Vol. 6, 1920 PHYSICS: THOMPSON, HICKMAN, RIFFOLT 169 
tion, only, are approximately correct for the actual case of location wherein 
both phase and intensity are altered. 
The work thus far clearly shows that with frequencies 100 to 1200 d.v 
phase is the chief factor in localization with pure tones. For the lower 
frequencies the intensity effect is practically nil, and at all frequencies 
within the range considered the effect must be small. These conclusions 
are quite contrary to the view commonly held and appearing in psycholog- 
ical literature and texts. But they are worthy of confidence because based 
on quantitative results, and not on personal judgment. There remains a 
great amount of experimental work covering an extension of the fre- 
quencies and the employment of complex tones. Strictly speaking the 
localization herein mentioned applies only to 90° either side of the median 
plane and in front. The removal of this limitation will probably prove 
relatively simple after the experiments are completed. The above report 
while covering a limited range, nevertheless does make a definite advance 
in our understanding of the relative importance of the factors, phase and 
intensity, in binaural localization, and furnishes a basis upon which the 
future experiments may depend. 
THE MEASUREMENT OF SMALL TIME INTERVALS AND 
SOME APPLICATIONS, PRINCIPALLY BALLISTIC 
By L. T. E. Thompson, C. N. Hickman and N. Riffolt 
BA1.LISTIC Institute, Clark University 
Communicated by A. G. Webster, February 25, 1920 
Precision in time calibration is of importance in many classes of re- 
search. Several types of apparatus have been developed for this purpose. 
Some of these, such as the Helmholtz pendulum, permit extremely ac- 
curate reproduction of a stated interval but le-ss exact absolute determina- 
tion. A method proposed and extensively used by A. G. Webster has 
given satisfactory, absolute determinations of intervals as small as a 
millionth of a second. ^ The recording apparatus is electrical and the cali- 
bration is accomplished by a careful measurement of the height of fall of 
a projectile. 
The Le Boulenge and the Aberdeen chronographs are in general use, 
the former having a falling projectile and the latter a rotating drum. 
With any apparatus involving a rotating drum, it is necessary to maintain 
the speed of the motor driving it within certain limits and to secure a de- 
pendable calibration, usually placed on the record as it is being made. 
Various instruments have been devised to accomplish this purpose.^ 
A tuning fork producing a photographic or mechanically inscribed record 
of a sinuous form is perhaps the commonest example. 
Means for applying discontinuously a record of a fork or other vibrator 
