Showing papers in "Journal of the Acoustical Society of America in 1957"
TL;DR: In this paper, it was shown that the linguistic information conveyed by a vowel sound does not depend on the absolute values of its formant frequencies, but on the relationship between the formant frequency for that vowel and the forman frequencies of other vowels pronounced by that speaker.
Abstract: Most speech sounds may be said to convey three kinds of information: linguistic information which enables the listener to identify the words that are being used; socio‐linguistic information, which enables him to appreciate something about the background of the speaker; and personal information which helps to identify the speaker. An experiment has been carried out which shows that the linguistic information conveyed by a vowel sound does not depend on the absolute values of its formant frequencies, but on the relationship between the formant frequencies for that vowel and the formant frequencies of other vowels pronounced by that speaker. Six versions of the sentence Please say what this word is were synthesized on a Parametric Artificial Talking device. Four test words of the form b‐(vowel)‐t were also synthesized. It is shown that the identification of the test word depends on the formant structure of the introductory sentence. Some psychological implications of this experiment are discussed, and hypotheses are put forward concerning the ways in which all three kinds of information are conveyed by vowels.
654 citations
TL;DR: In this article, the concept of the critical band, or Frequenzgruppe, is applied to loudness summation and it is shown that when the spacing between a group of pure tones is increased, loudness remains constant until a critical point is reached, after which the loudness increases, the same effect occurs when the width of a band of noise of constant SPL is made larger.
Abstract: The concept of the critical band, or Frequenzgruppe, is shown to apply to loudness summation. When the spacing between a group of pure tones is increased, the loudness remains constant until a critical point is reached, after which the loudness increases. The same effect occurs when the width of a band of noise of constant SPL is made larger. The critical band width at which loudness summation begins to depend on the spread of energy is approximately the same as the critical band width determined previously by methods involving thresholds, masking, and phase. The critical band as measured directly by these three methods (plus the method of loudness summation) is about two‐and‐a‐half times as wide as the critical band derived from the assumptions made by Fletcher, but its dependence on frequency is approximately the same. The relation of the critical band to other functions is noted.
623 citations
TL;DR: In this paper, the authors describe, and attempt to interpret, some of the research in that area, and describe how to identify acoustic cues which underlie the perception of speech.
Abstract: Recent experiments with synthetic speech have succeeded in isolating some of the acoustic cues which underlie the perception of speech. This paper describes, and attempts to interpret, some of the research in that area.
327 citations
TL;DR: In this paper, the elastic constants and static strains of a solid subjected to large hydrostatic pressures can be deduced from measurements of resonant frequencies (or transit times) for ultrasonic waves in specimens of suitable crystallographic orientations.
Abstract: The elastic constants and static strains of a solid subjected to large hydrostatic pressures can be deduced from measurements of resonant frequencies (or transit times) for ultrasonic waves in specimens of suitable crystallographic orientations. The pressure changes the specimen's size, shape, and density as well as the elastic constants, and all of the effects influence the resonant frequencies. An algorithm for separating out the effects due to variations in elastic constants from the effects due to static strains is presented and applied to cubic crystals and hexagonal crystals, these structures being of immediate interest to investigators concerned with the properties of metals. The results apply also to isotropic and transversely isotropic solids. The only measurements needed while the specimen is under hydrostatic pressure are resonant frequencies (or transit times). Also required are the size and density at zero pressure, or the elastic constants at zero pressure.
261 citations
TL;DR: In this paper, the authors attempted to correlate interaural intensity and time differences with the subject's localization response at a variety of frequencies using physical measurements of inter-aural time differences and intensity differences.
Abstract: In an earlier study, [Sandel, Teas, Fedderson, and Jeffress, J. Acoust. Soc. Am. 27, 842 (1955)] the writers attempted to correlate interaural intensity and time differences with the subject's localization response. The present paper is an extension of that work. It includes physical measurements of interaural time differences and intensity differences, and attempts to relate these differences to the localization response at a variety of frequencies. The stimuli to be localized were provided through earphones, and the subject was required to match the position (in his head) of a noise and a tone. The noise to one ear was delayed, and the tone presented with no time or phase difference. The subject adjusted the interaural level of the tone until it and the noise appeared to be in the same place. The two were presented alternately by means of a gate having a 150‐msec rise and decay time.Data from the localization judgments were then compared with the findings from the acoustical measurements of time and int...
