218 



SCIENCE. 



[N. S. Vol. XV. No. 371. 



German clocks is an interesting subject for 

 investigation. 



Within the last ten years a clock by 

 Riefler, of Munich, having certain novel 

 features, has come into notice. In the 

 Riefler clock the pendulum rod is a tube 

 filled with mercury by which the compen- 

 sation is effected. The pendulum is per- 

 fectly free, except that it receives its im- 

 pulse from the spring by which it is 

 suspended. The Riefler clocks have given 

 good results, and one of them has been 

 adopted as the standard clock of the Pul- 

 cowa Observatory at Odessa in Russia. 



Various devices have been used with suc- 

 cess at Greenwich, Pulcowa and elsewhere 

 for compensating clocks for variations of 

 barometric pressure. A newly discovered 

 alloy of 36 per cent, nickel with 64 per cent, 

 steel, which has a remarkably small coeffi- 

 cient of expansion, makes it possible to com- 

 pensate clocks more perfectly for changes 

 of temperature. 



The astronomical clock is a simple piece 

 of mechanism and the perfection of design, 

 excellence of workmanship and the effici- 

 ency of the various contrivances for com- 

 pensating for variations of temperature 

 and barometric pressure seem to have been 

 developed to a point beyond which no great 

 advance is to be expected along present 

 lines. Even if the effects of change of tem- 

 perature and air pressure on the pendulum 

 could be perfectly eliminated by compensa- 

 tion, we should still have their effects on 

 the clock train as well as the harmful in- 

 fluence of dust and moisture, unless the 

 clock-case affords protection from the 

 latter. 



The most obvious chance for future pro- 

 gress seems to lie in securing the greatest 

 possible uniformity of conditions. "With a 

 clock securely mounted, enclosed in an air- 

 tight case and kept at an invariable tem- 

 perature and barometric pressure, the only 

 conceivable cause for variations in its rate 



would be perhaps the imperfections in the 

 mechanism of the clock itself. It is neces- 

 sary for obvious reasons that the sides of 

 the air-tight case should be rigid. A con- 

 stant pressure cannot be maintained with- 

 out constant temperature, as may be seen 

 from the well-known formula connecting 

 the pressure, volume and temperature of a 

 body of gas, 



pv^=kt, 



in which, for our present purpose, v may be 

 regarded as constant. "We may therefore 



write, 



p=k't. 



In an air-tight case filled with air the 

 change of pressure due to a change of tem- 

 perature of 1° Centigrade is between 2 and 

 3 millimeters for pressures of 650 to 750 

 millimeters. 



The first successful attempt to mount a 

 clock in an air-tight case seems to have 

 been made by Tiede, of Berlin, who in 1865 

 installed for Professor Foerster in the base- 

 ment of the Berlin Observatory an electric 

 clock in an air-tight glass cylinder. This 

 clock, the escapement of which is a very 

 simple piece of mechanism, is described by 

 Professor Foerster in the ' Astronomische 

 Nachrichten, ' Nr. 1636. The impulses given 

 to the pendulum are independent of the 

 strength of the current, since they are pro- 

 duced by the falling of weights which are 

 lifted each second by an electromagnet. 

 The reason for adopting the electric clock 

 was that the winding of a clock run by 

 weights is attended by difficulties when the 

 clock is enclosed in an air-tight ease. "While 

 this clock does not run under ideal condi- 

 tions, being subject to a gradual change of 

 temperature and a consequent slight varia- 

 tion of barometric pressure during the year, 

 it is probably the best time-keeper in the 

 world. It has frequently run for periods of 

 two or three months with such accuracy 



