232 Dr. Alex. Hodgkinson on 



its correctness, with its absorption bands narrow and 

 defined as the lines of a radiation spectrum, admits of much 

 more accurate results being attained than is possible with the 

 broad banded, indefinite spectrum obtained in the ordinary 

 way. 



Having shown how the absorption line of maximum 

 intensity may be obtained from these converging absorption 

 bands, and the position of such primitive line be indicated on 

 the spectroscopic scale, we require to know, firstly, whether 

 the positions of these primitive lines are constant whatever 

 the strength of the absorbing solution, and, secondly,, 

 whether they are constant if the angle of the wedge-shaped 

 cell is varied ; since each of these conditions produces a pro- 

 found modification in the appearance of the spectra. 



Firstly, then, what is the effect of varying the strength of 

 the solution in the same hollow prism ? To ascertain this I 

 carefully constructed a small hollow prism with a re- 

 fracting angle of 34°4o'. This was placed in front of the 

 spectroscopic slit, with its refracting edge at right angles- 

 to the slit and so arranged that it could be moved 

 up and down, so that its edge could be made to 

 cross the slit at any point required. The spectrometer 

 employed on this occasion, one specially made for me 

 by Hilger, of London, is furnished with one whole and two 

 half prisms of calcite, and is graduated to 5' of arc reading 

 by means of a vernier to single minutes. Its dispersive 

 power is, therefore, considerable, and its mechanical construc- 

 tion is very good. It is, in fact, a twin instrument to the 

 one with which Messrs. Liveing and Dewar have done such 

 good work. Arranging the wedge-shaped cell, full of a strong 

 solution of blood, in front of the slit, the spectrum was 

 observed in the usual way, excepting that the eye-piece of 

 the spectrometer consisted of a filar micrometer, so arranged 

 that the cross webs moved up and down in the axis of the 

 slit, so that the decussation of the webs might be raised to 



