THE GENOTYPICAL RESPONSE OF THE PLANT 391 
II THE CONTROLLING EFFECT OF THE HABITAT FAC- 
TORS UPON THE GENOTYPICAL COMPOSITION OF THE 
SPECIES-POPULATION. 
In the above an attempt has been made to show to what extent 
the behaviour of a plant species in different habitats is the result of a 
direct response to the environment on the part of the individual plant, 
and to what extent it is due to the presence in the different habitats 
‘of different hereditary variations. While in some cases the observed 
characteristics were found to be purely modificatory, the differentiation 
of the species-population into different hereditary variations in the va- 
rous habitats was found to be the rule in the majority of cases. 
Although the conclusive proof of this latter proposition, viz. breeding 
experiments, has so far been presented in but very few cases, little 
doubt remains as to the hereditary nature of the characteristics seen 
in the cultures. The case of Hieracium umbellatum may be taken as 
an example. In the cases where plants have been raised from seeds 
(tables 14, 15, 16, 19, 27, 28) the results corroborate the view based 
upon the experiments with the transplantations. A further corrobora- 
tion of the facts is furnished by the series which show an increase in 
the prominence of the habitat characteristics when cultivated, which 
increase has been shown to. be present, in regard to the width of the 
leaves, in the case of the sea-cliff type from Hofs Hallar, Kristineberg 
and Stenshuvud. The opposite view, that this broad-leaved cliff type 
is for instance a habitat modification of the inland type, cannot possibly 
be upheld in the face of such facts. The behaviour in nature of the 
different types in places, where they come into contact with each other 
and more or less overlap, furnishes another proof. Mixed populations 
from such zones of contact have been discussed in the above and listed 
in tables 25 and 26. The argument increases in strength when it is 
seen that in a strictly intermediate area between two habitats of diffe- 
rent nature and populated by two different types no intermediate po- 
pulation is found but on the contrary individuals of the two definite 
types and bastards between them appear. A such neutral zone, into 
which habitats of different nature insensibly grade, has been located 
on the eastern coast of Scania at Kylsgärd (see above, pag. 291), where 
the drift-sand region of Sandhammar merges into the arenacious fields 
of the north. If the dune types of the drift-sand region and of the 
arenacious fields were but modifications, a third intermediate habitat 
modification would most probably be met with in this intermediate 
