14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 68 



THE SLIT 



The slit was lo cm. tall with jaws opening bilaterally. Because of 

 the great range of intensity to be observed, from one to a hundred- 

 thousand fold, from one end of the spectrum to the other, some means 

 had to be provided for altering and knowing accurately the amount 

 of alteration in the amount of energy flowing through the slit. It 

 was thought best not to depend upon the readings of the slit mi- 

 crometer screw for accurately measuring the widths of slits used. 



In the earlier part of the research relating to the upper portion 

 of the spectrum, 2 jx to 10 /x, when the lamp used had its filaments 

 vertical, a lateral change of the slit opening would have been bad 

 because of the irregular lateral changes in the intensity in the lamp. 

 Accordingly a constant slit width was used just wide enough, or less, 

 to be covered by the lamp image, and a series of horizontal grids or 

 diaphragms was used to lessen the intensity of the spectrum. This 

 scheme had the disadvantage of giving a low purity of spectrum 

 corresponding to the great slit width necessary in the least intense 

 part of the spectrum. It had the advantage of allowing the more 

 accurate determination of the energy reducing value of the dia- 

 phragm. For the change of deflection when the diaphragm was 

 inserted was not complicated by changes in the purity of the spec- 

 trum such as would attend change of slit width. Besides this, owing 

 to having a wide slit, the trouble from diffraction presently to be 

 described did not take place. 



For the observations reaching to the greater wave-lengths, a lamp 

 with horizontal filaments was used, the intensity of which was there- 

 fore uneven in the vertical direction. The largest slit width for a run 

 was set by closing the slit until the jaws just touched a cylindrical 

 template (4.99 mm., 5.82 mm., or 10.69 ^^^^- ^^ diameter). In the 

 more intense regions of the spectrum the slit was reduced by turning 

 up in succession in front of it two accurately cut slits the widths 

 of which had been measured on a comparator (2.36 mm. and 

 0.67 mm.) . Four rotating sectors could also be used for reducing the 

 energy, which according to their mechanically measured apertures, 

 reduced in the ratios, 0.333, 0.1054, 0.0501 and 0.00512 respectively. 



All this provision for making reductions of intensity with accuracy 

 proved somewhat illusory. It has been found here and by others* 

 that rotating sectors do not cut down the energy proportionally to 

 their mechanically measured apertures as would be expected. For 

 instance the 0.333 sector cut down the energy by 0.344 as determined 



* Coblentz, Bulletin Bureau of Standards, 4, p. 455, 1907. 



