Electrical Vibration associated with a thin Conductor. 97 



Some years ago I published an account * of an experi- 

 mental comparison, by a resonance method, o£ the periods of 

 the electrical vibrations associated with simple circuits. In 

 discussing the results, I assumed that the wave-length of the 

 oscillation connected with a narrow rectangular circuit was 

 equal to the perimeter of the rectangle, and was led to think 

 that the measurements supported Macdonald's theoretical 

 deduction as to the wave-length of the vibration on a straight 

 rod. The assumption, which affects the discussion only, is, 

 however, quite unjustifiable, as the wave-length depends on 

 the ratio of length to breadth as well as on the perimeter 

 of the circuit. Contrary to the opinion I then expressed, 

 the experimental results cannot be taken as supporting 

 Macdonald's theory. 



More recently, several values of the ratio of wave-length 

 to length of conductor, determined from interference ex- 

 periments, have been given in accounts of measurements of 

 electric waves from linear oscillators of small dimensions. 

 Writing \ = kl, where I is the length of the conductor, 

 Willard and "Woodman t deduce for h the value 2*48 ; Cole J, 

 2-52 ; Blake and Fountain §, 2 47 ; and Webb and Wood- 

 man ||, 2*3. Finally, Ives If, using a very high resistance 

 receiver with an interferometer arrangement, finds 2*04 as 

 the value of k for certain linear oscillators varying in length 

 from 5 to 10 centimetres. 



At the meeting of the British Association in S}'dney, in 

 1914, I mentioned that the interference experiments had 

 recently been repeated by students in the Physical Labora- 

 tory of the University, with results completely confirming 

 those of Ives. I now give a summary of the measurements. 



In the interferometer experiments Ives' general arrange- 

 ment has been followed, but a coherer was used as a receiver 

 instead of one involving a thermo-electric junction. At each 

 observation the coherer was isolated during the passage of 

 the waves, being connected to the galvanometer circuit, by 

 movable conductors working in mercury cups, only after 

 sparking had ceased. A coherer used in this way is un- 

 affected by disturbances not in its immediate neighbourhood. 



The positions of the nodes of the stationary wave system, 

 formed by the reflexion of an incident beam at a plane zinc 



* Pollock, Jouvn. and Proc. Roy. Soc. N. S. Wales, xxxvii. p. 198 

 (1903) ; Phil. Mag. vii. p. 635 (1904). 



t Willard and Woodman, Phys. Rev. xviii. p. 1 (1904). 

 r. Cole, Phys. Rev. xx. p. 268 (1905). 

 § Blake and Fountain, Phys. Rev. xxiii. p. 256 (1906). 

 || Webb and Woodman, Phys. Rev. xxixT p, 89 (1909). 

 1] Ives, Phys. Rev. xxx. p. 199 (1910) ; xxxi. p. 185 (1910). 



