220 HENRY A. KOWLAKD 



It is a question what the form of the magnet and the position of 

 these coils should be in order to get the greatest field with the least 

 xpenditure of energy. I have one or two propositions to make on this 

 subject which I think are of some interest. 



The first proposition I have to make is that a round magnet is better 

 than one of elongated cross-section. If the coils are long, and they 

 are usually long enough for the purpose, although the theory assumes 

 an infinite length, the magnetic force at any time acting on a round 

 iron core is exactly the same as on an elongated core. But the area 

 of a circular section is much greater than that of an elongated section 

 of the same circumference, and therefore the same amount of wire 

 which would be used to go around the elongated magnet, would, if 

 extended on a circular section of the same circumference, surround 

 much more iron. 



The principal object of making an elongated magnet is that it may 

 include the whole length of the armature. Most makers who adopt 

 this form think it better to elongate the cross-section than to have a 

 long pole piece. But we have seen that the round form is more efficient 

 in general than the elongated form, and the only question is whether it 

 will be more efficient in this particular case. I shall proceed in this 

 theory upon the known fact that we can consider lines of force as if 

 they were conducted by the iron and the air outside. The conductivity 

 of the iron for the lines of force is very great, much greater than that 

 of air. I experimented on it many years ago, and my idea is that it 

 Varies (according to the degree of magnetization) from several hundred 

 up to 5,000 times that of air. The conductivity for iron is very great, 

 especially for wrought iron; for cast iron it is probably less. Therefore 

 the lines of force will be conducted down through the iron from any 

 point over a circular cross-section very nearly as easily as they are from 

 an elongated cross-section, and the saving in the wire will be con- 

 siderable. 



I have another proposition to make with respect to the magnet, and 

 that is that one circuit of the lines of force is better than a number. 

 There is a loss from having a number of electro-magnets, even if they 

 are round. For this reason, that the same magnetic force is acting in 

 each of these coils provided there is the same number of wires per unit 

 of length; and the same wire will go more times around the same iron 

 concentrated in one magnet than when subdivided into several, and 

 will, therefore, act upon it with more magnetizing force. 



That proposition not only applies to this form of magnet (Fig. 1), 



