EXPERIMENTAL EVOLUTION 147 
as the starting point of a phylum. There is at present no clue whatever as 
‘to what calls forth this essential difference in behavior. This is not surpris- 
ing in view of the fact that there have been no comparative experimental 
studies oi stable and plastic species. Until these have been made, it is im- 
possible to do more than to formulate a working hypothesis as to the effect 
of stability, and an explanation of the forces which cause or control it is 
altogether out of the question. 
186. Variation and mutation. New forms of plants are known to arise 
by three methods, viz., variation, mutation, adaptation. The evidence in 
support of these is almost wholly observational, and consequently more or 
less inexact, but for each there exist a few accurate experiments which 
are conclusive. Origin by variation and subsequent selection is the essence 
of the Darwinian theory of the origin of species. According to this 
the appearance of a new form is due to the accumulation, and selection, 
through a long period, of minute differences which prove advantageous to 
the plant in its competition with others in nature, or are desirable under 
cultivation. Slight variations appear indiscriminately in every species. 
Their cause is not known, but since they are found even in the most uniform 
habitats, it is impossible to find any direct connection between them and the 
physical factors. In the case of origin by mutation, the new form appears 
suddenly, with definite characteristics fully developed. Selection, in the 
usual sense of the term, does not enter into mutation at all, though the 
persistence of the new form is still to be determined by competition. Muta- 
tions are known at present for only a few species, and their actual appearance 
has been studied in a very few cases. Like variations, they are indiscrimi- 
nate in character. The chief difference between them is apparently one of 
degree. Indeed, mutation lends itself readily to the hypothesis that it is 
simply the sudden appearance of latent variations which have accumulated 
within the plant. DeVries regards constancy as an essential feature of 
mutation, but the evidence from the mutants of Onagra is not convincing. 
Indeed, while there can be no question of the occurrence of mutation in 
plants, a fact known for many years, the facts so far brought forward in 
support of the “mutation theory” fall far short of proving “the lack of 
significance of individual variability, and the high value of mutability for 
the origin of species.”* Mutations do not show any direct connection with 
‘the habitat, but their sudden appearance suggests that they may be latent 
or delayed responses to the ordinary stimuli. Origin by adaptation is the 
immediate consequence of the stimuli exerted by the physical factors of a 
IDE Vries, H. Die Mutationstheorie, 1:6, 1901. 
