Paper - A model to illustrate the probable action of the tectorial membrane (1915)

From Embryology
Embryology - 29 Feb 2020    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Hardesty I. A model to illustrate the probable action of the tectorial membrane. (1915) Amer. J Anat. 18(3): 471-.

Online Editor  
Mark Hill.jpg
This historic 1915 paper by Hardesty describes tectorial membrane development. "Tectorial" means forming a cover.

See also by this author: Hardesty I. On the proportions, development and attachment of the tectorial membrane. (1915) Amer. J Anat. 18(1): 1-70.

Modern Notes: inner ear | pig

Hearing Links: Introduction | inner ear | middle ear | outer ear | balance | placode | hearing neural | Science Lecture | Lecture Movie | Medicine Lecture | Stage 22 | hearing abnormalities | hearing test | sensory | Student project

  Categories: Hearing | Outer Ear | Middle Ear | Inner Ear | Balance

Historic Hearing 
Historic Embryology: 1880 Platypus cochlea | 1892 Vertebrate Ear | 1902 Development of Hearing | 1906 Membranous Labyrinth | 1910 Auditory Nerve | 1913 Tectorial Membrane | 1918 Human Embryo Otic Capsule | 1918 Cochlea | 1918 Grays Anatomy | 1922 Human Auricle | 1922 Otic Primordia | 1931 Internal Ear Scalae | 1932 Otic Capsule 1 | 1933 Otic Capsule 2 | 1936 Otic Capsule 3 | 1933 Endolymphatic Sac | 1934 Otic Vesicle | 1934 Membranous Labyrinth | 1934 External Ear | 1938 Stapes - 7 to 21 weeks | 1938 Stapes - Term to Adult | 1940 Stapes | 1942 Stapes - Embryo 6.7 to 50 mm | 1943 Stapes - Fetus 75 to 150 mm | 1946 Aquaductus cochleae and periotic (perilymphatic) duct | 1946 aquaeductus cochleae | 1948 Fissula ante fenestram | 1948 Stapes - Fetus 160 mm to term | 1959 Auditory Ossicles | 1963 Human Otocyst | Historic Disclaimer

PubMed Search: tectorial membrane development |

Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

A Model to Illustrate the Probable Action of the Tectorial Membrane

Irving Hardesty

From the Deparlmenl of Anatomy of the Tulane University of Louisiana

Seven Figures


In some studies of the cochleae of the pig, the writer ('08 and '15) became convinced that the tectorial membrane and not the basilar membrane is at least the chief vibratory structure in the mechanism for hearing. With this conviction, the construction of a model was undertaken with the hope that it might illustrate something of the possible behavior of the mammalian tectorial membrane when energy represented by sound waves is imparted to the fluid in which it lies.

Of the many experiments with apparatuses and the several models described by others, none have attempted to reproduce the form and arrangements of the parts of the actual auditory apparatus and but little of the conditions under which these parts are supposed to act in the animal. It seemed to the author that more instructive indications of the process of hearing might be obtained with a model in which the outer, middle and inner divisions of the peripheral part of the auditory apparatus are represented and the forms and relative arrangements of the structures comprising each division imitated, and a model to which might be applied the disturbances giving rise to sensations of sound in the actual ear. The construction of such a model must necessarily be preceded by detailed studies of the structures of the ear and especially of those structures considered acted upon by sound waves in a way to arouse the various sound impulses in the auditory neurones. Also it seemed advisable that the studies of the structures be carried to the labyrinths of adult animals, rather than confined to those of foetuses, and the writer's paper of 1915 describes some observations upon adult cochleae, chiefly those of the adult hog.

The purpose of this paper is to describe a model and to mention some of the results obtained with it suggesting the action of the tectorial membrane when actual sound waves are appUed to the tympanic membrane. It is thought advisable to precede this description by a brief resume of the anatomical observations made in previous papers which have led to the conviction that the tectorial membrane is the structure best adapted for the vibratory activities required. The model lays no claim to completeness of plan. Both its form and excellence of construction leave much to be desired. Consideration of only the few simpler and most evident of the results obtained with it will be undertaken. Many of the phenomena it presented appeared so mixed and complex in the physics involved that, considering the great coarseness and imperfection of the model as compared with the cochlea itself, attempts to interpret them seemed needless as well as almost hopeless. Some figures are given for review of certain of the anatomical features mentioned as well as figures to illustrate the construction of the model. Citations to most of the literature consulted are given in the prevjious papers and but few of these will be repeated here.


It is considered established (1) that the cochlea is the essential organ of the peripheral part of the auditory apparatus, that it contains the neuro-epithelium in which auditory impulses are aroused and is generally thought to be so constructed as to be especially capable of serving, in conjunction with the central nervous system, in the analysis of sound; (2) that the auditory impulses imparted to the fibers of the cochlear nerve are aroused in the hair cells of the neuro-epithelium (organ of Corti) by impingement of the hairs of the hair cells against the basal surface of the tectorial membrane, and that the impingement results from vibratory motion induced by that form of atmospheric disturbances known as sound waves; (3) that sensations of sounds of the highest pitches of which the apparatus is capable are mediated by the basal end of the coil of the cochlea. For example, Munk destroyed the basal end of the spiral organ in dogs, operating through the fenestra cochleae (rotunda), and found deafness to high notes to result, and Baginski destroyed the cochlea of one side entirely and then operated to destroy parts of the cochlea of the other side, obtaining results confirming those of Munk.

The most commonly accepted theories of hearing may be divided into two: that elaborated by Helmholtz, comprising ideas involving phenomena of resonance purely, and the Telephone Theory, comprising ideas involving little or no resonance.

One of the earliest advanced ideas was that the hairs of the hair cells are themselves agitated selectively, or in resonance, by the different sound vibrations imparted to the cochlea. This was quickly abandoned as untenable on the ground that the hairs are neither suitably constructed, long enough, nor vary sufficiently in length to be effectively acted upon by the vibrations as transferred to the endolymph in the cochlear duct. The Helmholtz theory is a resonance theory wholly and was applied solely to the basilar membrane, that narrow, thinnest span of the membranous spiral lamina. Practically all who have tried to apply it to given phenomena have had to modify it. Based upon erroneous descriptions by others (Nuel, 1872, for example) of the basilar membrane as composed of independent, radially disposed fibers and upon the fact that the membrane increases in width (in length of the fibers) in passing from the basal to the apical end of the cochlea, the theory requires the sympathetic or selective vibration of these fibers in resonance with the various vibration frequencies imparted to the endolymph. The hairs of those hair cells resting over fibers of the basilar membrane with a length, or natural vibration frequency, corresponding to the vibration frequency of a given note, were thought to be thrown against the basal surface of the tectorial membrane by the selective vibration of such fibers. Thus the theory assumed that the basilar membrane is composed of fibers of lengths varying according to the vibration frequencies of all the sounds the organ is capable of appreciating, that mixtures of sounds are analysed in the cochlea by the resonant vibration of the fibers in different parts of it corresponding to the different tones contained in the mixture.

Aside from the many physiological difficulties met in applying it, the Helmholtz theory is not supported by the later found anatomy of the cochlea: (1) The basilar membrane is not composed of independent fibers. The structure given the name is composed of three layers of white fibrous tissue, one of which courses at right angles to the others. That part of it which is radially arranged is nothing more than a flat tendon, the fasciculi of which are abundantly connected with each other by collateral branches. (2) Were the basilar membrane composed of radial fibers capable of vibrating independently, it is blanketed on both its sides by continuous and thick layers prohibitive of any such action. On its apical or vestibular side is spread the spiral organ (of Corti) and the membrana propria of its epithelium; on its basal or tympanic side extends the continuous layer of epithelioidal syncytium with its blood vessels and the continuous endothelium lining the scala tympani. (3) Were the membrane as the theory assumes, the cells of the spiral organ are so closel}^ associated and cemented together that individual hair cells or groups of hair cells overlying the fibers concerned with a given note could not be made to impinge separately against the tectorial membrane by the resonant vibration of the fibers. (4) It is questionable whether the fibers, were they independent and free, are long enough for the resonant vibration assigned them by the theory, especially for the sound waves of the lower notes appreciated. Liberal measurements in the human cochlea give the supposedly vibrating part of the membranous spiral lamina a width of only fV mm. (304^4) at its apical end where it is broadest and f\ mm. (168At) at its basal end. Measurements of the same for the adult hog gave 258At as the average width at the apical end and 185^ as the average width at the basal end. Helmholtz himself appreciated the doubt whether fibers so short as the width of the basilar membrane can be thrown into vibration by sound waves. (5) Sometimes in the cochleae of the pig, and perhaps other mammals, a part of the organ of Corti in the basal end may rest upon the bony spiral lamina instead of the basilar membrane.

