78 THE PLANT WORLD 



forms of mutation is corroborated by the fact that species 

 that produce many bud variations, also produce many germ- 

 inal mutations; and these germinal and vegatative variations 

 are often similar, and many times are identical. Some of 

 these vegetative changes undoubtedly affect the germ cells, 

 for they are transmitted by sexual reproduction; others 

 appear to be changes only in somatic structures. In certain 

 bud mutations there seems to be an inertia of cell division 

 that keeps the mutation true when propagated vegetatively 

 for long periods of time, even when sexual reproduction 

 always induces a return to the normal form. Other changes 

 that have appeared many times in horticultural history, 

 have never been fi^ed by even the strictest selection. 



As long as there is no definite reason to believe that 

 bud mutations are specifically different from germinal muta- 

 tions, De Vries' grouping into progressive, degressive and 

 retrogressive mutations can here also be used. If then bud 

 variations belong to these three groups, the questions of 

 greatest practical importance are : What is the frequency 

 with which these groups occur in nature? What is the cause 

 of each? Can they be produced at will? 



If progressive mutations are relatively common, then 

 we may expect to obtain varieties with absolutely new 

 characters by this means. If, however, progressive mutations 

 are very rare, then bud mutations will give but few novelties 

 of real agricultural worth. Degressive and retrogressive 

 mutations are confined to rather narrow lines. They pro- 

 duce and reproduce certain definite variations in characters 

 having little economic value. Moreover these variations 

 show greatest similarity to existing biotypes of the species. 

 In floriculture, where slight differences in color and other 

 minor characters are important, such changes may be of 

 value; but it is unlikely that there will be changes in factors 

 which go to make up quality in agricultural products. 



I believe we should omit from discussion those varia- 

 tions that will probably admit of different explanation. For 



