IRON, STEEL, AND NICKEL TUBES IN THE MAGNETIC FIELD. 



547 



be for a fairly broad bar or tube, whose magnetisation is necessarily very far from being 

 uniform. In the hope of getting further insight into the real character of the strain, I 

 made some measurements with the tubes in fields, which, though symmetrical, were very 

 far from being uniform. 



By suitable combinations of the magnetising coils, seven distinct symmetrical dis- 

 tributions of field were obtained. There was (1) the usual combination of six coils 

 altogether ; (2) the combination of the four middle coils — symbol 040 ; (3) a second 

 combination of four, the two end and the two middle ones — symbol 10201 ; (4) the 

 third combination of four with the two middle coils left out — symbol 2002 ; (5), (6), 

 and (7), the combinations of pairs — symbols 020, 1001, 100001 respectively. Now, a 

 particular current passing through each of these combinations will produce a particular 

 distribution of field along the tube, a definite distribution of magnetisation in the tube, 

 and a corresponding change of volume. These can all be experimentally determined ; 

 or, as is more convenient, the field can be calculated. 



The distribution of field along the tube for one ampere flowing in the coils is 

 indicated by the following table, which gives the values of the field at three regions, 

 distant respectively 0. 10'8 and 2T9 centimetres from the middle of the coil. 





Fields due to 1 Ampere at distances 



Coil 







Combination. 















10-8 



21-9 



6 



172-5 



1711 



160-5 



040 



167-5 



158-6 



87-0 



10201 



150-2 



96-4 



87-3 



2002 



27-3 



87-2 



146-7 



020 



145-2 



83-9 



13-8 



1001 



22-3 



74-7 



73-2 



100001 



5-0 



12-5 



73-5 



From the next table some idea may be gained of the manner in which one of the 

 iron tubes becomes magnetised in these different fields. The quantities in the columns 

 headed B , B l5 B 2 are the inductions as measured ballistically in a secondary coil placed at 

 the distances 0, 10 "8, and 21*9 cm. respectively from the centre of the coil, when the tube 

 is inserted. The neighbouring columns contain the values of the currents in amperes 

 by which the corresponding inductions were produced. Three currents were used for 

 every combination of coils. The combination symbol is entered on the left. 



A similar table was made for the corresponding nickel tube ; but, as it presented 

 much the same features, it has not been thought necessary to give it. 



From a comparison of these tables we see almost at a glance how very different are 

 the laws governing the distribution of field and induction. As already pointed out, 

 the induction in the field produced by the 6 coils falls off more rapidly than the field as 

 the section is taken further from the centre. With combinations 040, 10201, and 020, 

 this leakage of the lines of induction is more marked, as, indeed, we should expect. 



