760 Dr. Fleming on the Determination of the 



the end of the inner condenser-tube stout wires are brought 

 to the ends of a vacuum-tube V. 



It will then be seen that if the handle is moved along, the 

 outer condenser-tube slides off the inner one, thus reducing 

 the capacity of the condenser by an amount which is almost 

 exactly proportional to the displacement of the handle. 



Experiment shows that the measured capacities of the 

 tubular condenser with the handle in various positions plot 

 out into a nearly straight line in terms of the displacement 

 of the outer tube. Again, the same movement of the handle 

 reduces proportionately the amount of inductance included 

 in the closed circuit, because the inductance of that portion 

 of the spiral included between the collar at the end b and the 

 semicircular clip K is almost exactly proportional to the 

 length of the spiral included. Therefore the oscillation con- 

 stant of the circuit or the quantity -y/UL for various positions 

 of the handle is proportional to the displacement, and can be 

 marked on the scale. 



In the instrument here described the scale covers a range 

 of oscillation constant from to 12, and it is almost exactly 

 equidivisional. 



The scale values have been determined experimentally by 

 measuring the capacity of the sliding condenser and the 

 inductance of the spiral for various positions of the handle, 

 by methods already described to the Physical Society*. 



If, then, we have any circuit, open or closed, in which 

 oscillations are taking place, we can determine their fre- 

 quency if we place the copper bar forming part of the 

 cymometer circuit near and parallel to the first-named circuit, 

 and move the handle until the vacuum-tube shines most 

 brilliantly. The scale-reading will then show us the oscillation 

 constant of the cymometer circuit in that position. Since 

 the frequency corresponding to that reading is related to the 

 constant as the abscissa? and ordinates of a rectangular hyper- 

 bola, the scale of the cymometer can also be graduated to show 

 frequencies as well. Again, if the circuit tested is an open 

 circuit which is radiating electric waves, there is a definite 

 relation between the length of the radiated waves and their 

 frequency expressed by Y = n\, where V is the velocity of 

 electromagnetic radiation. Hence the scale of the cymo- 

 meter can also be graduated to show wave-lengths directly, 

 and it becomes a direct-reading electric-wave meter for 

 measuring wave-lengths as used in wireless telegraphy. 



* See J. A. Fleming and W. C. Clinton, " On the Measurement of 

 Small Inductances and Capacities,'' Phil. Mag. ser. 6, vol. v. p. 493 

 (1903) ; also Proc. Phys. Soc. Lond. vol. xviii. p. 386 (1903). 



