136 Sar G. Greenhill, Note on Dr Searle's experiment, ete. 
On the assumption of constant w, and neglecting air drag, ine 
torsional period is the same for visible and invisible oscillation. 
But the circular correction is required for the equivalent | 
gravity pendulum. If its invisible oscillation is in tune with the 
torsion, then swinging through an angle of D° it would lose one 
beat in V = (* s ae =) beats, observable by the Method of Coinci-. 
dences, described in Maxwell’s Matter and Motion, p. 106; thus: 
D =T°.64 makes V = 3600, one second beat in the hour. 
Or if the pendulum synchronizes with the torsional oscillation | 
when it swings through D”, its equivalent length must be stretched 
D 
me (ae x 57:3 
oscillation ; this is 7, °/, for D = 7°.64. 
Air drag and buoyancy can then be measured incidentally : and | 
here is an opportunity of showing the Bessel pendulum at work, 
made of a length, say half a metre, of bicycle tube, filled with lead 
at one end. The theory is given in my Notes on Dynamics, 
183. 
re In a clock pendulum with a cylindrical bob, the air correction 
is about 7, °/, of its length /, and so would be felt in the next 
recorded figure, the fifth, of the experiment. 
To fasten a thin wire firmly in a supporting plate of copper, 
Mr C. V. Boys found the method useful described in La Science - 
amusante, par Tom Tit, where a needle enclosed in a cork is driven 
through copper to punch a hole, and the wire is then clinched in 
the hole and the copper hammered to hold it tight. 
With the copper plate screwed on to the end of a bracket, 
held by Six Point Contact with supports fixed firmly in the wall, 
the only defect of rigidity is in the elastic flinch of the bracket. — 
I should like to put in a plea here for Dr Schuster’s name 
Simple Vibration, instead of Harmonic Vibration, the word har- 
monic being reserved -for the overtones in a Fourier Series. 
i per cent, to give the equivalent length for invisible. 
: 
| 
