THE CHROMATOPHORES OF ANIMALS. 113 



are not clearly correlated with any chromatic function. 

 The entoplastic modifications shown by these pigment-cells 

 are in no way different from those which characterise 

 ordinary storage cells of reserve or waste materials. 



It would thus appear that the only pigment-cells dis- 

 tinctly specialised for chromatic purposes are those of 

 Vertebrata, of Cephalopod and certain Pteropod Mollusca, 

 and of Crustacea. To these pigment-cells the term chro- 

 matophore should be at present restricted, though it is very 

 probable that the mobile pigment-cells of certain Cteno- 

 phores and the branched pigment-cells of certain pelagic 

 Annelids will eventually find a place in the same category. 



We are now in a position to confront the question of the 

 origin of chromatophores. Have these remarkable elements 

 arisen independently of the remaining types of pigmented 

 cells, or have they been formed from pre-existing pig- 

 mented cells specialised for chromatic purposes ? Also, 

 do chromatophores exhibit any constant relation to the 

 germinal layers ? 



The answer to these questions can only be given after 

 a separate examination of the history of the chromatophores 

 in each group. Before proceeding to this examination it 

 may be well to point out the following considerations which 

 are not without weight as guides to the formation of an 

 opinion upon the subject. 



The usefulness of pigment for chromatic purposes could 

 only begin after the pigment had originated, and it is there- 

 fore probable that chromatophores have arisen by the 

 modification of pre-existing pigmented cells. Moreover, 

 since the prime necessity of chromatophores is that they 

 should be visible, it is probable that they would arise in the 

 external layer of the body, the ectoderm. This view is at 

 variance with current opinion, according to which chroma- 

 tophores are of the nature of connective tissue elements ; 

 but I shall show that there are many facts in the history of 

 chromatophores which are quite irreconcilable with this 

 opinion, and that the body of evidence strongly supports 

 the view which I am propounding. 



In the first place the branched contractile cells of 



