444 
Proceedings of Royal Society of Edinburgh. [sess. 
blindfolded ; in front was a table, and across the table was 
stretched a cord exactly in the median plane of the body ; he 
was not allowed to move the head. A second cord was tied 
round the subject’s neck below the ear. The sources of sound 
used were a watch and a tin whistle ; and when the sound was 
produced the subject was asked to point in the direction of the 
sound ; this was marked on the table, and the true direction 
of the sound was also marked. It was interesting to note the 
different degrees of accuracy with which the watch and the 
whistle were localised. Thus, with the watch an angle formed by 
the line of direction of the sound and the median plane of the 
body had a tangent of ff, but was given as angle of tangent |-|- ; 
an angle of tangent was given as tangent ; when held in 
the axial line of the ears it was given as angle of tangent -|- ; 
when held directly in the median plane of the body it was given 
correctly. With the tin whistle, on the other hand, all these 
were given correctly ; but, strangely enough, an angle of tan- 
gent behind the axial line of the ears was given as tangent 
in front. 
When one ear was closed the results of testing were very 
erratic, both with the watch and the whistle ; thus, the whistle 
was sounded in the axial line of the ears, and opposite the open 
ears; it was given as being in the axial line of the ears, but 
opposite the closed ear ! 
From these experiments, therefore, it will he seen that the 
difficulty of estimating the direction of a sound directly in front 
of the head is not always so great as some writers have found. 
It is not uncommonly stated that we are enabled to estimate 
the direction of a sound by the relative intensities with which it 
is heard by the two ears. But it must be remembered that this 
is only true for each particular sound, but it is incorrect to 
suppose that we can localise all sound with the same degree of 
accuracy according to the intensity with which they are heard 
by the two ears. Thus we may estimate the direction of a high 
note very accurately, while at the same distance we are not able 
to localise a sound of low pitch, even though the difference of 
intensity with which the sounds reach the two ears may be the 
same in both cases. Indeed, this may be said to be the crux of 
