338 The Unity of the Organism 



nucleus are "bearers of heredity", the effort to make the 

 facts presented by the protozoans support the general theory 

 of chromatin as the "hereditary substance" by comparing 

 the nucleus-blepharoplast-axial filament of the spermato- 

 zoan, is quite unwarranted. Indeed, due regard to all the 

 facts involved in this comparison finds in them very strong 

 evidence against the conception that chromatin is the exclu- 

 sive "hereditary substance." 



Evidence from Pigment Cells 



Another set of facts brought out by recent studies which 

 connect the nucleus directly with the production of definite 

 hereditary attributes, concerns the ontogenetic origin of 

 certain colors. A paper on an investigation in this field 

 published in 1915 is introduced by this sentence: "The more 

 recent work on the formation of melanin seeks to derive this 

 pigment from chromatin elements".^ One may remark the 

 form of expression here. Recent work "seeks to derive" 

 melanin from chromatin, rather than "seeks to learn whether, 

 if at all, and in how far melanin is produced from chroma- 

 tin". No biologist speaking as a proponent of the "scientific 

 spirit" will, I think, hesitate to admit that the latter mode 

 of stating the problem is more in accord with that spirit. 

 Yet when a specific situation arises, the tendency to depart 

 from the seeking-to-learn spirit and to assume that of seek- 

 ing confirmation for an adopted hypothesis is still well-nigh 

 irresistible even in science. The temper of the day in a 

 considerable section of biology is one of thorougli going 

 partisanship in behalf of chromatin. 



The "seeking to derive" things from the cliromosomes is 

 not by any means limited to melanin. Neurofibrils, muscle 

 fibrils, and glandular secretions are among the things which 

 so far have appeared as conspicuous claimants for such an 

 origin. And one familiar with tlie discussions of ontogenesis 

 in tlic ])eriod of the Gastrea theory, can hardly fail to recog- 



