516 



The Eev. T. R. Robinson's Experimental Researches on the 



I think it follows from what I have stated that there is little probability of 

 making any progress in the investigation of the law of inductive transmission 

 of magnetism, till that of tlie composition of magnetic forces has been de- 

 veloped. In the hope of giving some additional data which may aid in this, I 

 add results obtained with some magnets differing in construction and dimen- 

 sions from that which has been examined in such detail, — three of which have 

 already been noticed in Table xx. 



The first of these is that which I use for ordinary experiments; its cylin- 

 ders are 3' diameter and 12' long, their centres 6' apart; their base 3' by 2'-5. 

 The helices are lapped with copper wire No. 12, and contain 1070 spires.* 

 The keeper which I first used was 2'-5 high by 2' broad ; therefore, only 0'71 

 of the section of the cylinder. With it I obtain — 



Table XXVII. 



I suspected that the keeper was too small, from the intense display of free 

 magnetism ; and, therefore, two pieces of iron, half an inch thick, and 2'-5 high, 

 were carefully fitted to the sides of the keeper and face of the magnet, and 

 screwed firmly to the former while the magnet was powerfully excited ; they 

 now make the keeper 1-06 of the area of the cylinder. This addition gives a 

 considerable increase of force, as might be anticipated from the principles ex- 

 plained in (fl) and (c). I did not, however, expect to find so great a difference ; 

 as it is stated that a magnet lifts more powerfully when the keeper is slightly 

 rounded, so as to touch in a small surface. 



* It is a singular fact that though the -wire -was purchased at the same time and place, the 

 resistances of the two equal lengths ■\vere found to be 58-187 and 44-836 of my units, without 

 any sensible difference of quality. This anomaly has since been fully investigated by Dr. W. 

 Thomson. 



