134 CERTAIN INDICATING AND RECORDING APPARATUSi 
The above instrument, though working sufficiently well tp 
prove the correctness of the principles upon which it is made, 
had the serious defects which disk and roller mechanisms suffer 
from, and which render it difficult to give the disk sufficient 
speed, so as to obtain a reasonable range of variation of 
velocity. This led to an attempt to design an instrument which 
should measure a greater range. The following investigation 
resulted from this attempt, and though not yet put into a 
practical form, may possibly be so before long, and was therefore 
thought worthy of publication. 
In the equation 
when y = 0, 
y = R, 
w 
aj = 0, 
w = iv, 
which shews the range of velocity measured by the first form of 
disk and roller to be from 0 to iL. 
Suppose, now, the screw is driven by the clock, and the disk 
by the moving body, 
then 
II 
or 
io=-Ed,. 
y 
when 
o' 
11 
11 
tv — tv. 
The range is in this case from tv to ©o, but this is not a really 
convenient range. 
Suppose, now, instead of keeping the driving roller, C, at 
one position on the disk, it is connected with B, and moves with 
it as far as longitudinal motion is concerned, as shewn in the 
arrangement in Fig, 2. Then precisely the same action takes 
place as in the arrangement of Fig, 1 , except that the screw 
