20 ATLAS OF ABSORPTION SPECTRA. 



always associated with the same maximum depth, even when the minimum 

 thickness is unchanged, because several cells of different dimensions were 

 employed. 



Eleventh. Finally, a brief account of the most noticeable characteristics 

 of the absorption spectrum, between the limits o. 20//. and 0.63//, is furnished. 



The results of eye-observations of the absorption spectra come first and 

 serve as checks on the photographic records. The data obtained visually 

 are qualitatively reliable for all strong bands between 0.40// and 0.63//. For 

 cases of very weak, general absorption much less importance must be 

 ascribed to the visual results because, unfortunately, the cells were not con- 

 structed so as to present side by side, in the field of view, two spectra, the 

 one of the light after passing through the absorbing solution, the other of 

 the unabsorbcd light direct from the Nernst glower. 



When the solution is fluorescent, or decomposes when ultra-violet light 

 falls u])on it, or possesses a characteristic odor, etc., the facts are noted. 

 That the spectrograms are not distorted by the presence of fluorescent light, 

 but give as true records of the absorption spectra of fluorescent compounds 

 as they do for non-fluorescent solutions, was ascertained by direct experi- 

 ments. (In particular, sec the record for solution No. 107.) 



Lastly, the approximate wave-lengths of the maxima and minima of 

 absorption, as obtained from the spectrograms, are given, beginning near 

 o. 20/'. and continuing to 0.63//. When the wave-lengths of the "ends" of a 

 region of absorption are given they obviously have significance only under 

 the conditions of thickness of absorbing layer, of concentration, of length 

 of photographic exposure, etc., which prevailed at the time when the 

 spectrogram was taken. The maxima are not subject to the same limita- 

 tions. The fact that the Seed films can produce spurious absorption bands 

 in the green must be again emphasized. {See figure 102, plate 26.) 



When the end of the spectrogram, which marks the fading away of the 

 sensitiveness of the emulsion from the yellow to the orange, is practically a 

 straight line perpendicular to the length of the spectrogram it means that 

 there is no appreciable general absorption in this locality, but when the limit 

 just specified is approximately a right line inclined at an obtuse angle to the 

 positive direction of the axis of wave-lengths it signifies that appreciable 

 general absorption is present in this region. 



