THE CHROMOSOMES IN HEREDITY. 24! 



In their experiments on Mattliiola, Bateoon and Saunders 1 

 mention two cases of correlated qualities which may be explained 

 by the association of their physical bases in the same chromo- 

 some. " In certain combinations there was close correlation be- 

 tween (a) green color of seed and hoariness, (/>) brown color of 

 seed and grabrousness. In other combinations such correlation 

 was entirely wanting." Such results may be due to the associa- 

 tion in the same chromosomes of the physical bases of the two 

 characters. When close correlation was observed, both may be 

 supposed to have dominated their homologues ; when correlation 

 was wanting, one may have been dominant and the other reces- 

 sive. In the next paragraph to that quoted is the statement : 

 " The rule that plants with flowers either purple or claret arose 

 from green seeds was universal." Here may be a case of con- 

 stant dominance of two associated chromatin-entities. 



Dominance is not a conception which grows out of purely 

 cytological consideration. Cytology merely shows us the pres- 

 ence in a cell of two chromosomes, either of which is capable of 

 producing some expression of a given character, and it is left to 

 experiment in each case to show what the effect of this combined 

 action will be. The experiment'-' has shown that any one of the 

 three theoretical possibilities may be realized, viz : (i) One or the 

 other may dominate and obscure its homologue. (2) The result 

 may be a compromise in which the effect of each chromosome is 

 to be recognized. (3) The combined action of the two may re- 

 sult in an entirely new cast of character. In cases belonging to 

 the first category, the visible quality (allelomorph, chromatin- 

 entity) was described by Mendel as dominant and the other as 

 recessive, and the experiments of Bateson and Saunders and 

 others, as well as those of Mendel, have shown that in many 

 cases a dominant character tends to remain dominant during suc- 

 cessive generations if the environment is not materially changed. 

 Nevertheless, some experiments cited by Bateson z go to show 

 that dominance may be variable or defective. Furthermore, it is 

 not only conceivable, but highly probable that in most, if not all 



1 Bateson and Saunders, Experimental Studies in the Physiology of Heredity. 

 Reports to the Evolution Committee, I. London, 1902, p. Si, paragraphs II and 12. 



2 Cf. Bateson and Saunders, loc. cit. 



