86 A Method for Determining the Rate of Tuning-forks. 



Fig. 1. 



pendulum, 

 mercury globule, 

 gravity battery, 

 relay. 



bichromate battery, 

 induction coil. 

 Geissler tube, 

 fork to be rated. 

 " electric" fork. 



VHttk 



□ 



In practice only the even flashes are counted, the odd ones 

 being disregarded. Consequently if a be the number of these 

 per period, n the rate of the fork, and 2N the whole number 



nearest to 2n, then n = 1$ ±y— . To determine which sign is 



to be used, EUt 2 is weighted by a very small piece of wax. 

 If a is greater than before, the sign is + ; if a is less, the 

 sign is — . 



The electric fork may be dispensed with entirely by the 

 following method. The fork to be rated is placed vertical, so 

 that one edge is in the focus of a microscope provided with 

 cross-hairs. Behind the fork the Greissler tube is placed hori- 

 zontal. The appearances observed in the microscope at each 

 flash would then be as follows: — 



rig. 2. 



0123456789 10 



From these we can deduce the rate as before. 



This method would probably give results almost if not quite 

 as accurate as the preceding, and has the advantage of being 

 more direct. The nearest whole number may be found by 

 comparison with one of Konig's standards, or by the following 

 method. 



The fork is, in turn, compared with two pendulums, whose 

 times of vibration are t } and t 2 , t x — 1 2 being small. 



Let n l = number of vibrations of the fork in time t l ; 

 Let n 2 = number of vibrations of the fork in time t 2 ; 



n, and n 2 differing from a whole number by less than a small 

 fraction, e x . 



