276 



BULLETIN OP THE UNIVERSITY OF WISCONSIN. 



By this method the indices given in Table VIII, were calcu- 

 lated. 



Table VIII. 



These values of h- are indicated on Plate 16 by crosses and 

 the values obtained with the spectrometer by circles. 



To continue the work in the fall and winter, when cloudy 

 weather made sunlight unavailable, an electric arc formed 

 between an iron rod and a rapidly rotating iron disk, was sub- 

 stituted for the sunlight. The bright lines in the iron spec- 

 trum gave the same results as the Fraunhof er lines. With light 

 of wave length shorter than 368 p /u., passing through the 

 cyanin prism, no effect could be produced on the sensitive plate 

 although a five hours continuous exposure was made. To in- 

 crease the intensity of light when passing through the cyanin, 

 the prism was moved from H to D, eight centimeters behind the 

 slit, C (Fig. 5). 



With this arrangement of the apparatus photographs were 

 taken and the results are shown in figures 3, 4, 5, and 6 of 

 plate 19. Light coining by the direct path made the upper 

 set of lines, while the part coming through the cyanin formed the 

 lower set. In figure 3 the time of exposure for the upper set* 

 was forty seconds, and for the lower, seventy seconds. For the 

 longer waves the intensity of the two sets of lines in this pho- 

 tograph is almost the same, but beyond 368 ^ ^ no lines appear 

 in the lower set. The cyanin evidently absorbed the energy of 

 wave lengths shorter than 368 n /*. The presence of this ab- 

 sorption band in the cyanin and the approach to the absorption 

 band in the glass beginning at 335 p i*. is seen in figure 4 of 

 plate 19. The time of exposure in the photograph of the lat- 

 ter, by the direct ray, was forty seconds, and through the cyanin 

 five minutes. 



