3) and Hixon (20), each subject was seated in a comfortable position in order not to disturb the habitual carriage of the oral and pharyngeal structures. The relative position of the chair with respect to the head positioner was adjusted to make possible this comfortable posture .Marking 2t_ Structures None of the structures were marked 1n the present study. The rationale behind this lack of marking may be attributed 1n part to Buck (3). The experimental group in his study had the midline of the tongue, the m1dl1ne of the velum and hard palate, and the midline of the posterior pharyngeal wall covering Fassavant 1 s pad marked with a mixture ot barium and gum acacia. Severe gagging on the part of the control group members prevented markin g of the midline of the posterior pharyngeal wall. W ith this exception, similar markings were applied to the control group members. In an effort to discover the possible error resulting from a lack of marking of structures, Buck conducted an additional

PAGE 30

23 experiment in connection with his study. Eight normally structured individuals, who did not present severe gag reflexes, were X-rayed with and without the marking of structures. A comparison of measurements obtained from these X-rays revealed an average difference between the two sets of measurements of plus or minus 0.6 millimeters. X-ray Exposures~ Four lateral cephalometric plates were acquired on both the experimental and control groups: one at physio logic rest and one each while phonating the vowels [a], [1], and [u]. These vowels were selected because they are repre sentative of extremes of tongue positions used 1n phonation. The low front vowel [ae ], used in some X-ray studies, is not present 1n the language of the subjects of the present study. The fifth X-ray exposure was an antero-posterior view used to obtain lateral measurements of the skeletal structures of the face. It was not possible to secure all five of the X-ray plates on every subject. For the experimental group there were fifteen antero-poster1or plates, and of the lateral views: eighteen of the rest position, twenty of (a], eighteen of [1], and nineteen of [u]. Twelve rest exposures, twenty exposures each of the phonation of the three vowels, and twelve antero-posterior exposures were obtained for the control group. The remaining plates were not utilized due

PAGE 31

to faulty exposure or developing techniques. In addition, it must be remarked that of the one hundred and seventy-four plates avail a ble for this study, many were unavoidably over-exposed or underdeveloped, obliterating much of the soft tissue. Thia condition was 24 a res u lt of variations in electric out p ut, and in some instances the equipment appeared to be faulty. Various colored gelatins were used to clarify the structures illus trated by the under-developed or over-exposed plates. Straw, amber pink, and blue gelatins, or a combination of any two of these, were tried experimentally in viewing the p lates. It was found that frequent chan g ing of the gela tins was necessary in order to obtain the best results. This p rocedure tended to rectify the faulty exposure or development. The accuracy of the tracings was checked in various ways. Five subjects, randomly selected, were com pletely traced two times and the tracin g s were tested b y supe rp osition of one upon the other. At various times during the tracing process the researcher and his super visory chairman compared random tracings against the films from which they were drawn. In addition, all the tracings from th e lateral exposures of each subject were superimposed to assure internal consistency in the outlining of the bony structures Where discre p ancies were found, the films were

PAGE 32

re-examined and the faulty tracing was corrected. The small number of corrections necessary indicated a high degree of accuracy in the original tracings. Control .2f Vowel Phonationa, Intensity,~ Pitch The subjects in each group were instructed to phonate the vowels indicated in the preceding section. Whenever possible, a practice period was held prior to 25 the X-ray exposures. For the most part, the subjects had little difficulty producing good examples of these vowels. There is some evidence in the literature to indi cate that the positioning and movement of the ore-pharyngeal structures may be affected by vocal pitch and intensity {8, 21). Buck (3), however, reported that these two variables produced relatively minor effects on the positioning of the oral structures during vowel phonation, as long as the vari a tions 1n pitch and intensity were not large. The subjects in the present study were instructed to maintain comfortable pitch and intensity levels during phonation when the X-ray exposures were being made. No instrumental control was available to determine maintenance of the criteria. Measurement of the X-ray Plates The procedure for securing measurements from the X-ray plates was basically that used by H ixon {20) and

PAGE 33

26 Buck (3). Each plate was placed on an X-ray tracing table especially constructed for this study. Flashed-opal frosted glass and varyin g sources of light of fifteen, twenty-five, sixty, and one hundred watts were used. A clear acetate tracing paper and a hard-lead drawing pencil were used to trace the outlines of the structures. The sella turcica, the nasal bone, and that portion of the frontal bone just superior to the nasion were traced to be used as orientation lines when comparing different lateral exposures of the same individual. The outline of the hard and soft palates, the anterior nasal spine, the central incisors, the symphysis of the mandible, the posterior wall of the pharynx, the tubercle of the Atlas, the midline of the tongue, and the lips were traced. To construct the Frankfort line, the inferior margin of the bony orbit and the outline of the ear rod were traced. The Frankfort line was used as a base line for the contruction of all pe rpendiculars and parallels con structed for the various measurements of the study. The incisal reference line was drawn tan g ent to the superior margin of the central incisors of the mandible. This reference line intersected the posterior pharyngeal wall and was parallel to the Frankfort line. When the superior surface of the ear rod was obscured by the armature of the head positioner on some of the plates, it was necessary to

PAGE 34

27 reconstruct the actual dimension of the ear rod on the tracings. In a few instances, the tubercle of the Atlas was also partially or totally obscured by this armature. Whenever either of these effects was found, the obscured parts were constructed, by reference to the visible parts and by reference to the dimensions of unobscured ear rods and tubercles of the Atlas. Since the ear rods were of a standard eight millimeters in diameter, they were rela tively simple to construct. The tubercle of the Atlas can be expected to vary in size according to a normal distri bution. Therefore, ten of these structures which were clearly defined were selected at random and measured. The means of their heights and widths were taken, and those tubercles of the Atlas which were constructed, were con structed according to these means. Measurements on those subjects for whom the tubercle of the Atlas had to be constructed will be discussed in detail 1n a later section. Hagerty, tl ~(15), used a series of lines on the posterior pharyngeal wall to determine tissue movement at the level of the tubercle of the Atlas. The modifi cation of this technique used by Knobeloch (24) was also utilized 1n the present study,!~, a perpendicular line was constructed tangent to the most anterior projection of the tubercle of the Atlas. This line was extended to intersect the inciaal reference line and served as the

PAGE 35

28 posterior border of the area. A horizontal line, level 4, passed through the most anterior projection of the Atlas tubercle. At inter v als of one-half centimeter, other hori zontal lines were drawn superiorly and inferiorly to this level. The remaining lines were numbered 1n descending order, level l being the most superior line. Figure l is a master tracing of an X-ray plate of a normal subject. This figure shows the reference points and location of measu r em e nts used for all of the tracings 1n the study. The two reference lines already discussed are designated as follows: the Frankfort line, A'C', ex tending from the inferior mar g in of the bony orbit to the superior surface of the ear rod; and the 1ncisal reference, LP, extending from the mandibular incisors to the posterior pharyngeal wall. Measurements for the prese n t study will be classi fied as follows: linear measurements, measure m ents of openings, area measurements, measurements of movement from rest to phonation, and skeletal measurements of the face. They will be discussed 1n that order, with reference to Figure l. Linear Measurements The linear measurements that were made along the various lines include the following distances: FG. the distance from the termination of the hard palate to the

PAGE 36

29 I / ORBIT ~laj I 0Ni2-"'" / I MN y 5' -""------I:::+ -,-----~c'4-:>~-5U~SPINAL[ 2. / t? \ ) BODY 1--lYO I D fAR ROD ~ r '--'? G TVBf R Cl [ or THr I ATLA.S ) I I 0~ BONE. Figure 1. Master tracing of a typical lateral X-ray plate, shoving reference points and locati o h of measurements A'C' Frankf o rt line LP Incisal reference line FG T ermination of the hard palate to the posterior pharyngeal wall LP O ral cavity length MO Tongue length on the incisal reference line BI Frankfort line to the high point of the tongue HI Hard palate to the high point of the tongue IN Tongue height above the incisal reference line LN H orizontal position of the high point of the toncue OP Pharyngeal opening on the incisal reference line 1-6 Levels used for o btaining measurements of movement of the posterior pharyngeal wall

PAGE 37

30 posterior pharyngeal wall; LP, the length of the oral oavity; M O, the length of the tongue on the incisal reference line; BI, the distance between the Frankfort line and the high point of the tongue; H I, the distance between the hard palate and the high point of the tongue; IN, tongue height aboye the incisal reference line; and LN, the horizontal position of the high point of the tongue 1n the oral oavity. Each linear measurement was made twice. In case of a dis crepancy, the measurement was rechecked and recorded. Measurements .2f Open1n&s Measurements of openings are as follows: OP, pharyngeal opening on the incisal reference line; lip opening, the smallest opening between the lips; and the smallest pharyngeal opening, the smallest distance between the posterior border of the tongue and the posterior pharyngeal wall (above the body of the hyold bone and below the incisal reference line). Velo-pharyngeal opening is the smallest distance between the posterior surface of the velum and the posterior pharyngeal wall. Incisor opening is considered to be the distance between the parallel lines drawn tangent to the maxillary and mandibular incisors. In an effort to substantiate this measurement, the distance BN, extending from the Frankfort line to the 1ncisal reference line, was also measured. The procedure used to check linear measurements was employed in the case of measurement of openings.

PAGE 38

31 Measurements Area measurements for the study were made by means of a planimeter. The three areas measured were the total tongue area above the 1nc1sal reference line, the tongue area anterior to the high point of the tongue, and the area of the posterior pharyngeal wall. Each area was traced three times and the average of the three measurements was recorded. This precaution was taken in an effort to control variations occurring while the areas were measured. The planimeter is a sensitive instrument, and it was difficult to trace the outline of an area without varying from the lines. Measurements of Movement from Rest to ~honation ----------_.......,..,_,.,.,......,..,. Measurements of movement from rest to phonation were made along the parallel and horizontal lines constructed in reference to the tongue and the posterior pharyngeal wall. The vertical movement of the high point of the tongue was made on line BN, the distance IN being variable. Horizontal movement of the high point of the tongue was made on the incisal reference line, the variable distance being LN. Movement of the posterior pharyngeal wall was made on the lines constructed in the area of the tubercle of the Atlas. Unfortunately, the same area demarcated by the most superior and the most inferior of these lines was not visible on all of the lateral X-ray plates.

PAGE 39

32 Skeletal Measurements of the Face _____________________ ......, JK, the length of the maxilla was measured on the incisal reference line. This measurement is the distance between perpendicular lines dropped tangent to the notoh just inferior to the anterior nasal spine (the subsp1nale), and to the termination of the hard palate. LP', the length of the mandible, was also measured on the incisal reference line. This is the distance between the mandibular inoisors and the perpendicular dropped tangent to the most anterior projection of the tubercle of the Atlas (line CP' ), a measurement between stationary bony structures at rest. Figure 2 is a master tracing of a typical antero posterior X-ray plate. This view was used 1n the measurement of the width of the maxilla and the width of the mandible. These plates were not traced, but were placed on the illumi nated tracing table, and the measurements were taken directly from them. There were two reasons for making these measure ments directly from the X-ray plates: {l) since only two dimensions were required from them this was the simplest method of measurement, and (2) these dimensions were to be used as temporary estimates of maxillary width, until such time as the plaster casts made of the subjects should become available. Figure 2 is included only to remove any con fusion as to what measurements were taken. With reference to Figure 2, the width of the maxilla was measured between

PAGE 40

I I ,,__/ BON Y ORBiT ( "--::i__.c:---i~s-M A X It LA ---' X. -~ -, ~ I\ X. f ,,vt I(\'---J (~ \ y l _. 1 '(' ; \ ; __sMAN D I BLt / ------------Figure 2. t--19.ster tracing of a typical antero-poeterior X-ray plate, indicating the structures and the poi n ts fro m which all antero-posterior mea s urements were taken XX I Widt h of t he maxilla at the level of the hor i zontal plates YY. 1 Width of the mandible at the an g le of the ram u s 33

PAGE 41

34 points X and X 1 This represents the widest extent of its bony margin at the level of the horizontal plates of the maxilla. The mandibular width was taken as the distance between Y, the an g le of the ramus on the left, and Y', the an g le of the ramus on the ri g ht. The accuracy of these skeletal m easurements was checked in the same way as were the linear measurements of the soft tissues. Each rmasurement was taken twice, and the two measurements were compared. When a discrepancy appeared the dimension was rechecked and correctly recorded.

