60 DISCONTINUOUS VARIATION. 



i. e., is discontinuous, and does not occur gradually. 

 Such mutations may take place in all directions, but 

 probably they only occur from time to time, due, per- 

 haps, to the periodical action of fixed causes. They 

 are distinct from the slight differences observed in 

 local races and varieties, for these can be produced 

 gradually by artificial selection and changed condi- 

 tions of environment. Also Natural Selection can 

 only lead to the formation of such local races, it being 

 powerless to bring about true mutations. The varia- 

 tion which leads to the formation of new species, there- 

 fore, is essentially discontinuous, not continuous. 



In order to obtain evidence in support of his theory, 

 De Yries has cultivated over 100 different species of 

 plants, but only one of them, (Enothera Lamarckiana, 

 showed the desired mutations. This plant was origi- 

 nally brought to Europe from America, and kept under 

 cultivation. It has since run wild, and De Vries ob- 

 tained the stock of nine plants which formed his first 

 generation from a field near Hilversum. Unfortu- 

 nately, the true origin of the plant is obscure. In 

 Britton and Brown's recently issued " Flora of the 

 United States," no reference whatever is made to it as 

 a wild species. Hence it is probably a garden variety of 

 (Enothera biennis (Evening Primrose), and may be a 

 hybrid plant, whilst the mutations obtained by De 

 Yries may merely be partial or complete reversions to 

 the original ancestors of the plant. To obtain these 

 mutations, De Yries cultivated the plant through eight 

 generations, and during this time, obtained over 50,000 

 specimens. Of these, 834 showed characters which 

 sharply differentiated them from the normal 0. La- 



