136 Mr. H. F. Biggs on Decrease in the Paramagnetism 



the lower end o£ the tube g was fixed the drawn-out end of 

 a glass rod g\ whose weight was enough to keep the whole 

 system nearly vertical when suspended. To g' was fixed 

 an arm a of a narrow glass tube, through which passed the 

 leads of coil A slightly twisted together. The object in 

 having the leads twisted, as also in having the spiral Ph.Bz. 

 at the top instead of at the bottom (as in a galvanometer), is, 

 of course, that the linkages of the lines of the field with the 

 circuit of coil A shall be confined to the coil itself. The end 

 of the arm a was bent into the arc of a circle, fitting coil A, 

 which was glued to it and thereby stiffened. 



The image of the filament of a pocket flash-lamp in the 

 mirror was projected on a scale at a distance of about 60 cm. 

 The paper vane worked in a narrow metal chamber placed 

 on the bench. This arrangement confined the motion to 

 about 3° and provided the required damping. For the 

 highly paramagnetic uncharged palladium in the stronger 

 fields the equilibrium was found to be unstable ; closer stops 

 were therefore made by bending a tin wire so that its ends 

 nearly met on either side of the vane which projected 

 slightly above the top of the metal chamber. 



Pb is a bit of lead wire whose free upper end was bent so 

 as to bring the centre of gravity of the system on to the 

 axis of g' and g. c is a small glass cross-arm on which a 

 little lead rider was placed when the palladium was removed 

 from the coil, thus keeping the centre of gravity in the same 

 vertical. The arm c also served to clamp the system by 

 being raised slightly on a small lever while the palladium 

 was being inserted in, or removed from, the coil. 



It was found necessary to earth the torsion-head as well 

 as the core of the magnet to get rid of electrostatic effects. 



5. Winding and Measurement of the Coils. 



Coil A, the coil on the torsion-arm, and coil B, the 

 exploring-coil, were made as nearly alike as possible, being 

 wound as follows : 20 turns of No. 47 ('04 mm. diam.) 

 silk-covered copper wire were wound on a layer of thin 

 paper slightly waxed to a glass tube of diameter nearly 2 cm. 

 The coil was wound in 3 layers (of 10 turns, 6 turns, and 

 4 turns, respectively), each layer when finished being thinly 

 coated with seccotine. When dry it was slipped off the 

 tube, and the paper cut away from each side of the coil. 



To obtain the total area (mean area of a turn X number 

 of turns) of the coil, two points were marked on the parts of 

 the wire not wound on, and the distance between these two 

 points was accurately measured, (1) before the coil was 



