Both Bateson and Davenport appear to have tacitly 

 agreed that the dominance of absence over presence is 

 a difficulty for the "presence and absence" hypothesis, 

 for both have taken occasion to explain that what ap- 

 pears to be the absence of a character may really be the 

 presence of a positive inhibiting factor. Indeed, Daven- 

 port" lias taken the position that the positiveness of a 

 character determines its dominance, and. therefore, all 

 cases in which the absence of an external character domi- 

 nates its presence must be explained by the existence of 

 a positive factor in whose presence the given external 

 character can not be produced. Thus, he says: 



ontogenetic process, which depends on something having dropped out. 



While I recognize the probability that there are posi- 

 tive inhibiting factors, as well as factors which produce 

 specific structural and color characters, I think it can be 

 shown that such an assumption is not necessary for the 

 explanation of the dominance of the absence of a char- 

 acter over its presence. I will assume for the sake of the 

 discussion that the presence and absence hypothesis is 

 correct, and that the absence is real, having no internal 

 unit to represent it. This assumption seems to me, as 

 it did to Hurst, to l>e simpler and more practical than the 

 alternative idea that the internal units are paired in the 

 hoterozygote, having a representative for absence as well 

 as one for presence. I believe there is no fact on record 10 



•Davenport, C. B. Report of the Third International Conference on 

 Genetics, p. 139, 1906. 



10 Except perhaps so <\-i lie.] " spurious allrl'Hi'.nrphism." 



