ON THE EVOLUTION OF HEMOGLOBIN. 83 



the green blood of Sabella ventilabrum, though the exact 

 connection remains to be determined. 



Reviewing now the whole subject from the point of view 

 of the theory of evolution, it should apjDcar that the most 

 stable form of hsematin made its appearance as a bile pigment 

 in a state totally unfit for the purposes of respiration, since 

 it will not unite with loosely combined oxygen on exposure 

 to the air. By a slight change in its constitution it might 

 however perform this function, and perhaps to some extent 

 does in some of the Mollusca, or at all events supplies the 

 place of the hsEmoglobin which forms one of the constituents 

 of the red muscles of higher animals, Avhilst in some of the 

 Gasteropoda it is so combined as to give rise to a form of 

 luemoglobin. According to Lankester,^ Planorbis alone 

 amongst this group of animals has a red blood, coloured by 

 haemoglobin, perhaps,, as he suggests, because the necessity 

 for utilizing the wdiole of the small amount of oxygen avail- 

 able in the water of stagnant marshes, rendered the further 

 development of the respiratory fluid very advantageous. This 

 Avas brought about by the modified bile hsematin combining 

 with an albumenoid, so as to give rise to the modification 

 of hsemoglobin found in the blood, which is certainly a more 

 suitable oxygen carrier than haematin, since it can be deoxi- 

 dized and reoxidized without decomposition. Advancing to 

 higher animals, it appears to have been very advantageous to 

 have this oxygen carrier, not in solution, but in the form of 

 discs suspended in a nearly colourless serum, so that these 

 red discs might be filtered out, and it alone penetrate to 

 some parts. This was perhaps brought about or accom- 

 panied by the substitution of the original albumenoid by 

 another, Avhilst at the same time hsematin ceased to be formed 

 as a secretion of the liver. The green blood of Sabella per- 

 haps resulted from an analogous development in another 

 way, not yet fully understood. In all these changes, as we 

 advance from the most rudimentary condition, we find that, 

 though the fundamental coloured radical, which determines 

 the oxygen-carrying properties, remains the same, the 

 compound becomes more and more unstable, and is more 

 and more easily decomposed, which is in all probability 

 correlated to a greater and greater vital poAver, capable of 

 counteracting stronger and stronger chemical affinities. 



This advance from a very stable to a very unstable com- 

 pound, is in many resj^ects analogous to what I have found to 

 occur during the growth of orange-coloured flowers exposed 

 to varying light. When developed nearly in the dark the 

 1 ' Proc. of Roy. Soc.,' vol. xxi, p. 78. 



