Spectrum Lines for Monochromatic Illumination. 529 



The quartz plate has proved most satisfactory in every 

 respect. Placed between the two halves of a Dunoyer con- 

 denser o£ 10 cm. diameter and 12 cm. focus for parallel light 

 (L e. 24z cm. focal length when forming an image of a source 

 •of light equal in size to that of the source), it is possible to 

 utilize the entire 50 square cm. of the plate in forming an 

 image of the aperture (1x3 cm.) placed before the sodium 

 flame, containing only J) 1 or D 2 light. 



The arrangement of the apparatus is shown in the diagram. 



Double Image 

 Prism 



Double Image 

 Prism 



The double-image prism A is so oriented as to have the vibra- 

 tions of the two transmitted beams at 45° with the vertical. 

 The second double-image prism is oriented in the absence 

 of the quartz block, so as to give two images of the source 

 side by side, and separated by a distance equal to the 

 width of each image. Each image contains both D x and D 2 . 

 If now the quartz block is placed between the two halves 

 •of the condenser, all of the D 2 light (for example) leaves 

 the original images, and unites into a third image between 

 the other two. This is the image utilized. If it is desired 

 to have D x light in this image, it is necessary only to rotate 

 the block a degree or to, thereby increasing the optical path 

 in quartz. If we rotate the analysing double-image prism 

 the central image splits up into two, and the correct position 

 of the prism can be determined by bringing these two images 

 into exact coincidence. As I have said previously, this 

 method obviates the 50 per cent, loss of light which results 

 from the use of Nicol's prisms. As the double-image prisms 

 had apertures considerably less than that of the quartz block, 

 they were placed midway between the condenser lenses and 

 the image and source, as shown in the figure. With prisms 

 of too small aperture, however, placed close to the source 

 and image, the separation of the images would be insufficient. 

 To test the adjustment of the quartz block and the general 

 -efficiency of the apparatus, we have only to receive the three 

 images upon a piece of white paper, and project an image of 



