Jan. 6, J 888.] 



SCIENTIFIC NEWS. 



are the arbor and pivots. These could not exert any 

 perceptible influence unless very strongly magnetised, 

 when they might affect the lever or balance-wheel. It is 

 difficult to remove all traces of magnetism from the 



netisation have been devised, but none of them is so 

 certain as the separate demagnetisation of each part. In 

 carrying out this remedy, there are two conditions of 

 neutrality, both of which can rarely be attained. The 



Fig. I. Method of Making Glass Spring. Fig. 2. Flattening Coils of Spring. Fig. 13. Compensated 

 Balance-Wheel. Fig. 4. Regulator Arm. Fig. 5. Testing Magnetismof Lever with 

 Compass. Fig. 6. Shield to Prevent Magnetising of Watch. 



shoulder of a pivot, and it will be found that iron filings 

 will cling here when a delicate compass-needle shows no 

 indication of polarity. 



It is when a watch becomes magnetised as a whole 

 that harm is done. Several different methods of demag- 



one is to make the part under treatment unable to pick 

 up the smallest iron filing, and the other is to leave it so 

 that it will not affect a compass-needle more than a piece 

 of soft iron, which would feebly attract either pole. It 

 is as a rule better to aim at the first of these conditions. 



