Intelligence and Miscellaneous Articles. 391 



observations on bars and on needles by this method and by the 

 method of deflections. The results of these experiments exhibited 

 a suprising agreement among the various magnets, although the 

 number of experiments was not so great as to exclude an accident. 



We have therefore made a larger series of experiments of this 

 kind, the provisional results of which are here given. 



In these experiments we used : — 



(1) The deflection of a short needle by the magnet to be examined 

 in the ordinary way. 



(2) The simultaneous actiou of the magnet on two magnetome- 

 ters, between which the bar was placed, from two different positions. 

 This process is one of great accuracy, since the alterations in tem- 

 perature of the bar are without influence, and since the variatious 

 in the vertical component of the earth's magnetism completely, and 

 the horizontal at any rate partially, disappear. 



(3) The simultaneous action of currents in concentric circles of 

 different diameters on the needle to be investigated. If the dia- 

 meter of the coils is inversely as their number in each case, this 

 method may be made in a form which even exceeds in simplicity 

 and accuracy the preceding ; and is quite independent of external 

 disturbance. 



The factor with which the external length of the magnet must 

 be multiplied in order to obtain the distance of the poles, that is 

 to say the ratio of the ideal or reduced to the actual length, may 

 be briefly called the modulus of the polar distance, and may be 

 denoted in the following by Tc. 



Fourteen magnets were used, some of them in various conditions 

 of hardness. Where nothing else is said, they are cylinders mag- 

 netized in the direction of the axis. 



1. Six recently prepared small round bars of the hardness of glass, 

 50 centim. in length, 40 centim. in thickness, and 5 grammes in 

 weight, some magnetized to saturation, and some intentionally only 

 partially magnetized, so that the specific magnetism was between 6 

 and 24 C. Gr. units for 1 gramme of steel, gave the modulus 



k = 0-82 to 0-86. 



(After tempering in the lead-bath, the numbers increased, in 

 some cases partly to 0*90.) 



2. Two larger, externally equal magnets, which had long been in 

 use, and had been frequently magnetized, 786 grammes in weight, 

 dimensions 44 x 2*3 x 1*0 centim. and with the specific magnetism 

 39, gave the values 



Jc=0-82 and 0-83. 



3. A hard hollow cylinder, 15 centim. long, of 1*6 centim. ex- 

 ternal diameter, weighing 109 grammes, and with the specific mag- 

 netism 32, and which had been treated by Strouhal's method, having 

 been boiled during 30 hours since magnetiziug, gave 



£ = 0-85. 



4. A hard, massive steel cylinder recently hardened and mag- 

 netized, 16 centim. long, 1*48 centim. thick, and of the specific 