The idea involved in the telephone theory was first suggested by Rinne in 1865, or about thirty years before the elaboration of the Helmholtz theory. It was more fully worked out by Rutherford in 1886 and since modified by Waller in 1891, Meyer in 1898, Ewald in 1899, Gray in 1900, and others. Originally it assumed that the vibrations imparted effect the cochlea as a whole. Rutherford at first suggested that all the hairs of the hair cells are thrown into vibration by each note and the impulses thus aroused in the cochlear nerve are merely similar in frequency, intensity and quality to the vibration frequency, amplitude and quality of the notes acting upon the apparatus. Therefore, the analysis of sound would be wholly cerebral. This idea that the impulses are aroused by the hairs being acted upon directly was early abandoned by Rutherford and the telephone theory became applied to the basilar membrane and all the later modifications of it have applied it to this membrane. Waller and Meyer assumed that the vibrations as transferred from the tympanic membrane to the endolymph of the cochlea affect the basilar membrane as a whole. Meyer supposed that the wave motion produced by each note, as it passes in the scala vestibuli, affects an extent of the basilar membrane just in the proportion that the amplitude of the vibration is not decreased by the resistance to be overcome in its passing toward the apex of the cochlea. Thus certain wave motions will affect greater extents of the basilar membrane than others and therefore will cause the hairs of a greater number of hair cells to impinge against the tectorial membrane. Waves of lesser amplitudes (intensity), in overcoming the resistance met in passing from the basis of the stapes toward the apex, earlier become too faint to sufficiently agitate the basilar membrane, each note involving an extent of the membrane according to its amplitude or intensity. In this idea pitch depends upon the vibration frequency (the number of stimulations of the hair cells per unit of time) and intensity is expressed in the total number of hair cells (extent of membrane) irritated. Obviously some analysis of sound may thus be made in the cochlea.

The telephone theory differs from the Helmholtz theory in that the latter supposes the basilar membrane composed of fibers of varying length, those of given lengths vibrating in resonance with waves of given vibration frequencies, making entire analysis of sound by the cochlea possible, while the telephone theory as now modified assumes that the basilar membrane vibrates as a whole to every note in such extent as the amplitude of the wave motion and the resistance to its propagation will allow, the auditory neurones transmitting to the brain impulses of frequencies and intensities corresponding with those of the vibrations concerned.

Leave is here asked to submit below a modification of the telephone theory applied to the tectorial instead of the basilar membrane. Siebenmann ('98) seems to have first recognized the importance of the tectorial membrane in the auditory apparatus, noting that it begins in the animal series with the beginning of 'musical hearing' and that its absence or deformity results in deafness. He was the first to suggest that it may be thrown into vibration by sound waves. Von Ebner ('02) suggested that the tectorial membrane, especially its free zone, may serve as a mechanism for sympathetic vibrations. Following von Ebner, Kishi ('07) and Shambaugh ('07) attributed powers of resonance to the tectorial membrane but considered it attached to the spiral organ and composed of independently vibratory elements ('lamellae').

The anatomical studies undertaken by myself have had to do with the tectorial membrane of the foetal and adult pig and the adult ox, rat and human. Most of them have been made upon cochleae of the pig. My conviction that the tectorial membrane is by form, nature and position the best adapted of the structures in the cochlea for the vibratory activities required in the process of hearing is based upon the following observations:

(1) The tectorial membrane is strictly coextensive with the spiral organ (organ of Corti). It is developed in company with and from the same variety of cells as the spiral organ (both being of ectodermal origin). The so-called basilar membrane is not always coextensive with the spiral organ.

(2) In position, the tectorial membrane lies over the apical or vestibular surface of the spiral organ and thus may be first and more directly acted upon by the usual sound vibrations, since these are imparted by the basis of the stapes directly to the fluid in the scala vestibuli. The basilar membrane lies under the basal side of the spiral organ and next the scala tympani, the thick spiral organ, the tectorial membrane and the vestibular membrane intervening between it and the scala vestibuli.

(3) In the other sense organs, and all sensitive surfaces, the impulses are aroused at the peripheral or outer surface of the epithelium. The tectorial membrane lies over the peripheral surface of the spiral organ while the basilar membrane lies under the organ. The membranes of the maculae and cristae, whose origin is directly homologous with that of the tectorial membrane, arouse impulses at the peripheral surface of their respective neuroepithelia.

(4) When the two are teased out for examination in the fresh, the tectorial membrane is found to be far more flexible than the basilar membrane.

(5) The tectorial membrane is considered more adapted for application of the telephone theory, capable of being more freely agitated, in that one edge of its supposedly vibratory portion, its outspanning zone, is free, while the basilar membrane is merely a flat tendon, one edge continuous from the tympanic lip of the spiral lamina and the other continuous into the spiral ligament. Furthermore, not only are both edges of the basilar membrane attached or continuous with the wall of the labyrinth, but it is loaded on its two sides by thick layers of other tissues likewise continuous beyond its edges and thus it must be less readily agitated than is the tectorial membrane, certainly by the vibrations of lesser amplitude.

(6) Figure 1 is given to show the shape of the tectorial membrane of the adult hog viewed on the flat from the basal surface, its coil slightly opened to obviate the overlapping in its natural form. It is seen that its vibratory portion or outspanning zone (OZ, this figure and figure 2) increases gradually and regularly in width from its basal end (BE) to its apical end. Measurements of its width in its different turns recorded in my paper of 1915, show that this zone is about 7 times as wide in the apical as in the basal end. The ends terminate bluntly rounded. Liberal measurements of the width of the basilar membrane (the portion of the membranous spiral lamina supposed to vibrate) show that its apical end is only about 1.8 times wider in the human and averages 1.4 times wider in the hog than its basal end. Measurements of the attached axial zone of the tectorial

Explanation of Figures

Reference Letters

AM, 'Accessory membrane' AS, Adjusting screw AZ, Attached axial zone

B, Battery BE, Basal end

BL, Bony spiral lamina BW, Bony wall of cochlea BM, Basilar membrane BS, Basis of stapes

C, Cork

CD, Cochlear duct

CN, Cochlear nerve

CP, Copper plate

CW, Copper wire

ED, Endolymphatic duct

EM, External auditory meatus

FC, Fenestra cochleae (rotunda)

FV, Fenestra vestibuli (ovalis)

GP, Glass plate

GT, Glass tube

H, Horn

HT, Huschke's teeth (edge of vestibular lip of spiral limbus)

L, Boundary line between outspanning and attached axial zone. (Line of imprint of Huschke's teeth)

.1/, ^lanubriuni oi Malleus

MB, Metal Japan button

0, Auditory ossicles

OZ, Outspanning zone of tectorial

membrane PW, Platinum wire R, India rubber to make water-tight RC, Cover for organ reed RT, Rubber tube S, Switch SS, Spiral sulcus SC, Set screws SG, Spiral ganglion SL, Spiral lamina (basilar membrane) <S Lg, Spiral ligament SM, Signal marker ST, Scala tympani SV, Scala vestibuli TL, Tympanic lip of spiral limbus TM, Tectorial membrane TpM, Tympanic membrane WB, Heavy wooden block WS, Wooden strip VM, Vestibular (Reissner's membrane)

Fig. 1 The basal surface of the tectorial membrane from the left cochlea of the adult hog drawn with the coil opened slightly to obviate the overlapping of its edges existing in its natural coil in the cochlea. Drawing made from several teased out membranes, some whole but slightly damaged in places, others broken in removal and the pieces mounted in their order on the slide.

membrane of the hog (AZ, figs. 1 and 2) show that it varies very little in width throughout its entire length. It is slightly narrower in the basal turn. The free edge of the outspanning zone extends well beyond the hair cells in the apical coil but barely over the outermost of the outer hair cells in the basal coil. Section A of figure 2 passes some distance from the actual basal end of the cochlea.

Measurements of the thickness of the tectorial membrane, necessarily made from sections of dehydrated and stained specimens in which the membrane was no doubt somewhat shrunken, show that its outspanning zone likewise increases gradually and regularly in thickness in passing from the basal to the apical end and that at the apical end it is 3 times as thick as at the basal end. There is practically no variation in the thickness of the adult basilar membrane; an}' little variation shown is never regular. The usual variations in the thickness of the membranes are indicated in figure 2, which represents transverse sections of the 1st, 3rd, 5th and 7th half turns in a section of one of the cochleae of the adult hog used for the measurements recorded in the previous paper.

