with the object of having no moving parts to get out of order, and choice fell on a 

 combination of induction coil and vacuum tube. With this arrangement, the hammer 

 of the mechanical " break " for the induction coil is screwed up tight so that it cannot 

 move. If now the vacuum tube is joined up to the secondary of the coil, and a suitable 

 number of accumulators with appropriate resistance, capacity, and inductance to the 

 primary coil, every " break " in the primary current will cause a flash in the vacuum 

 tube, while the " make " shows no visible effect. With a properly designed coil, a 

 flash is given for a very low primary current and this is of advantage in that it reduces 

 the amount of current through, and of sparking at, the clock contacts. 



In these observations a small vacuum tube about 8 inches long filled with 

 helium was used as the source of light and gave a fine bright yellow light at every 

 break.* 



To show the accuracy of the observations with this form of apparatus, Table XLVIII 

 below is taken from the Christchurch 1913 observations. For comparison, two similar 

 observations of the coincidence interval, one from Potsdam I, and the other from Cape 

 Evans " C," are given in the same Table. 



TABLE XLVIII. 



The consistency of the results at Christchurch with this arrangement could hardly 

 be bettered, but an extended trial would be necessary before a statement of the relative 

 accuracy of the two methods could be made. 



* Note added, June, 1914. This arrangement, it is found, was used long ago by Dr. Horton in his 

 researches on the elasticity of quartz fibres at the Cavendish Laboratory. The technique is much improved 

 by the use of vacuum tubes filled with neon, as this gives a red spark of much greater brightness for the 

 same current density. A further improvement is introduced by the use of a very small vacuum tube 

 only 1 J inches long and with a constriction only i inch in length. Since our return, a small coil has been 

 made by Cox to my specifications. This works with a single accumulator and gives a fine bright spark 

 with only 0-03 amp. in the primary circuit. No external capacity is needed with this instrument and 

 the whole apparatus is exceedingly light and very simply set up. The set built by Cox has been sent to 

 the Indian Survey for trial at their field stations. 



62 



