Andrew F. D. Brown

Editor's Note: Andrew Brown is Assistant Professor of Music at Central Missouri State University at Warrensburg. He holds the DMA degree from the University of Iowa and this article is a condensation of his doctoral essay entitled "A Comprehensive Performance Project in Oboe Literature with a Cinefluorographic Pilot Study of the Throat While Vibrato Tones are Played on Flute and Oboe. "


Vibrato is generally believed to be produced on woodwind instruments by using one of three sets of muscles: those of the abdomen and diaphragm, of the throat, and of the lip of jaw. Some authors write of producing vibrato with the chest (intercostal) muscles, while others advocate use of more than one set in combination. Some authors, in writing about throat vibrato, erroneously speak of "constrictor muscles," "construction of the trachea," and "choking muscle."

Throat vibrato has received the least attention in the literature on woodwind vibrato. Much of that is of questionable value regarding the part played by the structures of the throat. Although the production of an acceptable throat vibrato does not depend on a knowledge of the anatomy and physiology of the throat, an understanding of the structures of the throat and their functions can help to correct misunderstandings and misinformation about vibrato production and can also help to provide better methods for teaching throat vibrato.

The purpose of this study is to demonstrate that cinefluorography is a valid tool for investigating the physiology of the throat during production of vibrato and vibratoless tones on flute and oboe.

The Throat Structures

It will be helpful to the reader to be acquainted with some of the structures of the throat (anatomy) and their functions (physiology). The principal throat structures which are of interest in woodwind vibrato are the pharynx, larynx, trachea, and esophagus. The first two are in the throat proper and the latter two extend from the larynx and pharynx into the upper chest cavity. Air passes through the nasopharynx and oropharynx into the larynx, directed there by the epiglottis. The epiglottis prevents food and other foreign objects from entering the larynx, and air from entering the esophagus.

The pharynx (figure 1) connects the nasal and oral passages with the larynx and the esophagus. It is a single tube that is common to both digestive and respiratory functions. The only muscles in the throat which are constrictor muscles are in the pharynx, and they are primarily involved in swallowing.

The larynx (figure 2) is the most complex of the throat structures. It is made up of a number of cartilages which are connected by muscle tissue: the thyroid cartilage, the cricoid cartilage, and the epiglottis. These are connected to each other and to the hyoid bone, from which the larynx is suspended. The tongue is also connected to the hyoid bone and causes the larynx to rise and fall. Within the cartilage structure are the arytenoid cartilages which are able to move in a variety of ways they can glide medially and laterally as well as rotate, and may slide forward and backward with restricted movements.

The vocal folds (or cords) are attached to the arytenoid cartilages at the back of the larynx and to the thyroid cartilage at the front. They are opened and closed by action of the arytenoid cartilages and are used both as a "valve" to stop air from escaping from the trachea and to produce speaking and singing sounds. The opening between the vocal folds is called the glottis. Above the vocal folds are the false folds. The space between the false and true folds is the laryngeal ventricle.

The muscles in the larynx are named by the cartilages or bones which they connect. The aryepiglottic folds, for example, connect the arytenoids and the epiglottis. The cricoarytenoid muscle connects the cricoid cartilages with the arytenoids. Four of the intrinsic laryngeal muscles will be of interest to us in later sections of this study. They are the posterior cricoarytenoid muscles, the lateral cricoarytenoid muscles, the cricothyroid muscle and the thyroarytenoid muscle (Figure 3). Individually, their actions are as follows:

The posterior cricoarytenoid muscles move the arytenoid cartilage in a rocking motion posterolaterally over the rim of the cricoid cartilage, thus abducting (opening) the vocal folds.

The lateral cricoarytenoid muscles draw the arytenoid cartilages anteriorly, shortening the vocal folds when they are held in adduction, and may cause slight rotation of the arytenoid cartilages, pressing the two vocal processes together.

The cricothyroid muscle lifts the anterior arch of the cricoid cartilage, tilting the top of the lamina posteriorly, thus lengthening the vocal folds. It may also move the cricoid cartilage posteriorly in relation to the thyroid cartilage.

The lateral portion of the thyroarytenoid muscle may act with the transverse arytenoid muscle in adduction of the arytenoid cartilages: may shorten the adducted folds for biologic and phonatory purposes. The medial portion may aid in adduction of the arytenoid cartilages and shortening of the vocal folds for biologic purposes: may aid in tension regulation for phonation. (3)

The reason for mentioning the trachea and esophagus is that they are connected to the larynx and pharynx. The trachea, or windpipe, is attached to the inferior side of the cricoid cartilage and connects the larynx with the bronchial tubes and lungs.


