ii6 



BOTANICAL GAZETTE 



[FEBRUARY 



mutations were actualh' unconnected with the parent form by 

 intermediates, which might have been overlooked in classifying 

 the young seedhngs. An endeavor has been made to forestall the 

 same criticism of the writer's work on 0. pratincola by the pubhcation 





Fig. 12. — Fj progeny of Lexington C, Oenothera praliiieolj ipan lo of the progeny 

 of C-:2); the only mutation shown is C-22-20, mut. mimmiilaria: the other plants 

 are typical 0. pralincola. 



of a series of photographs showing some of the pans in which 

 the mutations occurred. Each reader can judge for himself as to the 

 discontinuity of the mutations from the rest of the plants. It is 

 believed that no one has heretofore pubhshed so extensive a series 

 of photographs representing random samples of cultures from which 