246 citations
TL;DR: In this paper, the two major cues for stop consonants, the burst of the stop release and the formant transitions in the adjacent vowel, were investigated, and the energy density spectra of the bursts were prepared.
Abstract: The two major cues for stop consonants, the burst of the stop release and the formant transitions in the adjacent vowel, were investigated. Detailed energy density spectra of the bursts were prepared. The transitions were studied by means of sonagrams. Possible criteria for identification were developed and tested. In order to assess the efficacy of the two types of cue, perceptual tests were conducted with isolated segments that contained either stop bursts or vowel transitions alone. Common acoustical properties of bursts and formant transitions are noted; differences as well as similarities are discussed in the light of different varieties of pitch judgments.
243 citations
TL;DR: A model of the larynx has been constructed, and measurements of air resistance for this model have been made, and the applicability and the implications of the results to various problems of normal voice production are discussed.
Abstract: A model of the larynx has been constructed, and measurements of air resistance for this model have been made. The dc resistance R was measured for diameter d of the glottis from 0.1 to 3.2 mm, for subglottic pressures Ps up to 64 cm water and/or volume velocities v up to 2 liters/sec. To a good approximation, the resistance was equal to the sum of a frictional term, proportional to d−3, and a turbulence term, proportional to vd−2. The frictional term dominates at small diameters and/or small volume velocities, with v proportional to Psd3 and the differential resistance Rd=R. The turbulence term dominates at large diameters and/or large volume velocities, with v proportional to Ps12d and Rd= 2R. Over the major region the pressure at the outlet of the glottis was about −12PB, where PB is the kinetic pressure in the glottis, and the pressure over the entry of the glottis was about 1.37PB. The applicability and the implications of the results to various problems of normal voice production are discussed.
240 citations
TL;DR: In this article, the authors developed a method for the inspection of cardiac functions by recording these Doppler signals simultaneously with the electro-cardiograph and phonocardiograph, which made it possible not only to obtain direct informations for the valvular movement which could not have been ascertained up to present, but also to examine the transitional aspects of the myocardial excitation which is utterly uudetectable by the electrocardiography alone.
Abstract: When the continuous ultrasonic wave is sent forth towards the heart from the surface of the chest wall, the cardiac motion causes the Doppler effect upon the partial wave reflected from it. Therefore, an apparatus suitably constructed for sending and receiving ultrasounds is quite useful for the investigation of the movements of theatrium, ventricle, or valves, etc., through the analysis of the particular Doppler signals developed from the motion of the respective part.The author developed a method for the inspection of cardiac functions by recording these Doppler signals simultaneously with the electro‐ and phonocardiographs the oscillograph paper. This method made it possible not only to obtain direct informations for the valvular movement which could not have been ascertained up to present, but also to examine the transitional aspects of the myocardial excitation which is utterly uudetectable by the electrocardiograph alone.
220 citations
TL;DR: In this article, the authors used a standing wave tube to measure sound velocity and attenuation constants in a mixture of air bubbles in fresh water, using a standing-wave tube.
Abstract: Experimental measurements of sound velocity and attenuation constant in a mixture of air bubbles in fresh water, using a standing wave tube, are described. Bubble sizes in the mixtures were controlled between about 0.08‐ and 0.26‐in. diameter, concentrations ranged from 0.03 to 1%, and applied frequencies from 60 to 20 000 cps. Bubbles in each mixture were of a single uniform size except for one series of experiments in which mixtures of bubbles of two discrete sizes were used. Attenuation constants were obtained for each mixture through a range of frequencies, including natural frequencies of the bubbles in the mixture. Velocity measurements were not obtained near the natural frequencies of the bubbles because high attenuation prevented the establishment of standing waves.Data obtained in the tubes were reduced to mean, infinite conditions and compared with available theory. The measurements show that theory gives at least a good estimate of both velocity and attenuation constant in the region investigated.
219 citations
TL;DR: In this article, the reflection of a plane wave of sound at an interface between two perfect fluids is considered, and it is found that, in addition to total reflection in some range of angles of incidence at all finite, relative speeds, there exists the possibility of a reflection coefficient exceeding unity for sufficiently high, supersonic speeds.
