90 Mr W. N. Shaw and Mr PF. M. Turner, [Feb. 28, 
the ends of the magnet were close together. We have verified by 
means of iron filings the result to which we have been led by 
this reasoning (that the distance between the poles is very large 
compared with the distance between the ends); we found that even 
when the ends of the magnets were turned upon the lathe and 
then pushed together that the distance between the “poles” was 
more than a cm. 
If the coefficient of magnetic induction is small, p is very large, 
and most of the magnetism is on the flat end, in this case the cut- 
ting of the magnet would make little difference, so that this effect 
is probably much greater in proportion for iron than for any other 
metal. 
(2) On some measurements of the frequencies of the notes of 
a whistle of adjustable pitch. By W. N. SHaw, M.A, and F. M. 
TURNER. 
Among the anthropometric measurements proposed by Mr 
Galton is the determination of the limit of audibility of sound. 
For this purpose an adjustable whistle has been made by the 
Cambridge Scientific Instrument Company. A wire ‘73mm. in 
diameter forms a solid piston in the pipe of the whistle, and the 
pitch of the whistle can be altered by pushing the wire further 
in or pulling it further out. The distance between two parallel 
discs, one fixed to the sliding wire and the other to the pipe, is 
measured by means of a graduated wedge, and gives directly the 
length of the pipe of the whistle. 
In order to test the sensibility of a person’s hearing, the pitch 
of the note is gradually raised by pushing in the wire piston until 
a limit is reached beyond which the sound is inaudible, and the 
distance between the discs is then measured and gives the length 
of the pipe of the whistle. The sound is produced by suddenly 
squeezing between the thumb and finger a small indiarubber 
bladder attached to the whistle. 
From the measurement thus obtained we can deduce the 
frequency of the highest note audible by a particular subject if we 
assume (1) the velocity of sound in air at the time of observation, 
(2) that the wave-length of the note in air is four times the length 
of the pipe. In that case if 7 be the length of the whistle, v the 
velocity of sound in air (at the temperature of observation), V the 
frequency of the note, 
v 
N= AL 
This result would, if accurate, be independent of the particular 
whistle employed for the measurement, but there are well known 
reasons for regarding the assumptions mentioned above as only 
