34 Scientific Proceedings^ Royal Dublin Society. 



the same, the magnetic field round it was clianged, and a series of observations 

 taken tlie same as the first, and so on for sixteen different magnetic fields up 

 to 200 c.g.s. vmits. 



At the beginning and end of each series of observations, with a given 

 longitudinal load on the wire, the rate of subsidence of the oscillations was 

 observed when there was no magnetic field round the wire, the state of 

 no field being obtained by sending an electric current round the solenoid of 

 such a value and in such a direction as to annul the vertical component of the 

 Earth's magnetism. 



The method of obtaining the torsional oscillation without imposing any 

 pendulous motion on the wire was as follows : — The wire with a certain 

 magnetic field round it and a given load on its lower end was put in circuit 

 with a storage cell, rheostat, reversing key and plug-key, the circuit being 

 completed through the wire under test by means of a short piece of iron wire 

 fixed on the under side of the vibrator or load and dipping into a cup of 

 mercury. When everything was quite steady, a small electric current was 

 sent througli the wire and reversed in unison with the torsional oscillations 

 of the vibrator until a sufiicient amplitude had been attained ; the mercury 

 cup was then taken away, thus leaving the wire free and performing only 

 torsional vibrations. 



An extensive series of observations on torsional subsidence were made 

 under ,/b!«r different values of the longitudinal load ; and the results of a few 

 are sliown as curves in fig. 2. 



The results here shown were obtained wlien the load on the wire was at 

 tlie rate of TS x lO'^ grammes per square centimetre ; the numbers at the right- 

 hand end of each curve indicate the strength of the longitudinal magnetic 

 field wliioh was round the wire when it was being tested. 



Loads on the wire higher than 4 x 10^ grammes per sq. cm. were tried ; 

 but the damping of the oscillations was so rapid at the beginning of the 

 observations, and the amplitude became so small at the end of 70 vibrations, 

 that considerable error was introduced, and the observations were therefore 

 not recorded. 



The curves in fig. 2 show that the application of an external magnetic 

 field increases the internal viscosity of the wire up to a field of about 20 c.g.s. 

 units; then the viscosity begins to ffecrOTse when" higher fields are applied. 

 Witli a magnetic field of 80 units round tlie wire the amplitude, after 70 

 vibrations have taken place, is nearly the same as when there was no field 

 round the wire ; but the rates of subsidence in the two cases are diiferent, as 

 is shown by the slopes of tlie two curves. 



The rates of subsidence were taken for six ditfereut maguetio fields 