PAGE 42

CHAPTER III RESULTS The basic purpose of this study is to make a com parison between a group of normal subjects and an experi mental group having incomplete, unrepa1red cleft palates. A comparison will also be made between the experimental subjects of this study and the subjects with complete, unrepaired cleft palates reported by Knobeloch (24). 3 Facial and Oral Cavity Dimensions The measurement of the bony structu res ot the face and the oral cavity is an important concern of this study. Graber (12), Krogman (25), Slaughter and Brodie (35), and Buck (4) have al l presented evidence which tends to indi cate that early surgical repair of cleft palates may retard normal facial growth. The first section of this chapter will present a comparison of facial and oral cavity dimen sions between the unrepaired cleft palate group and the normally structured control gr oup. Other sections will 3 Throughout the remainder of this st~dy 1t 1s to be understood that the comparisons made with the measurements oonoern1ng complete, unre p aired clefts were those obtained from the analysis conducted by Knobeloch. 35

PAGE 43

36 present data from measurements in the following order: tongue size, horizontal tongue carriage, pharyngeal openings, vertical tongue carriage, tongue movement, velo-pharyngeal opening, posterior pharyngeal wall measurements, and lip and jaw opening. T he statistics used in the tables in this chapter include: (a) the arithmetic mean, (b) the standard deviation, and (a) the standard error of the mean. Also shown is the difference between the means for the two groups for each measurement, the standard error of this difference, and the critical ratio or "t" statistic. The values are given for "t" required for statistical significance at the 5.0 per oent and at the l.O per cent levels of confidence. The format for this presentation is t (df 38): l. ~ 2.711; 5. ~ 2.025. In this context df designates the degrees of freedom for a particular set of measurements. Differences below the 5.0 per oent level of confidence might be largely due to chance, and were not considered. Maxilla Length,!!! Width Table l presents data related to the skeletal measurements of the face. The mean length of the maxilla for the experimental group of this study was shorter than that for the control group by 5.04 millimeters, a dif ference which was significant beyond the 5.0 per cent level

PAGE 44

1 2. 3. 4. 5. 6. 37 TABLE 1 MEASUREMENTS OF MAXILLARY, MANDIBULAR, AND ORAL CAVITY STRUCTURES AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES (All measurements in millimeters) Normal Cleft Pa.late Differences Maxilla length M 50.08 M 45.04 M 5.04 SD 5.02 SD 5.62 SE 2.01 SE 1.51 SE 1.33 t 2. 507* t (df 29): 1.0% 2.756; 5.0% 2.045 Maxilla \ddth M 63.58 M 68.50 M 4.92 SD 5.94 SD 5.07 SE 2.24 SE 1.79 SE 1.36 t 2.196it t (df 25): 1.0% 2.78?; 5.0% 2.060 Mandible length M 80.92 M 77.19 M 3.73 SD 7.55 SD 7.98 SE 2.99 SE 2.27 SE 1.94 t 1.247 t (df 28): 1.0% 2.763; 5.0% 2.048 Mandible width M 97.1? M 102.43 M 5.26 SD 3.07 SD 7.69 SE 2.25 SE .92 SE 2.06 t 2.338* t (df 25) Oral cavity length M 83.66 M 78.72 M 4.94 on incisal reference SD 7.10 SD 6.10 SE 2.60 line SE 2.14 SE 1.48 t 1.900 t (df 28) Termination of M 27.83 M 28.00 M .17 hard r,e.la te to SD 3.90 SD 4.86 SE 1.65 posterior pharSE 1.17 SE 1. 78 t .103 yngeal wall t (df 28) *Significant beyond the 5.0% level of confidence

PAGE 45

38 of confidence. That is, there were fewer than five possi bilities out of one hundred that this difference was due to ohance. Graber (12) and Buck (4) found short maxillae among their subjects with repaired cleft palates. They felt this effect might be a result of early surgery. It should be noted, however, that the maxillary development of the present experi mental group does not appear to be like that of the repaired cleft palate subjects reported by Graber {12). He noted a concave facial angle (from nasion to subspinale to gnathion) which resulted from a ge nerally restricted maxilla. In con trast, the experimental group of the present study appeared to have flat facial angles like normal subjects. This would indicate that the shorter maxilla length of the group was probably not a result of a generally restricted development, but more likely a failure in posterior growth. For the cleft palate group, the mean maxillary width was 4.92 millimeters greater than that for the normal group, a difference which was significant beyond the 5.0 per cent level of confidence. Again, this indicates that there are fewer than five possibilities out of one hundred that this difference is due to chance. This is 1n direct contrast to the smaller maxillary width measurements found among those with repaired cleft palates (12, 4). It appears to support Krogman's (25) statement to the effect that unrepaired cleft structures can be expected to follow

PAGE 46

normal growth patterns, but modified by the anomaly. In other words, a hard palate with an unrepaired cleft may tend to grow to wider than normal dim e nsions. When the measurements of the inoomplete, unrepaired cleft palates 39 were compared with those ot the complete, unrepa1red cleft palates, there were no statistically significant differences. It was mentioned in Chapter II that one of the subjects in the experimental group had undergone surgical repair of the palate. She (Subject No. 206) was included because the surgical suture had pulled apart, indicating that the dy namic processes of facial growth had probably not been retarded by the operation. The measurements of maxilla width support this assumption. For the experimental group the mean maxilla width was 68.50 millimeters, with a standard deviation of 5.07. The maxilla of this subject was 67.0 millimeters, placing her 1n the mid-range of the unrepaired cleft palate group. Mandible Length~ Width The subjects of the experimental group did not differ significantly from the control group in mandibular length. This was also true of the subjects having complete, unrepaired palatal clefts. In contrast, the repaired cleft lip and palate subjects studied by Buck (4) were found t o have significantly shorter mandibles. This difference may have been due to restricting pressure from t h e re p aired

PAGE 47

40 ob1cular1s oris muscle which encircles the mouth (12). In the measurement of mandibular width, the experi mental group followed the trend seen in their maxillary width. The mandibles of the cleft palate group were wider, on the average, than those of the normal group by 5.26 millimeters, a difference which is significant beyond the 5.0 per cent level of confidence. Such measurement com parisons may be seen in Table l, item 4. A similar, though not statistically significant, trend was found among the mandible widths of the complete, unrepaired cleft palate subjects. It 1s a question whether the mandibles of the cleft palate subjects were wider due to mechanical pressures from the tongue and certain facial muscles (37), or due to neurological changes which may have been imposed by the cleft .Q!:tl Cavity Length The oral cavity length is defined here as the distance from the mandibular incisors to the posterior pharyngeal wall, measured along the incisal reference line. This dimension would be directly affected by the mandible length. Among the cleft palate subjects of this study, mandible length was not markedly different from that of the normal group. As expected, then, the measurement of oral c avity length revealed no significant difference between the normal and the experimental group This had also been true

PAGE 48

41 of the complete. unrepaired cleft palate group. An ter:_o-_posterior Measurements pf ~~-.er~_;'T mx The distance measured horizontally from the termi nation of the hard palate to the posterior pharyngeal wall was used as the antero-posterior dimension of the naso pharynx. For the two groups of the present study this di mension is strikingly similar. This similarity is especially remarkable when it is remembered that the cle!'t palate subjects had a significantly shorter maxilla. It mi c ht have been expected, then, that among the cleft palate subjects this measurement of the naso-pharynx would have been sig nificantly lar g er. It is possible that the shorter maxillae provided a more anterior attachment for the palatopharyngeus and superior constrictor muscles. This would have displaced the posterior pharyn~eal wall anteriorly. Thus, a naso pharyngeal opening of normal proportions might be mainta1nedo There were no significant differences on this dimension between the experimental group and the group with complete, unrepaired clefts. M easurements Related to Tongue Size and H orizontal Positi on of the H igh Point of the Tongue l'wo measure men ts of midsagi t tal tongue area were ma.de with a planimeter: (1) the total tongue area above the incisal reference line and (2) that part of the tongue area an terior to the high point of the tongue. As shown 1n Table 2,

PAGE 49

1 2. 42 TABLE 2 MIDSAGIT T AL T ON GUE AREA A N D H O RIZONTAL POSITION O F HIGH POINT OF T O NGUE ALONG INCI SA L R E FE R ENCE LINE AMO N G S lJ BJEC T S WITH N OR MA L DEVELOPMENT AND SUBJECTS WI'l 'H INCOMPLE'IE UNREPAIRED CLEFT PALATES Rest ( a] ( i] ( u] Total tongue Normal area above M 11.02 10. 7 8 9.,28 9o26 incieal referSD 2.,4.3 3.63 1.76 2.94 ence line; in SE 73 .83 .40 .67 square centiCleft Palate meters M 10.01 12.79 8064 10.96 SD 4.,29 3 68 2.93 3.,0.3 SE 1.04 084 71 .71 Dif f erences M 1.01 2.01 .64 1.70 SE 1.27 1.19 .81 .98 t .795 1.689 790 1.735 Tongue area Normal a nt e rior to M 6.,34 6.02 4.52 4.94 high point of SD 1.75 2.22 1.06 1.69 tongue; in SE 53 51 24 .39 s q uare cent i Cleft Palate meters M 5.41 7.15 4.,11 6.11 SD 2 .37 2.41 1.63 1.90 SE .58 .55 .40 .45 Differences M .,93 1.1.3 .41 1.17 SE 1.89 .75 .47 59 t .492 10507 .872 1.983 Rest t ( df 28) : 1.0 % 2 76 3; 5.0 % 2 0 4 8 [ a ] t (df .38): 1.,0% 2.711; 5.0% 2.025 [ i] t (df .36) : 1.0% 2.718; 5.0% 2.027 [ u] t (d f 37 ) : 1.0 % 2.718; 5.0 % 2.027

PAGE 50

43 TABLE 2 Continued Rest [ a) ( i] [ u ] 3. Horizontal Normal distance from M 43016 48.30 32.65 49.15 high point of SD 4.76 4.81 4.52 4.48 tongue to manSE 1.43 1.10 1.04 1.03 dibular central Cleft Palate incisors J in M 38094 49043 28072 52.55 millimeters SD 5.25 5.83 5.02 6024 SE 1.27 1.34 1.22 1.47 Differences M 4.22 1.13 3.93 3o40 SE 1.92 1.73 1.60 1.79 t 2.198* .653 2.456* 1.899 40 Tongue length Normal on incisal M 66.25 59e20 52.90 49050 reference SD 6.17 7.92 6.28 6.94 line; in SE 1.86 1.82 1.44 1.59 millimeters Cleft Palate M 62055 56030 55.17 49.82 SD 8.29 8.18 6.15 7.22 SE 2.01 1.88 1.49 1.70 Differences M 3. 70 2.90 2.27 .32 SE 2.74 2.61 2Q08 2.33 t 1 .350 lolll 1.091 .137 Rest t (df 28): 1.0% 2.763; 5.0% 2.048 [ a] t (df 38): 1.0% 2.711; 5.0% 2 .025 [ i] t (df 36): la0% 2. 718; 5.0% 2.027 ( u] t (df 37): 1.0 % 2. 718; 5.0% 2.027 *Significant beyond the 5.0% level of confidence

PAGE 51

44 there were no statistically significant differences found between the control group and the ex p erimental group re garding measurements of midsagittal tongue area. This re sult is essent ia lly the same as reported for the complete, unrepai re d cle ft palat e gr oup. As Knob e l o c h (24) pointed out, this is in contrast to the smaller m1dsag1ttal tongue areas found by H ixon (20) for normally structured nasal sp e akers, and by Buck {4) fo r re pa ire d cleft pal a te s peakers. Tongue Length~ H orizo n tal P osition Table 2 further shows that comparisons of the linear measurements of tongue length on the inoisal refer ence line between the control group and the experimental group revealed no statistically significant differences. H owever, in the horizontal carriage of the tongue it was found that the experimental subjects, like the subjects 1n the complete, unrepa1red cleft palate group, carried their tongues more anteriorly at the rest position and during the vowel [1]. Both of these differences were significant beyond the 5.0 per cent level of confidence. Also, like the complete, unrepaired cleft group, the cleft palate subjects of the present study exhibited a slight tendency toward a more posterior tongue position during the vowel [u]. The horizontal position of the tongue during the vowel [a] was quite similar 1n the experimental and the control g roups of the present study. The incomplete, unrepaired cleft palate