Abstract: The reflection of a plane wave of sound at an interface between two perfect fluids is considered. Previous analyses of Rudnick, Keller, and Franken and Ingard are found in error as a result of improper boundary conditions. It is found that, in addition to the possibility of total reflection in some range of angles of incidence at all finite, relative speeds, there exists the possibility of a reflection coefficient exceeding unity for sufficiently high, supersonic speeds; in particular, resonance may occur at one or more angles of incidence. The question of stability of the vortex sheet separating the two fluids also is discussed.
215 citations
TL;DR: In this paper, the reflection and transmission process for plane sound waves originating in air at rest and impinging obliquely on a plane interface with a moving stream is analyzed for a moving reference frame.
Abstract: The reflection and transmission process is analyzed for plane sound waves originating in air at rest and impinging obliquely on a plane interface with a moving stream. Use of a moving reference frame provides transformation to an equivalent aerodynamic problem of flows past a wavy wall—the rippled interface. The angles of incidence, reflection, and refraction are identified with the Mach angles. The angular relations and the amplitude relations (coefficients of reflection and transmission) are evaluated in closed form. In a graph three zones can be distinguished the plane of angle of incidence v. Mach number of the moving medium: ordinary reflection and transmission, total reflection, and amplified reflection and transmission. Included are three loci of infinite reflection: i.e., serf‐excited waves. The energy balance is examined, and the source of amplification is concluded to be the energy of the moving stream. In appendices the results are generalized (1) for the case of two moving media and (2) for di...
TL;DR: In this paper, the authors derived the final expression for the radiation force in terms of the asymptotic scattering function for the obstacle in the field of an incident plane wave by taking into account the interaction of the incident wave with the scattered wave.
Abstract: An expression obtained (1951) by the author for the radiation force on a scattering obstacle with arbitrary normal impedance is now shown to be valid for any scattering obstacle. The derivation of the final expression for the force in terms of the asymptotic scattering function for the obstacle in the field of an incident plane wave is accomplished by taking into account the interaction of the incident wave with the scattered wave. Thus the former assumption of a perfectly collimated beam (1951) is avoided by considering the incident plane wave to be of infinite extent. The result for the force in the direction of the incident wave is F∥ = c0−1{power scattered+power absorbed−∫Γs cosθdA} where c0 is the velocity of sound, Γs is the magnitude of the mean scattered intensity, θ is the angle formed between the incident and scattered waves. This is the same result as that obtained in 1951.The expression given in 1951 for the force perpendicular to the incident wave is correct only for an object which scatters ...
TL;DR: In this paper, a general orthonormality condition valid in all such cases together with the necessary rules for dealing with divergent normalizing coefficients are easily obtained by limiting procedures.
Abstract: In the theories of acoustical and electromagnetic vibrations of enclosures, as well as in field theory and electrodynamics, one quite commonly uses normal modes as generalized coordinates in Hilbert space. Here the method is extended to unlimited or partially limited mechanical media, essentially by first solving the problem for an enclosure and going to a limit while expanding all or some of the boundaries to infinity. This leads to a very useful technique, of somewhat more generality than analogous procedures used in field theory. Thus it is applicable to all non‐dissipative mechanical continua, for any boundary conditions, irrespective of the order or number of differential operators describing the continua and regardless of whether the coordinate systems are separable or not.A general orthonormality condition valid in all such cases together with the necessary rules for dealing with divergent normalizing coefficients are easily obtained by limiting procedures. An equally general formulation for arbitr...
TL;DR: In this article, it was shown that the threshold of intelligibility for a word in a wide-spectrum noise is a decreasing function of the frequency with which the word occurs in general linguistic usage (word frequency).
Abstract: The threshold of intelligibility for a word in a wide‐spectrum noise is shown to be a decreasing function of the frequency with which the word occurs in general linguistic usage (word frequency). The drop in threshold is about 4.5 db per logarithmic unit of word frequency. This rate is independent of the length of the word, although the thresholds for words of given frequency of occurrence are lower for long words.The effect of restricting the listener's alternatives in an intelligibility test to a specified number of words is calculated from this relationship. These calculations come within 1 db of published experimental data. Theoretical functions relating intelligibility threshold to word length are also calculated from the word‐frequency effect, on the assumption that listeners can discriminate the length of a word at levels too low for it to be identified. These functions are in general agreement with the experimental results.Implications for intelligibility testing procedures are discussed.
TL;DR: In this paper, a torsional transducer which has been constructed from n-type germanium is described, and the experimentally obtained voltage-torque characteristic given.
