580 Species According to the Theory of Mutation. 
variety; 1 or, as we have already expressed it in the first 
part of this volume (pp. 64, 71. etc.). the origin of new 
characters leads to specific differentiation, whereas the 
true derivative varieties arise by so-called retrogressive 
and degressive mutations without the formation of new 
characters. In my opinion this is, at any rate, the sim- 
plest view of the matter. 
This, however, is purely theoretical, for in practice 
our definition can, at present, only rarely be applied. Here 
however the principles of hybridization come to our aid ; 
for, expressed in hybridological terminology, our gen- 
eralization runs : Forms, all of the characters of which 
follow MENDEL'S laws in crosses, are to be regarded as 
varieties of the same species. This form of our generali- 
zation obviously admits of an immediate application in 
every case where the material can be dealt with experi- 
mentally. 
Obviously this generalization is at present too sweep- 
ing; nevertheless the best investigators 2 have regarded 
the study of hybridization as an empirical foundation on 
which this distinction may be based. Moreover the diffi- 
culties are not really so great as they seem to be at first 
sight; for as soon as the body of evidence will have at- 
tained a certain extent, definite laws will be detected 
which will fit the majority of cases by judging from anal- 
ogy. 
The species, however, which I am here distinguishing 
from varieties are the smaller or elementary species ; the 
delimitation of the larger or collective species is, from the 
very nature of the case, a question not for the experi- 
1 It should not be forgotten that varieties have been called in- 
cipient species and that from seed they are just as constant as species. 
2 See NAGELI, Joe. cit.. p. 396; FCCKE. loc. cit., pp. 488, 502; NAU- 
DIN, he. cit. p. 164 : ABBADO, loc. cit., p. 9, etc. 
