1G(J THE STRUCTURE OF TIIE NUCLKU8. 



into the box, h, tliroiigli a notch on top parallel to the axis of h, sufficiently closed 

 in adjustment by the ring, a. 



SEPARATION OF THE CHARGE OF THK INSULATED NEEDLE BY INDUCTION. 



4. Method of cJiarying. — The quadi-ants are charged directly with the water 

 batteiy as already ex[)lained. To convey the charge to the needle, which is now 

 freely suspended in air and insulated, it was my first endeavor to ionize the air in 

 the capsule, ece, by making the discs, e, e, of wire gauze, and securing pellets of 

 phosphorus to rear discs of mica. Thus the air in the caj).sule becomes a conductor 

 while the air without insulates iu view of the slow diffusion and absorption of 

 phosphorus ions. 



Phosphorus was used for experimental pui'poses. It is not practically suitable 

 since it delicpiesces on oxidation. Thorium and uranium 2)reparation j)laced on e 

 would be free from this difficulty. With this plan I succeeded but partially, as 

 the air could not be ke{)t dry enough and the emanating power of the phosphorus 

 is soon extinguished by the moisture of its oxidation. I shall refer to the results 

 below. 



It soon appeared that the ionizer was superfluous as the electrostatic induction 

 within the capsule, h, is nearly perfect. Thus if a positive charge is imparted to 

 the outside of tlie capsule as shown, it redistributes itself at once, forming a field 

 within the cajisule such that the inside of the capsule is positive throughout its 

 equatorial plane and the e^\gii of the disc, c, negative. Nearly the whole field is 

 closed within this equatorial zone, and but little induction is contributed by the 

 lateral discs, e, e, seeing that the edge of the vane has been brought close up to the 

 inner face of f> to secure good damping, and e is remote. Hence the needle, ?«., n, is 

 charged positively by an amount e(iual to the charge within h or on the edge of the 

 vane, c. 



An instrument is tlius ol)tained which is very siniihir to the electroscope, only 

 more sensitive and like it of low capacity. The needle of the apparatus being 

 freely suspended, no communication of permanent charge to it is possible for 

 reasonably low potentials if care be taken that no charge escapes by way of the 

 silk fibers of the bifilar suspension. The air within the cover of the apparatus 

 must therefore be kept scrupulously dr}', otherwise the zeio will shift sliglitly with 

 each charge. 



5. ()h><erv(itions. — The needle being free, it was found convenient to take 

 temporary instead of permanent deflections or to use the instrument ballistically, as 

 it were, su[)posing, however, that the contact is made during the whole swing. The 

 Itifilar suspension was about 12 cm. long, .02 cm. below, and .1 cm. above, the 

 needle being charged by a storage battery. Using silk threads of four (glued) 

 fibers each, their individual torsion is alread}^ a serious factor. 



Withdrawing tabulated statements for brevity, it suffices to indicate that both 

 the tempoiary and the permanent deflections varied w ith the potential. The zero 

 changed slightly, showing the undried silk threads to conduct. Since d^= aV^V^ 



I 