PAGE 52

45 group and the complete, unrepaired cleft palate group were similar in measurements of horizontal tongue carriage, tongue length, and midsagittal tongue area. Pharyngeal Openings The m1daagittal dimension of the oro-pharynx is a further indication of the horizontal carriage of the tongue at rest and during phonation. This can be understood when it is realized that the oro-pharyngeal opening is defined here as the distance between the posterior pharyngeal wall and the posterior border of the tongue. Thus, it would be expected that as the tongue moves backward this opening would grow smaller, and conversely, when the tongue moves forward this opening would grow larger. Table 3 gives two measurements of the oro-pharynx: the first is taken along the incisal reference line; the second is taken between the incisal reference line and the superior border of the body of the hyoid bone. For both measurements of oro-pharyngeal openings the experimental subjects exhibited smaller dimensions on the vowel [u] than did the normal subjects. For this vowel the two measurements were actually identical among the experimental subjects, because the smallest opening of the oro-pharynx was never below the 1ncisal reference line. The mean differences between the experimental group and the control group ware approximately four and one-half

PAGE 53

1. 2. 46 TABLE J THE PHARYNGEAL OPENINGS ON AND BELOW THE INCISAL REFE RENCE LINE AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEF!' PALATES (All measurements in millimeters) Rest [ a] ( 1] [ u] Pharyngeal openN ormal ing between the M 11.33 8.35 22.10 12.20 tongue and the SD 3.73 3o03 4. 54 4.24 posterior wall, SE 1.12 .70 1.04 .97 on the incisal Cleft Palate reference line M 11.11 7o90 22.72 7.76 SD 3.90 3.85 4o79 5.62 SE 095 .,88 1.16 lo32 Differences M .22 .45 .62 4.44 SE 1.47 1.12 lo56 1.64 t .150 .402 0397 2.707* Smallest pharNormal yngeal o pening M 8.41 7.15 19090 12.05 between the SD 4.58 J.20 5.21 4.25 incisal referSE lo38 73 1.20 .97 ence line and Cleft Palate the superior M 8.11 7o85 16.58 7.76 border of the SD 3.12 3.89 4.93 5.62 body of the SE .76 89 1.20 lo32 hyoid bone Differences M .30 0 70 3.32 4.29 SE 1.,50 lol6 1.69 1.65 t .200 .603 1.964 2.600* Rest t (df 28): 1 0% 2.763; 5.0% 20048 [ a) t (df 38): 1.0% 2.711; 5.0% 2.025 [ i] t (df 36): 1.0% 2.718; 5.0% 2.027 [ u] t (df 37): 1.0% 2. 718; 5.0% 2.027 *Significant beyond the 5.0% level of confidence

PAGE 54

47 millimeters. In both instances this difference was sig n ificant beyond the 5.0 per cent level of confidence. No other significant differences were found among the measure ments presented 1n this table. 1he control and the experi mental groups had very similar ore-pharyngeal openings for both the rest and the (a] pos1t1ons~ J This appears to c on firm a part of the data discussed in the previous section, that the experimental group tended to maintain a more pos terior placement of the tongue during the vowel [u], as well as a placement similar to that of the normal group during the vowel (a]. Like the experimental subjects of the present s tudy, the complete, unrepaired cleft palate subjects tended to have s malle r oro-pharyngeal openings for the vowel [u] than did the control group, although the tendency was not great enough to reach statistical significance. For the vowel [1], these subjects had a significantly larger oro-pharyngeal opening than did the control group. It is interesting to compare the measurements of oro-pharyngeal opening of the two unrepaired cleft palate groups during the phonation of [1]. In examinin g the inferior pharyngeal opening measurements, a difference significant beyond the 5.0 per cent level of confidence was found between the two cleft palate groups. The incomplete cleft palate group was more like the normal group than like the group with

PAGE 55

48 complete, Wlrepaired clefts. 'rhere appears, then, for the first time a real difference between the Wlrepaired cleft palate groups. They differ from the mean of the controls in opposite directions. An a p parent incongruity may be observed here, since it was previously noted that the high point of the tongue was significantly farther forward among the incomplete cleft palate subjects on the vowel [1]. It would be expected that the pharyngeal opening for this vowel might be wider rather than narrower. The probable explan ation is that during the phonation of [1], the subjects with incomplete clefts characteristically maintained two concen trations of tongue tissue. There was an anterior concen tration beneath the hard palate, and a posterior concentration projecting horizontally at the tongue base. Inspection of the X-ray tracings supports this statistical interpretation. It might be speculated that the function of this particular tongue shape was an involuntary result of anatomical and linguistic adjustments. It must be stated that oonolusions drawn from these measurements can only be tentative. M ention has already been made of the poor definition in some of the X-ray plates. At times the tracings could only be a compromise among two or more possibilities. It appears, however, that the tongue carriage cannot be described by a single measurement. For the vowel [a] no significant mean difference in tongue

PAGE 56

49 placement appeared between the groups in either Table 2 or Table 3. Finally, there was a definite tendency toward a more posterior tongue carriage among the experimental group for the vowel (u]. Vertical Tongue Position In order to adequately describe the vertical posi tion of the tongue at rest and during vowel phonation, the high point of the tongue was related to three different lines. First, a measurement was made of the maximum tongue height above the incisal reference line. Second, the dis tance was measured between the high point of the tongue and the inferior surface of the hard palate. Third, the dis tance from the Frankfort line to the high point of the tongue was measured. These three dimensions were expected to yield relatable, though not identical, information. Tongue Height Above~ Incisal Reference~ As seen in item l of Table 4, there was a tendency for U1e incomplete oleft palate group to display somewhat lower than normal tongue positions at rest and during the production of the vowel [1], and somewhat higher than normal tongue positions during the production of vowels [a) and [u]. These measurements for item 1 are measured verti c ally above the 1nc1aal reference line. The mean differences are sig nificant beyond the 1.0 per cent level of confidence for the

PAGE 57

1. 2. 50 TABLE 4 VARIOUS VERTICAL -rEASUREHENTS OF TONGUE CARRIAGE A M ONG SUBJECTS WITH NORMAL DEVELOFMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES (All measurements in millimeters) Rest [a] [ i) ( u ] Distance of Normal high point of M 23.00 24.80 23 .. 80 24.10 tongue above SD 3.13 5.78 2.98 5.14 the incisal SE .94 1.33 .68 lol8 reference line Cleft Palate M 21 .. 58 30048 21.61 27.63 SD 7.30 7.06 5.06 6006 SE 1.77 1.62 1.23 1.43 Differences M 1.42 5.68 2.19 2.93 SE 2.03 2.09 lo,4.0 1.85 t 0700 2.718** 1.564 1.584 Distance of Normal high point of M 6.66 12.85 3.60 5.05 tongue below SD 2.77 3.18 1.57 2.87 the hard SE .8.3 0 73 .36 076 palate Clett Palate M 9.47 15040 11. 50 9.08 SD 5o18 5.70 5.12 5.24 SE 1.26 1.31 1.24 1.24 Differences M 2.81 2.55 7.~ 4.03 SE 1.51 1. 50 1.29 1.40 t 1.861 1.700 6.124 3.474** Rest t (df 28)1 1.oi 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2. 711; 5.0% 2.025 [ 1] t (df 36): 1.0% 2.718; 5.0% 2.027 [ u] t (df 37)1 1.0% 2o 718; 5.0% 2.027

PAGE 58

51 TABLE 4 Continued Rest [ a] [ 1] [u] 3. Distance from Normal the Frankfort M JO.OS 41.05 31 45 33.40 line to high SD 3.40 5.61 3.97 5.10 point of SE 1.02 1.29 .91 1.17 tongue Cleft Palate M 34.39 39.78 37.00 33.76 SD 5.46 6.01 6.02 5 .. 71 SE 1.32 1.38 1.46 1.35 Differences M 4.31 1.27 5.55 .J6 SE 1.68 1.88 l. 72 1.78 t 2.566* .148 3.099** .202 Rest t (df 28): 1.0% 2.763; 5.0% 2.048 [a] t {df 38): 1.0% 2 .. 711; 5.0% 2.025 [ i 1 t (df 36): 1.0% 2.718; 5.0% 2.027 [ u] t (df 37): 1.0% 2.718; 5.0% 2.027 *Significant beyond the 5. 0 % level of confi d ence **Significant beyond the 1.0% level of confidence

PAGE 59

vowel [a]. Such results are in essential agreement with those reported for the complete cleft palate group. It 5 2 is interesting to note here that Hixon's (20) nasal subjects with normal structures had a tendency toward greater tongue height. In addition, the repaired cleft palate subjects studied by Buck (4) maintained a lower tongue than did the normal subjects with whom they were com p ared. Although all of these experimental groups may possess nasal voices, the nasality apparently cannot be explained solely by tongue usage without a careful description of the oral cavity. Distance~ the .fil:ah Point .2. Tongue Below~ Palate Item 2 of Table 4 represents the distance of the high point of the tongue below the hard palate. Thia prob ably should not be accepted as a reliable measure of abso lute tongue height. Towards the front of the mouth, the curve of the palate as it extends posteriorly would be ex pected to compli c ate the measurement of tongue height. And .for the back vowels, (a] and (u), the high point of the tongue was frequently behind the termination of the hard palate. The distance measured for [a] and [u] was an approximation therefore acquired by extending the line of the inferior palatal surface. This dimension is, however, useful as a statement of the relationship between palato lingual tissues. The present experimental group, as well as those with complete, unrepaired clefts, maintained a

PAGE 60

53 wider separation between tongue and palate than did the con trol group at rest and during all three vowels. For the experimental group, the difference from the control group was significant for the vowels [1] and (u] beyond the l.O per cent level of confidence It is interesting to note that Buck (4) reported similar findings in his study of sub jects with repaired cleft palates. At rest, and for the vowels [ae] and [u], his subjects also maintained a greater than normal separation between the tongue and the hard palate. Distanoe 2 Frankfort~ 12. .!!!sh Point ,2! Tongue Tongue height may also be measured by taking the dis tance between the Frankfort line and the high point of the tongue. This is perhaps the most reliable of the three dimen sions outlined 1n Table 4. The Frank.fort line does not move up and down as does the incisal reference line and it 1s not curved as 1s the hard palate. Me a surements of this dimension indicate that the cleft palate group of this study maintained a greater than normal distance between the Frankfort line and the tongue at rest and during the vowel [1]. As will be demonstrated in a later section, there was a slight tendency for this group to maintain a wider jaw opening for all posi tions studied. The effect of this wider jaw opening on the measurement of tongue height was offset by the tongue's being raised for [a] and [u], and increased by the tongue's being held lower for rest and [1], as shown in Table 4, item l.

PAGE 61

64 Thus it may be seen that the measurement of tongue hei g ht is not a simple one. It appears to be the confounding factor of jaw opening which accounts for the discrepancies between item land item 3 of Table 4. The two cleft palate groups were also compared on measurements of tongue height. During phonation of the vowel [u), a difference significant beyond the 5.0 per cent level of confidence was noted between the cleft palate groups. The subjects with incomplete clefts tended to maintain a greater distance between their tongues and their hard palates. For the vowel [1] the experimental group, with incomplete clefts, maintained ap p roximately five and one-half millimeters greater separation between their tongues and their hard palates than did the members of the g roup with complete clefts. This difference was significant beyond the 1.0 per cent level of confidence. This supports the observation made previously that the members of the ex p erimental group tended to have both an anterior and a posterior concentration of tongue tissue during the production of (1). It seems that because of the tissue concentration formed near the base of the tongue, less tissue would be available to form a tongue elevation beneath the hard palate. Hence, the hig h point of the tongue was lower for the incomplete cleft palate group. Similarly, the distance between the Frankfort line

PAGE 62

55 and the high point of the tongue was greater among the ex perimental group than among the group with complete, unre paired clefts for the vowel [1]. This difference was sig nificant beyond the 5.0 per cent level of confidence. No other statistically significant differenc e s were found between these groups on these measurements. Excursion of the Tongue from Rest to Vowel Positions The measurements of tongue movement, shown in Table 5, were secured by computing the means of the movement from rest to phonation for each subject. It will become immediately apparent upon examination of the table that relatively large standard deviations have been computed for the means. Buok (4) had observed a similar effect in his measurements of tongue movements. He felt that the measure ments were complicated by variations in the tongue positions at rest and by variable jaw positions during the production of vowels. The same complicating factors would be expected to increase the range of values for the measurements of both vertical and horizontal tongue movement. That is, these values vary in direction as well as in magnitude. Vertical Tongue Movement Item l of Table 5 indicates that the two groups in the present study varied little in the extent of vertical tongue movement from rest to the vowel [1]. For the vowels

