2 ATLAS OF ABSORPTION SPECTRA. 



made by Heraeus. The fact that the absorption of each substance in the 

 ultra-violet is recorded, makes the plates of especial value to anj^ engaged 

 in the preparation of screens for spectroscopic or photographic purposes. 



For the removal or transmission of one or more isolated lines some 

 other arrangement is often more useful than an absorbing screen. A 

 spectroscope with a slit placed in the focus of the observing telescope 

 (monochromatic illuminator) is frequenth' all that is necessary. But if more 

 light is required, the following device may be used. A block of quartz from 

 2 to 4 cm. in thickness, cut perpendicular to the optic axis, is mounted 

 between two Nicol prisms. The transmitted spectrum is crossed by black 

 bands, which result from the rotatory power of the quartz. By adjusting 

 the nicols and varying the thickness of the quartz it is often possible to get 

 rid of the spectrum lines which are not desired, and at the same time to 

 utilize the whole area of the source, which can never be done with the spec- 

 troscope. In this wa}^ with a quartz plate 45 mm. thick, the line 4809 of 

 the zinc arc in quartz can be completely removed and the two lines 4721 

 and 4679 transmitted. This method is especiall}' useful in the study of the 

 fluorescence excited in various bodies by monochromatic light. 



If it is necessar}' to separate radiations of very nearh' the same wave- 

 length, for example if we wish to work with the light of one of the two 

 sodium lines, the following arrangement can be used : A quartz plate about 

 2 cm. in thickness, cut parallel to the axis, is mounted between crossed nicols, 

 with its axis making an angle of 45 with the principal planes of the polar- 

 izing prisms. The source is placed behind a vertical slit 2 or 3 mm. in 

 width, and the light, after traversing the polarizing system is brought to a 

 focus by a lens. A number of concentric maxima and minima will be 

 formed, the light of Di and Dg being found in adjacent maxima. The 

 wave-length which is not desired can be stopped by a screen of suitable 

 dimensions placed at the focus of the lens. In this waj' it is possible to 

 obtain a source of Di or D2 radiation of sufficient intensity to show distinct 

 fringes in a Michelson interferometer. By a curious coincidence this method 

 occurred independently to the writer and to Professor Michelson on the same 

 day. It has been found to give excellent satisfaction. The thickness of 

 the quartz plates used in either of the above cases depends upon the close- 

 ness of the lines which it is desired to separate. 



We are under great obligation to the Actiengesellschaft fiir Anilinfabrika- 

 tion and to Meister, Lucius & Briining, both of which firms presented the 

 Johns Hopkins University with a large collection of aniline dyes. 