Considering that the usual vibrations are imparted at the basal end by the basis of the stapes, then these vibrations in passing toward the apical end must, according to their vibration frequencies and amplitudes, be damped out in overcoming the inertia of the membrane itself as well as the resistance offered by the walls of the labyrinth and the fluid contained. Thus, it may be argued that the tectorial membrane, varying far more in width and thickness than does the basilar membrane in passing from the basal to the apical end, is adapted for being affected by a far greater variety of vibratory activity, has a much greater possible scale of activity, than the basilar membrane.

Variations in width and thickness are but indications of variations in volume and variations in the volume of the membrane, or the load it carries, are the most important factors to be considered determining the extent to which it may be thrown into vibrations by given vibratory disturbances. Not only must a more voluminous segment absorb a greater amount of energy, offer greater resistance to being thrown into vibration sufficiently for the necessary impingement upon the hairs of the hair cells, than the less voluminous segment, but it must possess a different natural vibration frequency. The natural vibration frequencies of strips of material, or the vibration frequencies with which they may act in resonance, vary according to their volume or load as well as according to their length. Computations, given in the previous paper, based upon the areas obtained of its transverse sections, suggest that the volume of a given very short length of the basal end of the outspanning zone of the tectorial membrane may increase as much as 40 times in grading to the volume of the same length of the apical end. Calculated in the same way, the volume of the assumed vibratory portion of the basilar membrane at its basal end increases only about 1.4 times in passing to its apical end. The average thickness of the basilar membrane of the adult hog, measured under the spiral tunnel, was found to be 2.8/x. It varies irregularly from 1.8m to 3.7^. Usually it is found to be thicker in its basal instead of its apical end.

Fig. 2 Transverse sections of the spiral organ and its tectorial membrane taken from one side of a vertical section of the cochlea of an adult hog. A, through the 7th half-turn of the coil, represents a section near the basal end of the organ; B, C and D represent respectively sections through the 5th, 3rd and 1st half turns of the coil.

The length of the tectorial membrane of the adult hog is about 26 mm. It comprises nearly 4 turns.

(7) The shape and attachment of the tectorial membrane may suggest that it is of other use than that of a mere foreign body spanning over the spiral organ for impingement of the hairs against it upon vibration of the basilar membrane. Its outer edge is free in the developed mammalian cochlea; its most voluminous and thus most varying portion is over the organ; it suddenly thins toward its attached zone as though the thinner part of this side of the outspanning zone may serve somewhat as a hinge ; and the contour of the basal surface over the organ is alwaj^s parallel with the apical (peripheral) surface of the organ while other parts of the basal surface are not.

(8) The structure and consistency of the tectorial membrane suggest that it may be especially sensitive to vibrations in the fluid in which it lies and that it may express such disturbances almost wholly in motions vertical to the surface of the spiral organ. It is composed of fibrils imbedded in a gelatin-like matrix and its specific gravity appears practically the same as that of the endolyraph in which it lies. Its fibrils are not arranged in it transversely but course in it obliquely, their direction inclined toward the apical end. Those in its apical side are more inclined toward its apical end than those in its basal side, thus intercrossing in the structure of the membrane rendering it less fragile for, in teasing it out even after fixation, it never breaks transversely but always tears apart (fig. 3). In its natural state, it is most inconceivably flexible. In the cochlea while being teased out and in the water under which it is teased, it becomes tangled or adheres to the needle at the slightest touch with most exasperating readiness. In the fresh state especially, it is extremely sensitive to agitations of the fluid surrounding it. This flexibility is much greater in the transverse directions. While wafted about in the water of the dish it manifests sufficient elasticity for its apical turns to resume their normal coil before it settles upon the bottom of the dish. Its thinner basal turn will not do so. It manifests considerably more elasticity in the direction of its width, or against stress applied parallel to its length. Except where it may become twisted, even its very thin axial zone is never found folded upon itself in mounts of the whole membrane nor in pieces of it. With its thin axial zone adherent upon the vestibular lip of the spiral limbus, this latter elasticity is considered sufficient to hold the outspanning zone in its position in proximity to the hairs of the hair cells whatever the position of the head of the animal, especially since the membrane manifests a specific gravity but very little greater than the fluid surrounding it. It is suggested that this greater elasticity in its width may allow a delicate spring-like action of the thinner axial part of the outspanning zone, and the adjoining part of the attached zone which overlies the projecting edge of Huschke's teeth {HT, fig. 2, C). Such action may aid in controlling such undulatory motion as the outspanning zone may assume. The form of movement of this zone is suggested to be possibly coarsely represented by that of a flexible ribbon attached along one edge and immersed in water, plus the elasticity and varying proportions manifest in the tectorial membrane.

Fig. 3 Piece from the 5th half turxi of a teased out tectorial membrane of the pig viewed from its apical surface. Given to show more nearly the actual appearance of the membrane as seen over a black surface and to show the course of its fibrils embedded in the matrix. The right hand side of the figure represents the appearance presented where the membrane had been torn across in teasing. The cut surface shown in the left side was not drawn from this specimen but added onto the figure as suggested in the vertical sections of this turn of the coil. The bottom part of the figure represents the attached axial zone of the membrane. The line showing through the outspanning zone is 'Hensen's sti'ipe.'

In constructing the model, the following considerations were in mind.

The spaces within the cochlea are completely filled with lymph in the normal condition. The vibratory motion imparted to this lymph by the basis of the stapes must be of the form shown by experiment to be propagated in a column of water upon percussion at one end of the column, namely in the form of compression waves, or alternate phases of condensation and rarefaction, moving longitudinally. The pressure under which the endolymph and perilymph exists is probably the same as the general blood pressure of the animal. Should it be greater or less at any time, equilibrium may be regained chiefly by way of the endolymphatic duct and cranio-spinal fluid and by way of the blood vessels of the labyrinth. Considering the spiral lamina (osseous and membranous) the most rigid partition in the cavity of the cochlea, the two scalae, continuous with each other at the helicotrema, may be considered as a column of fluid along which at least the fainter of the vibrations imparted by the percussions of the basis of the stapes may pass apexward in the scala vestibuli and basalward in the scala tympani. Considering the lymph in the labyrinth as incompressible by such force as may be imparted to it by the basis of the stapes, pressure resulting from strong pulsations of the stapes may be compensated or relieved by the membrane over the fenestra cochleae and also, at need, through the endolymphatic duct by way of the ductus reuniens and sacculus. The membranes bounding the latter and the vestibular (Reissner's) membrane are considered so delicate as to allow scarcely appreciable differences in the vibratory movement in the fluid on their two sides, namely, in the perilymph and endolymph.

Sound waves afi"ecting the tympanic membrane are transformed by the ossicles, the amount of work transferred to the stapes being somewhat controlled by the muscles of the middle ear. The basis of the stapes fits accurately into the fenestra vestibuli (ovalis) and is joined to its bony walls in a piston-like joint. Taking into consideration the difTerence between the area of the tympanic membrane and the area of the basis of the stapes, together with the leverage afforded by the form and arrangement of the ossicles, it has been computed that the actual force of the motions of the tympanic membrane produced by the atmospheric disturbances may be increased thirty times in its transformation and transference to the perilymph, and that on the other hand the amplitude of the atmospheric vibrations may be reduced as much as seventy times. The amount of the increase in force and the decrease in amplitude depends upon the tensity of the tympanic muscles and also upon the pressure of the air in the tympanic cavity. Thus it may be borne in mind that the vibrations imparted to the lymph of the cochlea by the stapes must correspond to the atmospheric waves but resemble them only in frequency of vibrations and quality of vibration, for the quality of the vibrations transferred to the endolymph must depend upon the quality or form of the atmospheric vibrations acting upon the tympanic membrane.

The lumen of the scala vestibuli decreases in passing from the base to the apex of the cochlea and that of the scala tympani increases in passing from the apex to the base. The cochlear duct increases in diameter in passmg to the apex where it ends blindly. In the adult hog (fig. 4), the space on the apical side of the spiral lamina, namely the scala vestibuli and cochlear duct combined, increases in size in passing from base to apex. The lumen of the scala tympani increases in passing basalward much more than either or both the spaces on the apical side of the spiral lamina, enlarging very greatly in its basal or longest turn. These variations in the lumen of the lymph spaces must result in variations in the resistance offered by their walls to the vibratory motions passing from the base tow^ard the apex of the cochlea. Waves passing apexward in the scala vestibuli are thought to be imparted sunultaneously to the cochlear duct, the vestibular membrane between being considered only to control slightly or damp their force as affecting the tectorial membrane. In the cochlea of the adult hog, there appears to occur in the third turn of the coil a slight constriction of the scala vestibuli (fig. 4). This seems present but less apparent in the adult beef but not noticeable in the adult human and rat. No attempt was made to reproduce it in the model, though when present it must make an additional variation in resistance offered to transmission of wave motion toward the apex. The coiled character of the walls of the cochlea must give complicated results of resistance, some of them being no doubt phenomena due to reflections of the vibratory motions. In the model no attempt was made to reproduce the coil of the cochlea because of the difficulties of construction involved. It was decided that such evidences of vibration as could be obtained with the cochlea represented as a straight canal would be suggestive of actual activities though perhaps simpler.

