128 



Factors Affecting the Measurement of Absorption Bands. 

 By H. Haktkidge, M.A., Fellow of King's College, Cambridge. 



(Communicated by Prof. J. N. Langley, F.R.S. Eeceived November 1, 1912, — 



Read January 16, 1913.) 



(From the Physiological Laboratory, Cambridge.) 



In a previous paper a method was described* by whicb the percentage 

 saturation of haemoglobin with carbon monoxide can be estimated. This was 

 done by measuring with a special spectroscope the position of the absorption 

 bands of a solution of blood, since it was found that a definite relationship 

 exists between the percentage saturation with carbon monoxide and the wave- 

 length of the bands. The principle used in the instrument was one first 

 discovered by Zollner in 1870 and called by him the reversion spectroscope. 

 Two adjacent reversed spectra were obtained by passing beams through a slit 

 suitably placed in relation to a reflecting prism and a replica diffraction 

 grating, optical means being employed for shifting one of the spectra laterally, 

 so that corresponding points in the spectra might be adjusted into line. Since 

 first describing the method I have been able to investigate more thoroughly 

 its accuracy both in my own hands and also in those of other observers. Two 

 different classes of phenomena will receive attention, both of which tend to 

 introduce complications in the use of the method when absolute values for the 

 percentage saturation with carbon monoxide are required. These are : — 



(a) Variations in wave-length determinations made from time to time by 

 same observer on different samples of blood (personal variation). 



(b) Variations in wave-length determination by different observers on same 

 sample of blood (individual variation). 



Description of Personal Variation. 



As a result of wave-length measurements of the bands of the O2 and CO 

 compounds of haemoglobin, I found that those previously published differed 

 considerably from my own ; further, in these there was found from time to 

 time a small but quite definite divergence. Both these points are shown in 

 the table below. 



It will be seen from the table that by my measurements the change in 

 wave-length of the «-band, when 2 is replaced by CO in the haemoglobin 

 compound, is 53 - 3 Angstrom units ;f a value which is lower than that given 

 by the other observers quoted. This may have been due to dissociation of 



* STourn. Physiol.,' 1912, vol. 44, p. 1. 

 t Angstrom unit = 10~ 8 cm. 