Cinefluorographic films taken of six subjects using throat vibrato were analyzed for structural motion in synchronization with the vibrato. The films were viewed by the writer, an anatomist, two radiologists, and a speech pathologist. Films of tones with and without vibrato were traced, so that motion could be shown by measurement, and measurement graphically outlined.

The six subjects were faculty members and students of the University of Iowa School of Music. They included the oboe instructor, the flute instructor, a Master of Fine Arts candidate in oboe, one graduate, and two undergraduate flute students. Two were male and four were female. Their ages ranged from approximately twenty to forty years. All believed they used throat vibrato in their playing and indicated they could start and stop vibrato at will. Each chose his own vibrato rate.


The cinefluorographic equipment used in the study consisted of the following components: 1) a 300 milliampere generator, 2) a Rotalix 0-75/125 x-ray tube with a 0.6mm focal spot, 3) a fluoroscopic screen, 4) a nine-inch image intensifier tube, 5) a Cinevoice (model CM 72) sound-on film camera, 3) an Electrovoice (model 644) unidirectional microphone, and 7) an Auricon sound-on-film amplifier.

Kodak Plus X Reversal film No. 7276 was used and was processed by the University of Iowa Photography Laboratory. The contrast between light and dark was increased by making a print using color film.

The musical task was also recorded on tape at the same time that the films of each subject were made. This was done so that the musical task could be analyzed separately from the sound track on the cinefluorographic films. The tape recording was played through an oscilloscope. Earphones were attached so that the tape recording could be heard while the oscilloscope screen was being observed.

Preparation of Subjects

Each subject was seated in a dental chair. An adjustable head positioner consisting of ear rods and forehead bumper was suspended from the ceiling directly above the chair and attached to the subject's head to assure immobility. The subject was positioned with the x-ray beam approximately centered on the larynx. This allowed the larynx, pharynx, mandible, tongue, oral cavity, and cervical vertebrae to be viewed.

Each subject was instructed in the musical task and procedure. He played through the task before being positioned in the chair and played it again while in the chair, before being filmed. The beginning of the film of each subject was marked with a subject number, and a plastic rod containing 1/4'' ball bearings was photographed in the subject's mouth so that the films could be projected, traced, and measured at life size.


Each subject played three pitches--D, D', and D''. (Figure 4) He started each pitch without vibrato, began vibrato on the fifth beat, returned to a vibratoless tone on the ninth beat and ended the tone on the thirteenth beat. Speed of the beats was one per second (MM = 60).

In order to allow the subject to produce vibrato comfortably, no dynamic level was prescribed. It was felt that too soft a dynamic level might have encouraged the subject to use too small a vibrato, which would have been difficult to note aurally or visually. Too loud a dynamic might have encouraged the subject to produce an unnaturally wide vibrato, and might possibly have brought certain muscles into play which are not usually associated with throat vibrato.

Each subject was given four preparatory counts in time with the metronome, which remained audible during the entire task. The cinefluorographic equipment was started on the third preparatory count, so that the beginning of the first pitch could be clearly filmed. The subject then performed the task. Two beats elapsed between each pitch, and the subject was signaled when to begin and end a pitch and each section within a pitch.

This task was used in order to establish definitely that structural motion could be seen during vibrato production which did not occur during nonvibrato.


The film of each subject was marked off in seconds with a grease pencil. In the nonvibrato sections, the middle 24 frames marked A-B and E-F, were marked off for tracing, and in the vibrato sections, labeled C-D, the middle 48 frames were marked off for tracing. The middle frames were chosen so that the starting and stopping process of the vibrato could be avoided. Each tracing was marked with the subject number, A, C, or E for the sections which it represents, and with L, M, or H for the level of pitch which it represents.

The films were projected at life size using a Bell and Howell Specialist Autoload Projector (model 1552). Sheets of Clearprint 1000H tracing paper were attached to a clear plexiglass surface and the pictures were projected on them, one frame at a time. In order to facilitate later tracing and measurement, lines were drawn on the tracing paper from the superior point of the first cervical vertebra (atlas) to the lower edge of the upper incisor (w-x), and from the posterior point of the sixth cervical vertebra to the posterior point of the atlas (y-z) Figure 5). The two lines were traced on each piece of tracing paper used for a given subject, thus making a template for that subject. It was then placed on the plexiglass and set to coincide with the points of the vertebrae and incisor so that the structures which were traced would be the same distance from the lines and therefore more easily traced and compared.

