TIM: <"\vnn'TiON OF TIII-: KI.KCTRIC M-AI;K 





- inilliins. li tin- poles it reduces to alnuit 400 metres / second, which is not very 



different f'nuii tin- velocity of souml in air ;it the ordinary teni]>eratme. It is well to 

 realise this, for it shows that the sound-wave which is produced by the spark has 

 moved outwards through a distance of a few millimetres only by the time the metallic 

 molecules have readied the centre of the spark. 



In order to see whether the capacity of the jars and distance between the poles have 

 an appreciable effect on the result, the displacements of the /inc lines wen; measured 

 I'or sparking distances of approximately '5, TO, and 1'5 centims., and in each case the 

 capacity was altered by taking the discharge from 2, 4, and (! jars. Table II. emlxKlies 

 the result ; the velocity (V) given l>eing, as al*>ve explained, the average speed I >et \\een 

 the pole and a point at a distance of '2 millims. from it. 



* 



TABLE II. Average Velocity (V) in metres / second of Zinc Molecules. 



The first striking fact shown by the table is the uniformly slower speeds derived 

 from the doublet at 4925 as compared with that of the least refrangible line of the 

 adjoining triplet, and we have assured ourselves that there is no difference in displace- 

 ment between the two first components of the triplet. The third comjxment is too 

 weak and too near an air line to admit of satisfactory measurement. It was one of 

 the main objects of our investigation to discover, if possible, such differences in the 

 velocities as might indicate the presence of different kinds of molecules, but we 

 hesitate to ascribe the differences found to such a cause. The line 4925 is the least 

 refrangible component of a double line, which is wider and much stronger at the 

 base than in the centre of the spark. In order to measure the displacement, the cross 

 wire of the reading microscope has to be set on the edge of the displaced line, first 

 near the pole, and then on a corresponding point nearer to the centre of the sjwrk 

 gap. Where the line is strong, the edge of the line would be the least refrangible 

 edge of the least refrangible component, but when the line is weak, the strongest part 

 of the line would be that part where the two components begin to overlap, i.e., the 



2 D -2 



