V. CONCLUSION. 



Perhaps the most important general result of my ob- 

 servations on the origin of species in the genus Oenothera 

 is the proof which they afford that this phenomenon can 

 be dealt with experimentally. Hitherto the general opin- 

 ion has been that this extraordinarily important phenom- 

 enon was amenable neither to direct investigation nor 

 even to direct observation. The experience of horti- 

 culture is sufficient to demonstrate the fact that new 

 forms do sometimes appear and less rarely perhaps, than 

 is imagined. But when they appear it is too late to at- 

 tempt to discover how they arose. We may try to explain 

 how this happened, but it is no longer possible to deal 

 with the question experimentally. 



For this object it is necessary to have a plant which 

 happens to be in a mutation period, i. e., which has the 

 power of giving rise repeatedly to new species. Such 

 plants had hitherto not been found. 



The way to look for mutable plants is to make very 

 extensive sowings of various species. Seeds are collected 

 either from wild plants or from plants that have run wild 

 or lastly from cultivated plants which one has oneself 

 grown for a sufficient length of time to make certain 

 that they are free from the effect of any previous cross- 

 ings that may have taken place. I have either collected 

 the seeds in the field, or transplanted a few plants into 



