366 
De Vries on the Origin of Species 
The essential problem is therefore that of the origin of " elementary species," 
and the Mutationstheorie is a statement of the process by which this is believed to 
be effected. 
In an elementary species Professor de Vries recognises two distinct phenomena 
which produce differences between individuals : Vai'iation proper (Variabilitat im 
engeren Sinne, oder die individuelle Variabilitat) and Mutation. 
Variation proper is a phenomenon which regularly occurs in every generation, 
producing a series of differences between individuals such that the distribution of 
the various kinds of individuals in every generation obeys the laws of chance. Such 
variation can never lead to a permanent change in the mean characters of the 
species ; and if by stringent selection among such variations the mean character of 
the race is for a time changed, removal of the selection will be quickly followed by 
regression to the old mean of the species. 
Mutation, on the other hand, is a phenomenon which occurs intermittently, and 
has not been shown to obey any ascertained law of magnitude or of frequency. An 
individual which exhibits a mutation belongs already to a new species; and its 
offspring exhibit regression not to the old specific mean, but to a new one. The 
whole sum of the differences between two " elementary species," as enumerated in 
a long systematic diagnosis, may constitute a single mutation, and we are told 
" Dennoch hat man diese ganze Diagnose als den Ausdruck eines einzigen Merk- 
males zu betrachten, als eine Einheit, welche als solche entstanden ist, als solche 
verloren werden kann, deren einzelne Faktoren aber nicht voneinander getrennt in 
die Erscheinung treten konneu. Theoretisch habeu wir eine solche Gruppe von 
Eigenschaften gleichfalls als eine Einheit, als ein ganzes Merkmal zu betrachten." 
(p. 42.) 
Without mutation, therefore, no new species can be established ; when a 
mutation has occui-red a new species is already in existence, and will remain in 
existence, unless all the progeny of the mutation are destroyed. The only influence 
which natural selection can exert upon the course of evolution is that due to the 
total destruction of species. The phrase " survival of the fittest," as describing a 
process of evolution, ought to be replaced by " survival of the fittest species." 
The fundamental statements, on which the whole Mutationstheorie rests, are 
those concerning the regression of the offspring to one mean if their parents only 
vary, and to another if their parents exhibit mutations. 
The view of regression among the offspring of merely varying individuals is 
supported mainly by an appeal to experience. A summary of the results achieved 
in horticulture is held to show that a large number of florists' races have been 
obtained by crossing ; and where stable races have been certainly obtained without 
cross breeding their existence is attributed to mutation. The main part of the 
evidence for the asserted instability of forms produced by long-continued selection 
consists of facts which do not seem to me conclusive. Thus under the heading 
" Das Verhalten der veredelten Rassen heim Aufhoren der Selection" among the 
