144 DR. T. R. MERTON AND PROF. J. W. NICHOLSON ON 



almost indefinite period without any noticeable change, and there was no trace, on 

 the walls of the tube, of any metallic deposit from the electrodes. We have, in 

 addition, used tubes of the ordinary H pattern for the investigation of the mutual 

 action of Hydrogen and Helium, and of the effect of pressure on the spectrum of 

 Helium, and we are indebted to Sir HERBERT JACKSON, K.B.E., F.R.S., for a tube of 

 the conventional Pliicker form which contained Helium in a very high state of purity. 

 The tubes containing Hydrogen and Helium were filled in the same manner as those 

 with flat cathodes, with the exception that the preliminary exhaustion was effected with 

 an oil pump, the tubes being washed out repeatedly during the process of exhaustion 

 with pure Hydrogen. This was admitted by heating in a Bunsen flame a palladium" 

 tube which was sealed into a glass tube connected with the apparatus. After the 

 Helium had been admitted, the desired quantity of Hydrogen could be introduced 

 in this way. 



In all the experiments recorded, the tubes were excited by means of an induction 

 coil capable of giving a 10-inch spark in air, with a mercury jet interrupter by means 

 of which the discharge could be maintained uniformly over any desired period of 

 time. 



(V.) Tlie Helium, Spectrum as a Function of Cathode Distance. 



We now enter upon a discussion of a series of plates of the Helium spectrum, 

 taken at points whose distance from the cathode increased regularly. The tube was 

 filled with Helium containing a little Hydrogen and Mercury, but the exposure given 

 was not in any case sufficient to enable these impurities to appear on the enlarged 

 photographs. 



The essential features of the experimental arrangement are sufficiently evident 

 without the necessity of a diagram. The cathode was flat, and was arranged with its 

 length parallel to the plane of the slit of the spectrograph. By moving the vacuum 

 tube in a direction perpendicular to the plane of the cathode, light from any desired 

 region of the tube could be allowed to enter the slit and collimated in the usual 

 manner. It has not been possible to isolate the spectrum of each region with great 

 purity, but the slight overlapping of the effects of consecutive regions, which could 

 not be avoided, does not affect the conclusions subsequently reached. 



A series of eight photographs will be discussed. The first relates to the region 

 immediately in front of the cathode, and the others to regions at successive distances 

 of 1 mm. from this region. In the first five photographs the photographic intensities 

 are all directly comparable, as they were all taken on the same plate, with two hours' 

 exposure in each case. The other three were necessarily taken on a different plate, 

 and though directly comparable among themselves, are not necessarily so with regard 

 to the former set. At the same time, serious differences in the behaviour of the 

 two plates are not to be expected, for they were selected from the same batch of 

 plates. 