Abstract: The use of piezoresistive materials as strain gauges and in the measurement of displacement, force, and torque is discussed generally. A torsional transducer which has been constructed from n‐type germanium is described, and the experimentally obtained voltage‐torque characteristic given.
TL;DR: The mechanism of binaural fusion is discussed as a form of statistical operation based upon the brain's execution of running cross-correlation of the two ear signals as discussed by the authors, which results in curves which assess listener's judgments as to whether a source of sound lies in the right or left hemisphere of subjective space.
Abstract: The mechanism of binaural fusion is here discussed as a form of statistical operation based upon the brain's execution of running cross‐correlation of the two ear signals. The technique of measurement of the degree of binaural fusion, outlined in our earlier paper, is here extensively applied to sine waves, to chords of sine waves, to intoned vowels and to running speech; such technique results in curves which assess listener's judgments as to whether a source of sound lies in the right or left hemisphere of subjective space. An extended theory is also developed here explaining these judgment curves in detail. Finally, a discussion is made of the importance of having an understanding of the mechanism of binaural fusion, in any consideration of our binaural directivity sense and our speech discrimination sense (the “cocktail, party problem”).
TL;DR: In this paper, an analysis of dispersive properties of elastic plates in vacuo is presented, showing that for a given angle of incidence the propagating energy will be either chiefly shear or chiefly compressional, passage from one to the other corresponding to energy transfer between symmetric and antisymmetric modes or propagation.
Abstract: This is an analysis of dispersive properties of elastic plates in vacuo. For low modes there exists conclusive experimental verification of these properties. Model studies show prominent arrivals having the proper spectra and velocities for group velocity maxima and minima corresponding to several symmetric and anti‐symmetric modes. In addition, detailed calculations based upon exact formulas predict some new and as yet unconfirmed properties of plates, e.g., negative phase velocities. New results concerning high modes of propagation are also displayed. These modes are of considerable theoretical interest since they belong to the transition region between the domains of validity of the wave and ray theories. The structure of the symmetric and antisymmetric fiftieth mode dispersion curves illustrates that for a given angle of incidence the propagating energy will be either chiefly shear or chiefly compressional, passage from one to the other corresponding to energy transfer between symmetrical and antisymmetrical modes or propagation. Comparison with multilayered liquid wave guides are mentioned.
TL;DR: In this paper, an exact solution for the motion of the surface of a uniform elastic half-space due to the application at a depth H below the surface a concentrated vertical force is obtained.
Abstract: An exact solution is obtained for the motion of the surface of a uniform elastic half‐space due to the application at a depth H below the surface of a concentrated vertical force. The time‐variation of the applied force is assumed to be represented by the Heaviside unit function. The solution for the horizontal and vertical components of displacement cast in the form of single integrals over a fixed range, and these have been evaluated on the electronic computer of the Weizmann Institute (WEIZAC). The assumed source emits both S waves and P waves. Beyond a distance r1 from the epicenter, which is equal to H/√ in the case λ = μ, the original S wave is converted on reaching the surface into a diffracted SP wave traveling along the surface. At large ranges, the SP phase is more pronounced than the P phase. The S phase is marked by a finite jump for r r1. The coefficient of the logarithmic term is zero both at r=r1 and at large ranges, having a sharp maximum at r=1.004r...
TL;DR: In this article, the authors determined the surface of detectability of a pure tone partially masked by random noise in the plane of signal amplitude and duration in a two-dimensional space.
Abstract: The object of this study was to determine how signal amplitude and duration effect the detectability of a pure tone partially masked by random noise. If signal duration and amplitude are considered two dimensions in a space, the study attempted to determine the surface of detectability in this space. To accomplish this task three experiments were conducted with the same observers in each experiment. In the first experiment signal duration was held constant while amplitude was varied. In the second experiment signal energy was held constant while various pairs of values of signal duration and amplitude were tested. Finally, signal amplitude was held constant while signal duration was varied. A three parameter equation was determined which provided a reasonable fit to this surface of detectability in the plane of signal amplitude and duration. The equations are consistent with the data of previous research in this area. Finally, a comparison of the results and the predictions generated by a simple filter model is discussed.
TL;DR: In this paper, the propagation of free harmonic waves in the axial direction of a cylindrical shell of infinite extent is studied and discussed in the light of previous work on this problem.
