960 



SCIENCE. 



[N. S. Vol. XIX. No. 495. 



auditory capacity. ' An individual who hears 

 with difficulty ordinary conversation may, 

 nevertheless, pass a fair examination with the 

 watch tick or the tuning- fork. 



There is no douht that human speech, could 

 it be definitely controlled, would furnish the 

 most adequate and the most comprehensive 

 means of determining auditory acuity. But 

 there has always been the difficulty of standard- 

 izing so complex and so variable a thing as the 

 spoken word. This difficulty is serious; for 

 although speech has been somewhat widely 

 employed for this purpose by physicians, otolo- 

 gists and school and army examiners, the want 

 of a common unit of measurement makes it 

 impossible to compare the results obtained. 

 The results have, in consequence, only a local 

 interest. 



The first important step toward standard- 

 ization — the careful compilation of standard 

 test series, composed of a like number of repre- 

 sentative phonetic elements — was taken by Mr. 

 Andrews,* who likewise proposed, in the article 

 cited, an improved method of procedure.f 



The object of this note is to suggest a still 

 further advance in the perfection of the speech 

 test. Instead of employing directly the voice 

 of the investigator, and instead of relying upon 

 acoustic and organic conditions which vary 

 from experimenter to experimenter and from 

 place to place, it proposes to use permanent 

 phonographic records, which can be copied an 

 indefinite number of times and can be repro- 

 duced independently of local conditions. The 

 phonograph is especially available at present, 

 because recent improvements in construction 

 provide hard, durable cylinders which are 

 copies of a single master record. Thus it 

 should be possible to reproduce in any place 

 and under constant conditions the same test 

 series, inscribed upon a single cylinder by a 

 single voice. 



Through the courtesy of the National 

 Phonograph Company the writer has been per- 

 mitted to make preliminary records at the 

 company's works in Orange. These records 



* L. c, pp. 29-36 (final table of test words on 

 page 36). 



t L. c, pp. 53 ff. 



have since been subjected to trial on an Edison 

 phonograph in the Cornell laboratory. 



In reproducing the series of test words it is 

 necessary, of course, to control both pitch and 

 intensity of the sound. Pitch is easily con- 

 trolled by setting the phonograph at the rate 

 used in inscribing the record (e. g., 100 revolu- 

 tions a minute of the cylinder). Intensity is 

 controlled in two ways : (1) constancy of in- 

 tensity is obtained by the use of the new 

 'model C' reproducer (Edison), whose wri- 

 ting point is held in the wax groove by a con- 

 stant pressure automatically provided; (2) 

 reduction of intensity is obtained by a device 

 set into the rubber transmission-tube. The 

 reducing device consists of two telescoping 

 brass cylinders 15 cm. long. The outer cylin- 

 der is 1 cm. in diameter and is perforated with 

 35 holes of 4 mm. diameter, running from end 

 to end in a spiral pattern. The inner cylin- 

 der has closed walls. When the two are pushed 

 in together they form, therefore, a closed sec- 

 tion of the transmission-tube; but as they are 

 drawn graduallj' ajiart more and more of the 

 35 holes are uncovered, allowing a greater and 

 greater escape of the sound. When all the 

 holes have been exposed only a small fraction 

 of the sound reaches the ear ; when all the holes 

 are again covered the tube is completely closed. 

 Intermediate settings of the reducer (made by 

 scale readings on the inner brass cylinder) give 

 a wider range of intensities. To increase the 

 range still further, stops are inserted in the 

 smaller cylinder. The writer used three of 

 these stops ; one entirely closed, ' one with a 

 circular aperture of 0.5 nun., and one with an 

 aperture of 3.0 mm. diameter. The tests thus 

 far made indicate that only two of the stops 

 will be required. To avoid direct transmis- 

 sion of the sound, through the air, it is neces- 

 sary either to place the phonograph in a par- 

 tially sound-proof box or to conduct the trans- 

 mission-tube through a wall or the key hole 

 of a tightly-fitting door to a second adjoining 

 room. It is only necessary that no sound be 

 heard by the subject when the ear tubes are 

 inserted in the ears but disconnected from the 

 instrument. Several individuals can be ex- 

 amined at once by duplicating the ear tubes 

 or by substituting a megaphonic horn for the 