PAGE 63

1. 2. 56 TABLE 5 MOVEMENT OF THE TONGUE FROM REST TO VOWEL POSITIONS AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEFT PALATES (Negative numbers indicate a downward directiJn to vertical movement and a posterior direction to horizontal movement. All measurements in millimeters) [ a] [ 1) ( u] Linear Normal measureM 2o16 1.00 2Ql6 ments of SD 5.68 3.16 3.88 vertical Range -5.0 to 13.0 -6.0 to 6.0 -5 o 0 to lOoO movement SE 1.71 095 1.17 of the Cleft Palate tongue M 7o83 -062 5.65 SD 6028 6.(,0 6 55 Range -3o0 to 22.0 -15 5 to 10.5 -6.0 to 17.5 SE 1. 52 1. 70 lo64 Differences M 5o67 1.62 3.49 SE 2.29 1.95 2o01 t 2 0 476* 831 1.736 Linear Normal measureM -4.91 10.66 -5.41 ments of SD 5.37 3 94 4.23 horizonRange -15.0 to 7 .O 5o0 to 17o0 -11.0 to 2o0 tal SE 1 .. 62 1.19 1.,27 movement Cleft Palate of the M -10058 9.91 -13.91 tongue SD 5.68 4.89 5.53 Range -21.5 to 4.0 3o0 to 18o0 -21.0 to -400 SE 1.38 lo26 1.38 Dif'ferences M 5.67 .75 8. 50 SE 2.13 1.73 1.88 t 20662* .434 4. 521** ( a) t (df 28): lo0% 2.763; 5.0% 2.048 [ 1) t (df 26): 1.0% 2.799; 5.0% 2.056 ( u] t (dr 27): LO% 2. 771; 5.0% 2.052 *Significant beyond the 5.0% level of confidence **Significant beyond the 1.0% level of confidence

PAGE 64

57 [u] and (a] the experimental group tended to have greater vertical movement than the normal group. For [a] the dif ference was significant beyond the 5.0 per cent level of confidence. These results very olosely agree with those reported for the complete cleft group for this measurement. The greater upward tongue movement seen 1n the cleft palate subjeots for the vowel (a] may be understood in relation to other movements and positions. It will be recalled from the previous section that the two groups began at the rest position with approximately equal tongue height above the incisal reference line. It will also be recalled, however, that at the rest position the tongues of the experi mental subjects were significantly farther below the Frank fort line. As has been mentioned, there was a tendency for the experimental subjects to maintain a wider jaw opening for all positions studied. Thus the experimental subjects had to raise their tongues higher to approach the normal palato-lingual division of the oral cavity which they achieved. In absolute terms, the greater amount of vertical tongue movement displayed does not appear to be induced so much by the void in palatal structures, as Knobeloch (24) suggested, but by a tendency toward a wider jaw opening. An examination of the range of movements for the two groups of the present study reveals a consistently greater variability among the experimental subjects. This,

PAGE 65

58 as Knobeloch (24) indicated, is most probably directly af fected by the oral anomaly. It appears that persons must compensate for their abnormal structures in individual ways in order to produce acceptable speech results. H orizontal Tongue Movement Item 2 of Table 5 indicates the amount of chan g e 1n horizontal tongue position from rest to each vowel phonation. The experimental group and the normal group differ very slightly 1n this dimension for the vowel (1]. 'rhe experi mental g roup shows greater than normal posterior movement for the other two vowels, the difference for (a] being sig nificant beyond the 5.0 per cent level of confidence, and for (u] significant beyond the l.0 per cent level of confi dence. When the data presented earlier are recalled, these results are precisely what should have been expected. The cleft palate group carried their tongues farther forward at the rest position. Therefore, a normal amount of movement should leave their tongues farther forward for the [1] position. A somewhat greater than normal amount of posterior movement for (a] should leave their tongues normally placed for that vowel. A still greater than normal amount of move ment for [u) should be necessary to achieve a more pos terior p osition for that vowel. Such expectations assumed that the tongues of the two groups would tend to move in the same direction for each phonation, and this has

PAGE 66

59 apparently occurred. The range of movements for the two groups did not differ so greatly in the horizontal dimension as they did in the vertical dimension. The inoomplete, unrepaired cleft palate group was compared with the complete, unrepaired cleft palate group in measurements of tongue movement. No significant dif ferences were found here b etween these two groups. Measurements of the Naso-pharyngeal Structures No attempt was made to measure the vertical or horizontal shift of the velum from rest to phonation for two reasons. First, the exact outline of the velum was more poorly defined than that of most structures traced. Secopdly, such movement as could be identified among the experimental subjects was very slight and inconsistent from one vowel to another. The following discussion of velo pha.ryngeal openings, therefore, should be considered as highly tentative. Velo-pharyngeal Opening Item l of Table 6 shows that at the rest position the two groups of this study did not differ statistically in velo-pharyngeal opening. For all three of the vowels, however, the cleft palate subjects had consistently greater openings, differing from the normal group well beyond the 1.0 per cent level of confidence. Knobeloch (24), who

PAGE 67

1. 2. 60 TABLE 6 AREA OF THE POSTERIOR PHARYNGEAL WALL AT THE LEVEL OF THE TUBERCLE OF THE ATLAS AND OPENING OF THE VELO-PHARYNGEAL STRUCTURES AMONG SUBJECTS WITH NORMAL DEVELOPMENT AND SUBJECTS WITH INCOMPLETE, UNR E PAIRED CLEFT PALATES Rest [a ] [ 1 ] ( u ] ~llest veloNormal pharyngeal M 9. 83 1.00 0 55 .15 o pening; in SD 2ol7 1 6.2 1.19 .37 millimeters SE .65 .37 027 .08 Cleft Falate M 10.0J 6.22 8.78 7.39 SD 2.96 Jo02 3.67 J.58 SE .n .69 .89 .84 Differences M .20 5o22 8.23 7.24 SE 97 .78 .30 85 t .206 6.692** 27.433** 8. 518** Area of the Normal p o sterior M 1. 6 7 1.7 8 le60 1.5 8 pharyngeal S D .9 2 2.4 2 3 23 wall anterior SE .28 .06 0 5 005 to the tubercle C left Palate of the Atlas; M 1.59 1. 58 1. 51 1. 3 0 in square SD 84 79 23 .1 8 centimeters SE .20 .18 .06 .04 Differences M e08 .2 0 .09 .28 SE .35 .2.4 2.4 .2 2 t .228 833 .375 1.273 Rest t (df 28): 1 0% 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2.711; 5.0% 2.025 ( i ] t (df 36): 1.0% 2.718; 5.0% 2.027 (u] t (df 37): 1.0% 2. 718; 5.0% 2.027 **Significant beyond the 1.0% level of conf i dence

PAGE 68

61 found quite similar results for this measurement, felt that any velar movement could be primarily due to contact with the tongue. He observed that among his subjects for the vowels [a] and [u] the velum appeared to rest on the tongue resulting in a posterior displacement. For [1] the velum hung down ward very closely a pp roximating the rest position. An ex amination of the X-ray tracings of the experimental subjects of the present study yields similar observations in some instances. Some tracings, however, revealed velar movement independent of the tongue, though following the lingual pattern as described by Harrington (17) and Williams (40). Apparently, among some of the subjects, the muscular con tinuity of the palate was only parti a lly disturbed. One exceptional subject in the experimental group was able to achieve complete velo pharyngeal closure on all three vowels. This subject will be individually discussed 1n Chapter v. .f Posterior Pharyngeal !!,ll Hagerty, il .!.! (15) described a technique for measuring the area of the posterior pharyngeal wall at the level of the tubercle of the Atlas. For the present study a very similar area was measured, with the chief exception being that the location of the posterior margin of the area was changed. Whereas Hagerty,~.!!.! had measured forward of a perpendicular line through the mid-point of the tubercle of the Atlas, the present study closely follows Knobeloch's

PAGE 69

6 2 (24) design in placin g the posterior margin of the area tan gent to the most anterior p rojection of the tubercle. Inspection of ite m 2, Table 6 reveals no significant differences between the t w o groups of the present study in regard to this measurement. H agerty, ll & (15) re p orted a small increase in the area measured durin g p roduction of the vowel[~]. The vowel [a] used in the p resent study is roughly comparable, and amon g the control g roup a small in crease in area may be seen for this vowel. 1'he ex p erimental g roup, however, showed an insie;nificant decrease in area of .01 square centimeters. In g eneral, for both g roups the changes 1n posterior p haryn g eal wall area were quite small, No statistically significant differences were revealed whe n the two unrepaired cleft palate g roups were com p ared b y t h is measurement. Posterior Pharyngeal !ill M ovement Measurements of the movement of the p osterior pharyn g eal wall, like the measurements of its area, resulted from an adaptation of the technique em p loyed by H agerty, il .!:.! The horizontal lines alon g which they measured were drawn 1,0 centimeter a p art. They observed, however, that the greatest movements of the p haryn g eal wall fre q uently a pp eared to fall between the lines. Therefore, 1n order to measure the movements more p recisely, for the p resent study lines were drawn 0,5 centimeters apart h orizo n tally across t h e area

PAGE 70

63 to be measured. The most superior of these lines was desig nated level l. The lines below were consecutively numbered down to level 6. Table 7 displays the measurements made of the pos terior pharyngeal wall movements, level by level. Because of the differences in measurement procedures, a detailed com parison with the findings reported by Hagerty, tl !1. will not be made. It should be mentioned, however, that they found very little movement in an anterior direction and none in a posterior direction among their normal subjects. Move ments in both directions were found among the subjects of the present study; therefore, negative numbers were used to indicate posterior movement. No statistically significant differences were ob served at level 1. For [1], the normal group showed no mean movement. All other mean movements in both groups were in an anterior direction. The range or variability of movements for all vowels was greater among the no1 mal subjects. Again, at level 2, there were no statistically s1g n1f1oant differences observed. All mean movements among the experimental group tended to be 1n an anterior direction. Among the normal group the tendencies were towards posterior movement for [ i], no movement for [ u], and anterior movement for [a]. As before, tlie range or variability of movements among the normal group was greater for all vowels.

PAGE 71

l. 2. 64 TABLE? MEASUREMEN'T'S OF MOVEMENT Of THE POSTERI'1R PHARYNGEAL WALL AT THE LEVEL O F' THE 'T'TJBERCJ...E OF THE A 'l"LAS AMONG SUBJECTS WITH NORMAL DEVELO PMENT AND SUBJECTS WITH INCOMPLETE, UNREPAIRED CLEn' PALATES (Negative numbers indicate a posterior movement. All J!leasurernents in millimetere) [a] [ i] [ u] Levell Normal M .33 .oo .67 SD 3.88 3.90 4.32 Range -5.O to 600 -5.O to 5.0 -6.O to 6.0 SE 1.73 lo74 1.93 Cleft Palate M .37 .61 .10 SD 2.23 1.36 1.23 Range -3.O to 6.0 -1.O to 4.0 -3.O to 1.5 SE .58 .38 033 Differences M .04 .61 57 SE 1.83 1.78 1.96 t 0022 .34.3 .291 [a] t (df 2o)i 1.0% 2.845 ; 5.0% 2.086 [ 1 ] t (df 18): 1.0% 2,878; 5.0% 2.101 [ u] t (df 10,. 1 .. 0% 2.8611 5oO% 2.09.3 I Level 2 Norme.l M .8.3 -. 50 oOO SD J.41 2.58 2.41 Range -5.O to 7.0 -3.O to 6.0 -3 .. O to 5.0 SE 1.03 078 .. 73 Cleft Palate M 53 .43 .J4 SD 2.22 lo25 1.20 Range -3.5 to 4.5 -1. 5 to 3.0 -3.O to 2.0 SE 56 .33 .31 Differences M .30 0 9.3 .34 SE 1.17 .84 025 t .256 1.107 1 .360 (a] t (df 27): 1.0% 2.771; 5.0% 2.052 [ i] t (df 25): 1.0% 2.787; 5o0% 2.060 [ u] t (df 26): 1.O-i 2.779; 5.0% 2.056

PAGE 72

65 TABLE 7 Continued [ a] [ 1 ] [ u J 3. Level 3 Normal M .67 -.42 .oo SD 2.02 la44 1.35 Range -3o0 to 4.0 -3.0 to 2.0 -2.0 to 2.0 SE .61 043 .41 Cleft Palate M .33 .34 -088 SD 2.14 lo40 1.78 Range -3.0 to 5.5 -2.0 to 4.0 -4.0 to 1.5 SE 52 .36 .44 Differences M .34 .,76 .88 SE aOO 018 .. 60 t .425 40222** 1.467 4. Level 4 Normal M .83 -.08 .08 SD 1.70 lo93 1.93 Range -3.0 to 3.0 -4.0 to 3.0 -3 .0 to 3.0 SE 51 .58 58 Cleft Palate M .58 .31 -.32 SD lo50 1.27 1.25 Range -1. 5 to 3.0 -2.0 to 2o5 -3.0 to 2.0 SE .36 .33 .31 Differences M .25 .,39 .40 SE .20 .21 .21 t 1.,250 1.857 1.905 [a] t (df 28): 1.0% 2.,763; 5.0% 2.048 [ i J t (df 26): lo0% 2.779; 5.0% 2.056 [u] t (df 27): 1.0% 2.771; 5.0% 2.052

