Lifting Power of the Electro-Magnet. 



517 



Table XXVIII. 



The thermometer correction has not been specially investigated, but I use 

 tliat for the 2-inch magnet, -with cylinders of lO'-l, as the greater diameter will 

 compensate the greater length. The two sets of L are in a mean ratio of 1'135, 

 decreasing a little from the lower to the higher. As this is of the same circuit 

 as the hollow magnet of my first Paper, their actions can be compared witli 

 theory very easily, by Tables i. and ii. Since A in equation (4) is as r^, it is 

 obvious that if in two different magnets the radius of the magnet bears the 

 same ratio p to that of the spire, the series expressing the action of the latter 

 on the former must have the same value in each case. Therefore, it is only 



necessary to enter these Tables with ^ = -7, and multiply the result by r' . I 



thus find that the ratio of these two should be 3*139. Companng them, 

 however, by interpolating for the ^ of the large one from the Table of the other, 

 I get the ratios — 



129, 2-66. 



130 2-54. 



131 2-36. 



132, 214. 



133, 1-82. 



This seems to indicate that the central portion of the large magnet is 

 either inactive, or perhaps even in an opposite polar condition to the exterior 

 part.* 



The residual excitation A (that which remains when the current is with- 

 drawn before the keeper is raised) is constant for a -f of 916 and upwards, 



* It is probable that the entire of the 2-inch solid magnets is effective, at least with high exci- 

 VOL. XXIII. 3 Y 



