360 Dr. Barton and Mr. Lownds on Reflexion and 



a distant telescope and scale. The induction-coil is put on 

 end to avoid any magnetic effect on the electrometer-needle. 

 The leads also of the coil are kept close too ether, and, before 

 taking a set of readings, the electrometer is tested as to the 

 adequacy of these precautions against possible disturbance. 

 The disks and the ends of the needle are each about 2 cm. 

 diam. and 1 cm. apart. It will be shown later that the 

 reflexion of waves at the electrometer, regarded as a con- 

 denser on the line, is extremely small and quite negligible. 



8. TT' denotes the terminal bridge which is used to absorb 

 the waves on arrival at the end of the line. It consists of 

 graphite markings on a disk of ground glass, 3 inches 

 diameter, and before each set of readings is adjusted to the 

 resistance which corresponds to complete absorption ; namely, 

 for the line in question, 560 ohms*. 



DD' denotes one of the condensers in use to reflect the 

 waves. For the transmitted system they are removed from 

 DD' and inserted at FF'. They were all three made of 

 disks of zinc 1 cm, apart, their radii being respectively 15, 

 9, and 5 cm. The two disks were kept at the right distance 

 apart, and insulated by pieces of wood saturated in paraffin- 

 wax, notched and secured with shellac. 



9. The total length of the line from S to T is of the order 

 166 metres. The distance SE is 116 metres, SF is of the 

 order 60 metres. EJ) varies from about *75 metre to 

 50 metres. Exact details will be given in connexion with 

 the various experiments. 



10. The waves generated by the primary oscillator are of 

 the damped simple harmonic type 8"5 metres long, their 

 logarithmic decrement per wave being 0*6. Thus the wave- 

 train launched upon and propagated along the line with the 

 speed of light is a damped train with its large end or head 

 leading, the tail, after about 10 or 12 waves, being almost 

 negligible. As the waves proceed along the line they suffer 

 attenuation, also some loss on reflexion at the oscillator. 

 The constants expressing these effects were previously deter- 

 mined experimentally on the line now in use, and their 

 numerical values will be introduced into the theory later. 



Theory. 



11. Preliminary. — The plan adopted throughout the ex- 

 periments is to take (1) readings, of the electrometer with no 

 condenser on the line, but only the absorbing-bridge at the 



* " Absorption of Electric Waves, &c," Phil. Mag., January 1897. 



