ON THE EVOLUTION OF HAEMOGLOBIN. n 



perfect certainty. The physical facts have altogether out- 

 stripj)e(l chemical knowledge. They clearly point out what 

 kind of investigation is necessary, but at the same time it 

 would be very difficult, if not impossible, to obtain a suf- 

 ficient quantity of material to decide some of the most 

 important questions. In the present state of the subject 

 the only course open to us is to draw the most probable 

 analytical conclusions that we can by careful induction from 

 facts already determined by synthetical methods. 



The best illustration of this kind that I can give is the 

 difference in the spectra of haemoglobin, in which tlie 

 oxygen is replaced by other gasses without there being any 

 actual decomposition. Thus, for example, on replacing the 

 oxygen by nitric oxide, or by carbonic oxide, the general 

 character of the spectrum of oxidized haemoglobin is only 

 slightly changed, but the position of the bands is materially 

 different. When expressed by wave-lengths in millionths 

 of a millimeter, the centres of the absorption bands are as 

 follows : ^ 



It will thus be seen that the substitution of one gas for 

 another in loose combination with the same very complex 

 coloured radical hsemoglobin alters the wave-lengths of the 

 centres of the bands to an extent which can be easily 

 measured. This change in the spectra is accompanied by 

 most decided differences in chemical characters, so decided 

 in fact that, as is well known, the substitution of the oxygen 

 in the blood by carbonic oxide makes all the difference 

 between life and death. In all these cases, however, it is 

 easy to prove that there is one complex radical common to 

 all, since they can all be changed into identical products. 

 Now, supposing that these three substances were met with 

 in nature, and that their spectra were known, and also the 

 fact of their yielding common products, it is quite clear that 

 we should be justified in concluding that they were closely 

 related in containing a common radical, and at the same 

 time differed by the substitution of one substance combined 

 with this radical by another substance. I think that the 

 strict accuracy of this reasoning cannot be denied in this 



' Since tbe wave-lengtb method bas not yet been adopted by all observers, 

 it may be well to give the wave-lengths of a few of the Frauuhofer lines 

 that occur near the absorption bands described in this paper. 

 D. 589. E. 527. F. 486. 



