108 GOTE TURESSON 
The features dealt with belong strictly to the category of charac- 
ters usually called adaptations. As far as this term is employed to 
mean a Lamarckian mode of origin of the characteristics in question 
no support can be given it; if it is employed to express the idea that 
. certain hereditary forms of 
an ecospecies are favoured 
in a certain locality, while 
others are expelled in some 
way or other, much is to 
be said in support of it. 
The study of the coast 
plants just mentioned has 
clearly brought out the fact 
that the genotypical con- 
stitution of the ecospecies 
under discussion varies with 
the locality, and that the 
west coast forms of the 
plants represent throughout 
more extreme varieties than 
the east coast forms (when 
compared with the inland 
forms). When such a paral- 
lelism is found to exist be- 
tween the extremeness of 
type and the extremeness 
of locality there can be little 
doubt as to the profound 
importance of climatic and 
edaphic factors in the dif- 
ferentiation process of va- 
rieties and so-called adap- 
Fig. 4. Hieracium umbellatum, woodland variety tive forms 
(about !/, nat. size). 
; Before discussing any 
further the significance of such forms, mention should be made of the 
behaviour of an ecospecies distributed within a continuous area, where 
localities different in nature alternate with one another. The species 
which has been particularly studied from this point of view is Hiera- 
cium umbellatum, a typically cross-fertilizing plant (not apogamous) 
with an extensive range in Sweden. Material collected in different ha- 