PAGE 73

66 TABLE 7 Continueg [ a ] ( i ] [ u ] 5. Level 5 Normal M .82 .09 .27 SD 1.83 1.14 1.68 Range -3.O to 4,0 -3 0 to 1.0 4.0 to 2.0 SE 58 .36 5.3 Cleft Palate M -.06 -.56 -.88 S D 1.05 1.11 l.JO Range -2.0 to 2,0 -3.0 to 1.0 -4.0 to 1.0 SE 26 .29 .32 Differences M ~8 .65 1.15 SE .20 .14 .20 t 4,400 H 4.64 3** 5. 750** [a ] ( u ] t (df 26): 1.0% 2. 779; 5.0% 2.056 [ i ] t (df 25): 1 .0% 2.787; 5.0% 2.060 6. Level 6 Norml M .28 .14 .oo SD 2.22 1.46 1.97 Range 4.0 to 3.0 -3.0 to 1.0 -4.0 to 2.0 SE .91 06() .80 Cleft Palate M -.25 -.61 -1.32 SD 1.36 2.02 2o00 Range -2,0 tn 3,0 -5.5 t o 3.0 -5.5 to 2,0 SE .41 58 .63 D ifferences M 5.3 75 1.32 SE 1.00 .26 1.00 t .530 2.885** 1.320 [a ] t (df 17): 1.0% 2.898J 5.0% 2.110 ( i ] t (df 18 ) : 1.0% 2,878; 5.0% 2.101 [ u ] t (df 16h 1.0% 2.921; 5.0% 2.120 0 Significant be1ond the 1.0% lev e l of confictenc.;;

PAGE 74

67 At level 3 the experimental group displayed a slightly greater range of movement for all vowels. For [1] the normal group again showed posterior movement while the experimental group showed anterior movement. The difference here was significant beyond the 1.0 per cent level of con fidence. No other significant differences were noted. For (a] both groups showed anterior movement, while for [u] the normal group showed no movement and the experimental group showed posterior movement. The trend of movement for both groups at level 4 is very close to that seen at the previous level, although no significant differences were observed. The me ans of both groups indicated anterior movement for [a]. For [1] the normal group showed slight posterior movement and the experimental group continued to show anterior movement. A slight tendency toward anterior movement was observed among the normal group for [u], while the experimental group con tinued to display mostly posterior movement. A wider range of movements was once again seen among the normal group members. At level 5 the normal group showed anterior move ment, while the experimental group showed posterior movement for all vowels. For each vowel the group differences were significant beyond the 1.0 per cent level of confidence. The variability of movements tended to be slightly greater

PAGE 75

68 among the normal subjects. At level 6 the experimental subjects continued to display posterior movement during all vowels. For the normal subjects the movement was anterior for [a) and (1), and tended to be static for [u). For [1) the two groups were significantly different beyond the 5.0 per cent level of con fidence. The variability of movements displayed by the ex perimental group was greater during the vowels [1] and [u]. The normal group showed more movement for the vowel [a]. In summary, there appeared to be a greater range of movements of the posterior pharyngeal wall displayed by the normal subjects. A general tendency was observed for the norm.al subjects to narrow the velo-pharyngeal opening by anterior movement of the pharyngeal wall, while the ex perimental subjects tended to widen the opening by posterior movement of the pharyngeal wall. Even where the mean dif ferences between the groups were statistically significant, a large amount of overlap was seen in the ranges of move ment. In general, the similarities between these two groups seem greater than their differences in these dimensions This conclusion would seem to be supported by the previous table., which revealed no significant differences between the groups in the area of the posterior ph a ryngeal wall. An examination of the X-ray tracin g s revealed that only one or two of the normal subjects and only three of the

PAGE 76

69 experimental subjects displayed an obvious bulging of the posterior p haryngeal wall which could be called P assavants pad. In no instance was the phenomenon as marked as some of those pictured by H agerty, _!ll !!.! (15). Calnan (6) re ported that P assavant's pad may be an abnormal phenomenon in phonation, and probably more appropriately associated with deglutition. Little or no pharyngeal wall movement, for most cases, has been reported by Buck (5), H ixon (20), W illiams (40), Wolfe (41), and Hagerty,~.!! {15). The results of the present investigation seem to be consonant with their findings. Com p arisons were made between the complete, unre paired cleft palate group and the present experimental group regarding measurements of the naso-pharyngeal struc tures. There were no significant differences between these two groups Lip and M outh Opening Lip Opening Item 1 of Table 8 presents the measurements of the smallest lip openings of the subjects at rest and during phonation. A general tendency was seen in all positions studied for the experimental group to display wider lip openings. This tendency reached statistical significance beyond the 1.0 per cent level of confidence for (a]. The

PAGE 77

1 20 70 TABLE 8 LIP AND MOUT H OPENING AT REST AND DURING PHONATION A M O NG SUB.JE(;TS WIT H NCRMAL DEVELOPMENT AND SUBJECTS WIT H INCOMPLETE, UNREPAIRED CLEFT PALATES (All measurements in millimeters) Re st [ a ] [ i ] [u] Lip opening Normal M 2.66 14.15 7.20 1 .75 SD lo50 4.87 2080 1.25 SE .45 lol2 .64 .29 Cleft Palate M 2.80 19.42 7072 2.76 SD 4.36 6.25 3. 97 1.92 SE lo06 1.43 0% .45 Differences M .14 5.27 52 1 .01 SE 1.15 1. 82 1.,16 53 t .122 20896** 0448 1.906 R est t {df 28): 1.0% 2.763; 5.0% 2 .048 [ al t (df 38): 1.0% 2. 711; 5.0% 2. 025 [ 1] t (df 36): 1.0% 2.718; 5.0% 2.027 [ u ] t (df 37): 1.0% 2. 713; 5.0% 2.027 Incisor o pening Normal M 1.,00 1 2. 50 2.15 4.65 SD 4 o 37 6 e 20 3.73 4.69 SE 1.3 2 1 .. 42 086 1.08 Cleft Palat e M -.38 13.57 2.32 Jo?2 SD 2.67 5ol5 3o04 J.89 SE .67 1.22 76 .94 Differences M 1.38 1 .. 07 .17 .93 SE lo4 8 lo87 1.14 1.43 t .932 0 572 .149 .650 Rest t ( df 27): 1.0% 2. 771; 5.0% 2.052 ( a ] t (df J?): .LO% 2.7 13; 5.0% 2.027 ( 1 ) t ( df 35)s loC % 2 728; 5.0% 2.032 [ u ] t { d f 36 ) : 1.0 % .2. 713; 5.0% 2. 027

PAGE 78

71 TABLE 8 Continued Rest [a] [ i] [u] 3. Distance from Normal Frankfort line M 53.08 65.85 55c30 58 .. 10 to incisal SD 4.81 6.57 4.Jl 5.74 reference line SE 1.45 1. 51 .99 1.32 Cleft Palate M 55.97 70.25 58.61 61.39 SD 6.76 7.60 6.93 6.67 SE 1.64 1.74 1.68 1.57 Difference s M 2. 89 4o40 J.31 J.29 SE 2.19 2oJO 1.95 2o05 t lo320 1.913 1.697 1.605 Re st t (df 28): 1.0% 2.763; 5.0% 2.048 [a] t (df 38): 1.0% 2. 711; 5.0% 2.025 [ i] t (df 36): 1.0% 2.718; 5.0% 2.027 [u] t (df 37): 1.0% 2. 713; 5.0% 2.027 **Significant beyond the 1.0% level of confidence

PAGE 79

72 same general t endency was found among the subjects with com plete, unrepa1rad cleft palates. Knobeloch (24) felt that this tendency might be explained by the presence of unre paired lips among a majority of his cleft palate subjects. This explanation obviously will not hold, however, for any of the subjects of the present investigation. furthermore, since the two experimental groups were so nearly alike 1n lip opening, it mi ght be speculated that the same factors were operating in each of them. Mouth Opening Two dimensions were used to acquire data on mouth or jaw opening. The most direct measurement, the vertical distance between the mandibular and maxillary incisors, was apparently complicated by dental irregularities. The other dimension introduced was the vertical distance from the Frankfort line to the incisal reference line. It was ex pected that this second measurement would be only minimally affected by such irregularities, and consequently more valid. Ne1 ther 1 tem 2 nor item 3 of 'r able 8 shows any sig n1f1can t differences between the groups regarding mouth opening. But if the above assumption can be made, that item 3 is the more valid measure, than a slight tendency may be observed for the cleft palate group to have wider mouth openings in all positi ons. This tendency, however slight, would seem to be definite when consideration is given to

PAGE 80

observations made earlier on tongue position and movement (Tables 4 and 5). 73 The data from item 2 indicate a smaller mouth opening for the experimental subjects at rest and for [u], and larger mouth openings for the other two vowels. It would be difficult to attempt an explanation of the dif ferences between items 2 and 3 other than on the basis of dental irregularities. Analysis of Speech Samples Tape recordings were acquired of the members of both the normal and the cleft palate groups. The recordings of the normal subjects were used to provide a standard against which to analyze the recordings of the cleft palate group. With this standard in mind, the experimental group subjects were judged first, as to phonetic accuracy or ar ticulation, and secondly, as to deviations from normal oro nasal resonance. The articulation of each sound was scored as: O, correctly produced; 1, distorted; and 2, substituted or omitted. The total possible score, representing the max imum number of phonetic errors was 44. Any sound which was not correctly produced in all intraword positions was counted an error. The rationale behind this either-or sorting pro cedure was the judges' feelings that the sounds would be

PAGE 81

74 correct or incorrect due to the organic factor involved. Clinical observation indicates that there is a g reat deal of consistency in the miaarticulations of individuals with palatal inadequacies. The p ossibility remains, however, that if the errors had been noted in the s p ecific positions in which they occurred, inconsistencies might have s h ow n up indicative of functional errors. ttatings of vowel and consonant articulation by the judges did not differ with statistical significance. A high positive coefficient of correlation, .99, was calcu lated for the two sets of ratings. The articulation scores of the experimental group ranged from 3 to 20. Their mean score was 11 and their median score was 10. 5. !< or this same measurement, Knobeloch (24) reported a range of from 10 to 36, with a mean of 20. It 1s immediately apparent that the subjects with complete, unrepaired clefts tended to have more phonetic errors than did the subjects with incomplete, unrepaired clefts. The following symbols were used to represent the sounds peculiar to Spanish: [B], Bilabial fricative; (d], voiced dental stop or fricative; [t], voiceless dental stop; [n], dental nasal; (~]~ palatal nasal; [ll, dental lateral; [r], single ta p lingua-alveolar; [ rr], trilled lingua alveolar; [y], voiced palatal fricative; and [x], voiceless velar fricative. T he Spanish vowels approximate those of

PAGE 82

7 5 La tin. I' hey are pure v o w els, not di p h thongized. In E n g lish there is a tendency to d i p h thongize all vowels but [a]. An actual count of the number of times errors were noted yielded the followin g se q uence, in decreasing order of fre quency of occurrence: C onsonant Sound [rr] [k) [r] [ t] [ g ] [ t J] [ p ] [s] [ d] [B ] [ f] [ m ] (N] [n] [x] [l] [y] Frequency of E rror 20 l.3 13 12 11 10 10 8 7 6 6 5 3 l l 0 0 Although some vowel sounds seemed to border u p on distortion, they all a pp eare d to be sufficiently accurate p honetically not to be score d as errors A rank order correl a tion o f t h e p rece d in g consonant error count with a co mp ar a ble list re p orted for the com p lete, unre p aired cleft palate subjects yielded a coefficient of .85. T h at is, there w a s a high de g ree of corres p ondence betwee n the cleft palate grou p s in t h e relative frequency of misarticulations p er sound. Th e trilled lin g ua-alveolar [rr], heard in the

