the Means of augmenting indefinitely the Power of Magnets, 437 



It follows from this that the carrying-force is inversely as the 

 surface of contact — which is correct, provided that all free mag- 

 netism has disappeared, but ceases to be true if S falls below 

 certain proportions. It is on this account that cylindrical con- 

 tacts are generally employed, and not plane ones. Substituting 

 for M its value, the carrying-force P is 



4{ a + ne)*kn 3 

 S 



IX. The weight of the magnet is equal to that of a plate, 

 2aeld, multiplied by the number of plates, which is proportional 

 to their length, and may be expressed by ml; it is therefore 

 7r=2mael 9 d, Consequently the ratio of P to 7T, which measures 

 the carrying-force as a function of the weight of the magnet, 

 will be 



V = 2(a-hneY m 

 it Smaed ' 

 or approximately, neglecting ne against a, 



7T b 



This ratio will be proportional to the length and to the width 

 of the plate, and inversely as the surface of contact. 



X. There are two points which I have not examined in what 

 precedes : they are the question of armatures, and the influence 

 of the thickness of the plates. On the latter point I have ascer- 

 tained the following : — 



The power of a plate increases considerably with its thickness, 

 but less rapidly than the thickness, so that there is a limit 

 beyond which it remains stationary; but a plate of thickness 1 

 has less power than two others of \ thickness, which are much 

 less powerful than three plates the thickness of each of which 

 is \ ; and, generally, the difference increases with the number 

 of the layers of which a pile of a given thickness is composed. I 

 have thus been induced to employ ribands of steel; and as com- 

 merce supplies them abundantly and regularly, of excellent 

 metal, I had only to superpose them in sufficient number to con- 

 struct normal magnets and attain the extreme force, at the same 

 time considerably diminishing the weight. I have thus obtained 

 magnets carrying twenty times their own weight. I shall soon 

 exceed this limit, thanks to the kind cooperation of M. Breguet, 

 and thanks also to my excellent and devoted foreman, Cyprien 

 Bolle. 



