102 Scientific Proceed/ H(fs, Roj^ai Dublin Society. 



load has been shown by Honda and Terada in the paper ah-eady referred to 



above ; also the reason for repeating the measurements for direct longitudinal 



fields was that for the particular wire we used we should have a standard 



of comparison for the results obtained with the alternating longitudinal fields 



and the transverse fields. 



Section 2. 



Transverse Magnetic Fields. 



In this part of the investigation the arraiagement of the apparatus was 

 the same as in the previous section, with the exception that the solenoid was 

 replaced by a slotted iron tube. The transverse magnetic fields were produced 

 in the air-gap of this soft iron tube by means of electric currents goino- 

 through twenty -five copper wires running inside the tube. These copper wires 

 of No. 8 s.w.g. were joined in series, and were well insulated from each other 

 and from the iron tube. 



The tube had a .slot cut through the wall along its whole length of 

 51 cms. ; it was 4-8 cm. in external diameter, 4 cms. internal diameter, 

 breadth of face of slot 4 cm., and width of slot 0-45 cm. 



The magnetic field strength in the air-gap was measured in the usual way 

 by means of an exploring coil and earth inductor, and was found to be 29*96, or^ 

 say, 30 e.g.s., units per ampere, that is, for one ampere in each wire there were 

 25 amperes flowing inside the iron tube to give a field of 30 units. 



The loads on the wire used in this part of the work were fi'om 2 x 10^ to 

 4x10° grammes per sq. cm., so as to keep the wire stretched uniformly in the 

 slot. 



When the transverse magnetic fields were applied to the nickel wire it 

 had a tendency to go towards one or other of the faces of the slot. In order to 

 keep the wire in the centre of the slot and at the same time introduce as little 

 friction as possible, small glass tubes l*4m.m. in diameter were placed 

 vertically on each side of the wire at intervals of about 10 cms. 



The slight friction introduced by these smooth tubes was overcome by 

 gently tapping the iron tube before each reading of the microscope. The 

 highest magnetic field obtainable without introducing errors due to heat was 

 about 960 e.g.s. units, and up to this point the change in the twist of the wire 

 gave no indication of reaching a maximum. 



Before each set of measurements were taken the iron tulie was carefully 

 de-magnetized by putting through the wires inside it an alternating current 

 of value equal to the maximum direct current used in the previous experiment, 

 and gradually reducing it to zero. The tube was then allowed to rest for 

 about half an liour, so that it should attain the temperature of the room 

 before the next observations were made. 



On account of the heating of the nickel wire under test by the alternating 



