﻿704 Prof. R. W. Wood on Selective Reflexion, Scattering 



was focussed through the side of the bulb as close to the 

 centre of the flat bottom as possible. The adjustments were 

 made by means of a small piece of uranium glass, which 

 enables one to locate the path of the rays by its phosphor- 

 escence. It is most important to prevent the light which is 

 reflected from the walls of the bulb from getting at the photo- 

 graphic plate. This gave a good deal of trouble, but by means 

 of the device shown in PL X. fig. 4 a, it was practically elim- 

 inated. A wooden box was made, measuring 40 X 10 X 10 cm., 

 and a large circular hole cut in one end which was covered 

 with a cone of black cardboard made by cutting out a cir- 

 cular disk, cutting along a radius and pasting the cut edges 

 together, overlapping them about 2 cm. A small round hole 

 was burned through the apex of the cone, and this was placed 

 against the flat bottom of the resonance lamp. A quartz lens 

 mounted in a partition of the box rendered the rays which 

 came from the resonance lamp through the small hole parallel, 

 so that the intensity of the light after its passage through the 

 cell could be recorded close to the cell or at a distance from it. 

 The importance of doing this is apparent from the following 

 considerations. If we place a photographic plate close against 

 the double cell containing the vapour, it will be illuminated by 

 the primary beam which has traversed the cell and also by the 

 scattered resonance radiation. If, however, we place the plate 

 at a distance, say at the other end of the box, the primary rays, 

 being parallel, will reach it with undiminished intensity, while 

 the effect of the scattered radiation wall be negligible, since 

 its intensity diminishes according to the law of inverse squares. 

 No difference should be found with the plate in the two 

 positions for the light which has gone through the compartment 

 containing air and mercury vapour, since, as we have seen, the 

 presence of the air destroys the resonance radiation. The ex- 

 periment was made in the following way. One compartment 

 of the double cell was highly exhausted and the other to a 

 pressure of 3 cm. A strip of photographic plate 1*5 cm. 

 in width was mounted close to the cell and received the light 

 which had traversed the lower half of each compartment. A 

 larger plate was mounted at the other end of the box, and 

 received the light which passed above the first plate, and had 

 traversed the upper halves of the compartments. Thus four 

 records were obtained at once under precisely similar con- 

 ditions as regards exposure-time and development. Much 

 trouble was experienced in getting things adjusted so that the 

 intensity close to the lens and at the end of the box came out 

 the same with the cell removed, which is of course a necessary 

 preliminary experiment. It was finally found that the air of 



