apparatus, and being further corrected daily by the adjustment of weights to the pendulum. For the latter purpose the pendulum-rod is provided with a box, c (Fig. 1), for holding small weights; these are made of such shape that they can be easily put into the box or taken out by means of a small pair of pincers without in any way affecting the running of the clock. The box being placed above the center of oscillation of the pendulum, the addition of a weight makes the clock go faster and the removal of a weight retards it. By repeated experiment it has been ascertained what change each weight, under given conditions of atmospheric influences, will produce in a given time; so that the operator knows how to adjust the weights in every case, and the clock can be kept running on mean time with the greatest attainable accuracy.
This clock, beating seconds, closes for, say, one half second during each vibration an electric circuit along the line of which the secondary clocks are situated. This is done by means of the apparatus shown at the top of Fig. 1. To the upper end of the pendulum-rod are attached arms, V and V’, which alternately raise the levers, i and i’ as the pendulum vibrates, thus closing the contact of the electric circuit, one wire of which reaches the arms V and V’, while the other is attached to the levers i and i'. There are three levers at i and i', and three contact points on the arms V and V’, in order that the transmission of the current need not depend upon a single contact which some trivial circumstance—as, for example, the lodging of a grain of dust—might prevent.
The current thus transmitted is carried along wires placed in the city drains to the secondary clocks, which are controlled by the regulator at the observatory, as shown in Fig. 2; but the motive-power of each is a weight operating as in ordinary clocks. To the foot of the pendulum of each secondary clock is attached a piece of soft iron, which swings just above the poles of two electro-magnets in the