Abstract: A Timoshenko‐type theory of cylindrical shells, developed by the authors in a previous paper [G. Herrmann and I. Mirsky, J. Appl. Mech. 23, 563–568 (1956)] for the case of axial symmetry, is generalized presently to include nonaxially symmetric motions. As an application, the propagation of free harmonic waves in the axial direction of a shell of infinite extent is studied and discussed in the light of previous work on this problem.
TL;DR: It is shown that synthetically produced speech will fuse when the first formant is presented to one ear and the second to the other, but it will not do so if the formants are given different fundamental frequencies.
Abstract: A simple place theory of hearing raises the problem that several mixed harmonics may be attributed by the listener to their appropriate fundamental frequencies: the recognition of a vowel sound in the presence of other sounds requires that the formants of the vowel be detected as such and not classified with the other sounds. Thus, the neural message from a particular part of the basilar membrane probably conveys in some way information on the fundamental frequency, to a harmonic of which that part of the membrane is responding. The problem of fusion of sounds on the two ears is merely an extension of the problem of fusion of different frequencies in one ear.It is shown that synthetically produced speech will fuse when the first formant is presented to one ear and the second to the other, but it will not do so if the formants are given different fundamental frequencies. Even when both formants are given to the same ear, the latter condition fails to fuse. A further experiment with sustained formants shows...
TL;DR: In this paper, the properties of the two dilatational motions are studied and relations are given expressing the variation, in each, of phase velocity, amplitude attenuation, and specific loss with impressed frequency.
Abstract: In an isotropic, thermoelastic solid shear waves are unaltered by thermal effects. However, two distinct dilatational waves exist, both of which are dispersed and attenuated by the medium. One of the waves (denoted the Ewave) is close in character to the pure elastic wave, the other wave (denoted the Twave) is similar in nature to the pure thermal wave. The properties of the two dilatational motions are studied and relations are given expressing the variation, in each, of phase velocity, amplitude attenuation, and specific loss with impressed frequency. For the Ewave the result is verified that this disturbance propagates at the adiabatic velocity at low frequencies and at the isothermal velocity at very high frequencies. An explanation based on physical considerations is offered to account for this generally overlooked phenomenon. It is further found that the amplitude is attenuated exponentially as the square of the frequency at relatively low frequencies, but approaches finite value as the frequency increases without limit. The specific loss reaches maximum for the E wave, a minimum for the Twave, near the frequency whose period is equal to the relaxation time due to thermal currents. Finally, the ratios are computed of the amplitude of the temperature to the amplitude of the displacement in each of the two modes of motion. Numerical work indicates that, for metals at room temperature, the effect of coupling between elastic and thermal motions is very small.
TL;DR: In this paper, the real parts of the shear and compressional moduli of liquid glycerol were investigated using ultrasonic pulse techniques and the average compressional relaxation time was found to be equal to 0.81 times the average shear relaxation time.
Abstract: Relaxation of the real parts of the shear and compressional moduli of liquid glycerol was investigated using ultrasonic pulse techniques. Measurements of the decibel loss/echo for a glycerol‐quartz interface were made at frequencies from 15 to 85 megacycles per second over the temperature range from +70°C to −40°C. Measurements of the velocity of longitudinal waves in glycerol were made at frequencies from 1 to 55 Mc per sec over the temperature range from +50° to −80°C. The limiting values of the real parts of the shear and compressional moduli are studied and compared over the temperature range. The behavior of the relaxational parts of the real parts of the shear and compressional moduli cannot be quantitatively explained by either classical or single relaxation time theory. A distribution of relaxation times is shown to be adequate to explain both the shear and compressional relaxations. The average compressional relaxation time is found to be equal to 0.81 times the average shear relaxation time.
TL;DR: The constant ratio rule as discussed by the authors states that any two entries in a row of a submatrix are equal to the ratio between the corresponding entries in the master matrix, assuming that the only variables which differ systematically in obtaining the two matrices are the different sets of messages and the allowable responses.
Abstract: Three experiments are reported which give support to an empirical rule which may be used for predicting the entries in a closed confusion matrix for any subset of items drawn from a master set of items with a known confusion matrix. This rule, the constant‐ratio rule, states that the ratio between any two entries in a row of a submatrix is equal to the ratio between the corresponding two entries in the master matrix. For this statement of the rule it is assumed that the only variables which differ systematically in obtaining the two matrices are the different sets of messages and the allowable responses. This is an empirical rule which was formulated after examination of three 6×6 master matrices for CV's (consonant‐vowel syllables) and six 3×3 submatrices. Two more experiments using monosyllables and digits were then conducted to test further the rule. Although no direct experimental evidence is reported, the use of the constant‐ratio rule for predicting a master matrix given some of its possible submatrices is discussed.
