BUD VARIATIONS 83 



ever, reports that entire petals are dominant to laciniated 

 petals in Chelidoniiim majus. . This evidence is not con- 

 clusive, but it seems reasonable that the change is the loss 

 of something that enables the plant to produce entire leaves. 



A third group of bud variations, that occurs quite often, 

 is the loss of a prickly, hairy or felted condition. Such 

 characters have been shown to be Mendelian in several cases, 

 :n all of which the rough condition is dominant to its absence. 

 The evidence comes from such different plants as Lychnis, 

 Matthiola, Datura and wheat. 



The classical example of this type is the loss of the 

 hairy condition in the peach, which produces the nectarine. 

 The moss rose is a second case, although at first thought it 

 might be regarded as an exception to the theory. This rose 

 has been considered to have been derived from the Provence 

 rose by bud variation, but this is not definite. The moss rose 

 was imported into England from Italy, and its definite origin 

 is unknown. It may have originated as a germinal mutation. 

 It seems to be much more definitely recorded that smooth 

 variations have appeared from the moss rose rather than 

 mossy veriations out of the Provence rose. At least, the 

 change by loss of a character, is much more common. 



These data are merely presented as giving a reasonable 

 explanation of at least 75^^ of our recorded bud variations. 

 It may be that an increasing knowledge of Mendelism will 

 t^ en extend the proportion of cases to which it is applicable. 

 But even should this not be the case, it seems a rational 

 conclusion that comparativ^ely few bud varieties will come into 

 cultivation through real origin of useful characters. The 

 single idea explaining what is thought to be the common 

 method of hud variation is thought to be worthy of con- 

 sideration. 



Conn. Agr. Exp. Station, New Haven. 