PAGE 83

76 speech of all the control subjects, was misslng in the cleft palate subjects. The sovnds of (k] and [g], which normally re quire a functional veltur,, were fre auen tly misa.rtioula ted. It was somewhat surprisln g to find (r] and [t] more fre quently in error than (~ ], ::?inoe they do not de ; ,end so di rectly upon velar action. Part of the ex p lanation may be that these sounds require a greater implosion of air in the mouth than the experimental subjects were able to achieve, or that the contact required between the tongue and hard palate was difficult to achieve because of missing palate tissue. A partial explanation may be that some of these and other errors were functional. The deviations from normal oro-nasal resonance were rated on a five-point scale. A score of l represented normal resonance, and a score of 5 represented the greatest de viation from it. The judges observed other kinds of quality deviations among the experimental group such as harshness, hoarseness, and breathiness. However, they tried to consider only what might be called oro-nasal resonance in their ratings. The deviations which were heard could be classified as open or hypernasality, den.asality, and a combinat;ion of the two which might be termed cul~~ nasality. A tabu lation of the quality ratings follows: Rating Number of Subjects l 0 2 l 3 6 4 8 5 5

PAGE 84

77 The tapes were listened to continuously at several different times. No marked differences in numerical ratin g were noted. The final judgments yielded a positive coeffi cient of correlation of .92. The judgments did not reveal any statistically significant differences. The mean ratin g for the g rou p was 3.85. N one of the cleft palate subjects were judged to have normal quality, and only one of this group ap p roached a normal quality rating. Six of the ex perimental subjects received a rating that was half way between the two extremes. Thirteen, or over half, of the experimental group subjects were judged to have extreme de viations from normal oro nasal resonance. These ratin g s are much closer to those assigned the com p lete, unrepaired cleft palate g roup than were the scores for p honetic accuracy just discussed. The mean ratin g for the grou p with complete clefts was 3.7. Like the ex p erimental g rou p members of the present study, 65.0 per cent of the com p lete, unre p aired cleft palate subjects were assi gn ed ratin g s of 4 or 5, and none of them were assigned ratings of l. The ratin g s of ar ticulation and voice for these two cleft palate g roups are believed to be co m parable because the same two judges were used for both studies and the ratings were made under iden tical conditions within a few weeks of each other. S p riestersbach (36) and Sherman (34) have sug g ested that ta p e recordin g s of s p eech samples should be p layed

PAGE 85

7 8 baokward to give more valid ratings of quality. They found that articulation factors, which tend to in.t'luence quality judgments, are thus eliminated. Since neither of the judges was conversant with Spanish, it was felt that there was no need to reverse the tape recordings of the Spanish-speaking individuals. It should be said, however, that during the course of the analysis the judges did grow familiar with the Spanish words to which they listened. For this reason it may h ave been better to have played the tapes backwards. In order to investigate the influence of articulation on the judges' ratings of quality, a coefficient of correlation was com p uted between the ratings of articulation and quality. This oorrelation was -.35, which indicates that the articu lation of the Spanish-s p eaking subjects had no apparent effect on the judgments of quality. Co r rel a tio n.s betwe e n A c ou s t .t eal a nd J.> hysi c a l :, 1: 0as u re men ts The data to be presented in this section are intended to demonstrate some of the correlations found between the acoustic measurements of articulation and voice quality and the p hysical m e asurements made from the X-ray tracings of the ex p erimental grou p N o correlations were computed with physical measurements which were not statistically different from normal. A few other p hysical measurements were not used in the com p utations. These were:

PAGE 86

79 (1) the mandible width, which has been seen to follow the trend established by the maxilla, hence a correlation with maxilla width should suffice; (2) the pharyn g eal openin g between the incisal reference line and the superior border of the body of the hyoid bone o n [ u], because this was iden tical with the measurement made along the incisal reference line during the production of that vowel; (3) movements at cer t ain levels of the p osterior pharyngeal wall, those movements having so much variation that it was felt they would p robably not produce significant results. Among the experimental group the velo-pharyngeal o p enin g did not appear to vary in a demonstrably consistent p attern from one vowel position to another. Therefore, for each subject the velo pharyngeal o p enin g measured for all three vowels was avera g ed, and this average was correlated with the obtained acoustical scores. The significance of the correlation coefficients was teated by a method sug g ested b y Edwards (7). The cri terion of significance was established at the 5.0 p er cent level of confidence. An examination of Table 9 reveals a si gn ificant, positive c o rrelation between the Index of Articulation and

PAGE 87

1. 2. 3. 4. 5. 6. 7. 8. 9. 80 TABLE 9 COEFFICIENTS OF CORRELATION BETWEEN PHYSICAL AND ACOUSTICAL MEASUREMENTS AMONG SUBJECTS WITH INCOMPLETE, UNREPAIR~D CLEFT PALATES Index of Rating of Voice Articulation Quality Deviation Coefficients of Correlation Maxilla length .39 .02 Maxilla width 57* -.25 Horizontal distance from lo-wer incisors to high point of tongue: Rest .37 .22 Horizontal distance from lower incisors to high point of tongue: [i] -025 .47* Pharyngeal opening along incisal reference line: [ u] -.26 -.Cl Distance of high point of tongue above incisal reference line: [a] -.04 -.23 Distance of high point of tongue below hard palate: (1) .18 53 .. Distance of high point of tongue below hard palate: [ u] .28 .16 Distance of high point of tongue below Frankfort line: Rest ,24 .34

PAGE 88

81 TABLE 9 Continued Index of Rating of Voice Articulation Quality Deviation Coefficients of Correlation 10. Distance of h igh point of tongue below Frankfort line: [ i] .21 .45 11. Velo-pharyngeal opening, average for vowels [a,i,u] 018 .26 12. Lip opening: [a] -019 .17 Significant at or beyond 5.0% level of conf1don c 0

PAGE 89

82 maxilla width. That is, a large phonetic error count ap pears to be associated with maxillae which are wider. The correlation with maxilla width is not surprising, since presumably the width of the maxilla is directly associated with the severity of the cleft. It must be clearly pointed out, however, that no implication of causation is here in tended. Factors which yield a coefficient of .57 are pre sumably related, but both may be caused by some more basic factor. Further examination of Table 9 reveals significant correlations between voice quality deviation ratings and two measures of the tongue position for the vowel [i]. That is, those subjects who, during the vowel (1), tended to place the high point of the tongue more posteriorly, and those who tended to carry the tongue a greater distance below the hard palate also tended to be those subjects whose voices were judged to have the most abnormal oro-nasal resonance. I t was shown in Table 2 that the experimental group subjects t en ded to carry their ton g ues more anteriorly on [1). The correlation of .47 between a more posterior tongue placement and a higher rating of voice quality deviation would appear to indicate that those subjects who did not follow the trend toward more anterior placement tended to be those whose voices were judged to be more abnormal. T able 4 indicated that the cleft palate subjects

PAGE 90

83 tended to carry their tongues a greater distance below the hard pa late for (1). This tendency showed a .53 correlation with judgments of increased voice quality deviation. rhis should not lead one to conclude that a higher tongue on [1] will necessarily result in more normal voice quality. W ith out m ore evidence it can only be stated that these factors are correlated. It seems unlikely that there exists any cause-effect relationship between them. Insignificant correlations were found between the average velop haryngeal opening for the vowels and both of the acoustic scores. This does not mean that the velo pharyngeal o p ening was of no importance in contributing to these scores. It does mean that the velo-pharyngeal openings of this group were probably well beyond the critical point as described by M cDonald and Koepp Baker (26). Further correlatives of the physic -aco ustical re lationshi ps of th e se subjects and of those with com p lete, unrepaired clefts should be of interest, and will be made as the parent study continues. Direct measurements have still to be made of the pa latal casts, and these may p ossi b ly become the most valuable measurements yet considered for these subjects.

PAGE 91

C H APTER IV SUM M ARY AND CONCLUSIONS The experimental group of the present study was composed of twenty natives of Puerto Rico who possessed con genitally cleft palates. To be included in this group subjects had to have clefts which extended no farther for ward than the median border of the alveolar ridge, have had no prosthetic or successful surgical repair, and have reached physical maturity. A control group was made up of twenty adult P uerto Ricans who had normal facial structures, and who had the normal speech and hearing of Puerto Ricans. The two groups ranged in age from fifteen years to forty-nine years. The inclusion of the youngest members as adults is based on the study by Fleagle (9) which support s the popular conception that people tend to mature sooner in the tropics. Lateral X-ray exposures were made of all subjects at physiol ogic rest and during the phonation of the vowels [a], (1), and [u]. For the control group, twenty exposures of each vowel and twelve exposures of the rest position were obtained. The films available on the experimental group in cluded eighteen of the rest position, twenty of [a], eighteen of [1], and nineteen of [u]. These films were all traced and measurements were taken from the tracings. Antero-poaterior views of the subjects at physiologic rest were also made. 84

PAGE 92

85 'rwelve of these were obtained for the control group and fif teen for the experimental group. These films were not traced, but measurements were made directly from them. All measure ments were treated statistically to compare the control group with the experimental group. The experimental group was also compared with Knobeloch's (24) experimental group of indi viduals with complete, unrepaired cleft palates. Tape recordings were made of all experimental group member3 and some control group members. The recordings of the experimental group members were analyzed for phonetic ac CHrRcy of consonants and vowels and for deviations from normal oro-nasal resonance. Correlations were computed between acoustical and significant physical measurements. The following conclusions may be drawn from the data: l. Differences significant beyond the 5.0 per cent level of confidence occurred between the control and experi mental groups on measurements of maxilla length and width and mandible width. Measurements of mandible length, oral cavity length, and width of the naso-pharynx revealed no statisti cally significant differences (Table 1). No statistically significant differences were found to exist on the above measurements between th e ex pe rime nta l group and the subjects with complete, unrepaired cleft palates. 2. In measurements of the horizontal position of the

PAGE 93

86 tongue at rest and for [i], the experimental group differed from the control group beyond the 5.0 per cent level of con fidence, their tendency being toward a more anterior tongue placement. Other measurements related to horizontal posi tion of the tongue and ton g ue size revealed no statistically significant differences between the groups (Table 2). The experimental group did not differ significantly from the subjects with complete, unrepaired cleft palates on any of the above measurements. 3. Pharyngeal opening measurements on and below the incisal reference line differed with statistical signif icance for the vowel [u], but not for any other vowel position (Table 3). For the measurement below the incisal reference line on [1], the two unrepaired cleft palate groups differed from the control group in opposing directions, and were sig nificantly different from each other beyond the 5.0 per cent level of confidence. A posterior as well as an anterior con centration of tongue tissue during the sounding of this vowel among the subjects in the experimental group seems to account for this difference. 4. At rest the experimental group tended to hold their tongues somewhat lo wer below the hard palate and signif icantly lower below the Frankfort line than did the control group, but their tongues were similarly placed above the incisal reference line. For [a] the experimental group raised

PAGE 94

87 their tongues significantly higher above the incisal ref erence line, but their ton g ues remained normally distant from the hard palate and the incisal reference line. For [1] the experimental group showed a slight tendency to raise their tongues less above the incisal reference line than the control group. This difference was amplified to significance beyond the 1.0 per cent level of confidence 1n measurements from the tongue to the hard _palate and to the l<'rankfort line. Differences for [u] were not statisti cally significant except in the measurement from the ton g ue to the hard pa late. For this measurement the experimental group appeared to maintain a greater distance between the palato lingual surfaces ( rable 4). rhe apparent discrep ancies in the various measurements of tongue height may be accounted for by variations 1n jaw opening and by the fact that for [u] the measurement from the tongue to the hard palate was an approximation, actually made to a line ex tended from the hard palate. The experimental group and the group with complete, unrepa1red clefts did not differ sig nificantly on any but two of the above measurements. These were measurements during the vowel [i ] of the distance from the tongue to the h ard palate and from the tongue to the Frankfort line. These differences were significant beyond the 5.0 per cent level of confidence. They appear to be accounted for by the tendency already noted among the

PAGE 95

88 subjects of the experimental group to maintain a posterior as well as an anterior concentration of ton g ue tissue during this vowel. 5. Measurements of tongue movements (Table 5) follow the pattern which would have been hypothesized from the previous measurements of tongue position. Also, the greater variation seen in the vertical tongue movements of the experimental group may reflect the presence of the oral anomaly. The experimental group tended to move their tongues higher for [u] and [a], the differen c e on [ a] reaching sig nificance beyond the 5.0 per cent level of confidence. M ore posterior movement was seen on these same two vowels for the experimental group. The differences were significant beyond the 5.0 per cent level of confidence for [a] and beyond the 1.0 per cent level of confidence for (u]. No significant differences were noted for the vowel (i]. Comparisons be tween the experimental group and the group with complete, un repaired clefts on the above measurements revealed no statis tically significant differences. 6. Me asurements of velo-pharyngeal opening during phonation revealed very lar g e differences between the two groups of this study (Table 6). The ex pe rimental subjects had consistently larger openings. The mean group differences ranged from 5.22 millimeters to 8.23 millimeters, all of which were significant well beyond the l.O pe r cent level of