TL;DR: In this paper, the mutual nonlinear interaction of two plane waves of sound with each other is extended to encompass arbitrary directions of travel of one wave with respect to the other; an exact solution to the first-order scattering process is obtained.
Abstract: Earlier studies [J. Acoust. Soc. Am. 29, 199 (1957)] of the mutual nonlinear interaction of two plane waves of sound with each other are extended to encompass arbitrary directions of travel of one wave with respect to the other; an exact solution to the first‐order scattering process is obtained.
TL;DR: In this paper, a series of experiments with synthetic speech indicated that each of the voiced stops could be made to sound like its voiceless counterpart by eliminating the beginning of the first-formant transition.
Abstract: A series of experiments with synthetic speech indicated that each of the voiced stops could be made to sound like its voiceless counterpart by eliminating the beginning of the first‐formant transition. The amount of first‐formant “cutback” required to produce a voiceless stop varied according to the stop, the vowel which followed it, and the listener, but was in no case greater than 40 msec. It was also found that filling the first part of all formant transitions with noise (i.e., aspiration) contributed to the impression of voicelessness, but the importance of aspiration as a cue appeared to be considerably less than that of the first‐formant cutback. Many listeners showed great consistency in differentially indentifying the patterns as voiced or voiceless even when the stimulus difference was seemingly very small. In one experiment, for example, several listeners sorted the patterns perfectly on the basis of a 10‐msec difference in the starting time of the first formant. [This work was supported in part by the Carnegie Corporation of New York, and in part by the Department of Defense in connection with Contract DA 49‐170‐sc‐2159.]
TL;DR: The Lombard effect results in speech with characteristics different from those of speech that is normally produced as discussed by the authors, which can be demonstrated as an effective way to combat noise interference during reception, and also demonstrates the advisability of control of the production of speech by the speaker himself, in addition to that offered by equipment.
Abstract: The Lombard, or voice reflex, effect results in speech with characteristics different from those of speech that is normally produced. This change of characteristics can be demonstrated as an effective way to combat noise interference during reception. It also demonstrates the advisability of control of the production of speech by the speaker himself, in addition to that offered by equipment, during audiological evaluations.Fifteen naive speakers read words and sentences while noise was being delivered to their headsets. There were five noise conditions. Their speech was recorded, with the noise being kept out of the recording channel, and then limited. Noise was then added to the recording in such a way as to produce a constant speech‐to‐noise ratio. The result was played to 200 American listeners. Results indicate that at a constant speech‐to‐noise ratio of reception speech produced by a talker with masking noise in his cars becomes more intelligible as the masking level rises to a given value. The chang...
TL;DR: In this article, an equation for the radiated sound associated with body forces in a fluid is applied to the flow past a circular cylinder, and the sound field is found to be related to the oscillating lift and drag forces which act on the cylinder.
Abstract: The equation for the radiated sound associated with body forces in a fluid is applied to the flow past a circular cylinder. The sound field is found to be related to the oscillating lift and drag forces which act on the cylinder. Quantitative predictions are made of the directionality and intensity of the field. Some experiments were conducted in a subsonic wind tunnel, in order to test the theory. The microphone was located both inside and outside the tunnel. Over‐all sound pressure levels and sound spectra were measured with various cylinders mounted in the test section. There is qualitative agreement between the theory and the experimental results.
TL;DR: In this article, a mechanical model of the cochlea in which the skin of the arm is used as a sense organ was compared with the organ of Corti, and three phenomena were investigated: (1) directional hearing, (2) heats, and (3) harmonics.
Abstract: A mechanical model of the cochlea in which the skin of the arm is used as a sense organ was compared with the organ of Corti. Three phenomena were investigated: (1) directional hearing, (2) heats, and (3) harmonics. For directional hearing almost no differences were found between hearing and the sensation produced by the model. In the case of beats it was possible to confirm the earlier observations of Helmholtz by improving the frequency resolution of the model ear. Concerning the harmonics it was found that the method of best beats does not permit the measurement of overtones of the ear.