May 1 8, 1876] 



NATURE 



51 



clocks, including a seconds' relay, a, in the accompanying 

 sketch, are driven by the galvanic current, Isut the 

 standard clock further controls (by seconds' beats pas- 

 sing to London on a special wire from the seconds' 

 relay) other clocks in London, on a principle, introduced 

 nearly twenty years ago by Mr. R. L. Jones, in which 

 the galvanic force is used, not as the driving power, 

 but as an auxiliary, to keep right clocks already going 

 very nearly right, each by its own motive power. The 

 principle has assumed various practical forms, but that 

 proposed by Mr. Jones is generally employed, and is as 

 follows : — The ordinary bob of the pendulum to be con- 

 trolled being removed, a horizontal galvanic coil is sub- 

 stituted. At each swing of the pendulum the coil encircles 

 permanent bar magnets fixed to the clock-case, and the 

 galvanic current received at each second from the con- 

 trolling clock circulates through the wire of the coil. 

 Then (within certain rather wide limits), whether the 

 clock to be controlled tends to lose or gain, the magnetic 

 action produced between the coil and the permanent 

 magnets at the instant of passage of the current so acce- 

 lerates or retards the pendulum that the clock is main- 

 tained in perfect sympathy with the controlling clock. 

 Thus, at Greenwich various mean-time clocks within 



the Observatory, and several in London, depend on the 

 one pendulum of the standard clock at Greenwich. But 

 it is a condition that the clocks shall continue to show 

 exact Greenwich time, and as no pendulum will perform 

 with the necessary accuracy for any long period, it 

 becomes essential to provide convenient means of making 

 periodical correction. The plan used at Greenwich is as 

 follows : — To the pendulum of the Mean Solar Standard 

 is attached a slender bar magnet about five inches long, 

 carried parallel to the rod by an arm projecting forwards 

 from it. Immediately below, in a central and vertical 

 position, and supported by the clock-case, is placed a 

 hollow galvanic coil, the accelerating and retarding coil. 

 The lower end of the magnet passes closely over the 

 upper end of the coil. A galvanic current when passed 

 through the coil imparts to it magnetic properties, rever- 

 sion of the current reversing the direction of its mag- 

 netism. If the current be such as to cause attraction 

 between the adjacent ends of the swinging magnet and 

 fixed coil, the pendulum, carrying with it the whole sys- 

 tem of clocks, will be accelerated ; an opposite current 

 causing repulsion will conversely produce retardation. 

 The only caution to be observed is that correction must 

 not be made too rapidly, otherwise the controlled clocks, 



Fig. I.— Time Signal Apparatus in the Computing Room at the Royal Observatory, Greenwich. 



which are, as it were, merely guided by the controlling 

 current, might, so to speak, break away from control. 

 As at present arranged, to produce an acceleration or 

 retardation of one second, the current must remain in 

 action for about ten minutes. 



Having described the mean-time system of clocks, and 

 the magnetic appliance for correction of accumulated 

 error, we have now to show how at any time the amount 

 of correction required is determined. This makes it 

 necessary to turn our attention to the system of sidereal 

 clocks, and we shall now see how (as was stated at the 

 beginning of this article) the Sidereal Standard is the real 

 timekeeper of the country. This clock, with the system 

 of sidereal clocks in connection therewith, was so fully 

 described in the article already once referred to, that it 

 will only be necessary to repeat here that amongst other 

 things it galvanically registers its seconds on the paper of 

 the revolving cylinder of the chronograph, and drives 

 the sidereal chronometer b, situated on a certain desk 

 in the Computing Room. Without going into further 

 explanation it will be understood that, selecting a proper 

 star of the Nautical Almanac hst, the transit of which 

 over the meridian has been observed with the transit 

 circle and registered on the chronograph the times of its 



passing the several wires are extracted from the chrono- 

 graph record, and the mean taken, which being corrected 

 for the small errors of position of the instrument, and 

 also (as the observations are taken by various observers) 

 for " personal equation," the true clock-time of meridian 

 passage, reduced to one standard, is found. The diffe- 

 rence between this and the Natttical Almanac right 

 ascension of the star for the day gives the error of the 

 sidereal standard, which is also the error of the sidereal 

 chronometer b. Unlike the mean-time clocks (which are 

 required always to show true time), the error of the sidereal 

 clocks is allowed to accumulate, and correction applied as 

 necessary in any calculation in which time by one of the 

 sidereal clocks enters. 



Near to the sidereal chronometer b there is placed, on 

 the same desk in the Computing Room, a mean solar 

 chronometer, c, sympathetic with the Mean Solar Standard. 

 Between these chronometers is fixed a commutator, d, by 

 means of which a galvanic current can be thrown into the 

 accelerating and retarding coil of the Mean Solar Standard. 

 When the commutator index stands in the position shown 

 in the drawing no action takes place ; when turned to the 

 right the current accelerates the clock ; when turned to 

 the left it retards the clock. To ascertain at any time the 