The two scalae are relatively larger in the human cochlea than in those of the hog, beef and rat.

Fig. 4 Outline drawing of a vertical section of the cochlea of the adult hog showing variations in the scalae vestibuli andtampani and in the proportions of the tectorial membrane. For meaning of reference letters, see page 478.

Figure 5, A, is a line drawing of the model entire with the accessory parts attached as used in most of the work with it. Figure 7, A, shows that end of the model which represents the basal end of the cochlea and figure 6 represents in detail a transverse section of the model.

For the labyrinth of the cochlea, three pieces of wooden board 1 J inches thick and 44 inches long were so cut and joined together as to make a water-tight trough with inside dimensions of 2j inches square and 42 inches long when the ends, cut from the same board, had been put on. The joints, cut as shown in figure 6, were coated with a fresh xylol solution of asphaltum immediately before screwing them together. The spiral lamina, with its membranous portion and its vestibular and tympanic lips was represented b}^ a piece of wood 2| inches wide and trimmed out as indicated by SL, figure 5, C, and figure 6, the spiral sulcus, SS, being attained by cutting a lateral groove in the wood. The lamina was fitted as a water-tight partition along the trough, so slanting that the space above it, representing the scala vestibuli, SV, figure 6, increased in passing from the basal toward the apical end and thus the space below it, the scala tympani (ST) was smallest at the apex and increased in passing toward the basal end. At the basal end (BE, fig. 5) the lamina was fitted water-tight against the inner side of the end of the trough, spanning between the openings in this end representing the fenestra vestibuli above it and the fenestra cochleae below it. The basal end of the model showing these openings, FV and FT, closed by their respective membranes, is illustrated in figure 7, A. At the apical end the lamina was cut about 1^ inches shorter than the length of the trough, thus leaving a helicotrema or continuation here of the scala vestibuli with the scala tympani. The top side of the trough could be closed water-tight by means of a plate of glass {GP, fig. 6) cut to fit and placed upon strips of India rubber upon which it was pressed down firmly by means of wooden strips placed along its edges and held so pressed by means of metal Japan buttons (MB, fig. 5, A, and fig. 6) set at intervals along the wall of the trough.

Even an approximate imitation of the tectorial membrane was found verj^ difficult to attain. A material was desired with low^ specific gravity and a texture sufficientl}^ tough to hold when trimmed to the shape and proportions of the actual tectorial membrane. Gelatin cast in the desired shape and fixed with formaldehyde was tried but the thin, attached axial zone was not sufficiently resistant for the purposes of the experiments nor would the gelatin hold the wires which were decided to be passed through it at intervals for making electrical contacts. Chamois-skin has a specific gravity but little greater than water





1 ■ - ' n






^ ^ J


Fig. 5 A, line drawing of the model of the auditor}' apparatus with accessories as assembled for experiments; see reference letters on page 478 for names and purposes of parts. B, line drawing of the type of brass organ reed used for producing sound waves in most of the experiments. C, line drawing showing the artificial tectorial membrane as attached to wooden spiral lamina which fitted as a water tight partition dividing the trough of the model into scala vestibuli and scala tynipani.


Fig. ( of parts

) Line drawing of cross-section of model to show position and relation within; see reference letters on page 478 for names of parts.

Fig. 7 A, line drawing of 'basal end' of model to show the two fenestrae as represented and the position of the 'endolymphatic duct' and the support of the model. B, the form finally used to represent the auditory ossicles. BS, basis of stapes; M, manubrium of Malleus. For other reference letters, see page 478.


when wet and immersed. It could not be obtained thick enough to at all represent the thickness of the tectorial membrane. Some results were obtained in using a strip of chamois skin, one edge of which was turned over to represent the varying width of the outspanning zone of the membrane and stitched down along a line corresponding to the line of the edge of Buschke's teeth, while the remaining width of the strip represented the thin, attached axial zone. The long, tapering sac formed by the turned edge was then filled with a melted 8 per cent solution of gelatin, allowed to set, then fixed in 10 per cent formalin. When immersed, in water this appeared some heavier than was desired but when tacked down by its thin edge along upon that part which represented the vestibular lip and arranged under water in the adjusted model, it gave considerable evidence of vibratory activity. However, after the prolonged soaking necessary to remove the formaldehyde, this contrivance began to show inequalities from beginning disintegration of the gelatin after some time in the water in the model. It of course at no time could be allowed to dry out. Such final results as were obtained came with the use of an imitation but little more satisfactory than the preceding, except that it was more permanent. At a harness factory I found a piece of very thick leather which was said to be elk's hide. It was so tanned as to somewhat resemble the result obtained in chamois skin, being soft and easily wet, and more than usually flexible when wet. Any thick leather tanned by the same process would do as well probably, though there is apparent reason for the impression that the hides of the deer family are more porous than the beef hides. ^The strip of elk's hide was cut the length and width required and trimmed from the 'flesh side' to the approximate relative proportions of the tectoria' membrane. It was kept in frequently changed water for several days to remove any substances- soluble in water. During this washing it was often kneaded to increase its flexibility. Figure 5, C, gives an idea of it as tacked down by its thin edge (attached axial zone) along upon the vestibular lip of the spiral limbus. In this figure, drawn in larger scale than the model below, it is shown, to sug THB AMERICAN JOURNAL OP ANATOMY, VOL. 18, NO. 3


gest its shape in transverse section, as though two small segments had been cut out. When not surrounded by water in the model, its weight caused it to rest limply upon the spiral lamina below, but with the trough filled with water, its thin attached edge was sufficient to support it in a horizontal position. It may be noted that the outspanning zone is trimmed so as to gradually taper and is made much narrower in its basal end (BE).

With the spiral lamina fixed so as to divide the trough into the two scalae, pieces of platinum wire (PW) were forced through the outspanning zone of the leather tectorial membrane at six intervals. These wires were bent at right angles and joined by interlocking loops along the inner side of the trough with copper wires (CW) which passed through corks driven into holes bored in wall of the trough at the corresponding intervals (fig. 6). The platinum and copper wires were joined by looping them together to avoid as much as possible the additional rigidity that would be given the membrane by continuous wires, the loops being thought to act somewhat as hinges during movement of the membrane. That end of each of the platinum wires which passed through the outspanning zone of the membrane was arranged so that it could come in vertical contact with the surface of a small copper plate (CP) soldered onto the end of a heavier copper wire which passed through a small glass tube (GT, fig. 6) inserted in holes bored vertically through the spiral lamina and the bottom of the trough below. The passage of the glass tube through the bottom was rendered water-tight by forcing it in surrounded by a piece of rubber tubing, and, for the same purpose, the heavier copper wire passed through plugs of rubber which were forced into the glass tube {R, fig. 6) . Each of the six heavier copper wires (with the copper plates affixed upon their upper ends) was inserted below into a small brass block and held by set screw, and the lower end of each brass block had threaded into it an adjustment screw (AS, figs. 5 and 6) by which the copper plate above could be raised and lowered to adjust the contact between it and the end of the platinum wire. The adjustment screws were held fixed by attachment in a wooden strip (}VS) extending along beneath the model. To the lower


end of each of the heavier wires was attached a piece of copper wire of the same size as that interlooped with the platinum wire above. Thus, a battery interposed between these two copper wires would make possible an electric circuit which could be made and broken at the contacts between the platinum wire and the copper plate by vibrations of the artificial tectorial membrane. Figure 5 shows a battery (B) interposed in the circuit of the sixth interval, the circuit of the part of the model representing the apical end of the cochlea.

