of a Chronometer. 121 



creased from two to five degrees, and the intensity diminished 

 from 101.26 to 95.99. 



On examining the balance, the inner rims of the arcs of com- 

 pensation were found to be of steel, and so likewise were the 

 time-screws, which connected them with the transverse arm. 

 These parts were in a state of active magnetism, particularly the 

 time-screws, one having strong northern polarity, and the other 

 southern. The small wormed cylinders also, on which the ther- 

 mometer pieces moved, presented equal proofs of polarity, one 

 being a north pole, and the other a south. The time-screw and 

 thermometer piece having northern polarity, were on one side 

 of the balance, and those having southern on the other. The 

 balance-spring likewise exhibited vigorous polarity. 



When the north pole of a small bar magnet was placed near 

 the extremity of the wormed cylinder which possessed northern 

 polarity, the balance immediately receded a small quantity ; but 

 when the south pole was applied, the power was sufficient to 

 cause it to advance through a minute but sensible arc ; and simi- 

 lar effects were produced when the proper poles of the magnet 

 were presented to the extremity of the wormed cylinder having 

 southern polarity. On presenting a more powerful magnet, the 

 balance was drawn more than a quadrant from its quiescent po- 

 sition, and motion communicated to the Chronometer. 



By placing the time-screws in the direction of the magnetic 

 meridian, and bringing the north pole of a pocket-compass near 

 that which possessed southern polarity, no deviation was of 

 course perceptible in the compass-needle ; but when the balance 

 was moved through the arcs recorded in the first column of the 

 following Table, the deviation in the direction of the compass 

 amounted to the quantities entered in the second ; the inertia 

 of the needle being too considerable to admit of its inversion. 

 By employing a needle of a more delicate construction, an in- 

 version of its poles took place, the moment the time-screw had 

 VOL. x. P. i. Q 



