518 Dr. J. H. Vincent : Electrical Experiments 



C 

 L or F= f --i-K, where C and K are constants. This 



Li 



equation represented very closely the results obtained in 

 this series o£ experiments and of other series of the same 

 type. Changing the bore of the tube has very little effect 

 on the frequency. 



Experiments on the effect of change of pressure on the 

 frequency led to the result that the frequency was not pro- 

 portional to the square root of the pressure as in the necklace 

 effect; but was approximately proportional to a power of the 

 pressure greater than "5. With a narrow tube ('1 mm. bore) 

 the frequency varied very nearly as the pressure to the 

 power '86. Three series of experiments on the change of fre- 

 quency with pressure are shown in fig. 12 (PL VII.) . Curve A 

 gives the results for a quartz tube (4*9 cm., 6*4 mm., *23 mm., 

 105 volts, 51 ohms). Curve B shows those for another piece 

 of the same tube 10 cm. long, 103 volts, 37 ohms; while 

 curve C refers to a quartz tube (5 cm., 5 mm., "1 mm., 

 112 volts, 101 ohms). 



In the last case careful observations were made of the 

 appearance of the arcs on the screen after reflexion from the 

 revolving mirror as the pressure was changed. When the 

 pressure was much below 29 cm. the intermittent arcs tended 

 to give place to a steady arc which will be considered later 

 under the " ribbon " effect. When the pressure was 29 cm. 

 the appearance was that of fig. 9, the arcs being very regular 

 both in size and in periodicity. The period of luminosity was 

 1/10 of the whole period. As the pressure was increased, 

 the period of luminosity increased relatively to the whole 

 period, and became 1/9 at 44 cm. and 1/6 at 79 cm. 

 pressure. The fans meanwhile were getting less wide, lhatis, 

 the maximum arc length decreased. At 109 cm. the arcs 

 show a phenomenon which is only met with in very narrow 

 tubes : they go out while the threads of mercury are still 

 approaching. In this case the luminous fraction of the 

 period is about 1/4; more often they cease to be luminous 

 when the maximum width has been reached, and then the 

 fraction is about 1/5 ; but on this pressure the fans are 

 occasionally replaced by necklace figures in which the arc 

 breaks on growing as in fig. 3. As the pressure is increased, 

 the fans become less frequent and the broken necklaces become 

 more frequent until at 160 cm. the latter predominate. 



The effect of altering the electromotive force on the frequency 

 is shown in fig. 13 (PL VII.) , in which the curves a, b, c, d 

 refer to sets of experiments on different tubes. The dimensions 

 of the tubes and the magnitude of the external resistances 