The object of the similar arrangements of the wires at the 6 intervals was for the experimental determination of the behavior of the different regions of the tectorial membrane toward a given form of vibratory activity imparted to the fluid in the model at its basal end (BE). In addition to the necessary use of an ordinary switch (S, fig. 5) in the electric circuit, a simple electric signal-marker (SM, fig. 5) such as is used in physiological laboratories, and a telephone (TEL) were interpolated for use in observing results. A small piece of white paper was glued vertically on the end of the vibrator of the signal-marker that vibrations in it could be more easily seen. The noise of the horn (H) usually obscured its clicking to the ear of the observer. The telephone was found to be far more efficient than the signalmarker used. It gave evidence of vibrations of all frequencies whatever induced in the model with sufficient force and amplitude to make and break the circuit, while the signal-marker was found incapable of recording the higher vibration frequencies of which the model was capable. In other words, with the application of sounds of higher vibration frequencies, there could be heard a continuous buzz in the telephone when the vibrator of the signal-marker was practically motionless. An ordinary telegraphic 'sounder' was tried in the earlier stages of the experiments but had to be discarded because found capable of service only with vibrations of lower frequencies.

Before the use of the electric current was suggested, small mirrors were tried with the hope that vibratory activities could be read in the behavior of the small spots of light reflected by them. Thin circular cover-glasses were silvered on one side


and cemented with thin asphalt to very minute cork blocks stuck at intervals along the upper surface of the leather tectorial membrane. The apparatus was placed in dii-ect sunlight with the idea that evidences of vibration could be read in the movements of the spots of light reflected by the mirrors upon the ceiling of the laboratory, or on specially devised screens tried later, the long light-levers magnifying the movements. This plan, however, was found disappointing because of the very indefiniteness of its results. The light reflected from the surfaces of the mirrors had to pass through the w^ater in the apparatus and the glass plate of its top side and was so refracted that the boundaries of the spots upon the screen or ceiling were not sharply defined. Continuous vibration could not be distinguished from intermittent, vibrations of higher frequencies were doubtfully discerned and all results seemed exasperatingly confused. The mirrors gave evidence, however, of the first result obtained, namely, that vibrations may be induced by sound waves in a structure imitating the proportions and position of the actual tectorial membrane.

Distilled water was used to represent the lymph in the cochlea. The actual lymph of course differs in viscosity from distilled water and may be better adapted for the conduction of vibratory motion. But in using the electric current for indicating vibratory activity in the model, the fluid used had to be as nearly as possible a non-conductor of electricity. An oil with a low boiling point would of course have been most efficient as a nonconductor, but so far oil has not been used because of the difficulty with which its leakage could be prevented with a construction of the apparatus as- here employed and because it was considered that a difference in viscosity between an oil and the lymph would be greater than that of water. It is realized that the ordinarily used distilled water is not absolutely a non-conductor of electricity. Its conductivity is less than tap water and much less than the animal lymph. That the distilled water in use in the model might be as pure as possible, the trough was soaked in distilled water, frequently changed, for about two weeks before the experiments using the electric current began. Further,


the water was replaced with fresh at the beginning of each series of trials with the model. Thus since the work extended intermittently through a number of months, it is thought that the water used toward the last at least was fairly free of extracts from the wood and other material used in construction.

Instead of the copper plates employed for contact with the ends of the platinum wires, very small copper cups containing a drop of mercury were at first tried on the ends of the pieces of heavier copper wire. It was found, however, that the movements and changes in shape of the drops of mercury due to changes in surface tension when charged and uncharged seemed to render the contacts less definite and delicate and adjustment of the distance through which contact was made much more difficult than with the use of the copper plates. The plates were kept burnished.

The trough was always filled completely with water. The bent glass tube, ED, figures 5, 6 and 7, was put in the model to represent the endolymphatic duct of the labyrinth. It was considered a precaution for compensation in any changes in pressure produced by vibratory pulsations and by changes in temperature. By means of this tube the filling with water could be completed after the glass plate had been put on and pressed down watertight. Remaining air bubbles were gotten rid of through a small hole drilled in the glass plate at its apical end, the hole being closed with a plug of rubber.

The end of the model representing the basal end of the cochlea is shown in figure 7, A. The fenestrae were represented by cutting large holes through the wooden end piece and closing these holes with thin skin. To accomplish this cuffs of sheet copper were made to fit the holes and over one end of each of these was stretched and tied a piece of thin raw-hide of the sheep known as 'drum-head skin.' The end of the cufT carrying the skin was driven into the hole till the skin, or membrane, became flush with ihe inner surface of the trough. Driving in thus made the membranes tense. Then melted wax was poured around the cuffs making their insertion water-tight. The fenestra vestibuli (FV, fig. 7) was made oval in shape and larger than the


fenestra cochleae {FC) below it. These differences were probablyunnecessary'.

The external auditory meatus {EM, fig. 5, A) was represented by a wide funnel made from block-tin. A wooden collar was fitted upon the smaller end of this funnel that the tympanic membrane could be stretched over it and secured more safely and easily. For the tympanic membrane (TpM), a sheet of that thin preparation of white fibrous connective tissue known as 'Gold-beater's skin' was wet and stretched over the wooden collar of the meatus and bound firmly about it with both twine and rubber bands. The membrane became tense upon drying. The collar was so trimmed that only its periphery was in contact with the membrane, thus making the area of the vibratory part of the membrane greater than the area of the circle formed by the small end of the tin funnel. The area of this membrane was about 30 times the area of the skin closing the opening representing the fenestra vestibuli in the basal end of the model. A tympanic membrane represented by sheet-rubber was found useless in that it absorbed rather than transmitted its vibrations to the ossicle" pressed against it.

The ossicles of the middle ear were finally represented by a piece of wood trimmed as shown in figure 7, B, and figure 5, A (0), with an oval shaped piece attached to one end to serve as the basis of the stapes {BS, fig. 7). At first pieces of hard wood were cut and. joined together to imitate the three ossicles with their joints and lever arrangement and with rubber bands for the tensor tympani and stapedius muscles. This arrangement, however, was found difficult to keep in place during the experiments and it absorbed the vibrations of the tympanic membrane considerably more than the single piece shaped as shown. During the experiments, a straight single piece was tried placed perpendicular to the center of the tympanic membrane and to the membrane closing the fenestra vestibuli. This piece was of the same thickness as that illustrated in the figures and its "basis of the stapes" was similar, but its end representing the manubrium of the malleus was fan-shaped and was used extending in both directions across the center of the tympanic membrane


instead of having a single arm extending from the periphery of the membrane to its center as shown in figure 5. The planes of the two ends were cut at right angles to its long axis instead of in the form shown in figure 7. It was of interest to find that the vibrations of the tympanic membrane were transferred to the fluid in the model better and more definitely by the arrangement of a single arm extending from the periphery to the middle of the tympanic membrane, similar to the arrangement of the manubrium of the malleus in the mammalian ear, than by the straight piece pressing across the center of the membrane and extending at right angles to both its surface and that of the fenestra vestibuli. The form found most efficient was somewhat similar to the columella of birds.

Sound producers of various types were tried, including pipes giving different tones, trumpets (a cornet and a trombone), horns of different kinds, and finally an assortment of ordinary brass organ reeds. Organ reeds were used chiefly in the work for two reasons: first, instruments producing sound waves by means of vibrating tongues seemed to act upon the apparatus more definitely than any of those tried which gave in other ways the alternate phases of condensation and rarefaction, and second, the vibration frequency of each of the organ reeds was known, the name of the note it produced being stamped upon it. With the other instruments, the vibration frequency of a given sound had to be previously determined by means of a standard tuning fork and the kymograph, used with a writing arm registering the vibrations of the tympanic membrane. The form of the brass organ reed used is shown in figure 5, B. In order that its vibrations could be conducted directly into the external auditory meatus, a wooden box or horn (H, fig. 5, A) was constructed into the end of which fitted a removable cover for the reed (RC). This cover was so constructed that reeds could be interchanged at will. Air was blown through the reed by way of the rubber tube (RT). The whole sound producing apparatus was held in place by a heavy, ordinary filter stand and no part of it was allowed to touch the model nor the table upon which the model sat. The end of the horn was inserted into the large end of the


external meatus but could not be allowed to touch it during the experiments. Trumpets such as the cornet, held in the hand, were merely directed into the meatus.

The external meatus with the tympanic membrane had to be mounted upon a heavy w^ooden block (WB, fig. 5, A) and even then it was found necessary to brace it against a cleat fastened upon the table, for certain vibrations of low frequency would cause it to move gradually away from the ossicle pressed upon the membrane.

The adjustment of the distance between the copper plates and the ends of the platinum wires during the experiments was made to the point at which the least movement of either plate or wire would make or break the electric current. Usually a position just at the brink of the make of the current was employed. This could be determined by observing the vibrator of the signalmarker while manipulating the adjustment screws, the battery being interposed.


