142 SCIENTIFIC NOTES. 



pressure that, after the lapse of ten months, the internal pressure 

 was found to be very slightly less than at first, which Professor 

 Forster thinks arises from the deposition of some water vapour into 

 dew. The temperature is kept as uniform as possible by placing the 

 clock between thick walls below the ground, and having double 

 doors at the entrance. In reply to my question, Professor For- 

 ster said that, in spite of this uniform pressure and temperature, 

 the time-keeping for long periods was no better than under 

 ordinary conditions, but that he thought for measuring small 

 intervals of time there was a great gain, in fact that the clock-rate 

 did not vary so suddenly. Lord Lindsay, who has, at Dunecht, 

 near Aberdeen, perhaps the most complete observatory in Europe, 

 has constructed a special underground chamber for his clock, be- 

 sides the correction for pressure. Any possible variation of tempera- 

 ture is to be prevented by a most ingenious contrivance designed 

 by his astronomer, Mr. Gill. Outside the building is a boiler w it! i a 

 constant fire ; from this a pipe leads to a coil in the clock room 

 and thence back to the boiler, so that a constant circulation would 

 be kept up ; but at one point in the circuit the pipe is bent into a V 

 shape, and at this bend a mercurial thermometer of large size is 

 constructed with such a quantity of mercury that at any required 

 temperature, say 60°, it fills the lower part of the V bend and 

 stops the circulation ; if the temperature falls to 59° the mercury 

 contracts and allows the hot water to circulate again. 



Probably all these precautions will prove that great faults still 

 exist in the mechanism of clocks. With regard to chronometers, 

 Mr. Hartnup, the director of the Liverpool Observatory (estab- 

 lished specially for testing chronometers and making time signals), 

 has made a most important discovery, which, though it does not 

 remove the causes of irregular time-keeping at sea, makes the 

 effect of them a quantity which can be definitely known. I 

 have not time for a full description of what he has done, but will 

 endeavour in as few words as possible to indicate it. It has been 

 customary in testing chronometers to put them in ovens in which 

 the temperature was kept about a certain point, say between 60 

 degrees and 70 degrees, and then change that for another in which 

 a similar variation in temperature was allowed. In this way, 

 though the varying rates between high and low tempera- 

 tures could be ascertained satisfactorily, yet the law which governs 

 these changes was not indicated until Mr. Hartnup began 

 keeping the chronometers at a definite temperature long enough 

 to ascertain their rates at that point, and then changing it 

 to two others in succession, and at each determining the rate. The 

 temperatures chosen after various experiments are 50 degrees, 75 

 degrees, and 80 degrees ; and he finds that with the rates at these 

 points, he is able to state the law of the variation of rate for each 

 degree of temperature for each chronometer so closely that, in a 



