(Return to index)

Classics Editor's note: The numbers appearing in parentheses throughout the text refer to the items
in the bibliography at the end of the volume.
 

SENSIBILITY to pitch was the only aspect of hearing experimented upon. No attempt was made to find the threshold for hearing, because the laboratory was not provided with a sound-proof room. Three determinations of sensibility to pitch were made, as follows:

A. The upper limit.

B. The lower limit.

C. Discriminative sensibility, with the 512 fork as a standard.

The Galton whistle was the instrument used to investigate the upper limit of sensibility to pitch. The contrast between the shrill sound of the whistle where the pitch is perfectly distinct and the sound of the rush of air in the regions above the possible limit of pitch, was first given to the subject. He was then told listen carefully to each stimulation given him and whether he could distinguish the shrill pitch sound, or whether it was merely the rush of air. The number of vibrations was gradually increased until the subject [p. 70] lost the pitch, and was then decreased until he heard it again. In the regions near the limit of discrimination he was given from four to six stimulations for each turn of a division in the vernier scale, and the point at which he heard the pitch three-fourths of the time was fixed upon as the upper limit.

The diagram of results (Fig. 42) is made out for convenience in terms of the scale-readings of the whistle. It will be easily interpreted if it is borne in mind that the smaller numbers on the scale mean higher vibration rates. The following table gives the number of vibration rates per second for each scale-reading which appears in the table:

1.6=52,500 vibrations per second
1.7=49,411
1.8=46,667
1.9=44,210
2.0=42,000
2.2=38,181
2.4=35,000
3.0=28,000
4.0=21,000

The experiments on the lower limit of pitch were performed with the Appunn wire forks. There were eight forks in the series, ranging from twelve to fifty-six vibrations per second. Each fork is repre-[p. 71]sented in the abscissas of the curves of results (Fig. 43). Since the lowest fork (twelve vibrations per second) was not below the possible limits of pitch, the subjects could not in this case be given the contrast between pitch, and vibrations with no pitch. The experiments were begun with the forks of high vibration rate, and worked down to the limit. The subject kept his eyes closed during the test. Each fork was sounded  close to his ear several times. He was asked to tell each time whether the sound he heard could be called a tone or not. The lower limit of pitch is subjectively much harder to fix than the upper. As the vibration rate decreases, the smooth singing tone changes into a pulsating sound which still has a certain pitch quality. Many subjects found it very difficult to fix upon the point where the sound lost its pitch quality. An attempt was made to control the judgment by requiring discriminations of higher and lower in doubtful cases, but this proved to be impracticable; in the first place, because discriminative sensibility at the lower limit is so coarse; and in the second place, because the difference of vibration rate could be felt as the air struck the ear and an indirect judgment as pitch, based on vibration rate, was unavoidable. .The results therefore contain the source of error due to differences in individual standards.[p. 72]

The result (Fig. 43) seems to indicate a somewhat lower limit of pitch in the men. The limits of the two curves are the same -- from twelve to above the series of forks; the difference in them is in the number of lower limits falling on the twelve and sixteen forks. There are nine men and three women at twelve and vice versa at sixteen. Considering the source of error in the test, as small a difference as this is of doubtful significance.

The tests on pitch discrimination were made with two tuning forks with the pitch Ut 3 (512 single vibrations per second). One of the forks bore a rider by which its rate could be reduced as much as seven vibrations per second. The forks were mounted on wooden resonators. The subject sat with his back to the apparatus, at a distance of about six feet. The forks were sounded by tapping them with a rubber-tipped hammer. The chief source of error in the test was the inequalities of intensity incident to striking the forks by hand. Long practice reduced this to a minimum, and any tests where the differences of intensity were noticeable were discarded.

The usual directions for discriminative tests were given the subject. He was told that two tones would be sounded in succession and that he was to report each time which of the two was higher in pitch. The number of tests in each order was the same. The series began with the large differences -- six or seven vibrations per second -- and worked gradually down to the limit. Most of the subjects improved so rapidly with practice that it was found necessary to keep the [p. 73] time devoted to this test approximately constant. A few subjects came to a standstill before the end of the half-hour usually allotted to it and proved unable to go farther even after repeated trials. In these cases the test was stopped when improvement ceased. In all other instances the fineness of discrimination , reached at the end of the half-hour is what is recorded. It may not in all cases represent the limit of possible discrimination, but is a fair measure of the relative natural capacities.

The results (Fig. 44) are recorded in terms of the difference of vibration rate between the two forks at the limit of discrimination for each subject. The curves indicate finer discrimination in the women than in the men. The difference is shown principally by the greater number of women than men who could discriminate a difference of less than one vibration a second, and the preponderance of men who could not discriminate pitch at all within the limits allowed by these forks. The latter subjects seemed to have no clear idea of what the terms high and low meant with reference to pitch. Their attempts at discrimination were pure guesses, with no discernible regularities.[p. 74]

The only test on pitch comparable with ours is the one made in the series of tests at Columbia University (82). The method consisted in requiring the subject to find again, after the bridge had been shifted, a note sounded on a monochord. The result agrees with that of the present test in showing that the women have a finer pitch discrimination than the men.

The only other comparative tests on hearing are those by Lombroso, (51, chap. iii) and Roncoroni (72) on the limits of normal hearing. They both used as a measure the distance at which a watch could be heard. Although the number of persons tested was small, the method rough, and the results contradictory for the two ears, Lombroso comes to the conclusion that men's hearing is keener than women's. Roncoroni agrees with him.

Reik (71) reports an interesting anatomical investigation of the ears of 440 school children. He found abnormalities much more frequent in the ears of boys than in those of girls. He also made investigations on the physiological functions of the ear. Although he gives no report of his results on pitch discrimination, the presumption is that it would be poorer in the sex with the greater number of abnormalities -- a result which would be in accord with ours. The values he obtained for the upper limit of pitch agree very well with ours. Some of the children, however, could distinguish higher vibration rates than any of our adult subjects. He makes no comparison of sex with respect to the upper limit. Tests made under the direction of F. W. Smedley (71a) in the Chicago public schools revealed no great differences as regards defective hearing in boys and girls.[p. 75]

In the upper and lower limits of pitch the only difference of sex indicated was a possible lower limit for men. In pitch discrimination women are better than men. The tests on the threshold for hearing have been too few in number, too rough in method, and too contradictory in result to serve as a basis for any trustworthy generalization.