It was realized that an apparatus so infinitely larger, heavier and coarser in construction than the mammalian auditory apparatus could not be expected to give many results trustworthily representing the behavior of the actual apparatus. Many of the apparent phenomena with the model were so confused and seemed so complex that no attempt will be made to describe them on the ground that they were more the result of imperfections in the model than representing behaviors of the normal apparatus. Interpretations of only the few simpler results will, be undertaken and these interpretations will be limited.

I have described the construction of the model in considerable detail with the hope that its repetition with the possible improvements, may be suggested to other investigators and that the experuTients may be repeated and extended.

Of the results given by the model indicating the action of the tectorial membrane, the following are considered most directly suggestive.


(1) A structure imitating the tectorial membrane in shape, proportions and environment can be thrown into vibrations by sound waves so appHed that their energy is transferred to the fluid in which the structure lies and in a way similar to that employed in the actual ear.

(2) With sounds of low vibration frequency, at least, the tectorial membrane is thrown into vibrations of the same frequency as those of the sound giving rise to them. This fact was determined by allowing the vibrator of the signal-marker to write upon the drum of a kymograph, on which drum had been already traced the vibrations of a standard tuning fork. With the sound waves of higher vibration frequencies by which the model was capable of being acted upon, the signal marker either did not vibrate at all or acted irregularly.

(3) With the distance between the copper plates and ends of the platinum wires adjusted, a heavy step on the floor of the room, a slight blow upon the table or a hand laid upon the apparatus would produce vibrations of the artificial tectorial membrane sufficient to cause a series of makes and breaks of the current. This may be advanced as illustrating the possibility by which sensations of sound are aroused by way of other parts of the body than the tympanic membrane, stimuli explained as conveyed to the bony labyrinth, and thus to the structures within it, directly by way of the skull bones, teeth, etc. Sensations received in this way are among those of lower pitch.

(4) The model showed slightly less sensitiveness to vibrations imparted to the membrane covering the fenestra cochleae (rodunda) than to the same vibrations applied at the fenestra vestibuli. This slight difference is explained as due to damping out effects or absorbtion of the vibrations by the fluid in the scala tympani before they pass directly into the scala vestibuli by way of the helicotrema at the apex of the cochlea.

(5) Jarring of the model and notes of low vibration frequency applied to the tympanic membrane by way of the horn, were found to throw the entire tectorial membrane into vibration. This was determined by interposing battery, signal-marker and telephone at each of the regions of the membrane from 1 to 6,


(positions indicated in figure 5). It was found that all notes with vibration frequencies below that of the note a (concert pitch, 220 vibrations per second) resulted in constant vibration of all regions of the tectorial membrane sufficiently strong to make and break the current. This suggests that within the lower range of the natural auditory scale, all sound waves affect the entire spiral organ.

(6) The extent of the tectorial membrane that could be thrown into vibration by certain notes depended upon the intensity or amplitude of vibration with which the note was sounded. For example, the note g (196 vibrations per second), the next whole note below a, when sounded faintly into the external meatus, failed to result in vibrations of the apical end (region 6) of the membrane sufficient to make and break the current, while if sounded forcibly, this note caused vibrations in all regions. All sounds and notes with vibration frequencies below that of g resulted in vibration of the entire membrane, regardless of the amplitude that could be employed. Ewald ('99) applied a modification of the telephone theory to the basilar membrane. He constructed an arrangement in which narrow strips of thin sheetrubber were adjusted under water and observed with the microscope the behavior of these strips when sound vibrations were imparted to the water. His narrowest strips was 0.5 mm. broad. From the behavior of these strips he suggested that the basilar membrane vibrates throughout its entire length in response to every note applied. The above observation suggesting the action of the tectorial membrane, and the one following, do not fully support Ewald's conclusion for the basilar membrane as based upon the action of his rubber bands. His bands did not vary in thickness, or load carried, along their length. Meyer ('98) constructed an apparatus by which he applied the telephone theory to the basilar membrane. His apparatus was designed to indicate the results of varying amplitudes of vibration rather than varying pitch, using electrical contact and electric lamps as indicators. No attempt was made to imitate the basilar membrane or any other structures of the auditory apparatus. His results led him to conclude that the greater the amplitude of the vibration,


the greater the extent of the basilar membrane involved: that with a faint back-and-forth movement of the stapes only the basal or beginning part of the basilar membrane is set in motion; the greater the back-and-forth movement of the stapes, the further toward the apex of the cochlea will the motion of the basilar membrane extend. This suggestion for the basilar membrane agrees with that offered above as to the behavior of the tectorial membrane, namely, with a note of a given pitch or vibration frequency, a greater amplitude will throw into appreciable vibration a greater extent of the tectorial membrane than a lower amplitude, the extent beginning with the narrower basal end of the membrane and ascending, with the amplitude, along the gradually increasing membrane toward the apex of the cochlea. This action, shown by the model, is one to be expected, since the vibratory motion, imparted at the basal end of the cochlea, necessarily must be gradually absorbed in overcoming the resistance offered by the walls of the cochlea, the inertia of the fluid and the increasing inertia offered by the increasing bulk of the tectorial membrane itself in passing toward the apex of the cochlea. The variations in the diameter of the two sca.lae in passing from base to apex may result in additional decrease in the efficiency of the vibrations imparted. It is suggested that in the functional scale of any auditory apparatus there must be notes sounded whose amplitudes are gradually decreased by the resistance till the vibrations are completely damped out.

(7) It was further indicated by the model that frequency of vibration or pitch of a sound, considerably more than its amplitudes of vibration, determines the extent or length of the tectorial membrane that may be thrown into vibration. The note a (220 vibrations per second) produced with whatever amplitude, would not produce vibrations in the apical end of the tectorial membrane (region 6) sufficient to make and break the current, while it did produce vibrations in region 5. Note b (247 vibrations per second), the next whole note above a, also produced vibrations in region 5 when sounded with highest amplitude, but none in region 6. Both a and b produced vibrations in regions


1, 2, 3 and 4. The note c ( Wddle C/ 261 vibrations per second) produced vibrations in regions 1 and 2 but none in the other regions, regardless of ampUtude. Judging from the preceding, this note was expected to give vibrations in region 3 at least and I was unable to determine a reason for its not doing so. Also the notes d and e (294 and 329 vibrations per second respectively) gave vibrations in region 1 but not in the other regions, and the note / (349 vibrations per second) at greatest amplitude gave in region 1 only a slight hum in the telephone. Notes with vibration frequencies above that of / produced no evidences of vibration in the model. The note / and the three or four notes above / that were tried produced appreciable vibrations in the artificial tympanic membrane.

Beginning with the note h, the signal-marker ceased to be of service with the higher vibration frequencies. At times it would start vibrating at the first puff of the air through the reed and then become silent though a continuous hum might be heard in the telephone. The fact that notes c and d, e and / could not be made to give more graded results must have been due to imperfections in the construction of the model. Perhaps the intervals of the membrane arranged for were not short enough. However, such results as were obtained seem to indicate a relation between the vibration frequency of sounds and the extents of the membrane thrown into vibration by them. In explanation of this may be mentioned the familiar fact that of sounds given the same amplitude, those of lower vibration frequency are damped out less quickly in their transmission through a medium, or will travel farther, than those of higher frequency. For example, fog-horns of low vibration frequency or pitch are known to be more efficient for warnings at a distance. Also in atmosphere, sounds of high vibration frequency are not only sooner damped out in overcoming resistance than those of low frequency, but their speed of transmission decreases more rapidly as they become fainter. This damping out must occur much more quickly in transmission through a medium like the endolymph than in atmosphere and in overcoming the resistance


offered by the walls of the very small canals of the cochlea and the structures contamed withm them.