The hyoid bone, posterior pharyngeal wall, epiglottis, inferior line of the mandible and dorsum of the tongue were traced for all subjects. The arytenoids, aryepiglottic folds, laryngeal ventricle, and outline of the thyroid cartilage were traced for all subjects on which they were visible (Figure 6).

Measurements between specific points were chosen, based on observation of the films. A point near the posterior pharyngeal was located, near the laryngeal areas where motion was observed. Through this point, a line was drawn perpendicular to the superior template line (Figure 7, line w-x) and another line was drawn perpendicular to the posterior template line (Figure 7, line y-z). This was used as a template for measurements which were then made from the point where the perpendicular lines intersected, along the line parallel to the line y-z, to the aryepiglottic fold (Figure 7).

These distances were transferred from the tracings to graph paper with a divider, so that a frame-by-frame analysis could be made and outlined graphically (Figure 8).

Observations and Discussion

The cinefluorographic films were viewed and analyzed by the writer, an anatomist, two radiologists, and a speech pathologist. Motion in vibrato and nonvibrato sections of tone was compared.

The visual analysis and the graphs of tracing measurements indicated that there was structural motion in the throat, specifically in the larynx, which occurred when vibrato was used. One could clearly see movement of the arytenoids, aryepiglottic folds, and laryngeal ventricle on the films of some or all of the subjects. That motion did not occur during nonvibrato tones.

Arytenoid and Aryepiglottic Fold Movement

Arytenoid motion during vibrato production caused an anterior to posterior oscillation of the aryepiglottic folds. This was at the subject's vibrato speed for subjects one, two, three, and five. For subjects four and six (the two instructors), very little structural motion was noted and was almost entirely in the arytenoids. (4) Movement of arytenoids and aryepiglottic folds on all subjects was structural, from which muscular activity was deduced. According to the film observations of the radiologists, the anatomist, and the speech pathologist, based on structural motion within the larynx, movement of the posterior and lateral cricoartenoid muscles the cricothyroid muscle, and the vocalis muscle (a part of the thyroarytenoid) was noted. All of these are directly involved with opening, closing, tensing, or relaxing the vocal folds.

It would appear that pulsing of the air column was accomplished by the vocal folds by action of the arytenoids. This is the only place in the throat where the air stream can be modulated. It should be stressed that the vocal folds are not being used in a sound-producing capacity, but as mechanism to modulate and control the flow of air from the trachea into the oral cavity. A variation primarily of intensity would be the result of this type of air modulation.

Laryngeal Ventricle

Some changes in the shape of the laryngeal ventricle were evident on the films of three subjects (two, three, and five), indicating that the vocal folds were moving during vibrato. The changes were minute in all three subjects and not consistent enough for tracing and measurement.

Results of Tracing and Measurement

It was noted in the visual observation of the films that the most persistent movement visible during vibrato was the activity of the arytenoids and aryepiglottic folds. The arytenoid movement, which caused the aryepiglottic fold movement, is of primary interest. The arytenoids are difficult to trace because of a lack of clear outlines. They are within the thyroid cartilage and their outlines are not distinct because of the cartilage and the muscles surrounding it. The arytenoids could be seen on the cinefluorographic films, but were not always clear enough to trace. Consequently, the muscular movement which was traced and measured was that of the aryepiglottic folds.

Measurements of tracings of the six subjects showed the same results as the film observations of the writer, the radiologists, the anatomist, and the speech pathologist. The main movement for subjects one, two, three, and five was of the arytenoids and aryepiglottic folds. Arytenoid and aryepiglottic fold motion of subjects four and six was not shown on the tracings.

By studying the outline of measurements in the graphs, the amount and regularity of aryepiglottic fold movement in the vibrato sections (C-D) and nonvibrato sections were compared (Figure 9). Peaks in the graphs represent the greatest distance from the fixed point on the measuring template to the aryepiglottic fold. Troughs represent the shortest distance from the point.

Analysis shows that peaks occurred regularly every three to five frames in the vibrato sections, as opposed to less regular intervals in nonvibrato sections. This was true of all subjects measured (one, two, three, and five). The extent of the aryepiglottic fold movement in the vibrato sections of tone was greater than movement in the nonvibrato sections by three to five millimeters indicating greater arytenoid activity during vibrato.

The number of peaks (each representing a vibrato cycle) for each two-second time period (C-D) was between ten and fourteen. This would represent a vibrato rate of five to seven cycles per second.

