SPECTRO PHOTO METRIC CONSTANTS. 223 
sources of information. It is absolutely essential to work with Hiifner's 
spectrophotometer in a perfectly darkened room. 
Before commencing photometric measurements, the observer will ascertain 
whether the analysing Nichol is in the position in which it allows the polarised 
beam to pass unabsorbed. He will then fill the absorption-trough, and isolate 
and measure the spectral region for which the extinction-coefficient is to 
be determined. 
On now looking through the eyepiece two spectral strips will be seen, 
separated by a sharp horizontal line; these spectral strips will be of unequal 
brightness; the upper, being a portion of the spectrum of the polarised 
beam, will be much less luminous than the lower. The composite glass 
plate in front of the slit of the collimator is now moved inwards in the 
direction of the beam of the unpolarised light, so as to diminish its intensity, 
until the upper and the lower spectral strips appear of precisely the same 
brightness. 
The trough containing the coloured liquid under investigation is now 
brought into position, the upper surface of the glass cube in the trough being 
placed about 1 mm. below the plane passing through two horizontal angles of 
Albrecht's glass body. On now examining the spectra, it is at once seen that 
the lower of the two is darker than the upper. The analysing Nichol is then 
carefully rotated until equality in the intensity of the two spectral strips is 
attained; the angle through which the prism has been moved is then deter- 
mined ; several, say five, sets of readings being made in two opposite quadrants 
of the large divided circle. The mean of these readings gives the value 
of </>.! 
The spectrophotometric constants of oxyhemoglobin. — It was 
previously stated that it is usual to determine the photometric constants 
of colouring matters in two spectral regions, those, regions being chosen 
in which the variations in the absolution of light arc must rapidly affected 
by rariations in the concentration of the colouring matter. 
The reasons for determining in the first instance at least two values 
for A (which we shall distinguish as A and A'), and subsequently, each 
time that a determination is made, ascertaining the value of e in the 
same two regions (the two extinction-coefficients being distinguished 
as e and e, or in the case of oxyhemoglobin as e,, and e ') are the 
following: — (1) If we know the value of A and A' for any body, we are 
able to make two independent estimations when determining the 
concentration of a solution of the same body of unknown strength, the 
one estimate acting as a check on the other. (2) The knowledge of the 
value of A and A', for each of two colouring matters co-existing in 
solution, is a necessary condition to being able to determine spectro- 
photometrically the amount of each constituent when occurring together. 
(3) In the case of oxyhemoglobin, haemoglobin, and CO-haemoglobin, the 
quotient - is absolutely characteristic of each substance, and affords a 
valuable check on the purity of the colouring matter in solution and on 
the accuracy of the analysis. 
Hiifner's most recent determinations 2 of the spectrophotometric 
constants of oxyhemoglobin, made with his perfected spectrophotometer, 
have led to the results shown below. The two values of A are, as lias 
1 Hiifner's sp<:<-trophotoineter is constructed by, and can lie obtained from, tin- original 
maker, Herr Eugen Albrecht, Universitats-Mechaniker in Tubingen. 
2 Hufner, "Photometrische Constanten des Oxyhemoglobins, ''Arch. f. Physiol., Leipzig, 
1894, Physiol. Abth., S. 134 ct seq. 
