No. 511] "PRESENCE AND ABSENCE" HYPOTHESIS 415 



units I have nothing to suggest, for to that question I 

 am, like Professor liateson.'- inclined "to hold niy fancy 

 an a tight rein;" but there can he no doubt that the' risihle 

 Mendelian characters are always secondary, and but 

 little doubt that they are all dependent at some stage of 



This is too obvious to need discussion in the case of 

 color-characters, and in those structural characters 

 which involve only some by-]. redact of the metabolism of 

 the cells as, for instance, the starchy or sugary char- 

 acter of the endosperm in maize. It requires perhaps a 

 more daring flight at present to assert that such struc- 

 tural characters as hairiness, branching, lobation and 

 serration of leaves, production of horns, extra toes, dif- 

 ferent forms of comb, etc., which involve the number, 

 direction and succession of cell-divisions, depend like- 

 wise upon the intimate chemical nature of the proto- 

 plasts ; but even if it could be shown that physical prop- 

 erties of the protoplasm are to a certain degree 

 determining conditions of cell-division, the resulting 

 structures could hardly conceivably be permanent hered- 

 itary features, unless these physical properties are de- 

 pendent at last upon the chemical composition of the 



Having arrived at the conclusion that all the Mendel- 

 ian characters are dependent upon chemical relations, 

 we may return to the question of dominance, and the re- 

 lation between the two kinds of homozygotes and the 



may be' interpreted in terms of chemical experience. 



A fundamental principle in this connection is the law 

 that the extent of a reaction between two chemicals is 

 determined by the amount of that reagent which is pres- 

 ent in less relative quantitv, and not by the one which is 

 present in excess. When the positive homozygote, 

 AABB, and the heterozygote, ABB, are alike, i. e., when 

 there is complete dominance of presence over absence, it 