Oscilloscope Reading of Tape Recording

The oscilloscope showed a pronounced cycle of change in the amplitude of the wave during vibrato, while the waves remained quite stable during nonvibrato. The changes from low to high amplitude during vibrato sections of each subject were at the vibrato rate for that subject. This indicated a definite modulation of intensity during vibrato, which was probably a result of the air column being modulated by the vocal folds, changing the speed and volume of air into the oral cavity.

Certain changes were noted in the waveforms of the oboe tones during vibrato, indicating possible timbre variations. This was not as evident in the flute tones, which were seen on the oscilloscope screen as more pure waveforms than those of the oboe.

No frequency changes were noted; however, this may be due to the relatively unsophisticated oscilloscope and the writer's lack of training in this area.


In this pilot study, the author has demonstrated a method of using cinefluorography to study the throat while vibrato tones are produced on flute and oboe. It is the first such use of cinefluorography.

Indications are that 1) vibrato production in the throat can be observed with cinefluorographic techniques, 2) the air column is modulated by the vocal folds, and 3) a variation of intensity appears to be a resultant of throat vibrato.

While one cannot observe the muscle activity directly it can be deduced by observing the related structural movement. Therefore, the use of cinefluorography as a tool for investigating the throat while vibrato tones are produced depends upon deducing muscular activity from structural motion. Also, the usefulness and validity of cinefluorography is limited by the clarity with which one can see structures of the larynx, particularly the vocal folds, arytenoids, aryepiglottic folds, and hyoid bone.

The most conclusive evidence of structural motion during throat vibrato production was noted by visual observation. Less conclusive evidence was found on the tracings. They supported the findings of the visual observations and indicated the amount of structural motion.

Arytenoid and Vocal Fold Movement

Movement of the arytenoids and the accompanying motion of the aryepiglottic folds indicated that there was also movement of the vocal folds. The radiologists anatomist, and speech pathologist reported that the arytenoid movement indicated active participation of the lateral and posterior cricoarytenoid muscles, the cricothyroid muscle, and the vocalis muscle. All are involved with movement of the vocal folds. The aryepiglottic folds and the arytenoids were moving at the speed of each player's vibrato, and the vocal folds were also assumed to be moving at the player's vibrato rate.

Intensity Vibrato

It appears that the air column was modulated by movement of the vocal folds. This caused changes in the speed and amount of air reaching the oboe reed or flute head joint, and resulted in changes in the intensity of the sound. Assuming constant pressure from the abdominal muscles, the volume and intensity of the air would decrease as the vocal folds are approximated. As they are opened, the volume and intensity would increase.

The Oscilloscope reading showed a variation in the amplitude of the wave corresponding to the speed of vibrato cycles. This variation indicates a change in the volume of sound.

An oscilloscope study of vibrato might contribute significantly to the understanding of vibrato. Questions about the amount of pitch, intensity, and timbre modulation of vibrato tones might be answered. Coupled with further cinefluorographic research, the use of certain muscles or structures might be noted to coincide with certain types of vibrato.

It must be stressed that throat vibrato has not been isolated in this study. There may be connections between diaphragm movement and throat vibrato. It would be necessary to rule out any movement of chest, abdominal or back muscles in connection with vibrato, to be assured that the vibrato is being produced only by muscles of the throat. With electrodes attached to the chest, abdominal, and back muscles, it may be possible to ascertain how much muscular activity takes place in these regions. Coupled with a cinefluorographic study of the throat some interaction or lack of it might be established.

This article is based on the author's unpublished dissertation entitled A Comprehensive Performance Project in Oboe Literature with a Cinefluorographic Pilot Study of the Throat While Vibrato Tones are Played on Flute and Oboe (University of Iowa, 1973).

(1) Willard R. Zemlin, Speech and Hearing Science Anatomy and Physiology (Champaign, Illinois: Stipes Publishing Company, 1964), p. 225.

(2) William H. Saunders. "The Larynx," Clinical Symposia XVI (Summit, N.J.: CIBA Corporation, 1964), p. 74.

(3) David Ross Dickson and Wilma M. Maue. Human Vocal Anatomy (Springfield, Ill.: Charles C. Thomas, 1970). p. 106.

(4) One possible reason for this obvious lack of movement may be that the instructors' vibrato mechanism is more efficient that the students' and consequently shows less motion. This efficiency would probably be the result of less conscious muscular effort as the mechanism is used and relegated to habit.

Table of Contents