DE VRIES'S ANALYSIS OF VARIATION 95 



suddenly, but while " progressive steps are the marks of ele- 

 mentary species, retrograde varieties are distinguished by appar- 

 ent losses." Retrograde varieties usually differ from their parent 

 species by a single sharp character only, they have lost pigment, 

 or hairs, or spines, and so on ; while elementary species are dis- 

 tinguished from their nearest allies in almost all organs. More- 

 over, the same kind of retrograde variety occurs repeatedly in 

 different series of species, hence the long lists of unrelated varieties 

 called by the same varietal title e.g. alba, inermis, canescens, or 

 glabra. 



" Varieties differ from elementary species in that they do not 

 possess anything really new. They originate for the greater 

 part in a negative way, by the apparent loss of some quality, 

 and rarely in a positive manner by acquiring a character already 

 seen in allied species" (1905, p. 152). 



Ever- sporting Yarieties. De Vries uses this term to describe 

 cases like the striped larkspur, which for centuries has gone on 

 producing unstriped as well as striped flowers. " Its changes are 

 limited to a rather narrow circle, and this circle is as constant as 

 the peculiarities of any other constant species or variety. But 

 within this circle it is always changing, from small stripes to 

 broad streaks, and from them to pure colours. Here the vari- 

 ability is a thing of absolute constancy, while the constancy con- 

 sists in eternal changes ! " Plants with variegated leaves, with 

 double flowers, with fasciated branches, with peloric flowers, and 

 so on, often illustrate the " ever-sporting " tendency. The 

 common snapdragon (Antirrhinum majus) is a very good case, 

 the striped variety, for instance, cannot be fixed. There is some 

 inherent instability in the combination of unit-characters in 

 these ever-sporting varieties. 



Fluctuations. De Vries applies this term to the continually 

 occurring individual variations. "It is normal for organisms 

 to fluctuate to and fro, oscillating around an average type. 

 Fluctuations are linear, amplifying or lessening the existing 



