426 BOTANICAL GAZETTE [December 



0. Lamarckiana is about 2 per cent. Some of its mutations are 

 themselves mutable and give 4 per cent or even 6 per cent of 

 secondary mutations." The ordinary strains of 0. pratincola 

 produce about 2 per cent of mutations,' and the true 0. biennis, of 

 Holland, only about 0.45 per cent.* In strains which show mass 

 mutability, however, the number of mutations rises to 50 per cent 

 or even 100 per cent. This is the case with Lexington E. 



Lexington E differs from all the other strains of 0. pratincola 

 (i) in that it gives rise to a characteristic group of four mutations, 

 (2) in that these characteristic mutations occur in such large num- 

 bers as to justify the use of the designation mass mutation for the 

 phenomenon, and (3) in that it does not give rise to mut. num- 

 mularia and certain other mutations which are characteristically 

 produced by the strains which do not show mass mutation. 



Turning now to resemblances, we find absolutely no niior- 

 phological characteristics to indicate even the shghtest differ- 

 ence between Lexington E and the other strains of O. pratincola. 

 Moreover, Lexington E shows also the ordinary type of muta- 

 bility, in that it gives rise to small numbers of some of the same 

 mutations which the other strains produce. These mutations are 

 in no wise distinguishable morphologically from the same forms 

 occurring in the other strains, but all the evidence at hand shows 

 that when derived from the strain showing mass mutation the 

 non-characteristic mutations themselves show mass mutability. 

 The mutations characteristic of mass mutability are themselves 

 mutable and throw as secondary mutations other members of the 

 characteristic group. 



In view of the extraordinary interest of the phenomenon of 

 mass mutation, it is hoped that the reader will pardon the presenta- 

 tion of this confessedly preliminary report. Many of the genetic 

 relationships between the mutations and the parent species remain 

 to be worked out, and, as will be very obvious, a cytological study 

 of the whole subject is imperative. 



■• DeVries, H., Gruppenweise Artbildung. pp. 312-315. 



sAn estimate based upon the results tabulated in BoT. Gaz. 59:105-109. 1915. 

 ' DeVries, H., The coefiicient of mutability in Oenothera biennis L. Bot. Gaz. 

 59:169-196. 1915. 



