BOWMAN LECTURE. LXV 



If the disease be recessive the matings shown in c and 

 D will also be frequent, for then either one parent or 

 both will appear normal ; mating c will give only normals, 

 but half of them should carry the disease ; in mating D 

 one quarter of the offspring should show the disease if 

 sufficiently large numbers be taken, and one half should 

 carry it invisible or potential ; in mating E, as just stated, 

 the disease should appear in half and be carried by the 

 other half, if sufficient numbers be taken. 



Therefore if the simple Mendelian theory be applicable 

 to any human disease or defect we shall expect that, in 

 most cases, either one quarter or one half of the offspring 

 will show the condition (Fig. 6 D and E). 



The assumption is that dominance and recessiveness 

 are constant for the same character in all stocks and 

 families ; that the same character or disease cannot be 

 dominant in one pedigree and recessive in another. But 

 when a given character is linked with sex in such a way 

 as to be manifest only in one sex (the male), although 

 carried in an in completed and invisible state by the other 

 (female), the fact has been explained in Mendelian terms 

 by Professor Bateson on the assumption that the character, 

 although dominant in the male, becomes recessive in the 

 female. This hypothesis appears to explain some other- 

 wise difficult cases. For example it can be made to 

 account for the clinical fact invariable as far as we yet 

 know that a colour-blind woman transmits her defect to 

 all her sons, and that she herself has always had a colour- 

 blind father. But on the other hand it does not explain 

 the ordinary experience that a colour-blind father very 

 seldom has colour-blind sons. 



And in many other cases the experimental breeding of 

 plants and animals has given results which, in order to 

 bring them within the four corners of the Mendelian 

 theory, require the assumption of various modifying or 

 controlling influences. But this is not the time, nor am 

 I the person, to discuss the hypotheses dealing with such 

 subjects as (: dihybridism/' " gametic coupling," " rever- 



