MEASUREMEIsrT OF NOISE KAYE 181 



pitched and surprisingly penetrating, and can be heard in quiet sur- 

 roundings nearly 1,000 feet away. The masking range of audibility 

 was determined under a variety of conditions. In an airplane cabin 

 or near a pneumatic road breaker or a riveter, the range shrinks to 

 2 or 3 feet, while near an airplane engine the range is only a few 

 inches. On the somewhat doubtful assumption of the inverse square 

 law, the clicker confirmed the fact that the interior of a tulje train 

 (75 to 80 db.) is appreciably louder than that of an express train 

 traveling at about 60 miles per hour — even in the corridor with some 

 of the windows open (70 db.). It is, of course, common knowledge 

 that it is difficult to converse and listen in a tube train, but not dif- 

 cult in an ordinary train with the windows closed, particularly in a 

 first-class carriage, with its more generous upholstery. The cabin of 

 an airplane in a cross-channel flight was found to be at least one thou- 

 sand times (30 db.) noiser than an express train, although the ply- 

 wood cabin walls cut down the noise of the engine one hundredfold 

 (20 db.). The preference exercised by knowledgeable passengers for 

 seats in the rear of the cabin rather than in the region of the side 

 propellers was confirmed, there being some 10 decibels difference, 

 It was found that the customary practice of airplane passengers to 

 plug their ears with cotton wool resulted in a reduction of the noise 

 experienced by about 10 decibels. 



For the longer ranges of audibility the assistance was invoked of 

 a friend or any one else available, the interest of the general public 

 being at times a little embarrassing. 



TUNING-FORK MEASUREMENTS OF NOISE 



A very convenient and portable means of measuring noise has been 

 suggested and used by Davis at the National Physical Laboratory. 

 (Nature, January 11, 1930.) A tuning fork is struck in some conven- 

 ient standard manner — against the heel of the boot will do quite 

 well, and no unusual care is necessary. The fork is then held with the 

 flat of the prong toward the opening of the ear and as close as 

 possible without actually touching. The time of striking the fork is 

 noted, and the interval of time is observed until the loudness falls 

 to the level of the surrounding noise. If desired, the time interval 

 before the note of the fork is masked by the noise can also be meas- 

 ured. The rate of decay of the fork is calibrated in decibels by a 

 buzzer or other type of audiometer. As the decay of the loudness 

 of a fork is practically logarithmic, the calibration curve of decibels 

 against time is roughly linear. Keadings are facilitated in practice 

 if, as the fork is approaching the matching value, it is moved to and 

 from the ear, so that its sound is alternately louder and softer than 

 that of the noise. 