PAGE 96

89 confidence. It was felt to be unnecessary to statistically compare the two unrepaired cleft palate groups on this measurement, because of the similarity of the findings in these groups. 7. The experimental group did not differ signifi cantly from the control group in measurements of the area of the posterior pharyngeal wall ( rable 6). Since this had also been true of the complete, unrepaired cleft group, the two cleft palate groups were not statistically compared on this measurement. 8. Measurements along the lines constructed across the area of the posterior pharyngeal wall revealed only small amounts of movement for both groups. The only significant differences found were: at level 3 for [1], the control sub jects showed posterior movement, whereas the experimental subjects showed anterior movement; at level 5 for all vowels, the control group showed anterior movement, whereas the ex perimental group showed posterior movement; at level 6 for [i], the control group showed anterior movement, whereas the experimental group showed posterior movement. All differences noted were significant beyond the 1.0 per cent level of con fidence ( rable 7). A comparison between the unrepaired, cleft palate groups on these measurements revealed no statis tically significant differences. 9. A larger lip opening, significant beyond the 1.0

PAGE 97

9 0 per cent level of confi d ence, was found for the experimental group on the vowel [a]. N o other significant differences were found for measurements of lip opening, incisor opening, or distance from the Frankfort line to the incisal reference line. There appeared, however, to be a general tendency among the experimental g roup to maintain a slightly wider lip and jaw opening (Table 8). Com par isons between the un repaired, cleft palate groups for these measurements re vealed no statistically significant differences. 10. The speech of the experimental group subjects exhibited a great many phonetic error s including some which were inconsistently made. Those sounds inconsistently mis articulated d o not appear to be explained by the p resence of the oral anomaly. The subjects with co mp lete, unrepaired clefts pr oduced a great many more errors than the ex peri mental group subjects with incomplete, unrepaired clefts. 11. The voice quality analyses of the tape recor din g s indicated that the ex p erimental subjects d iffered markedly from the control subjects in ratings of oro nasa l resonance. H owever, the p attern of ratines for the two un repaired cleft palate g roups appeared to be quite similar. 12. The number of phonetic errors made by subjects of the experimental g roup was positively correlated with maxilla width (Table 9). 13. Among the ex p erimental g roup subjects,

PAGE 98

91 deviations from normal oro-nasal resonance were p ositively correlated with the distance of the hi g h point of the ton g ue below the hard p alate for [1] and also with the hori zontal distance from the lower central incisors to the high point of the tongue for [i] (Table 9).

PAGE 99

CH.APTER V DISCUSSION The two E rou p a of this study were found to differ significantly in the following skeletal measurements: the experimental group had maxillae which were shorter, and both maxillae and mandib l es which were wider than those of the control group. The cause of the greater mandibular width among the unrepaired cleft palate subjects of this study cannot be readily ascertained. It may be that the laterally spreading maxilla exerts an outward force on the mandible. But would not the more solid bony frame of the mandible tend to resist this force? It is possible that the mandibles of this group grew wider as a result of neurological changes which may have been imposed by the cleft condition. When the experimental group of this study, who had incomplete, unre p aired cleft palates, was compared with Knobeloch's (24) group who had complete, unrepaired cleft palates, no significant skeletal differences were found. The differences between the complete, unrepaired cleft palate group and the normal group, however, failed to reach significance except in maxilla width. It would appear that when a fetus fails to form a complete palate, whether that failure is partial or total makes little difference in the potential effect it has on the facial skeletal growth 92

PAGE 100

of the p erson. That effect, left undisturbed by surgery, appears to be most noticeable in the measurement of maxillary width. A question arises as to how directly related the width of the cleft is to the extra width of 93 the maxilla in the adult. T o what extent is the cleft condition exaggerated by a subnormal amount of bone tissue? The question provides a beginning point for future valuable research. When the incomplete, unrepaired cleft palate sub jects are compared with the repaired cleft palate subjects studied by Buck (4) and Graber {12) important differences are seen in skeletal dimensions. The repaired cleft palate subjects had significantly shorter vertical facial dimensions, together with concave facial angles at physiologic rest. Their faces were also shorter in antero-posterior width than the normal subjects with whom they were compared, and their maxillary and mandibular dimensions were smaller. In gen eral, it can be said that the unrepaired cleft palate sub jects were much more like normal subjects in facial skeletal development than they were like repaired cleft palate sub jects. The most outstandin g functional differences dis covered between the ex p erimental and the control groups was in the a.mount of velo pharyn g eal opening during phonation. As a group, the cleft palate subjects consistently showed a

PAGE 101

9 4 much greater than normal o p ening for all vowels, as did the subjects with c o mplete, unrepaired clefts and Buck's (4) subjects with clefts repaired at an early a g e. The generalization should not be drawn that sur g ical repair made no difference in the attem p ts of cleft p alate sub jects to achieve velo pharyn g eal closure. F or the re paired and the unre p aired cleft grou p s the patterns of velar movement were not sirrilar, exce p t that they both dif fered si g nificantly from normal grou p s. Nevertheless, t h e surgical aid t hat had been g iven Buck s subjects does not a p pear to h a ve left them better a ble to achieve velo pharyn g eal closure than they mi g ht have done without it. A word of caution is appropriate here, to the ef fect that the information obtained from lateral X rays of sustained vowels is neither c o mplete nor final. rhe clo sure o f the velo pharyn g eal openin g ap p ears to be reasonably well measured by the p r o cedure employed in this study. It must be continually borne in mind, however, that the o p enin g being considered is really three dimensional, and the p ic ture seen on the X ray p late may not reveal im p or t ant d etails. One subject, No. 149, will help illustrate this point. T he lateral X rays of this subject indicate complete closure of the velum a g ainst well develo p ed adenoid tissue for all three vowels. An examinat i on which included no more infor mation than this mi3ht lead to a conclusion th a t her

PAGE 102

vela p haryngeal function was normal. Further information based on visual inspection of the subject's palate indi cates that while she showed good excursion of the soft palate tissues, the velum was cleft. In addition, her voice received a rather poor rating of "4". Ap p arently then, the lateral X ray views must be accepted with their limitations in mind. 9 5 No other differences in function between the groups studied are so noticeable as the above. M easurements of tongue movement and placement reveal some minor differences. These measurements among the ex p erimental group were appar ently influenced by a slight, general tendency toward a wider jaw openin g particularly on [a]. Taking all these measurements to g ether, in the ex p erimental group the tonsue seems lower and flatter in the oral cavity for [1), about "normal" for [a], and somewhat more retracted and rounded for (u]. It may be that these differences occur to compen sate for the presence of the clefto Or are they better understood as a compensation for a wider jaw o p ening? This brings up the question of why there is a wider jaw o p ening. Is it a compensation for the cleft palate? Or is it a re sult of a wider mandible? As a partial answer to this last question, it may be significant that the com p lete, unre pa1red cleft subjects did not have as wide mandibles as the experimen ta l subjects of this study, nor did they show the

PAGE 103

same general tendency toward wider jaw o p enin g The general im p ression g ained from these compar isons is that oral pharyn g eal functionin g of normal p eo p le is more closely approximated by persons with unrepaired cleft pa lates than by persons with early sur gic al re p air 96 of cleft palat es. In addition, to the skeletal differences already discussed, Buck (3) found his repaired cleft palate subjects differed significantly from the norm in midsa g ittal measurements of tongue area, in toneue height within the oral c avity, and in velo-pharyn g eal o p enin g durin g p ho nation. The experimental g roup of the p resent st ncly like those with complete, unre p aired clefts, is similar to his in only one of the dimensions considered: velo-p h aryn s eal o p enin g. The high inci den ce of hearin g loss amon g cleft palate individuals has been frequently reported (10, 22, 33}. The con g enital deformity which affects t h e p alate may also affect the pharynge al o p enin g t o the E ust achian tub es, or the incom p lete p ala t e may fail to prot ect t he Eustachi an tubes sufficiently fro m col d dirt, or food parti c l e s. Of the twenty cleft palate subjects included i n t h e p resent study, seven or 35.0 pe r cent of the m ha d mild to m o d e ra te hearing losses which ap p e a red to be conductive. This is a high enough incidence to warr an t the inclusion of an audio metric test routinely in the examination of every p erson

PAGE 104

97 with a cle f t palate. The analyses of p honetic accuracy and oro nasal res o nance to which the ta p e recordings were subjected indi ca t ed that as a group, the cleft p alate subjects had a great many phone tic de via tio n.s a'1. d a g reat deal of deviation from n o rmal voice quality. A detailed study of the data, however, revealed that there were g reat individual differences in the severity of both types of abnormalities. 'rhere was also a sli g ht tendency toward an inverse relationship between pho netic deviations and voice q uality deviations. T hose with wider maxillae (and presumably this means wider clefts also) tended to h ave more articulatory errors and tended sli g htly to have m ore normal voices. rhose who kept the h i g h p o int of their tongues farther behind their lower incisors on [1] tended to have more deviant voices, and tended sli g htly to have better articulation. Perha p s there are questions for future research here. Wh at is the effect in the oral cavity of tne wider max i lla? Are the observed p honetic errors more a result of missing palatal tissue, or of the reduced ability to build u p intra o ral breath p ressure? Both factors would appear to be important. H ow im p ortant is the .25 corre lation or maxil l a width an d voice quality deviation? Does this mean that subjects with wi d e o p en clefts t e nd to have more normal voice quality? P erha p s narrower clefts more effectively s e p arate the oral and nasal cavities, thus

PAGE 105

98 em p hasizin g both an oral a..n.d a n asal resonance; while wider clefts tend more to produce one large oro nasal cavity. Examination of the palatal casts when they become available should throw valuable li 3 ht on t h ese questions. The si g nificant correlations between poor voice quality and lower and less forward tongue positions on one vowel, [i], would seem to implicate some more basic common cause of these two effects. It has been su gg ested that ton g ue positions peculiar to this group of subjects may possibly serve to compensate for the cleft palate condition. It would be interesting to learn whether the tongue positions of such a group might change if the cleft were prosthetically closed. The factor of functional misarticulation must be taken into account when considering the correlations between physical and acoustical measurements. The ta 1) e recordin 8S contained many articulatory inconsistencies, indicating t h at functional errors often overlaid organic errors. The incon sistencies would be ex p ected to reduce any correlations with ph ysical measurements. Furthermore., if some of the phonetic errors were a p roduct of wron g learnin g ., it i s conceivable that some of the devi a tion. in oro -nasal reaonence was also a pr oduct of wrong learnin g Apparently ar ; early pro g ram of speech tralnin g and psych ol ogical counseli ng would have been benefi cial to many of the p ersons in the cleft p alate group to h el p them pr o du ce the bes t speech p ossibleo

PAGE 106

In summary, the findings of this study would a p pear to lend support to the ar g ument advanced by Graber (11), Kro g man {25), Slau gh ter and Brodie (35), Buck (4), 99 and others that early surg ical repair of the palate tends to interfere with structural growth. It would seem to be worthwhile to consider delaying sur g ery until all or most of the facial g rowth had been completed. Such a course of action, it ap p ears, would p ermit the nearest approach to norrnai development and normal function. The abnormalities of the cleft itself and the virtually functionless velum could then be alleviated with a prosthetic device, and an early course of speech trainin g could help with its ef ficient use. Children would thus learn and be able to pro duce good speech, they would not be subjected to the adverse psycholoBical effects of facial deformity, an d their full physical growth would seem to allow more effective surgical closure at a later dat e Recent years h ave seen the formation of an in creasing number of cleft palate teams throughout the country, made up of members of various professions concerned with the p roblem. Besides a s p eech pathologist, these teams often include a p lastic surgeon, an orthodontist, a prosthedontist, and others. It is gr atifying to notice the increased cau tion with regard to cleft palate management these teams seem to sug g est. Increased u s e of pr osthetic devices .has made

PAGE 107

100 blanket prescription of sur g ical re p ai r no lon g er standard procedure. Indeed, it m ay be said that a standard procedure no longer exists, but eac h child is considered as an in dividual with an individual problem. Such a trend among cleft palate teams is extraordinarily valuable, bot h for the children the teams see and for the teams themselves. It is hoped that the findin g s presented in this study will be useful in addin g a note of caution as to t h e co mb ined p rofessional care o f p ersons with con g enital cleft pal a tes.

