510 Dr. J. H. Vincent : Electrical Experiments 



the resistance was 190 ohms. The first bead shows a 

 stationary drop of mercury left in the tube, which is re- 

 absorbed into one of the columns of mercury as the waist is 

 formed. The explosion which initiates the last bead but one 

 also leaves a pair of drops of mercury in the tube near the 

 cathode. These move outwards and are met by the return 

 column. 



All else being maintained as before, the resistance was 

 raised to 600 ohms and fig. 5 taken. The width of the beads 

 and waists is now greatly reduced. We are now approaching 

 the line DE (fig. 2), and the lamp would be extinguished by 

 raising the external resistance much beyond this limit. The 

 invariable characteristic of the arcs produced in regions in 

 the diagram indicated by points above AB and below DE, is 

 the presence of arcs that keep lit until the columns unite. 



Approximate straight lines like DE, AB, and AG form the 

 typical diagram showing the connexion between the external 

 resistance and the electromotive force in the necklace effect. 

 When the pressure on the fluid is raised considerably, the 

 lines are convex to the axis of volts. With the temperature 

 of the laboratory maintained constant, it is possible to obtain 

 very concordant series of readings for the circumstances 

 governing the production of the necklace effect in any given 

 tube, due care being taken to maintain the total hydrostatic 

 pressure on the fluid in the tube constant. This concordance, 

 in general, is only found among the numbers obtained from 

 a particular mounting of- the tube. Even when the attempt 

 is made to fill the perfectly dry tube with pure dry mercury, 

 and to remove all the air from the tube, differences in the 

 diagram are found with each filling. The most pronounced 

 difierence in the behaviour of a tube with separate fillings, each 

 performed with great care, is shown in PL VII. fig. 6, 10 cm., 

 3*9 mm., *36 mm., 81 cm. pressure, in which the lines DE, 

 AB, AC refer to one filling, and ab, ac refer to another. In 

 this latter case the line de was only indicated by the fact that 

 in the neighbourhood of the point a it was possible to get the 

 necklace to break at the waist without the lamp being extin- 

 guished altogether ; with higher voltages, the lamp went out 

 while still showing a good necklace when the resistance was 

 raised too high. 



From the diagrams (figs. 2, 6, & 7) we shall see that, 

 with different lengths of tube and at different pressures and 

 with different fillings, the perfect necklace region is always 

 bounded by two lines rising from left to right. From other 

 diagrams showing the behaviour of tubes of different bore, 

 not here reproduced, it was found that this applies also to 

 variation in internal diameter. 



