564 THE AMERICAN NATURALIST [Vol. XLVII 



of their behavior to that of (Enothera Lamar ckiana. 

 Thus the hybrids have thrown off marked variants of 

 new specific rank as does Lamar ckiana. Certain of these 

 new forms have held true and others have continued to 

 throw variants as do some of Lamarckiana' s "mutants." 

 One form {12.56x) appeared with a marked increase 

 over the normal chromosome number (14) and appar- 

 ently corresponds closely to the triploid "mutants" 

 from Lamarckiana or its derivatives (Lutz, '12; Stomps, 

 '12a). A most striking feature has been the production 

 in successive generations of classes of dwarfs, plants 

 which contrast sharply with the mass of the culture and 

 which are stable. 



This behavior of the hybrids appears to me to be of 

 quite the same character as the "mutations" of La- 

 marckiana, but the results, here concerned with crosses 

 between distinct species, are clearly of the sort that were 

 to be expected from their hybrid association. It is not 

 fundamental to my position that the various forms of the 

 variants in the F 2 and F 3 generations should match the 

 "mutants" from Lamarckiana. Since the F x hybrids 

 were not themselves the counterpart of Lamarckiana, 

 they should not be expected to give the same progeny as 

 this latter plant. It is sufficient for my purpose to point 

 out the essential parallelism between this hybrid be- 

 havior and that of Lamarckiana when it gives rise to its 

 "mutations." 



De Vries ('12, p. 30) has questioned the stability of 

 my grandiftora stock, apparently believing that my hy- 

 brids exhibit, at least in part, a mutating habit inherited 

 from the grandiftora parent. This view is based on the 

 appearance of two classes of hybrids (twin hybrids) in 

 the F, from the cross grandiftora B X biennis D. The 

 evidence, however, indicates that this peculiarity is con- 

 nected with the biennis parent, which may not have been 

 homozygous for the character of stem coloration at the 

 time the cross was made, although in later generations 

 the form has held true. 



