602 Validity of the Doctrine of Mutation. 



to some other one; and especially amongst American in- 

 vestigators the tendency has been in recent years to pro- 

 ceed as far as possible in this direction. 



If we look for a fixed point among these oscillating 

 opinions we may well choose the view repeatedly ex- 

 pressed by Darwin himself, that it is possible to imagine 

 that characters may originate by a slow process, but may 

 disappear all of a sudden.^ In combining this with the 

 distinction made in the first part of this volume between 

 progressive, retrogressive and degressive formation of 

 species, the proposition would run : Progressive forma- 

 tion of species may occur slowly and gradually, whilst 

 retrogressive and degressive specific differentiation is due 

 to mutations. Progressive differentiation consists in the 

 formation of a new character which was not previously 

 ])resent; whilst retrogressive and degressive differentia- 

 tion consists in the transference of internal factors, al- 

 ready present, from one condition to another. In the for- 

 mer case the active unit becomes latent ; in the latter the 

 latent becomes active, or the semi-latent semi-active ; but 

 the material vehicles of these characters remain funda- 

 mentally the same throughout ; nothing new arises in the 

 idioplasma.^ 



In horticulture, as we have seen, mutations are largely 

 of the retrogressive or degressive kind. Discontinuous 

 formation of species on the progressive line is much 

 rarer. Nevertheless I believe that my researches with 

 Oenothera have contributed instances Avhich may demon- 

 strate the occurrence of progressive mutations in this 

 species at any rate. Obviously there is a great need of 



^ With reference to this point see the valuable critique by L. 

 Plate, Uebcr Bcdcutung unci Tragzceitc des Dan^'iu'schcii Sclcclions- 

 p rill dps, 1900. p. ^7 and elsewhere. 



' See below, §§ 9-1 1. 