As noted above, the size and volume of the tectorial membrane so increases in passing from the basal to the apical end that the volume or load carried by a given short length of the apical end is about 40 times that carried by the same length of the basal end. This increase seems to occur uniformly throughout, or perhaps a little less rapidly through the basal turn of the coil. It is noted that these variations in the proportions of the tectorial membrane are far greater than can be claimed for any other structures of the cochlea, especially the basilar membrane. Increase of volume or the load carried by a structure means increase of its inertia. The much thinner and narrower basal end of the tectorial membrane extends in the region of the fenestra vestibuli (ovalis) , or the region at which the vibrations are imparted to the endolymph by the basis of the stapes. Obviously, a thin, narrow strip offers less resistance to agitation than a thicker, broader strip of the same material. Thus, the thin, narrow basal end of the outspanning or vibratory zone of the tectorial membrane may be thrown into vibration by notes of such high vibration frequencies as would be wholly damped out in overcoming the inertia alone of a more voluminous portion of the membrane. The volimie of the outspanning zone of the membrane increases with its distance away from the basis of the stapes and, therefore, among the appreciable sounds of higher vibration frequencies, those of different frequencies must be capable of throwing into vibration different lengths of the membrane, beginning at its basal end. The resistance offered by the endolymph and walls of the labyrinth, as well as the inertia of the membrane itself, contribute to their being damped out at their respective distances toward the apex of the cochlea. Since vibrations of higher frequency are damped out more quickly in transmission through a medium, there may be sounds whose vibration frequencies as imparted to the endolymph are such that they agitate the basal end of the membrane alone sufficiently for the required impingement against the hairs of the hair cells. And, of course, there must be vibration frequencies of similar


character but so high as to be damped out by the other structures of the auditory apparatus to an extent rendering them unable to produce effective stimulation of any of the hair cells at all. It is stated that the auditory apparatuses of certain individuals are capable of appreciating higher pitches (vibration frequencies) than others and that certain species of animals have an auditor^' range including high pitches not heard at all by other species. It is suggested by the model that the lower vibration frequencies may affect all auditory apparatuses. The range of the model extended no higher than the note / above 'middle C. Considering its relatively enormous size and coarseness of constuction, it is interesting that it had a range so high, even though the end of the horn was inserted directly into the external meatus.

The quality of a note is supposed to depend upon the form of the wave-motion producing it. The peculiar form of a vibration must be represented in its transmission to the endolymph as well as its amplitude and frequency, and the consequent fonn of the undulations of the tectorial membrane while striking the hairs of the hair cells must give rise to interpretations of quality by the central nervous system. Sensations of pitch must be determined by the number of stimuU appHed in a unit of time upon a unit area of sensory surface, that is, by frequency of vibration. Also, among the higher frequencies, it may depend somewhat upon the extent of the sensory surface involved by a given sound : the greater the extent, the lower the note. Intensity of sensation must depend upon the force of the impingement of the hairs of the hair cells (amplitude) and also upon the extent of the sensory surface involved. Of two sounds of the same vibration frequency , that of the greater intensity involves the greater extent of the tectorial membrane. It is suggested that sensations of intensity and quality of sound are produced in ways entirely comparable with those recognized for the other organs of special sense, namely, in all the interpretation of the sensations is correlated with the number, quality and intensity of the stimuli and the area of sensory surface involved.

(8) As to the question of resonance, very little could be determined with the model. The suggestion that, within the


higher part of the range of the organ, notes of varying vibration frequencies involve correspondingly varying extents of the tectorial membrane and that sensations of pitch are determined by the number of stmuli applied in a unit of time to a unit area of hair cells might be so elaborated as to satisfy some of the simpler requirements of the idea that analysis of sound is accomplished in the cochlea or peripheral part of the auditory apparatus.

It is known that the natural vibration periods of strips of material vary according to their proportions: that a thin, narrow strip has a higher natural vibration frequency than a thick, broad strip of the same length, and that a shorter strip has a higher natural frequency than a longer one of the same thickness and width. A load added to a strip lowers its natural vibration frequency. Of the number of musical instruments constructed on these principles, the xylophone may be preferably mentioned here. However, there is no available information as to what would be the behavior of a series of such varying strips joined continuous with each other end to end in their natural sequence, and especially if of a material soft and flexible as that of the tectorial membrane. The tectorial membrane is not only soft and very flexible but its vibrating part or outspanning zone is attached all along one side, and thus its vibratory behavior must necessarily be different from that of a metal or hard-wood strip shaped in the same proportions and lying free as to both its sides. Very possibly, being flexible and attached, the natural vibration frequency of each segment comprising the whole would be modified by that of the segments adjacent to it, that is, the natural frequencies of adjacent regions would overlap into each other as it were, and yet a given region of the whole membrane would have a natural frequency differing from that of another region. If such is possible, then a region would, in its center at least, have a tendency to vibrate in resonance with imparted vibration frequencies corresponding to its own natural frequency. Thus it would be possible that, when the apparatus is subjected to a given sound, a given region of the membrane may vibrate with sufficient excursion to impinge upon the hairs of the hair


cells, while regions both above and below it may not so vibrate. In other words, the regions immediately adjacent to that vibrating in resonance, having nearly the same natural periods, might be agitated but to a degree less effective and decreasing as the distance from the region most affected increases. Or, again, since loading a vibrating body lowers its natural period or pitch, if the load be uniformly distributed, the vibration frequency of all its components will be lowered, but if the load be placed at one end of the vibrating body then one of the complications resulting would be a lowering of the natural vibration frequency of that end more than the other. If the tectorial membrane may be considered as a vibrating body with a load gradually increasing till the load carried by its apical end becomes 40 times that of its basal end, then it may at least be assumed that the natural period of its different regions must be lowered as the apical end is approached and sound waves imparted to the endolymph may act upon it accordingly. Several waves, of course, may be transmitted in the same du'ection simultaneously through a medium.

Results with the model, however, show that sounds of lower vibration frequency throw the entire tectorial membrane into effective vibration.

It may be mentioned that in line with the above a possible suggestion of resonance with the higher sounds was offered by the model. In three cases a note seemed to produce stronger and more evenly continuous vibration of one region of the artificial membrane than in any other of the six regions. For example, the note a (220 vibrations per second), while causing vibrations in all regions except region 6, seemed to cause a more definite and more evenly continuous buzz of the signal-marker and telephone at region 4 than in other regions. In region 2, the indication in both signal-marker and telephone was more or less intermittent in contrast, regardless of the amphtude given the note. The note d, above a, gave the same in region 1, and, less to be expected, the note g below a gave the same in region 4, the same region as that in which the note a gave it. The fact that a and g gave the evidence in the same region leads to the


conclusion that all these suggestions might have been due to lack of uniformity in construction of the model or faults in arrangement of parts. Certainly the model showed no evidence of resonance in the definite way advanced by the theory of Helmholtz as applied to an assumed structure of the basilar membrane. Altogether, with the suggestions so far obtained, the question may be revived as to whether analysis of sound is not accomplished in the interpretations of the stimuli by the central nervous organ. It may be a question whether any actual analysis is accomplished by the cochlea at all other than to such extent as can be possible in the indications that, dependent upon the vibration frequencies and amplitudes of the sound waves applied, varying lengths of the tectorial membrane are thrown into efficient vibration, the lengths beginning at the basal end. Instead of its different regions manifesting selective resonance, the membrane may lie limply passive, agitated by given wave motions only in so far as the resistance offered by itself and the perilymph and endolymph will permit, and, just as in the sense of touch for example, differences in qualities, intensities and frequencies of the stimuli are interpreted by the central nervous system as qualities, intensities and pitch of sounds. A number of different stimuli applied to the unit area of the skin give rise, within the peculiar possibilities of skin innervation, to an interpretation different from one stimulus applied to that area, and stimuli involving varying extents of skin, above certain limits, are interpreted differently. The same may be urged in general for the other sense organs, within the variations of their special differentiation. Gray ('00) in applying a modification of the resonance theory to the basilar membrane, notes, among others, that a mixture of tones of closely approximate vibration frequencies cannot be analysed by the auditory apparatus. If of varying amplitude or intensity, that of the maximum intensity in the mixture alone is perceived. Such a mixture of notes must all involve so nearly the same extent of the vibratory mechanism and hair cells and with so nearly the same frequency of impingement upon the unit area that the different components are not separately interpreted, just as when two or more points



applied to the skin close enough together and simultaneously are interpreted as a single stimulus (the sum of the several stimuli) , but if one is given greater intensity, that one is appreciated by the brain, while the other images produced by the mixture may be neglected. Impulses mediated by given sense organs are distributed to given areas of the cerebral cortex and differences in the development of certain of these cortical areas are claimed in cases of absence or extraordinary development of the function of the corresponding sense organs. Auerbach ('11) found in his study of the brain of Bernhard Grossman, a famous musician whose specialty was the violincello, that, compared with the average type, the surface of the superior temporal gyrus of both sides was considerably greater than normal, with the middle temporal giyus taking part in the enlargement to a slight extent. With proficient musicians, the chief of the differentiations in the function of the auditory apparatus are the power of analysis and tone memory-. The supra-marginal gyrus of Crossmann's brain was described as being also extraordinarily developed. Helmholtz himself concluded that the power of analysis depends largel}^ upon 'attention.'

Whether a note produced vibrations in the artificial tympanic membrane could be easily determined by holding the finger in very delicate contact with the membrane while the note was being sounded. With one of the lower notes used this membrane seemed to manifest a sort of resonance, vibrating with greater excursion than with other notes sounded with approximately the same amplitude. The membrane vibrated apparently as a whole. When the 'ossicle' was removed, its vibrations were even audible. The vibrating diaphragms of the telephone must exercise some form of resonance for it is common experience that certain voices and certain tones are transmitted much more loudly and definitely than others.

