\ August 7, 1879] 



NATURE 



347 



B, which is pressing the tooth 5. B has been deposited 

 on b by the pendulum P, which is swinging to the left. 

 Presently the pendulum lifts the vertical arm of A, and 

 iinlocks the wheel. The wheel moves on, but B falls, and 

 the stop ^2 on b catches the tooth which is just above it. 

 The pendulum continues^ its swing, and on returning, 

 deposits A on the next succeeding tooth to a. It then 

 proceeds to b, and continues the performance as before. 



It is desirable that you should know that all your con- 

 trolled or driven clocks are performing properly. In the 

 circuit, alongside the standard clock, a galvanometer is 



m 



nr 



'^ 



a 



FiQ. 6. 



placed, and each of the controlled or driven clocks is 

 fitted with the following mechanism (see Fig. 6). 



A D, D B are two springs through which the current passes. 

 C is a lifter upon the axis of the escape wheel of one of the 

 controlled or driven clocks. Every minute C comes round 

 and breaks the circuit for one second, and anybody stand- 

 ing at the controlling clock, knows when this happens by 

 the needle of the galvanometer remaining stationary. We 

 arrange, for instance, that clocks A, B, c, D shall cut out 

 respectively the 3rd, 9th, 15th, and 40th seconds, and if 

 these seconds are regularly cut out, we know that our 

 controlled clocks are running with the standard. 



iW-jpr — r ^ ' 



VS^^^^i 





Fig. 7. 



It is a matter of importance that the system of making 

 electrical contact shall disturb the pendulum as slightly 

 as possible. The method described above (by the two 

 springs Nj N^) would scarcely suffice, when the most 

 accurate timekeeping was required. The standard sidereal 

 clock at Greenwich has a jewelled pin in the crutch rod 

 which in passing zero presses two weak springs together. 

 A better plan (Mr. Hartnup we are informed had also pre- 

 viously used it) seems one which has just been constructed 

 by Messrs. Dent for some other work. In this it would 

 appear that the resistance of the contact springs cannot 



affect the pendulum at all ; because the springs are lifted 

 during the drop of the escape wheel from one pallet to 

 the other (see Fig. 7). 

 A tooth has just passed from the impulse face of the 



Fig. 8. 



pallet Pj ; t^ now falls through the angle a^ on to the dead 

 face of the pallet Pi. During this interval the tooth / of 

 the smaller wheel mounted on the same axis as the escape 



Fia. 



wheel lifts the contact spring C, S, against the contact 

 spring c, $2, and the current passes. It will be seen that 

 when the tooth t^ goes on to give impulse to the pallet 



