﻿692 Prof. R. \V. Wood on Selective Reflexion, Scattering 



closed by end plates o£ the same substance which had been 

 ground fiat and polished. These plates were fused to the 

 ends of the tube, which had been flared out in order to 

 prevent spoiling the figure of the central portion of the 

 plates by fusion. The tube is shown in PL X. fig. 1, and 

 was designed for a study of the dispersion of the vapour by 

 an interferometer method, an investigation of which will 

 shortly be commenced. It was made by Heraeus and turned 

 out quite as well as I had hoped, for I have had no difficulty 

 in forming interference fringes with one beam of light 

 traversing the tube, showing that the process of fusing on 

 the end plates did not damage their optical quality. The 

 tube was highly exhausted and sealed, and the light of a 

 quartz mercury arc focussed along the axis of the tube. The 

 tube was now photographed from the side with a camera 

 furnished with a lens of quartz, which was constructed in a 

 few minutes from an old box used for storing photographic 

 negatives. A quartz lens of about 18 cm. focus was fitted 

 into a hole made in r,he bottom of the box, which was used 

 standing on its side. A strip of thick sheet brass was cut to 

 the same width as the plates for which the box was made. 

 To this strip was soldered another strip in which a circular 

 aperture 8 cm. in diameter had been cut. This strip stood 

 vertically in the box, and was provided with two spring clips, 

 or supporting clamps, to hold the plate in place against the 

 circular aperture. The camera was focussed by sliding the 

 strip along in the grooves which supported it. (These 

 grooves were made originally for holding the plates.) The 

 correct focus for the invisible rays with which we are 

 concerned in the present paper was found by pointing the 

 camera at the slit which had been mounted in the focal 

 plane of the quartz spectrograph (to convert it into a mono- 

 chromator), and observing the focus with a piece of uranium 

 glass. The distance from the lens to the slit from which the 

 ultra-violet rays diverged was then measured. This gives 

 the correct distance at which the object to be photographed 

 must be placed in order to be properly focussed. The 

 camera is shown in fig. la. I have found that much time 

 is saved by the use of a camera of this description, for the 

 plates can be cut carelessly and in a hurry, as they do not 

 have to be of an exact size, and no time is lost in loading 

 or unloading a plate-holder. In ultra-vio 7 et work of this 

 description, where the sources of error and all other un- 

 expected troubles and difficulties have to be located by 

 photography, the time saved in this way amounts to a good 

 deal. In the present work I have taken about six hundred 



