420 BOTANICAL GAZETTE [DECEMBER 
(italics mine). And (p. 23): “Die single variations sind zufallige, 
nur von Zeit zu Zeit auftretende, sprungweise die Formen verandernde 
Erscheinungen.” He says (p. 5): “Die Gesetze der Mutabilitit 
sind ganz andere als jene der Variabilitat;” but this clue to the dis- 
tinction fades when we read (p. 23) that “Die ‘single variations’ 
sind zufallige Erscheinungen, von deren Gesetzen man bis jetzt keine 
Erfahrung hat.” 
Calling single variations and saltatory variations and discontinuous 
variations synonymous with mutations does not tell what any of them 
are. The one criterion by which DE Vrirgs tries consistently to 
distinguish mutations is their giving rise to specific characteristics. 
This certainly does not admit of practical application, because we do 
not know how to identify a specific characteristic. It is a very tenable 
position at present that the species is a group of organisms with 
limits set by our convenience, and that many “valid” species—to put 
it moderately—are characterized by distinctions which are matters 
of degree. The specific characteristic can hardly be more clear-cut 
than the species it characterizes. If specific characteristics are in 
nature unstable and not exactly definable, this one means of identify- 
ing mutations is imaginary, in addition to being inapplicable. DE 
Vries holds that species, not necessarily with the usually recognized 
limits, are definable and never have merged, and that their individual 
characteristics are likewise definable and stable; but when he identifies 
these in turn by their origin by mutation, he brings his argument 
into a circle. 
The practical characteristic of mutations on which DE VRIES 
lays most emphasis is their inheritance: ‘“Solche sind fast stets 
entweder véllig oder doch in hohem Grade erblich” (p. 16). But, 
as he of course recognizes, the continuous individual variations are 
also hereditary. We see that on every hand. The most familiar 
examples are furnished by human beings. De Vries says explicitly 
that the differences between them have not arisen by mutation as he 
uses the term. Yet what characteristic of any species is more certain 
to be inherited than the straight hair or the black hair of a pure 
Chinese, or the complexion of an Ethiopian or an American Indian? 
Among the much less constant features of our own race We know 
how likely the color of the eyes and hair, and other physical peculiart 
