42 Rubens and Trowbridge — Dispersion and Absorption 



Measurement of the Absorption. 



Several measurements were made a short time ago by one of 

 us, touching on the infra-red absorption spectrum of rock salt 

 and sylvine. A diffraction grating served to produce the 

 spectrum. A zircon lamp was used as a heat source, the 

 rays of which had to pass, before their entrance into the spec- 

 trometer, through a thin lamp-blacked chloride of silver plate, 

 whereby the disturbing energy from rays of shorter wave 

 length was removed. In order to overcome the influence of 

 the spectra of higher order, the throws of the bolometer were 

 always observed in the following manner. A fluorite plate 

 6 mm thick was slid into the path of the rays, and after the com- 

 pletion of the throw, again removed. This fluorite plate 

 absorbed completely only the long waves of the spectrum of 

 the first order (12 to 20/x), while it let through the short waves 

 of the spectra of higher order unweakened, except for the loss 

 of strength due to reflection. The results on the absorption of 

 rock salt and sylvine plates, obtained in this manner, could, on 

 account of the exceedingly small amount of energy in the dif- 

 fraction spectrum, lay no claims to great accuracy and were of 

 service only as a forerunner of more careful determination. 

 On the other hand, it was possible for us, with the use of the 

 above mentioned acute-angled prisms made of rock salt and 

 sylvine, to produce a doubly dispersed, continuous spectrum 

 which was sufficiently intense and pure, to make possible the 

 measurement of absorption for both substances. To this end, 

 the grating g was removed from the table of the spectrometer 

 s i e i e i s v ai] d in its place the rock salt prism was set up, so that 

 it gave the minimum of deviation for the sodium line. In the 

 adjustment of the apparatus- for a given wave length, the 

 mirror e as well as the moveable arm of the spectrometer 

 s i e i e <2 s » was turned through the proper angle, the value of 

 which, in each case, was calculated from the dispersion of the 

 prisms and the constants of the apparatus. 



The plates, the absorption of which was to be determined, 

 were placed directly before the slit s in the path of the rays, 

 by means of a sliding holder. Near the slit s 1 was a shutter, 

 worked with cord and pulley, which could be set from the 

 observer's table so that it either let through or shut off from 

 the slit s } the rays coming from the mirror e . This shatter 

 consisted of a plane- parallel glass plate 2 mm thick, which let 

 through about 75^ of the total radiation of the zircon lamp, but 

 absorbed completely the rays in the field between X = 8/jl and 

 X = 21//-, in which we worked. The advantage which lies in 

 the use of such a glass plate instead of a metal shutter is that 

 the diffuse radiant heat of short wave lengths (the maximum 

 of emission of the zircon lamp lies at about 1'4/a) which may 



