220 



Mr. C. V. Boys on a 



to account for it, as when the field is fully grown the coil will, 

 on being forcibly drawn out, recover its shape much more 

 quickly. On breaking the magnetizing current, the impulse 

 in the opposite direction, being the same as before but lasting 

 for so much shorter a time, is far more evident. 



My first experiments were made with a view to determine 

 whether the impulse was proportional to sin 2u when other 

 things remained the same. I therefore cemented a disk of 

 metal (a half-crown) to an ebonite rod carrying a glass index 

 and hung by a torsion-wire of platinum. The glass index 

 travelled over a card divided into degrees ; and the wire to 

 which the upper end of the torsion-wire was soldered also 

 carried a pointer, the position of which could, if desired, be 

 read on a divided card. The disk was suspended between two 

 parallel polar faces of iron. The upper index was turned until, 

 on making and breaking the current, the lower index showed 

 no sign of motion. It could thus be placed within a small 

 fraction of a degree, so that a was either 90° or 0°. The lower 

 card was then adjusted, and the upper index turned so that 

 the lower rested successively at 5°, 10°,*15°, &c. up to 90°. 

 The negative impulse on making, and the positive impulse 

 on breaking, the current were observed by the throw of 

 the lower index. These are given in the second and third 

 columns of the following Table. In the fourth column is 

 a series of numbers in the proportion of sin 2a, the largest 

 number being made to agree with the observed positive throw. 



Table I. 



Position of 

 rest. 



Negative 

 throw. 



Positive 

 throw. 



Calculated 

 positive 

 throw. 



5 



1 



45 



46 



10 



2 



9 



91 



15 



3 



13 



132 



20 



4-5 



17 



171 



25 



5 



20 



20-3 



30 



6 



22-5 



23 



35 



7 



24-5 



25 



40 



7-8 



26-2 



26 2 



45 



8-8 



26-5 



26-5 



50 



8-5 



26-2 



26-2 



55 



8-2 



24-5 



25 



60 



7-8 



22 



23 



65 



75 



20 



20-3 



70 



6-5 



16 



17*1 



75 



5-5 



12-5 



13-2 



80 



35 



9 



9-1 



85 



2 



4 



4-6 



The close agreement of the other numbers shows clearly 



