392 Intelligence and Miscellaneous Articles. 



magnetism 26, had 



&=0-81. 



5. Finally, two bars of the same hard steel, 1*03 eentiin. thick, 

 30 and 20 eentim. in length, and weighing 200 and 133 grammes 

 respectively, the corresponding magnetizations being 42 and 28, 

 gave 



& = 0-82. 



If we compare with these twelve bars, which together gave 

 numbers between 0*81 and 0-86, the values previously found by 

 Schneebeli for two bars, 0*84 and 0*86, together with the numbers 

 observed by me for two magnetic needles, and, finally, the values 

 recently given by von Helmholtz*, 0-84 and 0-86; if, further, we 

 consider the comparatively great difficulties and sources of error of 

 such measurements, without which the differences would be still 

 less, we may maintain with some certainty that the value 



& = 0-83, or 5/6 



is very near for ordinary long magnets. In other words, 



The effective poles of a magnet are about -^ of the distance from 

 the ends. 



On this assumption, we need never fear any errors amounting 

 to 5 per cent, of the length ; and hence in many magnetic measure- 

 ments there is no necessity for a separate determination of the 

 distance of the poles. 



Eiecke has observed f that that distribution of the magnetism in 

 a bar which had resulted from the experiments of Van Rees leads 

 to a modulus of the polar distance of a similar magnitude to the 

 above. 



From their importance as magnetic needles, especially in galva- 

 nometers, we have examined a circular magnet and a ring-magnet, 

 and, as observed above, by the simultaneous action of circular cur- 

 rents of the same diameter. Calling cl the external diameter of 

 the disk or of the ring, we found that in the factor of reduction of 

 the tangent-compass which contained the member 3/16 \ 2 /H 2 , A is 

 to be introduced as the " length of the needle." 



For the circular disk (diameter =2*2 eentim., thickness 0*17 

 eentim., mass =4-6 grammes, and specific magnetism =7), 



\ = 0-80d; 



and for the ring (external diameter 2*3 eentim., internal diameter 

 1*9 eentim., thickness = 0*22 eentim., mass = 2 grammes, specific 

 magnetism =14), 



A=0-88d. 



Until a sufficiently large number of round magnets shall have 

 been investigated, these numbers may afford at any rate an 

 approximation. — Wiedemann's Annalen, No. 7, 1884. 



* Von Helmholtz, Sitzungsber. der Berliner Academic, 1883, p. 405. 

 Compare also Topler, ibid. p. 2040, who confirms the value 5/6. 

 t Ibid. p. 319. 