(9) With the notes by which the model was capable of being affected, two further phenomena may be mentioned. (1) Occasionally a note would produce a constant buzz in both signalmarker and telephone in one region of the artificial tectorial membrane while in another region the buzzing would manifest


arising and falling in intensity and, more rarely, the result was actually intermittent while the sounding of the note by the horn was continuous. (2) With some of the low notes employed, while a low intensity would result in a continuous buzzing at all regions, if the intensity was greatly increased, certain regions of the membrane seemed to 'buckle' or be lifted away from contact of the platinum wire with the copper plate. Or, in some other cases the current remained made in a region, indicating that the membrane in the region had been forced in closer contact than the adjustment made for a make and break of the current. Ewald and Jaderholm ('06), experimenting with an apparatus producing sounds of known vibration frequency, formulated the conclusion that "all intermittent noises give intermittent tones." This conclusion I think is necessarily correct, but also, could the construction of the model be trusted, the above would suggest that there may be evenly continuous forms of vibration (noises) which can give sensations of intermittent tones. Obviously also the manifestations of rise and fall of intensity or throbs in the buzzing of the telephone suggest the phenomena described as beats and perhaps overtones as well. As is known, the vibrations of the tongue of the organ reed do not give pure tones.

Tinnitus aurium could be produced by a region of the tectorial membrane being so forced down upon the hairs of the hair cells as to remain adherent to them temporarily, or for a long period as sometimes happens, after very violent auditory stimulation. In origin, the tectorial membrane is directly analogous to the otolithic membranes. These latter are nonnally loaded with calcarious deposits. Under pathological conditions and with advancing age the tectorial membrane may become so loaded, as are other tissues of the body. A light load or a load in a part of the membrane, increasing its inerita, might give rise to 'tone islands' among the higher pitches; a heavier load equally distributed could contribute to the 'failing ears' of old age.

It is not intended here to claim that the tectorial membrane is the only possibly vibratory structure in the cochlea, but, that it is the most adapted in every way and must be the chief vibra


tory structure contained. In my paper of 1915 attention was called to the fact that the spiral organ (of Corti) or the load carried by the basilar membrane increases appreciably in both width and thickness in passing from the base to the apex of the cochlea. This increase in proportions is, however, by no means so great as is the case with the tectorial membrane. The possibility that wave energy imparted to the fluid in the cochlea may also result in some vibration of the membranous spiral lamina, or basilar membrane, as a whole is not denied. While it is a flat tendon stretched with both edges attached instead of having a free edge and lying limph^ passive, the fact that it is a little wider at its apical end (averaging 1.4 times) and that its chief load, the spiral organ, increases nearly two times in size in passing to the apical end seem suggestive of an adaptation for vibratory activity. But it must offer greater resistance to undulatory movement than the tectorial membrane, being far less flexible and of different proportions as well as of different structure and position. It is possible that vibrations of low frequency and high amplitude may throw both the tectorial and the basilar membrane into vibration, while the lesser amplitudes may affect the tectorial membrane only. If the two were equally affected by given vibrations, then their undulations w^ould be parallel and, the two remaining the same distance apart, the required stimulation of the hair cells would not occur But, if the position and greater flexibility of the tectorial membrane allow in it vibrations of greater excursion than in the basilar membrane, then the hair cells may be stimulated and with the frequency of the wave motion imparted. It might be advanced that in case of very violent stimulation, the undulation of both membranes is a necessary and economical arrangement, for in such cases injuriously forcible impingement of the tectorial membrane, which alone stimulates the hair cells, may be avoided while yet the tectorial membrane is kept within working distance of the hairs. The increasing load (spiral organ) upon the basilar membrane must act as suggested for that of the tectorial membrane, making it possible, by its increasing resistance with the increasing distance from the basis of the stapes, that varying extents of it may be affected according to the intensity and vibration


frequency of the wave energy transferred to the surrounding fluid. In the waves of less amplitude and higher frequencies, vibrations must be reached which are unable to agitate the lamina at all but may agitate the tectorial membrane alone, and then those which agitate decreasing extents of this and with decreasing excursion or amplitudes of vibration up to the functional upper limit of the auditory apparatus.


Discussion of most of the usually described physiological phenomena of hearing is not undertaken in this paper. It is felt that the model, as compared with the actual auditory apparaatus, is so crudely constructed and so gross in size that attempts, based upon its behavior, to explain many of the seemingly complex phenomena would be not only difficult but unprofitable at the present stage of the study. The discussion is confined to the more evident and more probably trustworthy suggestions offered by the behavior of the model :

(1) That the tectorial membrane is a vibratory structure and from its consistency, shape, proportions and position, is the chief vibratory structure in the cochlea.

(2) That it may vibrate in such a way that a modification of the telephone theory of hearing may be applied to it.

(3) That sound waves of low vibration frequency produce vibrations in the entire tectorial membrane regardless of amplitude while those of higher frequency produce vibrations in varying extents of it, depending upon their amplitude.

(4) That among the sound waves of the higher vibration frequencies capable of being appreciated, the extent or length of the tectorial membrane in which efficient vibrations may be induced depends more upon the vibration frequency or pitch than upon the amplitude of the sound waves. The higher frequencies being damped out more readily than the lower, varying extents of the membrane are affected according to the vibration frequencies of the waves applied, the highest frequencies affecting the thin, basal end of the membrane alone.

(5) Some evidences of what may be considered resonance in the tectorial membrane were indicated but were thought question


ably trustworthy as such. No resonance of the form demanded by the Helmholtz theory is possible. It is suggested that no resonance or analysis of sound is accomplished by the cochlea in itself other than such as may be possible in the evidences that, beginning with its basal end, different extents of the tectorial membrane may be thrown into vibration by different sound waves, depending upon the frequency and amplitude of vibration. (6) It is admitted that waves of low frequency and high amplitude may induce vibrations also in the less flexible and otherwise less adapted basilar membrane as a whole. Attention is called to the fact that its chief load, the spiral organ, increases in size in passing from the base to the apex of the cochlea and that this suggests a vibrator^" behavior possible in it similar to but not so comprehensive as that claimed for the tectorial membrane.


AuERBACH, S. 1911 Zur Localisation des Musikalischen Talentes im Gehirn

unci am Schadel. Dritter Beitrag. Archiv fiir Anat. und Entwick.,

H. 1, 2, p. 1. EwALD, J. R. 1899 Zur Physiologie des Labyrinths. VI. Mittheilung. Eine

neue Hortheorie. Archiv fiir die gcsammte Physiologie, Bd. 76, p. 147. EwALD, J. R., und Jaderholm, G. A. 1906 Auch alle Gerausche geben, wenn

sie intermittert werden, Intermittenztone. Archiv fiir die gesammte

Physiologie, Bd. 115, p. 555. VON Ebner, V. 1902 KoUiker's "Handbuch der Gewebelehre des Menschen,"

Bd. 3, p. 917. Gray, A. A. 1900 On a modification of the Helmholtz theory of hearing. Jour.

Anat. and Physiol., vol. 34, p. 325. Hardesty, I. 1908 On the nature of the tectorial membrane and its probable

role in the anatomy of hearing. Am. Jour. Anat., vol. 8, p. 109.

1915 On the proportions, development and attachment of the tectorial membrane. Am. Jour. Anat., vol. 18, p. 1. KisHi, K. 1907 Corti'che Membran und Tonempfindungstheorie. Archiv

fur die gesammte Physiologie, Bd. 116, p. 112. Meyer, Max 1898 Ueber die Function des Gehororgans. Verh. der Physik.

Ges. zu Berlin, 17 Jahrg., no. 5, p. 49. NuEL, P. 1872 Beitrag zur Kenntniss der Siiugethierschnecke. Archiv fiir

mik. Anat., Bd. 8, p. 200. Shambaugh, G. E. 1907 A restudy of the minute anatomy of structures in the

cochlea with conclusions bearing on the solution of the problem of tone

perception. Am. Jour. Anat., vol. 7, p. 245. SiEBERMANN, F. 1898 Das innere Ohr. Bardeleben's Handbuch der Anat. des

Menschen. Bd. 5, Abth. 2, p. 313.

Cite this page: Hill, M.A. (2020, February 29) Embryology Paper - A model to illustrate the probable action of the tectorial membrane (1915). Retrieved from

What Links Here?
© Dr Mark Hill 2020, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G