PAGE 108

LIST OF REFERENC E S crrED 1 Brader., A.C ., The application of the principles of ceph al o metric lami n agraphy to studies of the frontal planes of the human head. !,_ m erican Journal of Orthodontics, 35, 1949., 249-268. 2. Brodie., A.G., Some recent observations on the growth of the face and their im p lications to the orthodontist. m erican Journal of Orthodontia and Surger;r., 26., 1940., 741-757. 3. Buck., M.W ., An X-ray study of cleft p alate oral and pha ryngeal str u ctures and their functionin e; during vowel phonation. Ph.D. Dissertatio n ., State University of Iowa., 1 95 1. 4. Buck., M., Facial skeletal measurements and tongue carriage in su b jects with repaired cleft p alates. Journal .2.f. Sneech H earing D isorders, 1 8 1953., 121-132. 5. Buck., M ., P ost-operative vela-pharyngeal m oven~nts in cleft palate cases. Journal S!,f_ Speech~ Hearing Disor d ers, 19., 1954, 28 8 -294. 6 Calnan, J., The error of Gustav P assavant. P lastic and~ constructive Surgery, 13, 195 4 27 5 -28 9 7. Edwards, A.L., Stati s tical Analysis (Rev. E d.). N ew York: R inehart and Compan y I n c., 19 55 8. Fairbanks, G., A physiolo g ical correlative of vowel inten sity. Speech Monograph~, 17, 1950., 390-395. 9. Fleagle, F. K ., Social P roblems in P orto R ico. B oston, N ew Yor k an d Chicago: D.C. H eath and Company, 1917. 10. Gaines, F. P ., Frequency and effect of hearin g loss in cleft palate cases. JOU n al of Speech D isorders, 5., 1940., 141-14 9 11. G raber, T .M., Changin ~ philosop h ies in cleft palate man ageme n t. Journal of P e d iatrics, 37., 1 9 50, 4004 15. 12. Gra b er, T.M., C raniofacial m or p holo g y in cleft p alate and cleft li p d e f ormities. Surge~y, Gynecology, an d Obstetrics, 88, 1949, 359-369. 101

PAGE 109

102 13. Hagerty, R. and H offmeister, F.s., Velo pharyngea l clo sure; an index of s p eech. P la stic and Reconstructive Surgery, 13, 1954, 290 -2 9 [ 14. Hag e rty, R.F. and H ill, N .J., Pharyngeal wa ll and pal atal movement in p o stop erative cleft pa l at es and norma l palat es. Journal E.f S peech~ H earing R ese arch, 3, 1960, 59 -66. 15. Hagerty, R. F., H ill, M J., Pettit, H S., and Kane, J.J., Pos terior pharyngeal wall movement 1n normals. Journal 2.f. Speech~ Hearing Research, l, 1958, 203-210. 16 Hagerty, R .F., Hi ll, M .J., Pettit, H .S., and Kane, J.J., Soft palate movement in normals. Journal of Speec h and H earing Research, 1, 1958, 325-330. 17. Ha r rington, R., A note on a lingua-velar relationship. Journal 2.f. Speech Disorders, 11, 1946, 25. 18. Harrington, R., A study of the mechanism of velopha ryngeal closure. Journal .2 Speech Disorders, 9, 1944, 325-345. 19. H erzberg, F. and E olic, R., An anthro p olo g ic study of face hei g ht. American Journa l of Orthodontics and Oral Surgery, 29, 1943, 90-100. 20. Hixon, E. H ., An X-ray study comparing oral and pharyn g eal structures of individuals with nasal voices and 1nd1v1duals with superior voices. M .S. Thesis, State Uni versity of Iowa, 1949. 21. H olbrook, R. r. and Carmody, P.J., X-ray studies of speech articulation. University of California P ublications~ M odern Philol o gy, 20, l937-;-No. 4. 220 H olmes, E. M and Reed, G.R., Hearing and deafness in cleft p alate patients. Archives .2f Otolaryngology, 62, 1955, 620-624. 23. Howell, E. T ., An investigation of cinefluoro g raphic tech niques, p rocedures, and equipment pertinent to the study of speech structures and their mo vement. M .A. Thesis, University of Florida, 1955. 24. Knobeloch, F.x.c., An X-ray study of unrepaired, complete cleft palate oralp haryngeal structures and their func tioning durin g vowel p h onation. Ph.D. Dis sertat ion, Uni versity of Florida, 1959.

PAGE 110

103 25. Krogman, W.M., The p roblem of the cleft palate face. Plastic~ R econstructive Surgery, 14, 1954, 370-375. 26. M cDonald, E.r. and Koepp-Baker, H ., Cleft palate speech: an integration of research and clinical observation. Journal .2.f. Speech~ H earing Disorders, 16, 1951, 9-20. 27. Norris, M .A., X ray studies of vowel production as it is related to voice. M .A. Thesis, State University of Iowa, 1934. 28. Parmenter, C.E. and Trevino, S. N ., Vowel positions as aho'WO by X-ray. Quarterly Journal .2 Speech, 18, 1932, 351-369. 29. Ricketts, R. M ., The cranial base and soft structures in cleft palate s p eech and breathing. Plastic~ Recon structive Surgery. 14, 1954, 47-61. 30. Russell, G.o~ First preliminary X-ray consonant study. Journal of the Acoustical Society of America, 5, 1934, 247-251. 31. Russell, G.O., Seech Voice. New York: MacMillan, 1931. 32. Russell, G.O., The mechanism of speech. Journal of the Acoustical Society .Qf. America, l, 1929, 83-109. 33. Sataloff, J. and Fraser, M., Hearing loss in children with cleft palates. Archives of Otolaryngology, 55, 1952, 61-64. 34. Sherman, D., The merits of backward playing of connected speech in the scalin g of voice quality disorders. Journal of Speech and Hearing Disorders, 19, 1954, 312321. 35. Slaughter, W .B. and Brodie, A G. Facial clefts and their surgical mana g ement in view of recent research. P lastic Reconstructive Surgery, 4, 1949, 311-332. 36. Spriestersbach, D C. Assessing nasal quality 1n cleft palate speech of childre n. Journal of Speech and H earing Disorders, 20, 1955, 266-270:37. Subtelny, J. D ., An X-ray study of width of nasopharynx and related anatomical structures in normal and pre operated cleft palate children. American Association f2!: Cleft P alate R ehabilitation Newsletter, 3, 1953, 3-7.

PAGE 111

104 38. Subtelny, J.D., The significance of early ortho d ontia in cleft palate hab111tat1ve planning. Journal of Speech~ Hea ring D isorders, 20, 1955, 135-147.39. Townshend, R H ., The formatio n of Passavants bar. The Journal .!2f Laryngology O tology, 55, 1940, 154-165. 40. Williams, R.L., A serial radiographic study of velo pharyngeal closure and tongue position in certain vowels. Northwestern University Bulletin, 52, 1952, 9-12. 41. Wolfe, W G ., X ray study of certain structures and movements involved in nasopharyngeal closure. M.A Thesis, State University of Iowa, 1 942. ADDITIONAL REFERENCES Beers, M .D., and Pru zansky, s., The gr owth of the head of an infant wi th mandibular micro gna thia, g lossoptosis, and cleft p alate following the Beverly Douglas oper ation. Plastic!!!! Re construotive Surgery, 16, 19 55, 189 193. Bloomer, H. Observations on palatopharyngeal movements in speech and de g lutition. Jour na l of Speech~ H earing Disorders, 1 8, 1953, 230-246. Dorrance, G. M. The Operative Story of Cleft Pa late. Philadelphia and7:'ondon: W P. Saunders Co., 1933. Graber, T M An appraisal of the developmental de formities 1n cleftp alate and cleft-lip individuals. Northwestern University Medical School O uarterly letin, 23, 1949, 153-169. Halm, E., Speech therapy for the pre-school child. Journal of Speech and H earing Disorders, 23, 1958, 605-609.Harkins, c.s. and Koepp-Baker, H ., -;' wenty-five years of cleft palate prosthesis. Journal .!2f Speech~ H eari ng Disorders, 13, 1948, 23-30. H oldsworth, w.G., Cleft Lip and Palate. New York: Grune snd Stratton, 1953.

PAGE 112

105 Kaltenborn, A .F., An X -ray study of velopharyngeal closure in nasal a~d non-nasal speakers. M .A. Thesis Northwestern University, 1948. Kantner, C. E ., Dia gn osis and prognosis in cleft palate speech. Journal of S peech~ H earing D isorders, 13, 1948, 23-30. Kelly, J.P., Studies 1n nasality. Archives .f. Speech, 1, 1934, 26-42. M acCollum, D.W. and R ich a rdson, s o ., M anagement of the p atient with cleft lip and cleft ~a.la te. P ediatrics, 20, 1957, 57358 3. M asland, M w ., Testin r, and correctL~ g cleft palate s p eech. Journal of So eech Disorders, 11, 1946, 309-320. M c W illiams, B.J., Some factors in the intelligibility of cleft palate speech. Jour na l .Q, Speech~ F earing Disorders, 1 9 1954, 524-527. Nusbau m, E.A. Foley L., and Wells, c., Experimenta l studies of the firmness of the velarph aryn g eal o clusion during t he p roduction of the Englis h vowels. Speech M onographs, 2, 1935, 71-80. P almer, J. M ., The pharyng eal fla p operation: role of the speech th e ra p ist. Journal of Speech and Hea ring Disorders, 23, 1958, 601-604. Pruzan sky, s., The role of the orthodontist in a cleft palate tea m P lastic~ R econstructive Surgerx 14, 1 954 10-2 9 Ruess, A.L., The clinical psycholoeist in the habili tation of the cleft p alate patient. Journal of Speec h and H earing Disorders 23, 1958, 56 1-57 6. Sherman, D., Spriestersbach, D.C., and Noll J. D ., Glotta l stops in the s p eec h o f c h ildre n with cleft palates Journal .2.f Speec h !ill H earin5 Disorders, 24, 19 59, 3742 Slaughter, W.B.,and P ruzansky, S., The rationale for velar closure as a p rimary procedure in the re p air of cleft p alate defect s P lastic and Re constructive Surgery, 13, 1 954 341 -3 57.

PAGE 113

106 Travis, L.E. (Ed.), H andbook .Q, Speech Pathology. New York: Appleton-Ce n tury-Crofts, Inc., 1957. Trusler, H M ., Bauer, r.B., and rond.ra, J. M ., The cleft lip-palate problem. P lastic~ Reconstructive Surgery. 16, 1955, 174-188. Van Riper, c. and Irwin, J.v., Voice and Articulation. Englewood Cliffs, N .J.: 1958. Williamson, A. B ., D iagnosis and treatment of eighty four cases of nasality. c;uarterly Journal..! Speech, 30, 1944, 471-479.

PAGE 114

BIOGRAPHICAL NOTE Joseoh Jossions Ke enan was born in Carrollton, Miss.. 1ssi_:1pi, on October l, 1~)28. He gr aduated from Parmvllle i. U 6 h Sc h ool in Farmville; V ir 6 1nia in 1945. He receivod the Bachelor of Arts deeree froin Indiana University in 19 50 and the Mas ter of Arts degree from the University o f V ir g inia in 1957. H e served in the Army of the United States from No vember, 1950, to Se_ ;) torober, 1954. Durin g h is last year of military service he was stationed in the Republic of Korea, where he attained the rank of First Lieutenant as an aviator with the Second Un ited States Infantry Division. On September 3, 1955 he married Na nn ette Marie Weimar. Their son, Alain Christopher, was born December 15, 1956, and their daughter, Cecille Deveaux, was born Februar y 6, 1959. From September, 1959, to the tine of this writing, Mr. Keenan has held the position of Assistant Director of the Hearing and Speech C enter, N orth Carolina Baptist H os p itals, Inc., W inston-Salem, N orth Carolina. Prior to this appoint ment, he held the followin g positions: Clinical Assistant, Speech and Hearing Clinic, University of Florida, 1957-59; graduate assistant in Speech and Drama, University of Virginia, 1956-57; and free-lance actor, 1954-56. Mr. Keenan is a member of Sigma Alpha Eta, national honorary speech and hearing fraternity; and Theta Alpha Ph i, 107

PAGE 115

100 national drama.tics honorary fraternity. He is also a mem ber of the American Speech and H earin g Association the Speech Association of America, and the North Carolina Speeoh and Hearing Association.

PAGE 116

This dissertation was prepared under the directi o n of the chairman of the can d idate's supervisory committee and has been approved by all members of that committee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. June 6, 1960 Sciences Dean, Graduate School SUPERVISORY CO MM ITTEE: