166 DRS. GUY BARLOW AND H. B. KEENE ON THE ANALYSIS OF SOUND. 
is approximately -fT, where T is the period when undamped. 
A phase change of + 
71 
8 
may be allowed at either end of the interval, which we may take as extending from 
t = — |T to t — -j-jT. Then since n — is the phase change per sec. at t, the required 
condition is expressed by 
therefore 
Hence the limiting rate is ~j sec. Thus in the experiments, with T = 3 sec., unit 
range 1 /sec. can be covered in 4^ sec. But since the phase at synchronism may be 
unfavourable, the rate must be much slower than the above limit—probably about 
X T S the rate. 
The investigation shows that the limiting rate is the same at all frequencies, and it 
is inversely as the square of the galvanometer period. The result is of importance in 
considering the possibilities of making a photographic record of a spectrum by a con¬ 
tinuous sweep through the whole range of frequency. 
BRITISH l 
MUSEUM 
25 MAR 22 
