SPECIES AND VARIETIES. 201 



one developmental process and therefore on several characters, 

 such as the gene studied by Herbert Nilsson as influencing the 

 colour of the veins in Oenothera. But here we are concerned 

 with the average effect of the presence or absence of one gene. 

 New species on the other hand, arise in a way, fundamentally 

 different from varieties. A new species is a group of individuals 

 originated from a limited number of plants or animals in some 

 way isolated from the body of the species. Such a group must 

 automatically become pure for its own type. The group can be 

 said to constitute a new species, if the type is a new one, in 

 other words if the total potential variability of the isolated 

 group admitted of such a new set of genes. The formation of a 

 new species out of one old one must be rare, the total potential 

 variability of an old species not being sufficiently large. New 

 species will mostly originate by isolation of a group of animals 

 or plants belonging to a species, of which the total potential 

 variability has been recently heightened by a cross (with some 

 other sub-species). New species, will, with rare exceptions, differ 

 from already existing ones in several characters, because the 

 genot5^ic difference is one in several genes. 



It follows from what is known about the action of genes, 

 their distribution and stability, and about the causes for geno- 

 variability and specific purity, that varieties differ from the 

 species to which they belong and among each other in single 

 characters, whereas species differ from each other in groups of 

 characters. 



Without the necessity of breeding-experiments or physiolog- 

 ical tests being required to decide whether a few aberrant in- 

 dividuals constitute a variety or a species, we have a very sim- 

 ple morphological test, applicable to dried and flattend out 

 herbarium-specimens and empty skins of animals. But far from 

 being new, this criterion is a very old one. It is the common 

 systematist's criterion of what constitutes a species and what 

 a variety. To resume, species are realities, and they are stable, 

 not changing. Further, we beheve that those individuals which 

 are seen to differ in one striking point only from the members 



